The Curtis Center 170 S Independence Mall W 300E Philadelphia, Pennsylvania 19106 SPINE SURGERY: TECHNIQUES, COMPLICATION AVOIDANCE, AND MANAGEMENT Copyright 2005, Elsevier, Inc. All rights reserved.
ISBN 0-443-06616-7
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. Permissions may be sought directly from Elsevier’s Health Sciences Rights Department in Philadelphia, PA, USA: phone: (+1) 215 238 7869, fax: (+1) 215 238 2239, e-mail:
[email protected]. You may also complete your request on-line via the Elsevier homepage (http://www.elsevier.com), by selecting ‘Customer Support’ and then ‘Obtaining Permissions.’
NOTICE Surgery is an ever-changing field. Standard safety precautions must be followed, but as new research and clinical experience broaden our knowledge, changes in treatment and drug therapy become necessary or appropriate. Readers are advised to check the product information currently provided by the manufacturer of each drug to be administered to verify the recommended dose, the method and duration of administration, and contraindications. It is the responsibility of the treating physician, relying on experience and knowledge of the patient to determine dosages and the best treatment for the patient. Neither the Publisher nor the editor assumes any responsibility for any injury and/or damage to persons or property. First edition 1999.
Library of Congress Cataloging-in-Publication Data Spine surgery: techniques, complication avoidance, and management / [edited by] Edward C. Benzel.– 2nd ed. p. ; cm. Includes bibliographical references and index. ISBN 0-443-06616-7 1. Spine–Surgery. I. Benzel, Edward C. [DNLM: 1. Spinal Diseases–surgery. 2. Spinal Diseases–complications. 3. Spine–surgery. WE 725 S7599 2005] RD768.S684 2005 617.5′6059–dc22 2004045517
Acquisitions Editor: Rebecca Schmidt Gaertner Developmental Editor: Agnes Hunt Byrne Publishing Services Manager: Joan Sinclair Project Manager: Daniel Clipner
Printed in the United States of America. Last digit is the print number:
9
8
7
6
5
4
3
2
1
This book is dedicated to my wife, Mary; my children, Morgan, Jason, Brian, and Matthew; and my mother and father, Anna and Carl, for their past and present encouragement, tolerance, and guidance. A special thanks to Christine Moore for her energy, diligence, and effort, as well as her continued and perpetual commitment to this project; to all the contributors for their masterful “works” and for the privilege of their friendships, both new and old; and to prior residents and fellows, who have provided energy, encouragement, and guidance.
Contributors
Giancarlo Barolat, MD Thomas Jefferson University Hospital, Philadelphia, Pennsylvania H. Hunt Batjer, MD Department of Neurosurgery, Northwestern University Medical School, Chicago, Illinois
Mark F. Abel, MD Professor, Department of Orthopedics, University of Virginia School of Medicine, Charlottesville, Virginia
Thomas W. Bauer, MD, PhD Departments of Orthopaedic Surgery and Pathology, The Cleveland Clinic Foundation, Cleveland, Ohio
Kuniyoshi Abumi, MD, Dr Med Sci Professor, Department of Orthopaedic Surgery, Health Administration Center, Hokkaido University, Sapporo, Japan
James R. Bean, MD Neurosurgical Associates, Lexington, Kentucky Brion J. Beerle, MD Chair, Department of Anesthesiology, Alaska Regional Hospital, Anchorage, Alaska
Mark S. Adams, MD Saginaw Valley Neurosurgery, Saginaw, Michiga
Gordon R. Bell, MD The Cleveland Clinic Foundation, Cleveland, Ohio Cleveland Clinic Spine Institute
Cary D. Alberstone, MD Ventura County Neurosurgery Associates, Oxnard, California
Gregory J. Bennett, MD Clinical Director-Neurosurgery, Erie County Medical Center, Buffalo, New York
Joseph T. Alexander, MD Assistant Professor, Department of Neurosurgery, Wake Forest University School of Medicine, Winston-Salem, North Carolina
Edward C. Benzel, MD Chairman, Cleveland Clinic Spine Institute, Vice Chairman, Department of Neurosurgery, The Cleveland Clinic Foundation, Cleveland, Ohio
John A. Anson, MD, FACS Las Vegas, Nevada Ronald I. Apfelbaum, MD Professor of Neurosurgery, Department of Neurosurgery, University of Utah Health Sciences Center, Salt Lake City, Utah
Darren Bergey, MD Loma Linda University Medical Center, Loma Linda, California Marc L. Bertrand, MD Assistant Professor and Director, Neuroanesthesia, Department of Anesthesiology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
Paul M. Arnold, MD Professor of Neurosurgery, Department of Surgery, University of Kansas Medical Center, Kansas City, Kansas
Mark H. Bilsky, MD Memorial Sloan Kettering Cancer Center, New York, New York
L. Brett Babat, MD Premier Orthopaedics and Sports Medicine, Nashville, Tennessee
Barry D. Birch, MD Department of Neurosurgery, Mayo Clinic Scottsdale, Scottsdale, Arizona
Julian E. Bailes, MD Professor and Chairman, Department of Neurosurgery, West Virginia University, Morgantown, West Virginia
Robert S. Biscup, MS, DO, FAOAO Cleveland Clinic Florida Spine Center, Weston, Florida
Jamie Baisden, MD Assistant Professor of Neurosurgery, Medical College of Wisconsin, Milwaukee, Wisconsin
Kevin Blaylock, CPA CFO, Neuroscience Specialists, CEO, Oklahoma Spine Hospital, LLC, Oklahama City, Oklahoma
Nevan G. Baldwin, MD, FACS Clinical Associate Professor, Texas Tech University, Lubbock, Texas
Oheneba Boachie-Adjei, MD Hospital for Special Surgery, New York, New York
Perry A. Ball, MD Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
Maxwell Boakye, MD Assistant Professor of Neurosurgery, Stanford School of Medicine, Stanford, California vii
viii
Contributors
Scott D. Boden, MD Professor of Orthopaedics, Director, The Emory Spine Center, Emory University School of Medicine, Atlanta, Georgia
Joseph S. Cheng, MD Assistant Professor and Co-Director, Section of Spinal Surgery, Department of Neurosurgery, Vanderbilt Medical Center North, Nashville, Tennessee
Henry Bohlman, MD Professor, Case Western Reserve University and University Hospitals, Cleveland, Ohio
Yong-Jun Cho, MD Department of Neurosurgery, Stanford University, Stanford, California
Michael Bolesta, MD Associate Professor, Department of Orthopedics, Southwestern Medical School, Dallas, Texas
Tanvir F. Choudhri, MD Assistant Professor and Director, Neurosurgery Spine Program, Department of Neurosurgery, Mount Sinai Medical Center, New York, New York
Mary B. Bondy, MBA Cleveland Clinic Spine Institute, The Cleveland Clinic Foundation, Cleveland, Ohio Christopher M. Boxell, MD Oklahoma Spine & Brain Institute, Tulsar, Oklahoma Keith H. Bridwell, MD Washington University School of Medicine, St. Louis, Missouri Darrell S. Brodke, MD Associate Professor, Department of Orthopedics, Director, Spine Service, University of Utah, Salt Lake City, Utah James Butler, MD Resident, Department of Neurosurgery, The Cleveland Clinic Foundation, Cleveland, Ohio David W. Cahill, MD (deceased) Department of Neurosurgery, College of Medicine, University of South Florida, Tampa, Florida Robert C. Cantu, MA, MD, FACS, FACSM Chief, Neurosurgery Service and Director, Service Sports Medicine, Emerson Hospital, Concord, Massachusetts; Adjunct Professor, Exercise and Sport Science, University of North Carolina, Chapel Hill, North Carolina Allen L. Carl, MD Professor, Division of Orthopaedic Surgery, Albany Medical College, Albany, New York John A. Carrino, MD, MPH Assistant Professor of Radiology, Harvard Medical School, Clinical Director, Magnetic Resonance Therapy Program, Co-Director, Spine Intervention Service, Boston, Massachusetts John R. Caruso, MD Neurosurgical Specialists, LLC, Hagerstown, Maryland Andrew G. Chenelle, MD, MS DuPage Neurosurgery, S.C., Glen Ellyn, Illinois
Frank Conguista, MD Resident, Division of Orthopaedic Surgery, Albany Medical College, Albany, New York Edward S. Connolly, MD Professor of Neurosurgery, Ochsner Clinic Foundation, Louisiana State University School of Medicine, New Orleans, Louisiana Paul R. Cooper, MD Professor of Neurosurgery, New York University School of Medicine, New York, New York Jean-Valéry C.E. Coumans, MD Department of Neurosurgery, Massachusetts General Hospital, Boston, Massachusetts Albert E. Cram, MD Professor, Department of Neurosurgery, University of Iowa Hospitals and Clinics, Iowa City, Iowa H. Alan Crockard, MD The National Hospital for Neurology and Neurosurgery, London, United Kingdom Richard Crownover, MD The Cleveland Clinic Foundation, Cleveland, Ohio Bryan W. Cunningham, MD Director, Biomechanics Laboratory, Union Memorial Hospital, Baltimore, Maryland William T. Curry, Jr, MD Department of Neurosurgery, Massachusetts General Hospital, Boston, Massachusetts Joseph F. Cusick, MD Professor, Medical College of Wisconsin, Madison, Wisconsin Scott D. Daffner, MD Thomas Jefferson University, Philadelphia, Pennsylvania
Contributors
Mark D. D’Alise, MD, FACS Neurosurgical Associates, Complex and Reconstructive Spinal Surgery, Lubbock, Texas
Sanford E. Emery, MD, MBA Professor and Chairman, Department of Orthopaedics, West Virginia University, Morgantown, West Virginia
Vinay Deshmukh, MD Carolina Neurosurgery and Spine, Charlotte, North Carolina
Nancy E. Epstein, MD Clinical Professor of Neurological Surgery, The Albert Einstein College of Medicine, Bronx; Attending in Neurosurgery, Winthrop University Hospital, Mineola; and The North Shore-Long Island Jewish Health System, Manhasset, New York
Denis DiAngelo, MD Associate Professor, School of Biomechanical Engineering, University of Tennessee Health Science Center, Memphis, Tennessee Curtis A. Dickman, MD Associate Chief, Spine Section, Director, Spinal Research, Division of Neurological Surgery, Barrow Neurosurgical Associates, Phoenix, Arizona Thomas B. Ducker, MD Maryland Neurological Institute, Annapolis, Maryland Scott T. Dull, MD, FACS Neurosurgical Network, Toledo, Ohio Stewart B. Dunsker, MD Emeritus Faculty, Mayfield Clinic and Spine Institute, Cincinnati, Ohio
Jennifer Erdos, MD Resident in Orthopedic Surgery, Allegheny General Hospital, Pittsburgh, Pennsylvania Thomas J. Errico, MD Associate Professor of Orthopedic and Neurologic Surgery, New York University School of Medicine, Chief of the Spine Service, Department of Orthopedic Surgery, NYU/Hospital for Joint Diseases, New York, New York Tom Faciszewski, MD Chairman, Department of Orthopedic Spine Surgery, Marshfield Clinic, Marshfield, Wisconsin
Michael J. Ebersold, MD Professor of Neurosurgery, Mayo Clinic College of Medicine, Department of Neurologic Surgery, Luther Middlefort Clinic—Mayo Health System, Eau Claire, Wisconsin
Michael G. Fehlings, MD, PhD, FRCS(C) Professor of Neurosurgery, Krembil Chair in Neurological Repair and Regeneration, University of Toronto; Director Krembil Neuroscience Program, Heed Spinal Program, Toronto Western Hospital, University Health Network, Toronto, Ontario, Canada
Jason Eckhardt Spine Research Laboratory, The Cleveland Clinic Foundation, Cleveland, Ohio
Lisa A. Ferrara, MS, PhDc Spine Research Laboratory, The Cleveland Clinic Foundation, Cleveland, Ohio
Bruce L. Ehni, MD Neurosurgical Group of Texas; and Clinical Associate Professor, Baylor College of Medicine, Houston, Texas
Richard G. Fessler, MD, PhD Professor and Chairman, Section of Neurological Surgery, University of Chicago Hospital, Chicago, Illinois
Matthew Eichenbaum, MD Spine Research Fellow, Department of Orthopaedic Surgery, Thomas Jefferson University and the Rothman Institute, Philadelphia, Pennsylvania Kurt M. Eichholz, MD Resident, Department of Neurosurgery, University of Iowa, Iowa City, Iowa Marc E. Eichler, MD Division of Spine Surgery, Department of Neurosurgery, Brigham and Women’s Hospital, Boston, Massachusetts Samer K. Elbabaa, MD Resident, Division of Neurosurgery, Department of Surgery, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
ix
Kevin T. Foley, MD Associate Professor, Neurosurgery, University of Tennessee, Memphis, Tennessee Robert M. Galler, DO Assistant Professor of Neurosurgery, Stony Brook University Medical Center, Stony Brook, New York John W. German, MD Albany Medical College, Department of Neurosurgery, Albany, New York Alexander J. Ghanayem, MD Associate Professor and Chief, Division of Spine Surgery, Department of Orthpaedic Surgery, Loyola University Medical Center, Maywood, Illinois
x
Contributors
Zoher Ghogawala, MD Assistant Clinical Professor of Neurosurgery, Yale University School of Medicine, New Haven, Connecticut
Blaine I. Hart, MD Professor, Department of Radiology, University of New Mexico School of Medicine, Albuquerque, New Mexico
Vijay K. Goel, PhD Professor and Chair, Department of Bioengineering, University of Toledo, Director, Spine Research Center, Department of Orthopedics, Medical College of Ohio, Toledo, Ohio
Robert A. Hart, MD, MA Chief, Spine Section, Assistant Professor, Department of Orthopaedic Surgery, Oregon Health and Science University, Portland Shriner’s Hospital, Portland, Oregon
Jan Goffin, MD, PhD Department of Neurosurgery, Catholic University of Leuven, University Hospital Gasthuisberg, Leuven, Belgium Ziya L. Gokaslan, MD Professor of Neurosurgery, Johns Hopkins University, Baltimore, Maryland Sohrab Gollogly University of Utah, Salt Lake City, Utah Jorge Gonzalez-Martinez, MD Resident, Department of Neurosurgery, The Cleveland Clinic Foundation, Cleveland, Ohio James E. Greensmith, MD, PhD Department of Anesthesia, Saint Elizabeth Hospital, Appleton, Wisconsin Jeffrey D. Gross, MD Comprehensive Spine and Wellness Center, Ladera Ranch, California
Robert F. Heary, MD Associate Professor of Neurological Surgery, University of Medicine and Dentistry-New Jersey, New Jersey Medical School; Director, The Spine Center of New Jersey, Neurological Institute of New Jersey, Newark, New Jersey Fraser C. Henderson, MD Associate Professor of Neurosurgery, Director of Neurosurgery of the Spine and Craniocervical Junction, Georgetown University Medical Center, Washington, D.C. Patrick W. Hitchon, MD Professor, Department of Neurosurgery, University of Iowa Hospitals and Clinics, Iowa City, Iowa James P. Hollowell, MD Integrated Spine Care, Milwaukee, Wisconsin Paul J. Holman, MD Department of Neurosurgery, Cleveland Clinic Foundation, Cleveland, Ohio
Regis W. Haid, Jr, MD Atlanta Brain and Spine Care, Atlanta, Georgia
John K. Houten, MD Associate Professor, Department of Neurological Surgery, Montefiore Hospital, Albert Einstein/Jaboni Hospital, Bronx, New York
Andrea L. Halliday MD, PA Center for Neurological Disorders, Fort Worth Brain and Spine Institute, Fort Worth, Texas
Robert E. Isaacs, MD Head, Section of Minimally Invasive Spine Surgery, Cleveland Clinic Florida Spine Institute, Weston, Florida
Allan J. Hamilton, MD Head of Surgery, University of Arizona Health Sciences Center, Tucson, Arizona
Manabu Ito, MD, Dr Med Sci Assistant Professor, Department of Orthpaedic Surgery, Hokkaido University Graduate School of Medicine, Sapporo, Japan
Fadi Hanbali, MD Assistant Professor of Neurosurgery, University of Texas Medical Branch, Galveston, Texas Jürgen Harms, MD Center for Spine Surgery, Department of Orthopedics and Traumatology, Klinikum Karlsbad-Langensteinbach, Karlsbad-Langensteinbach, Germany James S. Harrop, MD Assistant Professor, Division of Spinal Surgery, Department of Neurosurgery, Neurosurgical Director, Delaware Valley Spinal Cord Injury Center, Thomas Jefferson University, Philadelphia, Pennsylvania
John A. Jane, Jr, MD Department of Neurological Surgery, University of Virginia, Charlottesville, Virginia J. Patrick Johnson, MD Co-Director, Cedars-Sinai Institute for Spinal Disorders, Los Angeles, California Christopher Kager, MD Lancaster Neuroscience and Spine Association, Lancaster, Pennsylvania
Contributors
Iain H. Kalfas, MD Head, Section of Spinal Surgery, Department of Neurosurgery, Cleveland Clinic Foundation, Cleveland, Ohio
xi
Donlin M. Long, MD Johns Hopkins University Medical School, Baltimore, Maryland
George J. Kaptain, MD Loma Linda University Medical Center, Loma Linda, California
Mark G. Luciano, MD, PhD Head, Section of Pediatric and Congenital Neurosurgery, Department of Neurological Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio
Saad Khairi, MD Spine Fellow, Cedars-Sinai Institute for Spinal Disorders, Los Angeles, California
Charles A. Luevano, BS Component Reliability Development, General Motors Corp., Milford, Michigan
Daniel H. Kim, MD Associate Professor and Director, Spinal and Peripheral Nerve Surgery, Depatment of Neurosurgery, Stanford University Medical School, Stanford, California
Parley M. Madsen III, MD, PhD Microsurgery and Brain Research Institute, St. Louis, Missouri
David H. Kim, MD Attending Spinal Surgeon, The Boston Spine Group, Boston, Massachusetts Thomas A. Kopitnik, Jr., MD University of Texas Southwestern Medical Center at Dallas, Dallas, Texas Robert J. Kowalski, MD, MS, PE Resident, Department of Neurosurgery, The Cleveland Clinic Foundation, Cleveland, Ohio Ajit A. Krishnaney, MD Department of Neurosurgery, Cleveland Clinic Foundation, Cleveland, Ohio John A. Lancon, MD Associate Professor of Neurosurgery and Pediatrics, Blair E. Batson Hospital for Children, The University of Mississippi Medical Center, Jackson, Mississippi Giuseppe Lanzino, MD Associate Professor, Department of Neurosurgery, University of Illinois College of Medicine at Peoria, Peoria, Illinois
Dennis J. Maiman, MD, PhD Professor of Neurosurgery, Medical College of Wisconsin and Clement J Zablocki VA Medical Center, Milwaukee, Wisconsin Jacek M. Malik, MD, PhD Peninsula Neurosurgical Associates, Salisbury, Maryland David G. Malone, MD Oklahoma Spine & Brain Institute, Tulsa, Oklahoma Joseph C. Maroon, MD Clinical Professor and Vice Chairman, Department of Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Eric M. Massicotte, MD, MSc, FRCS(C) Assistant Professor of Neurosurgery, University of Toronto, Toronto, Ontario, Canada Shunji Matsunaga, MD Assistant Professor, Department of Orthopedic Surgery, Kagoshima Graduate School of Medical and Dental Sciences, Kagoshima, Japan
Sanford J. Larson, MD, PhD Medical College of Wisconsin, Department of Neurosurgery, Milwaukee, Wisconsin
Daniel J. Mazanec, MD, FACP Vice Chairman, Cleveland Clinic Spine Institute, The Cleveland Clinic Foundation; and Associate Professor of Medicine, Cleveland Clinic Lerner College of Medicine, Cleveland, Ohio
Jorge Lastra-Power, MD Assistant Professor, Neurosurgery Section, University of Puerto Rico, San Juan, Puerto Rico
Paul C. McAfee, MD Chief of Spine Surgery, St. Joseph’s Hospital, Baltimore, Maryland
Nathan H. Lebwohl, MD Voluntary Associate Professor, Clinical Orthopaedics and Rehabilitation University of Miami School of Medicine, Miami, Florida
Bruce M. McCormack, MD Clinical Faculty, University of California San Francisco Medical Center, San Francisco, California
Isador H. Lieberman, MD, MBA, FRCS(C) Department of Orthopaedics and The Cleveland Clinic Spine Institute, The Cleveland Clinic Foundation, Cleveland, Ohio
Paul C. McCormick, MD, MPH Professor of Clinical Neurosurgery, Columbia University College of Physicians and Surgeons, New York, New York William E. McCormick, MD Schwartzapfel Novick, West Islip, New York
xii
Contributors
Robert A. McGuire, Jr., MD University of Mississippi Medical Center, Jackson, Mississippi
Sait Naderi, MD Associate Professor, Department of Neurosurgery, Dokuz Eylul University, Izmir, Turkey
Robert F. McLain, MD The Cleveland Clinic Spine Institute, The Cleveland Clinic Foundation, Cleveland, Ohio
Dileep Nair, MD The Cleveland Clinic Foundation, Cleveland, Ohio
Nagy Mekhail, MD, PhD The Cleveland Clinic Foundation, Cleveland, Ohio D. Mark Melton, MD Resident, Department of Neurosurgery, College of Medicine, University of South Florida, Tampa, Florida Carole A. Miller, MD Ohio State University, Department of Neurological Surgery, Columbus, Ohio Jared H. Miller, BA Case Western Reserve University, Cleveland, Ohio Sung Min, MD The Cleveland Clinic Foundation, Cleveland, Ohio William Mitchell, MD Attending Neurosurgeon, JFK Medical Center, Edison, New Jersey Junichi Mizuno, MD, PhD Associate Professor, Department of Neurological Surgery, Aichi Medical University School of Medicine, Aichi, Japan Michael T. Modic, MD The Cleveland Clinic Foundation, Cleveland, Ohio
Hiroshi Nakagawa, MD, PhD Professor and Chairman, Department of Neurological Surgery, Aichi Medical University School of Medicine, Aichi, Japan Jaime H. Nieto, MD Divisions of Neurosurgery and Orthopedic Surgery, Maimonides Medical Center, Brooklyn, New York Russ P. Nockels, MD Associate Professor, Neurological Surgery, Orthopaedic Surgery and Rehabilitation, Loyola University Medical Center, Maywood, Illinois Bruce E. Northrup, MD Professor Emeritus, Department of Neurosurgery, Thomas Jefferson University, Philadelphia, Pennsylvania Chima Ohaegbulam, MD Resident, Department of Neurosurgery, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts Tunc Oktenoglu, MD VKV Amerikan Hastanesi, Nisantasi, Istanbul, Turkey Bernardo Jose Ordonez, MD Neurosurgical Associates, Norfolk, Virginia Jeffrey H. Owen, PhD Sentient Medical Systems, Cockeysville, Maryland
Howard W. Morgan, Jr, MD University of Texas, Southwestern Medical School, Dallas, Texas
A. Fahir Özer, MD American Hospital, Istanbul, Turkey
Robert J. Morlock, PhD Pfizer Incorporated, Worldwide Outcomes Research, Ann Arbor, Michigan
Stephen M. Papadopoulos, MD Director of Surgical Navigation, Barrow Neurosurgical Institute, Phoenix, Arizona
Michael A. Morone, MD, PhD Deaconess Billings Clinic, Billings, Montana
Christopher G. Paramore, MD Lake Norman Neurological and Spine Surgery, Mooresville, North Carolina
Wade M. Mueller, MD Assistant Professor of Neurosurgery, Medical College of Wisconsin, Madison, Wisconsin Praveen V. Mummaneni, MD Assistant Professor, Department of Neurological Surgery, Emory University School of Medicine, Atlanta, Georgia John S. Myseros, MD Assistant Professor, Division of Neurosurgery, Cincinnati College of Medicine, Staff Pediatric Neurosurgeon, Cincinnati Children’s Hospital
Robert S. Pashman, MD Director, Scoliosis and Spinal Deformity, Cedars-Sinai Institute for Spinal Disorders, Los Angeles, California Warwick J. Peacock, MD Professor Emeritus, Department of Neurosurgery, University of California, San Francisco, San Francisco, California Stanley Pelofsky, MD Neuroscience Specialists, Oklahoma City, Oklahoma
Contributors
Noel I. Perin, MD, FRCS, FACS Clinical Associate Professor, Department of Neurosurgery, St. Luke’s/Roosevelt Hospital Center, New York, New York Christopher J. Pham, DO Department of Neurosurgery, Stanford University Hospital School, Stanford, California Rick J. Placide, MD, PT West End Orthopaedic Clinic, Richmond, Virginia Branko Prpa, MD All Saints Healthcare, Racine, Wisconsin Gregory J. Przybylski, MD John F. Kennedy Medical Center, Edison, New Jersey Ashraf A. Ragab, MD University of Mississippi Medical Center, Jackson, Mississippi Y. Raja Rampersaud, MD, FRCS(C) Assistant Professor, Division Orthopedic Surgery, University of Toronto; Spinal Program, Krambil Neuroscience Center, Toronto Western Hospital, University Health Network, Toronto, Ontario, Canada Peter A. Rasmussen, MD Department of Neurosurgery, The Cleveland Clinic Foundation, Cleveland, Ohio Richard B. Raynor, MD Clinical Professor of Neurosurgery, New York University School of Medicine, New York, New York Gary L. Rea, MD, PhD Department of Orthopaedics, Ohio State University Hospital East, Columbus, Ohio Glenn R. Rechtine, MD Dunspaugh-Dalton Professor of Spinal Surgery, Department of Neurosurgery, University of Florida, Gainesville, Florida John Regan, MD Medical Director, Institute for Spinal Disorders, Cedars-Sinai Medical Center, Los Angeles, California Setti S. Rengachary, MD Associate Chairman and Professor, Department of Neurological Surgery, Wayne State University, Detroit Medical Center, Detroit, Michigan Daniel K. Resnick, MD Associate Professor, Department of Neurosurgery, University of Wisconsin Medical School, Madison, Wisconsin Laurence D. Rhines, MD Assistant Professor and Director, Spine Program, Department of Neurosurgery, University of Texas MD Anderson Cancer Center, Houston, Texas
xiii
Albert J. Rhoton, MD Chairman Emeritus, Department of Neurosurgery, University of Florida – Gainesville, Gainesville, Florida Donna J. Rodriguez, MS, RD, CNSD Former Visiting Professor/Lecturer, College of Education, Nutrition/Dietetics Program, University of New Mexico; Professional Healthcare Writer, Lovelace Healthcare Innovations, Lovelace Health Systems, Albuquerque, New Mexico Gerald E. Rodts, Jr, MD Associate Professor and Director of Neurosurgery Spine, Department of Neurological Surgery, Emory University, Atlanta, Georgia Michael J. Rosner, MD Department of Neurosurgery, Walter Reed Army Medical Center, Washington, DC Alexander Sah, MD Resident, Harvard Medical School Combined Orthopedic Residency Program, Boston, Massachusetts Jared P. Salinsky, DO Resident, Department of Orthopedic Surgery, NSUCOM/Parkway Regional Medical Center, North Miami Beach, Florida Paul Santiago, MD Assistant Professor, Department of Neurological Surgery, Washington University School of Medicine, St. Louis, Missouri Mehdi Sarkarati, MD Assistant Clinical Professor, Department of Physical Medicine and Rehabilitation Tufts University School of Medicine, Healthsouth N.E. Rehabilitation Hospital, Woburn, Massachusetts Richard L. Saunders, MD Professor Emeritus of Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire Paul D. Sawin, MD Orlando Neurosurgery, Orlando, Florida Edward H. Scheid, MD Senior Resident, Thomas Jefferson University, Department of Neurosurgery, Philadelphia, Pennsylvania Meic H. Schmidt, MD Assistant Professor and Director, Spinal Oncology, Department of Neurosurgery, University of Utah Medical Center, Salt Lake City, Utah Michael Schneier, MD Neurological Surgery, Philadelphia, Pennsylvania Dilip K. Sengupta, MD, Dr Med Professor, Department of Orthopaedics, DartmouthHitchcock Medical Center, Lebanon, New Hampshire
xiv
Contributors
Christopher I. Shaffrey, MD Professor, Department of Neurological Surgery, University of Virginia School of Medicine, Charlottesville, Virginia Mark E. Shaffrey, MD Department of Neurosurgery, University of Virginia Health System, Charlottesville, Virginia
Loretta A. Staudt, MS, PT University of California Los Angeles, Los Angeles, California Michael P. Steinmetz, MD Resident, Department of Neurosurgery, The Cleveland Clinic Foundation, Cleveland, Ohio
Alok D. Sharan, MD Albany Medical College, Albany, New York
Charles B. Stillerman, MD Clinical Professor of Surgery, University of North Dakota School of Medicine, Minot, North Dakota
Ashwini D. Sharan, MD Assistant Professor, Department of Neurosurgery, Thomas Jefferson University, Philadelphia, Pennsylvania
Kota Suda, MD, Dr Med Sci Spine Surgeon, Center for Spinal Injury and Disorder, Bibai Rosai Hospital, Bibai, Japan
Christopher B. Shields, MD, FRCS(C) Professor and Norton Hospital Chairman, Department of Neurological Surgery, University of Louisville, Louisville, Kentucky
Sonia Suys, MD General Surgeon, Virginia Beach, Virginia
Frederick A. Simeone, MD Department of Neurosurgery, Thomas Jefferson University, Philadelphia, Pennsylvania Kern Singh, MD Department of Orthopedic Surgery, Rush Presbyterian St. Luke’s Medical Center, Chicago, Illinois
George W. Sypert, MD Southwest Florida Neurosurgical Association, Fort Myers, Florida Charles H. Tator, MD, PhD, MA, FRCSC, FACS Director, Canadian Paraplegic Association Spinal Cord Injury Research Centre, Toronto Western Hospital, Toronto, Ontario, Hospital
Ran Vijai P. Singh, MD Neurological Associates, Norfolk, Virginia
Nicholas Theodore, MD Director, Neurotrauma, Division of Neurological Surgery, Barrow Neurosurgical Associates, Phoenix, Arizona
Donald A. Smith, MD Associate Professor, Department of Neurosurgery, College of Medicine, University of South Florida, Tampa, Florida
Ajith J. Thomas, MD MeritCare Neuroscience, Fargo, North Dakota
Maurice M. Smith, MD Semmes-Murphey Clinic, Germantown, Tennessee Volker K.H. Sonntag, MD, FACS Vice Chairman and Chief, Spine Section, Division of Neurological Surgery; Barrow Neurosurgical Associates; and Director, Residency Program, University of Arizona, Phoenix, Arizona Ivan J. Sosa, MD Chief Resident, Neurosurgery Section, University of Puerto Rico, San Juan, Puerto Rico Micheal J. Speck, MD The Cleveland Clinic Spine Institute, The Cleveland Clinic Foundation, Cleveland, Ohio Robert F. Spetzler, MD Director, Department of Neurosurgery, Barrow Neurological Institute, Phoenix, Arizona Sudhakar T. Sridharan, MD Research Director, Department of Rheumatic and Immunologic Diseases, The Cleveland Clinic Foundation, Cleveland, Ohio
Nicholas W.M. Thomas, MD Department of Neurosurgery, Kings College Hospital, London, England Robert E. Tibbs, Jr., MD Mercy Health Center, Oklahoma City, Oklahoma Daisuke Togawa, MD, PhD Department of Orthopaedics and The Cleveland Clinic Spine Institute, The Cleveland Clinic Foundation, Cleveland, Ohio Frank J. Tomecek, MD Oklahoma Brain & Spine Institute, Tulsa, Oklahoma Richard M. Toselli, MD University of North Carolina at Chapel Hill, Chapel Hill, North Carolina Vincent C. Traynelis, MD Professor, Department of Neurosurgery, The University of Iowa, Iowa City, Iowa Gregory R. Trost, MD Assistant Professor, Departments of Neurosurgery and Orthopaedic Surgery, University of Wisconsin Medical School, Madison, Wisconsin
Contributors
Eeric Truumees, MD Staff Spine Surgeon, William Beaumont Hospital, Royal Oak, Michigan; Orthopedic Director, Gehring Biomechanics Laboratory; and Adjunct Faculty, Wayne State University Biomechanics Center, Detroit, Michigan Gary W. Tye, MD Department of Neurosurgery, Virginia Commonwealth University Health System, Richmond, Virginia Abm Salah Uddin, MD John F. Kennedy Medical Center, Edison, New Jersey Alexander R. Vaccaro, MD Professor of Orthopaedic Surgery, Co-Chief of Spine Surgery, Jefferson Medical College, Thomas Jefferson University, The Rothman Institute; and Co-Director, Delaware Valley Regional Spinal Cord Injury Center, Philadelphia, Pennsylvania Ceslovas Vaicys, MD Division of Neurosurgery, Memorial Healthcare System, Hollywood, Florida Alex Valadka, MD Baylor College of Medicine, Texas Medical Center, Houston, Texas Arnold B. Vardiman, MD Neurological Associates of San Antonio, San Antonio, Texas Anthony A. Virella, MD Chief Resident, Division of Neurosurgery, University of California Los Angeles Medical Center, Los Angeles, California Elizabeth Vitarbo, MD University of Miami, Miami, Florida Todd W. Vitaz, MD University of Louisville, Louisville, Kentucky Dennis G. Vollmer, MD Division of Neurosurgery, University of Texas Health Science Center at San Antonio, San Antonio, Texas Jean-Marc Voyadzis, MD Resident, Department of Neurosurgery, Georgetown University Hospital, Washington, D.C.
xv
John K. Webb, FRCS Chairman, Center for Spinal Studies and Surgery, University Hospital, Nottingham, United Kingdom Philip R. Weinstein, MD University of California San Francisco School of Medicine, San Francisco, California Martin W. Weiser, PhD Technical Product Manager, Johnson Matthey Electronics, Spokane, Washington William C. Welch, MD, FACS Associate Professor, Department of Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Simcha J. Weller, MD Director, Neurosurgery Spinal Disorders Program, Beth Israel Deaconess Medical Center, Boston, Massachusetts L. Erik Westerlund, MD CORE Orthopaedic Medical Center, P.C., Encinitas, California Jonathan A. White, MD University of Texas, Southwestern Medical Center, Dallas, Texas Melvin D. Whitfield, MD The Cleveland Clinic Spine Institute, The Cleveland Clinic Foundation, Cleveland, Ohio Gregory C. Wiggins, MD David Grant Medical Center, Travis Air Force Base, California Jack E. Wilberger, MD Acting Chairman, Department of Neurosurgery, Allegheny General Hospital, Pittsburgh, Pennsylvania William S. Wilke, MD Department of Rheumatic and Immunologic Diseases, The Cleveland Clinic Foundation, Cleveland, Ohio Diana Barrett Wiseman, MD Department of Neurosurgery, Naval Hospital, Okinawa, Japan W. Putnam Wolcott, MD Neurosurgeon, Newport News, Virginia
John D. Ward, MD Professor and Vice Chairman, Department of Neurosurgery, Chief, Pediatric Neurosurgery, Virginia Commonwealth University Health System, Richmond, Virginia
Eric J. Woodard, MD Assistant Professor and Chief, Section of Spine Surgery, Department of Neurosurgery, Harvard Medical School, Brookline, Massachusetts
Joseph Watson, MD Assistant Professor, University of California Davis School of Medicine, Davis, California
Philip Yazback, MD Neuroscience Group of NE Wisconsin, Neenah, Wisconsin
xvi
Contributors
Narayan Yoganandan, PhD Department of Neurosurgery, Medical College of Wisconsin, Milwaukee, Wisconsin
Seth M. Zeidman, MD University of Rochester Medical Center, Rochester, New York
Kenneth S. Yonemura, MD Assistant Professor, SUNY-Upstate Medical University, Syracuse, New York
Barry M. Zide, MD Professor, Department of Surgery (Plastic), New York University Medical Center, New York, New York
Kazuo Yonenobu, MD Department of Orthopaedic Surgery, Osaka University Medical School, Osaka, Japan
Mehmet Zileli, MD Professor of Neurosurgery, Ege University Faculty of Medicine, Department of Neurosurgery, Bornova, Izmir, Turkey
Hansen A. Yuan, MD Professor, SUNY-Upstate Medical University, Syracuse, New York
Preface
It is important to be objective, fair, and nonjudgmental when assessing complications, particularly those of others. There are clearly many ways to effectively and safely accomplish a task. Very infrequently, if ever, does a single method always work. What works for one surgeon may not work for another, and vice versa.
Nomenclature
This, the second edition, is bigger and (I think) better than the first. This preface is as appropriate for the second edition, as I feel it was for the first. Therefore, it is presented again with minimal modification. The purpose of this book is to assist the spine surgeon with the avoidance, identification, and management of complications. This differs little from a presentation of operative technique and medical management. Therefore, this book in many respects is a techniques book. To achieve its purpose, an understanding of history, decision making, medical management, differential diagnosis, ethics, and even discussions of problems associated with related disorders (such as peripheral nerve injury and metabolic bone diseases) is mandatory. This process requires an understanding of the fundamental basic science components of spine surgery (e.g., anatomy, biomechanics, and physiology). The book’s style was created by its authors, as well as their interactions with each other. In most cases, two or more senior authors were assigned to most chapters. These senior authors were not necessarily chosen on the basis of their philosophical compatibility with their coauthor(s). In fact, coauthors with opposing viewpoints were chosen in many circumstances. Authors were also selected on the basis of experience, educational adeptness, and communication and writing skills. The greater-than-usual number of expert authors contributing to each chapter achieves continuity within the text itself that would not be possible otherwise.
In this book, some standardization of nomenclature was thought to be important. For example, ventral (according to Dorland’s, pertaining to the belly or to any venter; denoting a position more toward the belly surface than some other object of reference) is used instead of anterior (according to Dorland’s, situated in front of or in the forward part of an organ toward the head end of the body; a term used in reference to the ventral or belly surface of the body).1 Similarly, dorsal is used instead of posterior; rostral instead of cephalad; and caudal instead of caudad. Exceptions to this are situations in which the term designates a structure or concept that is clearly established (e.g., anterior longitudinal ligament).
Repetition We learn most effectively by having data presented in a repetitive manner, often from different perspectives, using differing techniques (e.g., written, mathematical, or visual). Truly understanding a concept involves a spiral, which often involves multiple exposures to information, so that a solid data base is acquired. New data (raw data) are then added and assimilated. This “expanded” knowledge base can then be applied to, and enhanced by, additional basic science and clinical applications. This entire process is perpetually refined by new experiences, such as clinical encounters or through reading (Figure 1). Repetition is good.
Standard of Care
Risk Taking
Before undertaking the definition of a complication, the issue of standard of care deserves consideration. S. Haines (personal communication) illustrates the discrepancy between the legal definition (the degree of care a reasonable person would take to prevent injury to another) and the evidence-based definition (a generally accepted principle for patient management reflecting a high degree of clinical certainty) of the standard of care. Emphasis on “certainty” by clinicians and its de-emphasis by the legal system are most certainly noteworthy.
Surgery is a risk-taking process. The patient places himself or herself in the hands of the surgeon, and the ensuing decision-making process involves the resolution (or the attempts at such) of many technical and quality-oflife–related issues and dilemmas. A surgical procedure may be warranted if the sum of the costs (both financial and personal) and risks is less than the sum of the benefits. This risk/benefit analysis should be of paramount concern and should be emphasized by the surgeon and realized by the patient. This book is designed to help surgeons achieve their goals, by minimizing the risk-taking component of this “equation.”
What is a Complication? Peter McL. Black addressed the issue of defining a complication of neurological surgery in the front matter to Brain Surgery: Complication Avoidance and Management.2 Much of what Black addressed is pertinent to spine surgery. The unique nature of spine surgery, however, dictates the examination of complications from a slightly different approach. The nuances and complexities of the spine and spine surgery are associated with unique concerns. In order to more clearly understand the nature of a spine surgery complication, each author (of the first
Opinion and Dogmatism This book is intended to be used as a textbook, to serve as a reference, and to function as a resource for information. Much information is available in the pages that follow. Perspectives, pearls, conventional wisdom, and objective data are presented, as is opinion. Great care has been taken to identify opinion when presented, and to avoid its extreme dogmatism. xvii
xviii
Preface
4. EXPERIENCE
5.
EXPERIENCE EXPERIENCE CLINICAL APPLICATION BASIC SCIENCE APPLICATION RAW DATA
BASELINE KNOWLEDGE
Figure 1 The Learning Spiral.
edition) completed a questionnaire that incorporates some of the concepts and questions used in Black’s questionnaire. It was specifically designed to facilitate an understanding of a complication. The results of this questionnaire are reported and discussed for our enlightenment regarding the complications of our “trade.” There was a 98% response rate by spine surgeon authors. Fourteen authors, who do not participate significantly in clinical decision making (i.e., engineers, nurses, anesthesiologists, and physiologists), did not complete the questionnaire. What is particularly interesting and revealing is the commentary by the majority (73%) of the respondents. The first section (Questions 1 and 2) explores surgeon and institutional policy regarding complications. The second section (Questions 3 through 9) concerns the specifics of defining a complication and involves surgeon opinion to a significant degree. Nevertheless, this section provides a baseline for those wishing to derive an “operational definition(s)” of complications. The final question (Question 10) provides a philosophical and speculative slant. The Responses Question 1 reveals that spine surgeons do not have an adequate definition of what constitutes a surgical complication. Question 2 implies that in teaching institutions, there is variability regarding what is to be presented at a morbidity and mortality conference. D. Long provides a pertinent response that is worth sharing. He outlines seven rules for morbidity and mortality conferences: 1. All mortality must be discussed. The case must be presented by the residents and the responsible faculty surgeon must be at the meeting to summarize the mortality. 2. All postoperative neurological change, expected or unexpected, must be presented. The responsible surgeon must be present to discuss the case. 3. All postoperative infections are discussed. Those with serious patient implications are discussed in detail.
6. 7.
The others, which are more routine, are listed. For instance, a urinary tract infection is merely listed, but a wound infection or meningitis is discussed. Any surgical or postsurgical event that prolongs the expected hospitalization must be discussed. Any failure of equipment must be discussed. This includes shunts, stimulators, and internal fixators. All serious events involving non-neurosurgical organ systems are discussed. All postanesthetic complications are discussed in anesthesia morbidity and mortality conference but also must be listed at the neurosurgery morbidity and mortality conference.
Excessive cost is treated as a complication. If there is an outlier, when the cost is extreme, that patient case is discussed. D. Long reported that after the monthly meeting at his institution, chief residents who present the cases produce a written summary that goes into the minutes of the conference. This details the complication that occurred and the discussion that centered on prevention. Question 3 shows that the majority of respondents thought that a complication was an adverse event that occurred within 1 month of surgery. This has constituted traditional thinking, but without an objective rationale. Question 4 “spreads out” this field even further. The majority of authors who responded to Question 5 believed that the younger and healthier the patient in whom a pneumonia occurs after routine surgery, the greater the probability that the pneumonia should be considered a complication. Perhaps this should apply to selected other complications as well. Regarding Question 6, the authors were split on whether a dural tear that was successfully repaired during surgery signifies a complication. Clinical manifestations of this “tear,” however, were relatively consistently considered complications by the authors. Question 7 revealed an interesting response by the authors. Generally, a pedicle screw fracture was not considered to be a complication if asymptomatic and associated with a solid fusion. However, if there was an associated pseudarthrosis, with or without back pain, this same fracture was generally considered to be a complication. Finally, if the back pain persisted but the fusion was solid, no complication was believed to be present. Therefore, in the authors’ opinion, the acquisition of fusion was the main determinant regarding definition of a complication in this circumstance. Similarly, the responses to Question 8 depict a relatively clear demarcation between symptomatic and asymptomatic pseudarthroses after anterior cervical discectomies. The case of a symptomatic patient with a pseudarthrosis was regarded by the majority of respondents to be a complication, whereas those cases in which a pseudarthrosis was asymptomatic, no complication was thought to be present. Regarding Question 9, a complaint of back pain at the rostral implant insertion site was not believed to be indicative of a complication by the majority of authors, almost regardless of circumstance. Question 10, although theoretic, elicited consistent responses. The majority of respondents thought that the quality of our medical practices and medical reporting
Preface
would be improved by a more accurate definition of what constitutes a complication. They also believed that quality assurance would be enhanced. They, however, had mixed feelings about whether the medical-legal climate would be improved. Author Commentary The majority of authors not only completed the questionnaire but added comments that were of significant relevance. Although most thought that the questionnaire was of some utility, the responses in this regard varied. They ranged from statements such as “Nice try, Ed, but I think this standardization would be difficult to do and really does not matter. . .” (anonymous) to “. . . a very thought-provoking questionnaire that illustrates well the ‘gray areas’ involved with the definition of complications.” (M. Gallagher) “You touch on a very important point in that standard agreement of what constitutes a complication is highly subjective. We need to develop an analogue of the Glasgow Outcome Scale for spine that can be more universally applied.” (E. Woodard) It is in this vein that perhaps the most clear and concise response regarding the definition of a complication was derived. P. McCormick defined a complication as “an untoward unanticipated adverse event.” Although this is clear and concise, it is made more specific by J. Lancon. He defined a complication as “any perioperative event which results in persistence or recurrence of symptoms or the development of new symptoms.” B. Northrup defined a complication as “a preventable symptomatic maloccurrence that requires additional treatment. . .” This adds specificity to Lancon’s definition while broadening McCormick’s. A. Hamilton defines a complication “. . . as any untoward event occurring to a patient while on the neurosurgical service. These are usually subdivided into errors in diagnosis, errors in management, or errors in technique. By this definition, therefore, the development of a urinary tract infection in a patient who had Foley catheter is considered a complication.” J. Bean provided a particularly relevant medical-legal slant to the definition. He stated, “A complication has no generally agreed meaning until it is operationally defined. A medicolegal definition will of necessity vary from a ‘scientific’ or clinical definition because the responses to and uses of the information will be different. One focuses on culpability, one on the modification of clinical behavior or decision making.” The most pertinent question, perhaps, is not necessarily what constitutes a complication but what constitutes an avoidable versus non-avoidable complication. More importantly, what constitutes negligence? In this vein, R. Apfelbaum responded, “Complication to me doesn’t imply failure to provide adequate care—although it can mean that. Most often, it reflects the fact that we work with a biologic system with multiple variables acting concomi-
xix
tantly. We therefore need to evaluate ‘complications,’ which can be anything short of a perfect result . . .” In the future, we should perhaps focus on a further breakdown of the term “complication.” Perhaps a simple discussion of the concept of “avoidable” and “nonavoidable” complications is the place to start. The use of refined outcome assessment tools will clearly facilitate this process in the future.
The Canadian Thistle What is a complication? We still have not answered this question. Perhaps we never will. It, indeed, has different meanings and implications, depending on both its definer and the circumstances. Perhaps an analogy is in order. The Canadian thistle is considered a weed on the eastern Washington farm on which I grew up. This weed is a destroyer of wheat crops and the enemy of the farmer. Its mere presence in a field, in a sense, is a complication of farming; a manifestation, perhaps, of inadequate control measures and of less than optimal surveillance, early detection, and eradication techniques. If left unchecked, the Canadian thistle is associated with serious detrimental consequences. In Albuquerque, New Mexico, where my family had made our home in years gone by, the Canadian thistle is annoying to some, unnoticed by others, and considered a flower by many. Beauty is in the eye of the beholder, and, perhaps more relevant here, ugly is clearly a matter of perception and perspective, as is fault and blame. To the spine surgeon, the patient, and the attorney, a complication has different meanings, and often different consequences. Whereas postoperative pain (as subjective as it may be) may not be considered a complication by the surgeon, and occasionally may be annoying and a source of distress to the patient, it may be a source of revenue, and therefore joy, for the attorney. Beauty is clearly in the eye of the beholder, and without question, ugly is indeed a matter of perception and perspective.
xx
Preface
The Components And Factors Involved With The Definition Of A Complication Of Spinal Surgery: A Survey 1. Does your service have an explicit written (or “understood”) definition of what constitutes a surgical complication? If so, could you append it to this questionnaire? If not, would you briefly describe your own concept of a complication under comments below? Yes 9% No 88% No response 3% 2. How do you decide what will be presented at a Morbidity and Mortality conference? Residents decide 31% Faculty decides 24% Faculty/residents decide 35% Other 10% 3. Do you consider a complication as an adverse event occurring (check all that apply): a. Within 48 hours of surgery 45% b. Within a week of surgery 46% c. Within a month of surgery 75% d. While the patient was in the hospital 43% e. With reasonable assurance as a result of the surgical manipulation 75% f. None 23% 4. Do you consider recurrent sciatica, that is identical to the preoperative lumbar laminotomy pain pattern, to be a complication if it occurs (check all that apply): a. Within 48 hours of surgery 35% b. Within a week of surgery 36% c. Within a month of surgery 41% d. Within 2 months of surgery 25% e. Within 6 months of surgery 35% f. None 1% 5. Do you consider the occurrence of pneumonia to be a complication if it occurs (check all that apply): a. In a 25-year-old postoperative ventilated cervical quadriplegic patient 57% b. In a 65-year-old postoperative lumbar fusion patient with chronic obstructive pulmonary disease 13% c. In a 25-year-old nonoperated ventilated cervical quadriplegic patient 45% d. In a 40-year-old healthy nonsmoker on day 2 following a routine lumbar discectomy 96% 6a. Do you consider a dural tear, that is successfully repaired during surgery and that has no adverse sequelae, to be a complication of surgery? Yes 43% No 57% 6b. Do you consider this same dural tear to be a complication of surgery if (check all that apply): a. It is associated with 2 days of severe positional headaches 74% b. It is associated with CSF leakage through the wound and that requires lumbar drainage to successfully manage 96% c. It requires reoperation to manage 96% 7. Do you consider pedicle screw fracture at 6 months following surgery to be a complication if (check all that apply): a. It is asymptomatic and associated with a solid fusion 24% b. It is asymptomatic and associated with a pseudarthrosis 65% c. It is associated with persistent back pain and a solid fusion 38%
d. It is associated with persistent back pain and a pseudarthrosis 86% e. None of the above 13% 8. In a patient who has undergone an anterior cervical discectomy with fusion, do you consider a pseudarthrosis (without excessive movement on flexion/extension x-rays) to be a complication if (check all that apply): a. It is asymptomatic 24% b. It is associated with neck pain 80% c. It is associated with radicular pain 80% d. None 15% 9. One year following a fusion and the placement of a hook-rod system for an unstable L1 fracture in a patient without neurological deficit, the patient complains of back pain at the rostral implant insertion site. Do you consider this a complication if (check all that apply): a. You successfully manage the pain with an exercise program 12% b. Narcotic analgesics are required to manage the pain 30% c. The pain is managed successfully by surgical removal of the spinal implant 40% d. None 52% 10. If we as surgeons could more accurately define what constitutes a complication (check all that apply): a. Would the quality of our practices be improved? Yes 63% No 28% No response 9% b. Would medical reporting in the literature be enhanced? Yes 92% No 6% No response 2% c. Would quality assurance be enhanced? Yes 76% No 14% No response 10% d. Would the medico-legal climate be: Worsened Improved No effect No 29% 38% 29% response 4% e. And if we could simultaneously standardize which complications were the result of negligence, would the medico-legal climate be: Worsened 20% Improved 50% No effect 23% No response 7%
The definition of a complication is not as clear as outsiders (e.g., the lay public and the legal system) often believe is the case. With all this in mind, and in the best interest of our patients, we should attain and maintain objectivity. We should not be swayed by uneducated or undeserved accolades from the medically naive, or by threats from entrepreneurs. Complications must be defined, avoided, identified, and aggressively managed. Their avoidance, identification, and management should not be charged with emotion and anger but attacked with an armamentarium of logic, thoughtfulness, science, and objectivity. The avoidance, identification, and management of the complications of spine surgery are addressed in the pages that follow by experts in the field. These experts themselves are not infallible. They address complications with which they have had first-hand experience. We must seize the opportunity to benefit from their wisdom and experience. A wise person can learn from the observations and mistakes of others.
Preface
Like a Canadian thistle, a complication means different things to different people. We must put complications in their appropriate perspective by clarifying their definition. Then we should actively avoid them and aggressively identify and manage them when they occur.
Final Comments Chapter 1 by Robert Grossman, in the sister volume to the first edition of this textbook, is worthy of careful review.3 His “Maxims Concerning Complications” are repeated here, to emphasize their value. MAXIMS CONCERNING COMPLICATIONS Some aphorisms that are useful in guiding one’s practice to avoid complications are as follows: 1. There is no such thing as a simple neurosurgical (spine surgery) operation. 2. It is easier to stay out of trouble than to get out of trouble. 3. The time expended in avoiding complications will be more than compensated by the time saved in not having to treat them. 4. The patient’s well-being is paramount. A neurosurgeon (spine surgeon) should never hesitate to request consultation, or assistance, during surgery.
xxi
5. Surgeons should always operate with the meticulousness that they would wish for if they were the patient. It is a salutary exercise for surgeons to think of their own feelings and reactions if they had to undergo the procedures being carried out.
REFERENCES 1. Dorland’s Illustrated Medical Dictionary, ed 26. Philadelphia, WB Saunders, 1981. 2. Black P McL: What is a complication in neurological surgery? A practical approach. In Apuzzo MLJ (ed): Brain Surgery: Complication Avoidance and Management. Supratentorial Procedures. New York, Churchill Livingstone, 1994, pp xxv-xxviii. 3. Grossman RG: Preoperative and surgical planning for avoiding complications. In Apuzzo MLJ (ed): Brain Surgery: Complication Avoidance and Management. Supratentorial Procedures. Churchill Livingstone, New York, 1994, p 9.
Conflict of Interest
In order to minimize bias, the disclosure of potential conflicts of interest is imperative. The following contributors to this book have disclosed financial relationships with industrial partners. These relationships could bias the author’s opinion and, therefore, should be considered accordingly. Author Ronald I. Apfelbaum, MD Giancarlo Barolat, MD Edward C. Benzel, MD Scott D. Boden, MD Joseph S. Cheng, MD Curtis A. Dickman, MD Thomas J. Errico, MD Jan Goffin, MD, PhD Regis W. Haid, Jr., MD Allan J. Hamilton, MD Robert A. Hart, MD, MA Robert F. Heary, MD Patrick W. Hitchon, MD Iain H. Kalfas, MD Lawrence G. Lenke, MD Alan D. Levi, MD Isador H. Lieberman, MD Paul C. McAfee, MD Nagy Mekhail, MD, PhD Hiroshi Nakagawa, MD, PhD Daniel K. Resnick, MD Gerald E. Rodts, Jr., MD Christopher B. Shields, MD, FRCS(C) Vincent C. Traynelis, MD Kenneth S. Yonemura, MD
Industrial Partner Aesculap Instrument Co, Medtronic Advanced Neuromodulation System (ANS), Medtronic DePuy Spine, NuVasive, Orthovita, & Spinal Concepts Medtronic, Centerpulse, Osteotech, & Wright Medical Medtronic Sofamor Danek Johnson and Johnson & Medtronic Fastenetix, Medtronic, & Synthes Spine Spinal Dynamics, Medtronic Sofamor Danek MedTronic; Johnson & Johnson; Biogen & Guilford Pharmaceuticals Stryker Howmedic, MedTronic Sofamor Danek; Depuy Spine; Kyphon; Synthes/AO; Orthovita DePuy Spine DePuy Spine, Medtronic Sofamor Danek, Synthes Spine, Spine Tech Medtronic Sofamor Danek DePuy Spine & Medtronic Kyphon, DePuy Spine, Orthovita DePuy Spine Pfizer, Merck, Medtronic Ammtec (Tokyo, Japan) NuVasiv, Orthovita Medtronic Norton Healthcare Medtronic Sofamor Danek Smith & Nephew – Spine
xxx
C H A P T E R
sis, scoliosis, spinal dislocation, and spinous process fracture. He addressed the relationship between spinal tuberculosis and gibbus. According to Hippocrates, spinous process fracture was not dangerous. However, fractures of vertebral body were more important. He described two frames for reduction of the dislocated spine, including the Hippocratic ladder and the Hippocratic board.2 The details of Hippocratic treatment were recorded by Celsus (25 BC-50 AD). Aristotle (384-322 BC) focused on kinesiology. His treatises—“parts of animals, movement of animals, and progression of animals—described the actions of the muscles.” He analyzed and described walking, in which rotatory motion is transformed into translational motion. Although his studies were not directly related to the spine, they were the first to address human kinesiology and, in fact, biomechanics. Galen of Pergamon (130-200 AD), another physician of the antique era, worked as a surgeon and anatomist. He studied the anatomy of animals and extrapolated his findings to human anatomy. His anatomic doctrines became the gold standard for more than 1200 years. He used the terms kyphosis, scoliosis, and lordosis, and he attempted to correct these deformities. He also worked as the official surgeon of gladiators in amphitheaters. Because of this position, he was accepted as “the father of sports medicine.” He confirmed the observations of Imhotep and Hippocrates regarding the neurological sequences of cervical spine trauma. Nevertheless, to the best of our knowledge, he did not operate for spinal trauma.76 Oribasius (325-400 AD), another antique period physician, added a bar to the Hippocratic reduction device and used it for both spinal trauma and spinal deformity.113 One of the most important figures dealing with spinal disorders during the end of this period is Paulus of Aegina (625-690 AD). He collected what was known from the previous 1000 years in a seven-volume encyclopedia. Paulus of Aegina not only used the Hippocratic bed, but also worked with a red-hot iron. He is credited with performing the first known laminectomy. This was performed for a case of spinal fracture resulting in spinal cord compression. He emphasized the use of orthoses in spinal trauma cases.42
1
History Cary D. Alberstone, Sait Naderi, and Edward C. Benzel
The evolution of spinal surgery has revolved around three basic surgical goals: decompression, surgical stabilization, and deformity correction. To emphasize their importance, these surgical goals form the framework for this chapter. However, other related fundamental arenas, such as anatomy, biomechanics, nonsurgical treatment modalities, contributed to the development of surgical concepts as well. Although the main advances in spine surgery occurred in the nineteenth and twentieth centuries, their roots date back several thousand years. Without understanding and appreciating the past, it is not possible to understand and appreciate the advancements of the last two centuries. Therefore, before touching upon the last two centuries’ spine history, a short history of spine medicine of the antique period, medieval period, and Renaissance is presented.
The Antique Period and Spine Surgery There is no evidence of surgical decompression and stabilization, nor the surgical correction of deformity, during the antique period except for laminectomy in a trauma case reported by Paulus of Aegina. However, it is known that the antique period’s physicians were, to some extent, able to evaluate patients with spinal disorders. They in fact did use frames for reduction of dislocation and gibbus and applied some of the knowledge gained from human and animal dissections. Srimad Bhagwat Mahapuranam, an ancient Indian epic (3500-1800 BC), depicts the oldest documentation of spinal traction. In a passage from this document, it is described that Lord Krishna applied axial traction to correct a hunchback in one of his devotees.63 The Edwin Smith Papyrus (2600-2200 BC) is the most well-known document on Egyptian medicine. This document reports 48 cases. Imhotep (2686-2613 BC), an antique surgeon, authored this papyrus. Six cases of spinal trauma were reported. Hence, nearly 4600 years ago, vertebral subluxation and dislocation and traumatic quadriplegia and paraplegia were described.12 Recently, it was reported that Egyptian physicians described the “spinal djet column concept.”65 Antique medicine was also influenced by the GrecoRoman period physicians.13 Hippocrates (460-375 BC) addressed the anatomy and pathology of the spine. He described the normal curvatures of the spine, the structure of the spine, and the tendons attached to the spine. He defined tuberculous spondylitis, posttraumatic kypho-
The Medieval Period and Spine Surgery The studies and reports of Paulus of Aegina are the most important source of information regarding this period of medicine. This age was followed by the dark age in Europe. Whereas Western medicine showed no progress during the dark age, the Eastern world developed the science. The early Islamic civilizations realized the importance of science and scientific investigation. The most important books of the antique age were translated into Syrian, Arabic, and Persian. Therefore, using the Western doctrines, the Islamic civilizations discovered new information and were able to contribute further. In terms of spine medicine, several important contributors, including Avicenna and Abulcasis, added to this movement. Avicenna (981-1037 AD), a famous physician of this era, worked in all areas of medicine (Figure 1-1). His famous
1
2
Chapter 1 ■ History
Figure 1.1 Avicenna.
book, the Canon, was a seminal textbook until the seventeenth century in Europe. He described the biomechanics-related anatomy of the spine. He also described flexion, extension, lateral bending, and axial rotation of the spine.86 Avicenna also used a traction system similar to the system described by Hippocrates. Abulcasis (936-1013 AD), a famous Arabian surgeon of the eleventh century, wrote a surgery treatise, “At-Tasnif.” He described several surgical disorders, including low back pain, sciatica, scoliosis, and spinal trauma. He recommended the use of chemical or thermal cauterization for several spinal disorders. He also developed a device to reduce dislocated spine.99 Serafeddin Sabuncuoglu (1385-1468 AD), a Turkish physician of the fifteenth century, wrote an illustrated atlas of surgery.85 He described scoliosis, sciatica, low back pain, and spinal dislocations. He described a technique for reduction of spinal dislocations, using a frame similar to that designed by Abulcasis.
Renaissance and Spine Surgery The dark age in Europe evolved into the Renassiance. During this period actual academic centers were established in Europe. In addition, centers for the translation of documents, similar to centers established in Islamic regions, were founded. The antique-age classics were translated into Latin from Arabic in such translation centers. This infusion of scientific information contributed to the “character” of the Renaissance. During this time, the Western world spawned disciplines, including art, medicine, physics, and mathematics. The works of Leonardo da Vinci (1452-1519 AD) are of importance in this regard. da Vinci worked on the philosophy of mechanics and on anatomy in De Figura Humana. He described spine anatomy, the number of vertebrae, and the joints in detail. By studying anatomy
in the context of mechanics, da Vinci gained some insight into biomechanics. He considered the importance of the muscles for stability of the cervical spine. However, his work was unpublished for centuries, and his brilliant daydreaming had a limited scientific impact on biomechanics.87 Andreas Vesalius (1514-1564 AD), an anatomist and physician, wrote his famous anatomy book, De Humani Corporis Fabrica Liberi Septum, and changed several doctrines described by Galen. Actually, it took several centuries for the world to accept that Galen had made errors that were corrected by Vesalius. Because he described and defined modern anatomy, he is commonly accepted as the father of anatomy. He described spine, intervertebral disc, and intervertebral foramina. His biomechanical point of view regarding the flexion extension of the head was similar to that of Avicenna.10 The early anatomic studies and observations were followed by biomechanical advancements. In this regard Giovanni Alfonso Borelli (1608-1679 AD) deserves significant credit. Borelli described the biomechanical aspects of living tissue. He is the founder of the “iatrophysics” concept; a term that subsequently became known as biomechanics. He is accepted as the “father of spinal biomechanics.” His book, De Motu Animalium, describes the movements of animals. He wrote that the intervertebral disc is a viscoelastic material. He reported that the intervertebral disc carries loads. This is so because he observed that muscles could not bear the loads alone. Therefore, he concluded that the intervertebral discs should have function during the load bearing. He was the first scientist to describe the human weight center (center of gravity).75,82 The studies and accomplishments of the Renaissance period were not limited to the aformentioned. Many scientists contributed to the body of the literature in this period. The advancements from this period resulted in the formation of early modern surgery, beginning in the nineteenth century.
The Early Modern Period and Spine Surgery Spinal Decompression and the Early Modern Period Although an open decompression of the spinal canal for spinal cord compression was recommended by some surgeons between the sixteenth and eighteenth centuries (e.g., Pare, Hildanus), there is no evidence of successful intervention except for two cases reported by Paulus of Aegina and Louis prior to the nineteenth century. Spinal decompression in the early modern period was primarily via laminectomy. Throughout most of the nineteenth century, laminectomy was developed and its utility debated as the only surgical approach to all spinal pathologies, including tumor, trauma, and infection. At the dawn of the twentieth century, the indications for laminectomy were extended to the decompression of spinal degenerative disease, an understanding of which had eluded nineteenth-century surgeons because they failed to appreciate the connection between its clinical and pathologic manifestations.
Chapter 1 ■ History
During the nineteenth century, spinal surgery was performed almost exclusively for neural element decompression. Numerous nonoperative approaches to deformity correction were attempted over the centuries, but the surgical approach to deformity correction was a twentieth-century development. Nor were the techniques of spinal stabilization a product of the nineteenth century, because both spinal fusion and internal fixation did not appear until after or around the turn of the century. Moreover, a failure to recognize the implications for treatment of degenerative spinal disease, including spondylosis and degenerative disc disease, meant that the solution to these problems had to await the new century. Thus, during the nineteenth century, the indications for spinal surgery were limited to the treatment of tumor, trauma, and infection. Although each of these conditions posed unique clinical and surgical problems, they shared the need for surgical decompression. Throughout the early modern period, surgical decompression of the spine was the single most common reason to undertake the risks of spinal surgery, and laminectomy was the most commonly used technique to achieve it. The Birth and Development of the Laminectomy H.J. Cline, Jr. and the Argument Against Spinal Surgery At the beginning of the nineteenth century, the prospects for spinal surgery appeared grim. The dismal results of a well-publicized operation for a traumatic spinal injury stimulated a heated debate over the “possibility” of spinal surgery that persisted for nearly a century. At the center of this debate was H.J. Cline, Jr., a little-known British surgeon. In 1814, Cline performed a multilevel laminectomy for a thoracic fracture-dislocation associated with signs of a complete paraplegia (Figure 1.2).23 The patient was a 26-year-old man who fell from the top of a house. “He was bled previous to his admission” to the St. Thomas’s Hospital in London, “and some imprudent attempts were made to relieve him by pressing the knees against the injured part, which only increased the pain and inflammation.”23 Upon admission to the hospital the patient was examined by Cline, who “ascertained that some of the spinous processes . . . were broken off and were pressing upon the spinal marrow . . . [and] who resolved to cut down and remove the pressure from the spinal marrow.”23 The patient was observed overnight in the hospital, and on the day following admission, Cline performed his proposed operation. Although the operation was performed within 24 hours of injury, Cline was unable to reduce the dislocation or to achieve a complete decompression of the neural elements. The patient survived for 3 days after surgery, with increasing pain and a steadily increasing pulse. On postoperative day 4, however, the patient died, “and on an examination of the body by Mr. Cline, it was found that the spinal marrow was entirely divided.”23 Despite the severity of the neural injury, and the complexity of the fracture-dislocation, the untoward outcome of this unfortunate case would remain a topic of conversation for
Figure 1.2 First page of H.J. Cline Jr.’s historic laminectomy, as
reported by G. Hayward. (From Cline HJ Jr [cited by Hayward G]: An account of a case of fracture and dislocation of the spine. New Engl J Med Surg 4:13, 1815.)
almost a century, providing ample ammunition for the opponents of spinal surgery. Of course, the case of Cline was not an isolated mortality. In 1827, for example, Tyrell102 reported a 100% mortality for a small series of patients with spinal dislocation and neurologic injury treated surgically. Other reports (e.g., Rogers92 in 1835) were often equally discouraging. Looking back on these early years of the debate about spinal surgery, the early twentieth century British surgeon Donald Armour7 described the controversy this way: This [Cline’s operation] precipitated and gave rise to widespread and vehement discussion as to its justification. This discussion, often degenerating into bitter and virulent personalities, went on many years. Astley Cooper, Benjamin Bell, Tyrell, South, and others favored it, while Charles Bell, John Bell, Benjamin Brodie, and others opposed it. The effect of so eminent a neurologist as Sir Charles Bell against the procedure retarded spinal surgery many years—the operation was described with such extravagant terms as “formidable,” “well-nigh impossible,” “appalling,” “desparate [sic] and blind,” “unjustifiable,” and “bloody and dangerous.”
Of course, surgical fatalities in this period were due as much to septic complications and anesthetic inadequacies as they were to surgical technique. The lack of an effective means of pain control during surgery intensified the problem of intraoperative shock and made speed essential. Furthermore, the problems of wound infection and septicemia were both predictable and frequently fatal. These hindrances to surgery were not ameliorated until the introduction of general anesthetic agents (i.e., nitrous oxide, ether, and chloroform) in the mid-1840s and the adoption of Listerian techniques (using carbolic acid) in the 1870s.
3
4
Chapter 1 ■ History
A.G. Smith and the First Successful Laminectomy Despite these risks, a little-known Danville, Kentucky surgeon named Alban G. Smith performed a laminectomy in 1828 on a patient who had fallen from a horse and sustained a traumatic paraplegia. To Smith’s credit, his patient not only survived the operation but achieved a partial neurologic recovery. The operative technique and surgical results were reported in the North American Journal of Medicine and Surgery in 1829 (Figure 1.3).97 Smith’s procedure comprised a multilevel laminectomy through a midline incision, involving removal of the depressed laminae and spinous processes, exploration of the dura mater, and closure of the soft tissue incision. Although the report of this landmark case appears to have attracted little attention at the time, it is a significant technical achievement and places Smith among the pioneers of the early modern period in spinal surgery. Laminectomy for Extramedullary Spinal Tumor During the half century after Smith’s historic operation, the primary indication for laminectomy was spinal trauma. In the latter part of the nineteenth century, the indications
Figure 1.3 Title page of journal that contains the first successful report of a laminectomy. The surgeon, and the author of the report, was Alban G. Smith of Danville, Kentucky. (From Smith AG: Account of a case in which portions of three dorsal vertebrae were removed for the relief of paralysis from fracture, with partial success. North Am Med Surg J 8:94-97, 1829.)
for laminectomy were extended to tumor and infection. The first and most celebrated surgical case for spinal tumor in the nineteenth century, which was also the first successful one, played an important role in the rehabilitation of the laminectomy as a safe and effective procedure. This was the case of Captain Gilbey. Captain Gilbey was an English army officer who suffered the misfortune of losing his wife in a carriage accident in which he also was involved. Although Gilbey himself escaped serious injury, he soon began to experience progressive dull back pain, which he attributed to the accident. As the pain became relentless, Gilbey sought the advice of a series of physicians, all of whom were unable to identify the source of his pain. Eventually, Gilbey was referred to the eminent London neurologist, William Gowers, who elicited from the patient a history of back pain, urinary retention, paraplegia, and a thoracic sensory level (Figure 1.4).41 The neurologist’s diagnosis was immediate and unequivocal: the cause of Gilbey’s symptoms was located in his spine, where a tumor was causing compression of the thoracic spinal cord. Although no intraspinal tumor had ever been resected successfully, Gowers referred the patient to his London surgical colleague, Victor Horsley (Figure 1.5). After all, Gowers had himself asserted, in his authoritative textbook, Manual of Diseases of the Nervous System, that removal of an intradural spinal cord tumor was “not only practicable, but actually a less formidable operation than the removal of intracranial tumors.”
Figure 1.4 William R. Gowers.
Chapter 1 ■ History
Horsley acted quickly. Within 2 hours of the initial consultation, a skin incision was made at 1 PM, June 9, 1887 at the National Hospital, Queens Square, London. Despite his precipitous decision to undertake this dangerous operation, Horsley did not approach the operation unprepared. Although the Act of 1876 made it a criminal offense to experiment on a vertebrate animal for the purpose of attaining manual skill, Horsley had repeatedly practiced the proposed procedure in the course of his surgical experimentation. Despite some initial difficulty in locating the tumor, an intradural neoplasm in the upper thoracic spine causing compression of the spinal cord was identified and safely resected. The pathologic diagnosis was “fibromyxoma of the theca.” Follow-up 1 year later revealed almost complete neurologic recovery. The patient was ambulating without assistance and had returned to his premorbid work schedule. He remained well, with no evidence of tumor recurrence, up to the time of his death from an unrelated cause 20 years later. Laminectomy for Intramedullary Spinal Tumor In 1890, Fenger attempted to remove an intramedullary spinal tumor in an operation that resulted in the patient’s death.22 In 1905, Cushing28,29 also attempted to remove an intramedullary spinal cord tumor but decided to abort the procedure after performing a myelotomy in the dorsal column. To Cushing’s surprise, the patient improved after surgery. In 1907, von Eiselsberg109 successfully resected an intramedullary tumor.
The unexpected improvement that was observed in the patient reported by Cushing attracted the attention of the New York surgeon Charles Elsberg. Elsberg34 described the technique of Cushing, which he aptly named the “method of extrusion.” The technique was intended to remove an intramedullary tumor by spontaneous extrusion of the tumor through a myelotomy made in the dorsal column. The rationale for this method was predicated on the theory that an intramedullary tumor was associated with an increase in intramedullary pressure. Release of this pressure by a myelotomy that extended from the surface of the spinal cord to the substance of the tumor was expected to provide a sufficient force to achieve spontaneous extrusion of the tumor. According to Elsberg, the advantage of this procedure over a standard tumor resection was that it required minimal manipulation of the spinal cord and therefore minimal spinal cord tissue injury. Because the spontaneous extrusion of an intramedullary tumor occurred slowly, Elsberg performed these procedures in two stages. In the first stage, a myelotomy was fashioned in the dorsal column, extending from the surface of the spinal cord to the tumor (Figure 1.6A). When the tumor was identified and observed to begin to bulge through the myelotomy incision, the operation was concluded, the dura mater left opened, and the wound closed. In the second stage of the procedure, which was performed approximately 1 week after the first stage, Elsberg reopened the wound and inspected the tumor (Figure 1.6B). Typically, the tumor was found outside the spinal cord, and the few adhesions that remained between the spinal cord and the tumor were sharply divided. After the tumor was removed, the wound, including the dura mater, was closed. Variations in Laminectomy Technique By the last decade of the nineteenth century, after the case of Captain Gilbey, the possibility of safely performing a spinal operation was embedded in the collective surgical consciousness. Furthermore, new anesthetic techniques and aseptic methods had become available to most practicing surgeons.5 All of these factors served to increase the appeal of the laminectomy to surgeons and to widen its range of application. For example, after Horsley’s widely publicized surgical success for spinal tumor, many similar operations were soon described in the literature,* and in 1896, Makins and Abbott74 reported 24 cases of laminectomy for vertebral osteomyelitis. Although the safety and efficacy of the laminectomy had convinced many proponents of the utility of the procedure, surgeons toward the end of the century began to worry about postoperative instability. Advances in operative technique and perioperative management meant that more and more patients survived the operation and ultimately became ambulatory, which further heightened concern about stability. In 1889, Dawbarn31 described an osteoplastic method of laminectomy that addressed this concern about postoperative spinal stability. Instead of a midline incision, Dawbarn described two lateral incisions that were carried down to the
Figure 1.5 Sir Victor Horsley.
*References 1,19,27,39,66,93.
5
6
Chapter 1 ■ History
transverse processes. The lateral incisions were connected in an H-like fashion, and a superior and inferior flap— including skin, muscle, fascia, and bone—was then turned. In closing the wound, the intact flaps were reflected back and reapproximated in their normal anatomic positions. Although not all surgeons subscribed to the osteoplastic method, many turn-of-the-century surgeons were largely preoccupied with modifications of this procedure.77 At the same time, however, a more important innovation in laminectomy technique, the hemilaminectomy, was developed independently in both Italy4,108 and the United States.100 In 1910, A.S. Taylor of New York described the hemilaminectomy: a midline incision, a subperiosteal paraver-
tebral muscle takedown, and the removal of a hemilamina with a Doyen saw. The advantages of the hemilaminectomy over the cumbersome osteoplastic method were obvious, and Taylor argued that compared with the laminectomy, the hemilaminectomy interfered less with the mechanics of the spine. Despite such detractors as Charles Elsberg, who responded that the field of view was narrow and the effect of laminectomy on spinal mechanics negligible, Taylor successfully championed its use. Charles A. Elsberg: The Laminectomy in Stride Charles A. Elsberg was one of the most influential writers on spinal decompression (Figure 1.7). Working at the Neurological Institute of New York, which he had helped to found, Elsberg36 published his first series of laminectomies in 1913. In 1916, he published his classic text, Diagnosis and Treatment of Surgical Diseases of the Spinal Cord and Its Membranes.35 Although these publications represent landmarks in the history of spinal surgery, they comprise more of a culmination than an innovation in spinal surgery. Elsberg’s work on spinal surgery, coming as it did at the end of a century of evolution of the decompressive laminectomy, effectively codified nineteenth and early twentieth century developments. In his textbook, Elsberg outlined the surgical indications and contraindications for laminectomy. He noted the beneficial effects in his own large series of laminectomies and puzzled over the benefits that may occur in the absence of
Figure 1.6 (A) The first stage in an intramedullary spinal cord
tumor resection by the extrusion method. Note that the tumor is bulging through the myelotomy incision. The wound was subsequently closed. (B) The second stage in an intramedullary spinal cord tumor resection by the extrusion method, 1 week after the first stage. Note that the tumor has spontaneously extruded since the first operation, and now may be removed easily. (From Elsberg CA, Beer E: The operability of intramedullary tumors of the spinal cord. A report of two operations with remarks upon the extrusion of intraspinal tumors. Am J Med Sci 142:636-647, 1911.)
Figure 1.7 Charles A. Elsberg.
7
Chapter 1 ■ History
evident increased intradural pressure, such as in patients with multiple sclerosis. He argued that the primary indications for operation were cases of tumor, trauma, and infection that were associated with symptoms localized to a spinal level. Patients with progressive symptoms should be operated on quickly, in the absence of contraindications such as metastatic cancer or advanced Pott’s disease. Given the exhaustive scope of these early Elsberg publications—which, in addition to tumor, trauma, and infection, also review the management of congenital spine disease—conspicuously little is said about the most common late twentieth century indication for laminectomy: degenerative spine disease. The tardy development of a treatment for degenerative spine disease should be understood in the larger context of nineteenth and early twentieth century knowledge of spinal pathology. Unlike degenerative disease, tumor, trauma, and infection were already well-known in antiquity. Although the concept of localization of function in the nervous system was undeveloped during the nineteenth century, the diagnosis and localization of tumor, trauma, and infection, particularly in their late stages, was not especially difficult. Degenerative disease, on the other hand, possessed a more subtle pathophysiology that was not as easily characterized, especially without the help of radiography. Thus recognition of degenerative spine disease eluded the nineteenth-century surgeon. This tardy appreciation for the clinical, surgical, and pathologic importance of degenerative spine disease deserves further mention. Figure 1.8 Rudolph Virchow.
Laminectomy for Intervertebral Disc Herniation. Intervertebral disc pathology was first described by Rudolph Virchow108 in 1857 (Figure 1.8). Virchow’s description of a fractured disc was made at autopsy on a patient who had suffered a traumatic injury. In 1896, T. Kocher,61 a Swiss surgeon identified and described a traumatic disc rupture in an autopsy of a patient who had fallen 100 feet and landed on her feet. Although Kocher recognized that the L1-L2 disc was displaced dorsally, no clinical correlation was suggested. The first transdural intervertebral discectomy was reported by Oppenheim and Krause89 in 1908. However, they reported the disc as “enchondroma.” In 1911, George Middleton,81 a practicing physician, and John Teacher, a Glasgow University pathologist, described two cases of ruptured intervertebral disc observed at autopsy. Like Virchow and Kocher before them, however, Middleton and Teacher, although they described the pathology, failed to postulate its connection with radiculopathy or back pain. In 1911, Joel Goldthwaite39 made this connection. In an article on the lumbosacral articulation, Goldthwaite described and illustrated how weakening of the anulus fibrosus could result in dorsal displacement of the nucleus pulposus. The nucleus pulposus, he argued, could in turn result in low back pain and paraparesis. What eluded Goldthwaite and the surgeons before him, however, was the connection between a herniated disc and radiculopathy. In a 1929 issue of the Archives of Surgery, Walter E. Dandy30 published a description of two cases of herniated lumbar discs causing a cauda equina syndrome (Figure 1.9). Dandy correctly described how “loose cartilage from
Figure 1.9 Walter E. Dandy.
8
Chapter 1 ■ History
the intervertebral disc” produced the symptoms of cauda equina compression that were relieved after surgical decompression. He considered that in the second decade of the twentieth century, more than 20 years after the first spinal fusion operations, intervertebral disc disease could be added to the list of indications for decompressive laminectomy. Despite the several aformentioned publications on intervertebral disc herniation, the concept of disc herniation and its relationship to radiculopathy was defined by Mixter and Barr. While several studies were performed in North America, an anatomic, radiologic and microscopic study was performed on 5000 human spines in Dresden Pathology Institute by Schmorl and Junghanns. The results of this study were published in a book entitled “The Human Spine in Health and Disease.” In 1932, Barr, an orthopedic surgeon from Massachussets General Hospital, was assigned to write a critique of this study. In June of 1932, Barr attempted to treat a patient with an extruded disc herniation. Following a 2-week unsuccessful course of nonoperational treatment, Barr consulted with Mixter. Mixter recommended a myelogram. The myelogram revealed a filling defect. Mixter operated on the patient and removed the “tumor.” Barr studied the “tumor” specimens. Because Barr contributed to Schmorl’s study published in German, he remembered the microscopic appearances of Schmorl’s study and realized that the specimen from this index patient was the nucleus pulposus. After this finding, Mixter, Barr, and Mallory (pathologist) reevaluated all the cases that were diagnosed (or misdiagnosed) as chondroma in recent years at Massachusetts General Hospital. They retrospectively diagnosed most of these cases as ruptured intervertebral discs. Mixter and Wilson operated on the first ruptured disc herniation diagnosed preoperatively on December 31, 1932. Mixter and Barr reported the case in New England Surgical Society in September 30, 1933.83,113 In the late 1930s Love69 from Mayo Clinic reported on an extradural laminectomy technique. In 1967, Yasargil114 used the microscope for discectomy. The first results of the lumbar microdiscectomy were reported by Yasargil114 and Caspar.20
Further demonstration of the efficacy of decompressive laminectomy for spinal stenosis came from Sachs and Frankel94 in 1900. They published an account of a man aged 48 years with neurogenic claudication and spinal stenosis whose symptoms improved after a two-level laminectomy. Recognition of the degenerative nature of the clinical entity of spinal stenosis was established by Bailey and Casamajor8 in 1911 in a report on a patient who was successfully decompressed by Charles Elsberg. In his 1916 textbook, Elsberg35 later wrote, “a spinal operation may finally be required in some cases of arthritis or spondylitis on account of compression of the nerve roots or the cord by new-formed bone. . . .” In 1945, Dr. Sarpyener, a Turkish orthopedic surgeon, described congenital lumbar spinal stenosis.95 This report was followed by a report on adult spinal stenosis from Dr. Verbiest.105 In 1973 Hattori48 described the technique of laminoplasty.
Approaches to the Spine Dorsolateral Approaches to the Spine In 1779, Percival Pott described a condition involving spinal kyphosis and progressive paraplegia in a now-classic monograph titled “Remarks on that kind of palsy of the lower limbs which is frequently found to accompany a curvature of the spine and is supposed to be caused by it; together with its method of cure; etc.” (Figure 1.10). For the management of this condition, which now bears his name, Pott recommended the use of a paraspinal incision to drain pus from the invariably present paraspinal abscess. For almost a century, this simple surgical procedure became a standard part of the treatment of Pott’s paraplegia.
Laminectomy for Cervical Disc Herniation. In 1905 Watson and Paul110 performed a negative exploration for cervical spinal cord tumor. They found an anterior extradural mass in the intervertebral disc at autopsy. This may be the first reported case of cervical disc herniation. The first dorsal approach was performed by Elsberg37 in 1925. He found a “chondroma” in a quadriparetic patient. Laminectomy for Spinal Stenosis. Unlike the herniated intervertebral disc, the stenotic spinal canal was described comparatively early in the nineteenth century. Portal,90 in 1803, observed that a small spinal canal may be causally related to spinal cord compression, leading to paraplegia. No clinical reports of this entity were published, however, until 1893 when William A. Lane64 described the case of a woman aged 35 years with a progressive paraplegia and a degenerative spondylolisthesis. The patient improved after a decompressive laminectomy.
Figure 1.10 Percival Pott.
Chapter 1 ■ History
By the late nineteenth century, however, the laminectomy had received widespread acceptance as a safe and effective method of spinal decompression. This was in part related to the decrease in surgical mortality associated with the adoption of the Listerian methods beginning in the 1870s, and it was only natural then that the laminectomy would play a role in the management of Pott’s disease. As in many of its applications, however, disenchantment arose with the results of laminectomy, and alternative approaches were therefore sought.115 The most promising of these approaches was the so-called “costotransversectomy” of Ménard. Ménard’s Costotransversectomy Like many surgeons at the turn of the century, Ménard79 was disappointed by the surgical results of the laminectomy. In 1894, he described the costotransversectomy as an alternative method to achieve the goal of Pott, namely, drainage of the paraspinal abscess. The advantage of the costotransversectomy over the laminectomy lay in the improved exposure that it provided of the lateral aspect of the vertebral column. The procedure was also known as the “drainage latéral,” emphasizing that the goal of the procedure was to drain the lateral, paravertebral tubercular abscess. As described by Ménard, the costotransversectomy involved an incision overlying the rib that was located at the apex of the kyphos. The rib was then skeletonized and divided about 4cm distal to the articulation with its corresponding vertebra, from which it was disarticulated and removed. These maneuvers provided access to the tuberculous focus, which was exposed and then decompressed directly (Figure 1.11). Ménard did not intend to totally remove the lesion, but rather to simply decompress the abscess. The surgical results of Ménard’s costotransversectomy far surpassed the results obtained with the
laminectomy. There were several successes among his first few cases, including significant motor improvement among his first 23 cases.98 Regrettably, these promising initial surgical results began to sour with time, as it became increasingly clear that two major complications were occurring with increasing frequency. These complications comprised the postoperative development of secondary infections and the postoperative formation of draining sinus tracts, both of which were created as a result of opening up the abscess. Because no antitubercular chemotherapeutic agents were available at the time, the consequences of the infections that ensued after surgery were frequently disastrous, resulting in significant surgical mortality. As Calot16 grimly put it in 1930, “The surgeon who, so far as tuberculosis is concerned, swears to remove the evil from the very root, will only find one result waiting him: the death of his patient.” The operation of Ménard thus fell into disrepute, and in time even Ménard abandoned it. Capener’s Lateral Rhachotomy Like Ménard, Norman Capener of Exeter and Plymouth, England, attempted to find a surgical solution to the problem of Pott’s paraplegia. Capener modified Ménard’s costotransversectomy in a procedure that he developed and began using in 1933, which was first reported by H.J. Seddon96 in 1935. Departing from the emphasis of Pott and Ménard, who simply decompressed the tubercular abscess, Capener attempted to directly remove the lesion, which typically consisted of a ventral mass of hardened material. To achieve his more radical goal of spinal decompression, Capener required a more lateral or ventral view of the vertebrae than was afforded by Ménard’s costotransversectomy. Capener’s solution was to adopt Ménard’s costotransversectomy with this difference: whereas Ménard approached the spine via a trajectory that was medial to the erector spinae muscles, Capener18 transversely divided the muscles and retracted them rostrally and caudally (Figure 1.12). He named his new approach the lateral rhachotomy to distinguish it from Ménard’s costotransversectomy. The simple change in dissection planes that distinguishes the
Figure 1.11 Drainage of a tubercular abscess via the
Figure 1.12 Dorsolateral exposure via Capener’s lateral
costotransversectomy of Ménard. (From Ménard V: Causes de la paraplegia dans le mal de Pott. Son traitement chirurgical par l’ouverture direct du foyer tuberculeux des vertebras. Rev Orthop 5:47-64, 1894.)
rhachotomy. Note that the exposure requires a transverse division of the paraspinal muscles. (From Capener N: The evolution of lateral rhachotomy. J Bone Joint Surg 36B:173-179, 1954.)
9
10
Chapter 1 ■ History
costotransversectomy from the lateral rhachotomy produces a significantly different trajectory and surgical exposure. Although the operation was designed for the surgical treatment of Pott’s paraplegia, Capener later drew attention to the versatility of the approach, and its appropriateness for a variety of pathologic processes, including “the exploration of spinal tumors, the relief of certain types of traumatic paraplegia, and the drainage of suppurative osteitis of the vertebral bodies.”18 It was perhaps unfortunate that for 19 years the only description of Capener’s lateral rhachotomy was in a single case report published by another surgeon.96 Not until 1954 did Capener himself describe the procedure, and even then he still chose not to publish the results of his 23 cases.18 In the interval between Seddon’s 1935 description of the lateral rhachotomy and Capener’s 1954 report of the same operation, the emergence of a new treatment, antitubercular chemotherapy, was to transform the history of the treatment of Pott’s paraplegia. In 1947, streptomycin first became available for clinical use. This was followed by the introduction of para-aminosalicylic acid (PAS) in 1949 and isoniazid (INH) in 1952. The effect of the introduction of these new chemotherapeutic agents on the treatment of tuberculosis was spectacular. With the addition of streptomycin alone, the average relapse rate of tuberculosis was decreased by 30% to 35%. Although the effect of antitubercular chemotherapy was not as substantial for the treatment of spinal tuberculosis as for the pulmonary form, its mere availability raised new questions about the optimal management of Pott’s paraplegia and, in particular, about the indications for surgical intervention.
Spinal Stabilization and Deformity Correction The history of surgical stabilization and deformity correction must include a description of the birth and evolution of spinal fusion and spinal instrumentation. Special emphasis must be given to the role of spinal biomechanics and its influence on the development of internal fixation. Many factors hindered the development of surgical approaches to the decompression, stabilization, and deformity correction of the ventral spine. The development and mastery of the special techniques that were required to safely manage ventral spinal pathologies did not appear until after the turn of the century, in part because they depended on advances in anesthetic techniques and a more sophisticated approach to perioperative management. Except for degenerative disease, the technique and indications for the decompressive laminectomy were well established by the turn of the century. The idea of spinal decompression, previously the exclusive province of surgical pioneers, had demonstrated its clinical utility with results that fully justified its acceptance into standard surgical practice. However, the idea of decompression, which had dominated spinal surgery during the nineteenth century, did not exist alone. Indeed, before the dawn of the twentieth century, attention had already turned to another surgical idea: spinal stabilization. Of course, many attempts at surgical stabilization of the unstable spine had been made during the nineteenth century and before. However, the ancient admonition that vertebral fractures comprised an “ailment not to be treated” was reinforced by the surgeon’s singular lack of success. And, thus, despite early attempts at spinal stabilization in the latter part of the nineteenth century, spinal decompression
Larson’s Lateral Extracavitary Approach In 1976, Sanford J. Larson and his colleagues67 at the Medical College of Wisconsin published an influential article that helped to popularize Capener’s lateral rhachotomy, which they modified and renamed the lateral extracavitary approach (Figure 1.13). This approach has been used more for trauma and tumor than for tuberculosis. The technical difference that distinguishes the lateral rhachotomy from the lateral extracavitary approach lies primarily in the treatment of the paraspinous muscles. Whereas the procedure of Capener involves transversely dividing these muscles and reflecting them rostrally and caudally, the procedure of Larson involves a surgical exposure with a trajectory ventral to the paraspinous muscles, which are then reflected medially to expose the ventrolateral aspect of the spine. Later in the procedure these muscles are redirected laterally to provide access for instrumentation of the dorsal aspect of the spine using the same surgical exposure as that during the ventrolateral approach. Although neurosurgeons, as spine surgeons, had traditionally emphasized spinal decompression over spinal stabilization, an essential aspect of the significance of Larson’s overall contribution to the discipline of spinal surgery lies in the fact that, as a neurosurgeon, he dedicated his career to the advancement of reconstructive spinal surgery.
Figure 1.13 Sanford J. Larson.
Chapter 1 ■ History
remained the primary indication for surgery of the spine, until World War II. Recognition of the idea that compression of the neural elements, in cases of tumor, trauma, and infection, could be responsible for neurologic compromise was the crucial first step needed to develop the idea that spinal decompression could improve neurologic outcome. The invention of a technical means to achieve decompression, namely by laminectomy, represented the next necessary step in bringing this concept to clinical practice. Similarly, the idea of spinal stabilization arose from the observation that the unstable spine was at risk for the development of progressive deformity and that surgical intervention might prevent this. Of course, bringing this concept into practice depended on achieving an adequate technical means. And indeed, two technical advances were developed around the turn of the century that provided a means for spinal stabilization that would revolutionize the practice of modern spinal surgery.
The Birth and Development of Spinal Fusion and Spinal Instrumentation Both spinal fusion and spinal instrumentation were born around the turn of the century as methods of stabilizing the unstable spine. For many years, these two technical advances were developed and applied essentially independently, with results that were often complicated by pseudarthrosis. Early attempts at spinal instrumentation in particular failed to gain popularity because of their inability to maintain more than immediate spinal alignment. Spinal
fusions were often used to achieve stabilization, but these also frequently suffered a similar fate: pseudarthrosis.101 By the 1960s, however, a half century of experience with spinal fusion and instrumentation suggested the concept of the “race between bony fusion and instrumentation failure.” The improved surgical results that arose from the application of this important surgical concept provided support for the successful strategy of combining spinal instrumentation with meticulous fusion. Spinal Fusion The idea of using spinal fusion for stabilization is attributed to Albee3 and Hibbs,49 who, in 1911, independently reported its use (Figure 1.14). Although these early operations were performed to prevent progressive spinal deformation in patients with Pott’s disease, the procedure was later adopted in the management of scoliosis and traumatic fracture. The method of Hibbs, which was most frequently used, comprised harvesting an autologous bone graft from the laminae and overlaying the bone dorsally. Despite later improvements in this technique, however, such as the use of autologous iliac crest graft, the rate of pseudarthrosis, particularly in scoliosis, remained unacceptably high.11 In the 1920s Campbell17 described trisacral fusion and iliac crest grafting. In 1922 Kleinberg59 used xenograft for spinal fusion. ALIF was described by Burns14 in 1933, and PLIF was performed by Cloward24 in 1940. In the late 1990s TLIF was described. In 1977 Callahan et al.15 used bone for lateral cervical facet fusion. Several ventral cervical fusion techniques were described in 1950s. Robinson and Smith91 described their
Figure 1.14 (A) Fred Albee. (B) Russell Hibbs.
11
12
Chapter 1 ■ History
technique in 1955, and Cloward25 described his cervical fusion technique in 1958. Spinal Instrumentation and Clinical Biomechanics Like spinal fusion, internal fixation was first applied around 1900. These early constructs comprised tensionband fixators that were applied dorsally, primarily in cases of trauma. The limitation of the constructs, however, soon became apparent because the metals they contained were subject to the corrosive effects of electrolysis. With the introduction of vitallium by Venable and Stuck104 in the 1930s, a metal was found that was previously used successfully as a dental filling material and that had proven resistant to electrolysis (Figure 1.15).103 Further attempts at internal fixation during the 1930s and 1940s included fixed-moment arm cantilever constructs. These also failed to maintain alignment.58,77,111 F.W. Holdsworth In the 1950s, the British orthopedic surgeon Sir Frank W. Holdsworth51 performed perhaps the first large systematic study of the problem of internal fixation for the treatment of posttraumatic fracture. Although the constructs he used, which comprised cantilever beams attached to the spinous processes, were traditional, Holdsworth’s emphasis on patient selection brought the process of surgical spinal stabilization to a new, more sophisticated, level. His rationale for patient selection was based on a biomechanical definition of instability that Holdsworth derived from a study of a large number of spinal-injured patients. In 1963, Holdsworth52 published his results and proposed a classification scheme of subaxial spinal fractures based on a two-column model of spinal stability. Four categories of fractures were identified on the basis of the mechanism of injury and on the presence or absence of
spinal stability. The latter determination rested significantly on the integrity of the dorsal ligaments. Holdsworth categorized the fractures as follows: 1. Pure flexion: A pure flexion mechanism is usually associated with an intact dorsal ligamentous complex and no evidence of spinal instability. The vertebral body absorbs the greater part of the impact, and the result is a wedge compression fracture (Figure 1.16A). 2. Flexion-rotation: A rotation or flexion-rotation mechanism causes disruption of the dorsal ligamentous complex and results in an unstable fracture-dislocation. It is usually associated with paraplegia (Figure 1.16B). 3. Extension: An extension mechanism, which is usually stable, most frequently occurs in the cervical spine. It may be associated with a fracture of the dorsal elements, with an intact dorsal ligamentous complex (Figure 1.16C). 4. Compression: A compression or “burst” fracture is caused by forces transmitted directly along the line of the vertebral bodies. All of the ligaments are usually intact, and the fracture tends to be stable (Figure 1.16D).
Figure 1.16 (A) Wedge compression fracture of the vertebral
Figure 1.15 Radiograph showing no bone changes in dog limb
about vitallium screws (right), but erosion of bone around steel screws (left). (From Venable CS, Stuck WG, Beach A: The effects on bone of metals; based upon electrolysis. Ann Surg 105:917-938, 1937.)
body. Pure flexion mechanism. Note that the posterior ligamentous complex is intact. (B) Rotational fracturedislocation of the lumbar spine. The posterior ligamentous complex is disrupted. This is a very unstable injury. (C) Extension injury. The anterior longitudinal ligament is ruptured. The posterior ligamentous complex is intact. (D) Burst fracture. All ligaments are intact. (From Holdsworth FW: Fractures, dislocations, and fracture-dislocation of the spine. J Bone Joint Surg 45B:6-20, 1963.)
13
Chapter 1 ■ History
In his 1891 report of a case of interspinous wiring for cervical fracture, Berthold Hadra43 considered in what circumstances his newly described procedure would be indicated (Figure 1.17). Hadra concluded that his procedure might be indicated for “any deviation of a vertebra.”43 Despite the prescience of his innovation, the substance of Hadra’s comment is remarkable, not so much for what it contains, as for what is missing from it; namely, any hint of consideration of biomechanical principles. When one considers the importance of biomechanical principles in Holdsworth’s 1963 classification of spinal fractures, Hadra’s turn-of-the-century approach to spinal stabilization serves to underline how much progress was made in the interval. The significance of this new (biomechanic) approach to spinal stabilization, which was heralded by Holdsworth, was brought home in the 1960s with the work of the father of modern spinal stabilization, Paul Harrington (Figure 1.18).
In 1945, after military service in World War II, Paul Harrington47 entered into orthopedic practice in Houston, Texas. Within 2 years of beginning his practice in Houston, Harrington was faced with the orthopedic problems of a large population of poliomyelitis patients, which at that time had reached epidemic proportions. The involvement of the trunk, which afflicted many of these patients, often resulted in scoliotic spinal deformity in association with cardiopulmonary compromise. The presence of cardiopulmonary compromise in a patient with scoliosis often meant that the standard cast corrective measures could not be applied safely. Furthermore, in 1941, the American Orthopaedic Association6 published a report on the results of treatment in 425 cases of idiopathic scoliosis. The report was quite discouraging. Among those patients treated by exercises and braces, but without spinal fusion, 60% progressed in their deformity, and 40% remained unchanged. In another group of patients who underwent surgical correction and fusion, 25% (54 of 214) developed pseudarthrosis and 29% had lost all correction. Among the entire group, the end result for 69% was considered fair or poor, and only 31% were rated good to excellent. It was against this backdrop of dismal results from nonoperative treatment and dorsal spinal fusion that Harrington began his seminal work. After an initial (unsuccessful) trial of internal fixation with facet screw instrumentation,46 the method was abandoned in favor of a combination of compression and distraction hooks and rods made of stainless steel. The advantages of these instruments in the establishment of deformity correction became obvious: for the first time in
Figure 1.17 Berthold Hadra.
Figure 1.18 Paul Harrington.
Holdsworth’s classification was important, as he himself observed, not as a biomechanical theory (although it was this too), but because it had implications for treatment. At around the same time that Holdsworth’s article appeared, several other classifications of spinal fractures were proposed. With the introduction of modern spinal biomechanics, a new era in spinal surgery had begun.9,57
Paul Harrington and the Birth of Modern Surgical Stabilization
14
Chapter 1 ■ History
the history of spinal stabilization, spinal instruments provided compression, distraction, and three-point bending forces, which proved equally useful in deformity correction as it did in the maintenance of post-traumatic stability. There were 19 patients observed during the early phase of Harrington’s investigation of dorsal instrumentation. The results of this investigation were published in 1962.47 The longevity of Harrington’s spinal instrumentation system, which remains in use today, is a testimony to both its safety and its efficacy. Nevertheless, despite a frequent and gratifying correction of the poliomyelitis curvature, the loss of that correction was commonly discovered within 6 to 12 months after surgery. In part, the failure to maintain the alignment achieved at surgery was the result of frequent instrument failure, most commonly instrument fracture and disengagement of the hooks. However, more fundamentally, Harrington recognized that the concept of a dynamic correction system was inherently flawed: the complication of instrument failure would be far less significant if a spinal fusion could maintain the deformity correction achieved by the placement of the implant. The underlying principles that emerged from Harrington’s early failures, then, became clear: (1) because spinal instruments fail over time, they should be applied as a strictly temporary measure; and (2) after instrumentation failure, a successful spinal fusion will maintain stabilization. As a corollary to these principles, Harrington acknowledged that there is a “race between instrumentation failure and the acquisition of spinal fusion.” It stands to reason that if fusion is attained before instrumentation failure, the maintenance of deformity correction and stabilization will have been achieved. An understanding of the importance of a successful fusion in an instrumented spine is one of Harrington’s most significant contributions to spinal surgery and marks the birth of the modern era of spinal stabilization and deformity correction.
Ventral Approaches to the Spine Dorsal decompression via the laminectomy had become well-established by the turn of the century and was codified by Charles Elsberg in his 1916 textbook, Diagnosis and Treatment of Surgical Diseases of the Spinal Cord and Its Membranes. Interestingly, whereas the turn of the century marked the culmination of dorsal decompression in spinal surgery, it also signified the beginning of procedures for dorsal stabilization and deformity correction, as pioneered by Hadra (1891), Albee (1911), and Hibbs (1911). The groundwork for further development in this area was laid with the classification scheme of spinal fractures by mechanism and stability, as initially proposed by Holdsworth in 1963. This introduction into clinical practice of the principles of spinal biomechanics is also found in the work of Harrington in the 1950s and 1960s in his development of a novel system of dorsal thoracolumbar instrumentation. Although Harrington later recognized the need to supplement his instrumentation with meticuluous spinal fusion, and many modifications and innovations have since been made in dorsal instrumentation, successful outcomes in dorsal decompression, stabilization, and deformity correction had been achieved by the 1960s.
Nothing, however, has been said so far about the achievement of these goals in the ventral spine, where a significant portion of spinal pathology is located. As it happens, the first successful interventions for stabilization of the ventral spine were achieved in the same time frame as the dorsal ones (that is, in the first half of the twentieth century). What is peculiar about surgery of the ventral spine is that a decompressive procedure must be accompanied almost invariably by simultaneous stabilization, which often includes measures taken to obtain deformity correction. Therefore the history of the major goals of ventral spine surgery—that is, decompression, stabilization, and deformity correction—has been one of parallel developments, not serial ones, as was the case for the dorsal spine. In other words, the history of stabilization and deformity correction of the dorsal spine developed in the half century following the establishment of dorsal decompression. All three goals, in the ventral spine, were achieved during the same 50 years. Ventral Decompression and Stabilization The primary difficulty in applying ventral techniques to the spine was in the surgical approach. The relative technical ease and low morbidity associated with a dorsal approach to the dorsal spine provided ample opportunity for the early development of dorsal spinal techniques. By contrast, ventral approaches to the ventral spine required transgression of the abdomen or chest, which (similar to the head) up until the 1880s remained sanctuaries not to be opened, except by accident.78 In part, the late development of abdominal and thoracic surgery was a product of the problem of infection: cognizant of the morbidity and mortality related to hospital-acquired gangrene, few of those who entered a surgical ward in the nineteenth century did so with the hope of leaving alive. The reluctance to adopt the principles of antisepsis as first enunciated by Lister68 in 1867 and a slowness to accept its theoretical foundation—the germ theory of disease—meant delays for the development of abdominal and thoracic surgery. However, even after the practice of antiseptic surgery became generally accepted, turn-of-the-century surgeons still approached abdominal surgery with trepidation. Anyone who would contemplate surgically violating the thoracic cavity had to face the technical problem of the pressure relationships in the chest.80 Beginning in 1903, Ferdinand Sauerbruch of Breslau conducted a series of experiments that led to the development of an apparatus in which negative pressure for the open thorax could be maintained, and around 1910, endotracheal or insufflation anesthesia became available (Figure 1.19). This alleviated one of the major technical difficulties confronted by would-be thoracic surgeons, but even then, good control of respiration by a reliable apparatus was not widely available until the late 1930s. W. Müller The first report of a successful attempt to approach the ventral thoracic or lumbar spine is attributed to Müller.84 In 1906, Müller performed a transperitoneal approach to
Chapter 1 ■ History
the lumbosacral spine in a patient with a suspected sarcoma. At operation, Müller found tuberculosis. After curetting the infected bone, Müller applied iodoform powder and closed. The surgical result was excellent. Notwithstanding the success of this initial operation, however, later attempts at the same procedure failed miser-
Figure 1.19 An early version of Sauerbruch’s negative-pressure
chamber.
ably. After several misadventures that ended in disaster, Müller was forced to abandon further attempts at a ventral exposure. B.H. Burns Perhaps the next published report of a successful ventral exposure did not appear until 1933, when B.H. Burns14 of Great Britain performed a ventral interbody fusion of the lumbosacral spine for an L5-S1 spondylolisthesis (Figure 1.20). Before the Burns procedure, the only method available to stabilize an unstable spondylolisthesis was a dorsal fusion. However, the results of dorsal fusion for ventral instability, as Burns himself learned firsthand, proved unsound both in theory and in practice. Faced with a high incidence of failed dorsal fusions, Burns chose to take a transabdominal, transperitoneal approach to the lumbosacral spine, which he first investigated on three cadavers before operation. The first operation involved a 14-year-old boy who presented with low back pain and neurogenic claudication after jumping from a height. A radiograph of the lumbosacral spine showed an L5 spondylolysis and a Grade II, L5-S1 spondylolisthesis. A tibial autograft was taken and was tamped into a hole drilled obliquely from L5 to S1. Convalescence was uneventful and pain relief was achieved, even on ambulation at 2 months postoperatively.
Figure 1.20 (A) Lateral radiograph of lumbar spine showing the graft placement in B.H. Burns’
operation for spondylolisthesis. (From Burns BH: An operation for spondylolisthesis. Lancet 1:1233, 1933, with permission.) (B) Illustration of Burns’ operation. Ventral view.
15
16
Chapter 1 ■ History
Ito and Others Like the landmark operations of Albee and Hibbs, the first reported series of ventral spinal operations comprised a group of surgical treatments for spinal tuberculosis. In their 1934 article, “A New Radical Operation for Pott’s Disease,” Ito and others54 observed that the surgical stabilization procedure described by Albee and Hibbs did not differ significantly from nonoperative immobilization; the goal in both instances was to rest and unload the diseased spine. Ito, on the other hand, a professor of orthopedic surgery from Kyoto, Japan, proposed a decompressive procedure, which he believed provided a definitive surgical treatment. Of course, the obstacles that Ito confronted in devising a ventral approach to the spine were considerable. In addition to the obvious anatomic obstacles, all turn-of-thecentury spine surgeons faced the seemingly intractable problem of infection. Although postoperative infections posed major difficulties for the development of (clean) abdominal and thoracic surgical procedures, these difficulties were compounded when the surgical indication was infection, as in the case of Pott’s disease. Indeed, previous attempts to surgically decompress tuberculosis of the ventral spine via a lateral approach (i.e., a costotransversectomy) met with a high incidence of complications from postoperative secondary infection, permanent fistulas, or persistent spinal tuberculosis resulting from incomplete removal of infected bone.60,79,106,107 In part, these operations failed because they were performed prior to 1910, in the age of antiseptic, rather than aseptic, surgery. Perhaps they also failed in part because they predated the introduction of antimicrobial chemotherapy. However, the unsatisfactory results that these operations yielded was also importantly attributed to the poor surgical exposure of the vertebral bodies that the lateral approach provided. Recognizing this, Ito proposed a decompression operation that would adequately resect infected vertebrae in order to fully eradicate the presence of tuberculosis in the spine. Drawing on experience with the transabdominal approach, which he had previously used for another purpose, Ito reported his operative technique and surgical results on 10 patients with moderately advanced Pott’s disease. The possibility of approaching the ventral spine occurred to Ito and his colleagues after repeated operations using their original technique for lumbosacral sympathetic ganglionectomy. In 1923, Ito and his colleagues53 originated this technique for the purpose of improving lower extremity circulation and reported their results to the Japanese Surgical Society in 1925. The technique was subsequently modified to provide an extraperitoneal approach to the lumbar spine and was adopted for their radical operation for Pott’s disease (Figure 1.21). The significance of the work of Ito was beneficial for several reasons. First, Ito and his colleagues recognized the need to address the pathology directly, despite the technical difficulties that such an approach presented. Second, at a time when the major surgical treatment for Pott’s disease was dorsal fusion, Ito proposed a radical new surgical therapy: decompression. An attempt to eradicate spinal infection by surgical decompression represented an
Figure 1.21 Extraperitoneal exposure of the body of the lumbar
vertebra and resection of the body with a chisel. (From Ito H, Tsuchiya J, Asaini G: A new radical operation for Pott’s disease. J Bone Joint Surg 16:499-515, 1934.)
alternative approach to the standard stabilization procedure originated by Albee and Hibbs. In another sense, the idea of decompression harked back to the nineteenth century laminectomy for Pott’s disease, which was largely abandoned because of disappointing results, after the introduction of dorsal spinal fusion.88 Finally, Ito recognized the need, and established the technique, to stabilize the spine, which if not already unstable, was certainly rendered unstable by resection of the major load-bearing element. He accomplished this goal by fashioning a ventral interbody fusion, which both provided significant stability and facilitated spinal fusion (Figure 1.22). However, despite Ito’s successes—all except two of Ito’s ten cases showed a healing by primary intention and despite his acknowledgment of the inadequacies of the dorsolateral approach—Ito himself used the costotransversectomy approach in the two cases of thoracic Pott’s disease included in his series. Hodgson and Stock Thus, it fell to another group of surgeons treating Pott’s disease to develop a true ventral approach to the thoracic spine. In 1956, Hodgson and Stock50 published their first report on ventral spinal fusion for Pott’s disease. These authors acknowledged the contributions of Ito and his colleagues, and they repeated Ito’s assessment of the restricted field of view afforded by the costotransversectomy. They noted that this field of view provided insufficient exposure to determine the extent of the lesion or to
Chapter 1 ■ History
incontrovertible; it facilitated decompression, stabilization, and deformity correction through a single incision and surgical exposure, providing excellent neurologic and anatomic results. The authors took account of the unique anatomic features of the cervicothoracic and thoracolumbar junctions, where the approach was appropriately modified.
Ventral Deformity Reduction and the Development of Ventral Instrumentation
Figure 1.22 Schematic illustration of the insertion of ventral
bone graft. (From Ito H, Tsuchiya J, Asaini G: A new radical operation for Pott’s disease. J Bone Joint Surg 16:499-515, 1934.)
confidently undertake its complete resection. What is more, the limited exposure of the costotransversectomy left no room to accurately insert a ventral bone graft, which they considered offered the best chance for fusion because the bone graft would be placed in a compression mode. Hodgson and Stock also joined Ito and others in emphasizing decompression, rather than simple stabilization, as a method to arrest further vertebral destruction (which may be responsible for neural element compression and progressive kyphotic deformity), and as a means to eradicate the spinal focus of disease. Their approach to the thoracic spine via a thoracotomy, the first significant series of such an approach described, was facilitated by developments in the medical management of tuberculosis, including the introduction of chemotherapeutic agents (not available to Ito and others), and safer, more effective anesthetic techniques. The benefits of this approach, then, despite its technical difficulties, were
The contributions of Burns, Ito and associates, and Hodgson and Stock were seminal in the history of spinal surgery. They opened new vistas in the management of spinal pathologies, and their techniques were later applied to an increasingly wide range of pathologic conditions, including tumor, trauma, disc disease, and spinal deformity. The methods of Ito and associates were particularly prescient. They accomplished, with a single incision, the goals of both decompression and spinal stabilization, and they achieved both of these goals in the most effective possible manner. The establishment of deformity correction was addressed in the report by Hodgson, who confronted the problem of severe kyphotic deformity causing cardiopulmonary compromise. On a larger scale, however, the problem of progressive spinal deformity did not receive the attention of these early authors, and no method of ventral internal fixation was yet available to spinal surgeons who wished to establish and maintain a deformity correction via a ventral approach. As mentioned, Paul Harrington addressed the problem of scoliotic deformity by the development of dorsal thoracolumbar distraction rods in the 1960s, and in doing so he initiated the modern instrumentation revolution. Harrington’s method of scoliosis reduction was based on the principle of lengthening the short (concave) side of the curve. After the introduction of a meticulous fusion technique to supplement the immediate rigid internal fixation achieved by the implant, the Harrington instrumentation system proved both a safe and effective corrective measure, an assessment that is corroborated by its long and successful history of clinical application. Nevertheless, the principle of simple dorsal distraction had its drawbacks. First, the Harrington method requires that the fusion be extended at least two levels above and below the extent of the spinal curvature, thus decreasing mobility in otherwise normal spinal motion segments. Second, in most instances, the distribution of force application with the Harrington instrumentation system is uneven, such that the total force applied is borne only by the two vertebrae attached to the upper and lower hooks. Finally, for patients who require a simultaneous ventral decompression and dorsal stabilization procedure, this could be accomplished only through a two-stage operation involving two separate incisions and surgical exposures. Thus, the arrival of a ventral instrumentation system, introduced by Dwyer32 in 1969, proved an important addition to the spinal surgeon’s surgical armamentarium.
17
18
Chapter 1 ■ History
A.F. Dwyer
Summary
A.F. Dwyer was an orthopedic surgeon from Australia who appears to have originated his method in an effort to provide an alternative to the Harrington technique of scoliotic deformity reduction. In his initial report of 1969, Dwyer described a method of ventral instrumentation in which compressive forces are applied to the convex side of the curve at each segmental level. The technique comprises excision of the discs at the motion segments involved, and the insertion of vertebral body screws into the convex aspect of the curve. A titanium cable is then threaded through the heads of the inserted screws and a tension force is applied, providing corrective bending moments at the intervertebral spaces. The tension is maintained by swaging the threaded cable on the screw heads (Figure 1.23). In a follow-up article published in 1974, Dwyer and Schafer33 reported their results of treatment in 51 cases, which demonstrated a generally favorable record of deformity correction and only a 4% rate of pseudarthrosis.3 Furthermore, some of the disadvantages of the Harrington dorsal instrumentation system were overcome—fusion could be restricted to the motion segments of the curve only; the load borne by the instrumentation device was evenly distributed over the curve; and the exposure necessary for ventral decompression, stabilization, and deformity correction was achieved using a single incision. Although the initial enthusiasm for the Dwyer device was later diminished by the recognition that it encouraged the tendency of the spine toward progressive kyphosis and that it provided no resistance to axial loading, the generally successful application of this ventral instrumentation system stimulated the development of additional ventral implants, such as the instrumentation systems of Zielke and Pellin 116 and Kaneda and associates.56
The technical accomplishment of performing surgery on the ventral spine provides perhaps a useful marker for the endpoint of the history of “early modern” spinal surgery. By 1970, it may be argued, the basic groundwork had been laid for the subsequent advances, particularly in spinal instrumentation, that have been made over the last 25 years. These advances include an emphasis on location-appropriate decompression; the development of segmental spinal instrumentation by E.R. Luque in the early 1970s70-73; the refinement and proliferation of pedicular instrumentation techniques, first described by Harrington in 196945,112; the introduction of universal spinal instrumentation by Cotrel and associates26; the further development of ventral thoracolumbar instrumentation by Zielke, Kostuik,62 and Kaneda; the introduction of ventral cervical instrumentation by Caspar and associates in 198921; and most recently, the application of endoscopic techniques.55 In conclusion, this essay has sought to organize and present the history of spinal surgery as a series of attempts to improve the surgeon’s ability to more safely and effectively achieve spinal decompression, stabilization, and deformity correction—the three major goals of spinal surgery. The occasionally formidable obstacles encountered by those surgeons who have participated in this centurylong odyssey were frequently managed, if not overcome, by concentrated and indefatigable effort. Alas, many of the same obstacles that faced the early spinal surgeons— including blood loss, pseudarthosis, instrumentation failure, and neurologic injury—continue to challenge and vex even the best-equipped contemporary spinal surgeons.
Figure 1.23 Dwyer’s ventral short segment fixation device.
(From Dwyer AF, Schafer MF: Anterior approach to scoliosis. Results of treatment in fifty-one cases. J Bone Joint Surg 56B:218-224, 1974.)
REFERENCES 1. Abbe R: Spinal surgery: Report of eight cases. Med Rec 38:85-92, 1890. 2. Adams F: The Genuine Works of Hippocrates. Baltimore, Williams & Wilkins, 1939. 3. Albee FH: Transplantation of a portion of the tibia into the spine for Pott’s disease. JAMA 57:885-886, 1911. 4. Alessandri R: Laminectomia della terza e quarta vertebra lombre per lesione della cauda equina. Riv Patol Nerv 10:86-92, 1905. 5. Alessandri R: Processo osteoplastico modificato di laminectomia, con due casi operati. Arch Ed Atti Soc Ital Chir 18:135B154, 1905 6. American Orthopaedic Association Research Committee: End-result study of the treatment of idiopathic scoliosis. J Bone Joint Surg 23:963-977, 1941. 7. Armour D: Surgery of the spinal cord and its membranes. Lancet 1:423-430, 1927. 8. Bailey P, Casamajor C: Osteoarthritis of the spine as a cause of compression of the spinal cord and its roots. J Nerv Mental Dis 38:588-609, 1911. 9. Bailey RW: Fractures and dislocations of the cervical spine: orthopedic and neurosurgical aspects. Postgrad Med 35:588-599, 1964. 10. Benini A, Bonar SK: Andreas Vesalius 1514-1564. Spine 21:1388-1393, 1996.
Chapter 1 ■ History 11. Boucher HH: A method of spinal fusion. J Bone Joint Surg 41B:248-259, 1959. 12. Brasted JH: The Edwin Smith Papyrus. In: Wilkins RH (compiled): Neurosurgical Classics. Park Ridge, Ill, American Association of Neurologic Surgeons, 1992, pp 1-5. 13. Brok AJ: Greek Medicine. J.M. Dent & Sons. London, 1929. 14. Burns BH: An operation for spondylolisthesis. Lancet 1:1233, 1933. 15. Callahan RA, Johnson RM, Margolis RN, et al.: Cervical facet fusion for control of instability following laminectomy. J Bone Joint Surg 59(A):991-1002, 1977. 16. Calot T: Sur le meilleur traitement localdes tuberculoses doses articulations et ganglions lymphatiques. Acta Chir Scand 67:206-226, 1930. 17. Campbell WC: An operation for extra-articular fusion of sacroiliac joint. Surg Gynecol Obstet 45:218-219, 1927. 18. Capener N: The evolution of lateral rhachotomy. J Bone Joint Surg 36B:173-179, 1954. 19. Caponotto A, Pescarolo B: Estirpazione di un tumore intradurale del canale rachideo. Riforma Med 8:543-549, 1892. 20. Caspar W. A new surgical procedure for lumbar disc herniation causing less tissue damage through a microsurgical approach. Adv Neurosurgery 4: 77-80, 1977. 21. Caspar W, Barbier DD, Klara PM: Anterior cervical fusion and Caspar plate stabilization for cervical trauma. Neurosurgery 25:491-502, 1989. 22. Church A, Eisendrath DW: A contribution to spinal cord surgery. Am J Med Sci 103:395-412, 1892. 23. Cline HJ Jr (cited by Hayward G): An account of a case of fracture and dislocation of the spine. New Engl J Med Surg 4:13, 1815. 24. Cloward RB: The treatment of ruptured intervertebral disc by vertebral body fusion. Ann Surg 136: 987-992, 1952. 25. Cloward RB. The anterior approach for removal of ruptured cervical disks. J Neurosurg 15:602-617, 1958. 26. Cotrel Y, Dubousset J, Guillaumat M: New universal instrumentation in spinal surgery. Clin Orthop 227:10-23, 1988. 27. Cushing H: Intradural tumor of the cervical meninges with early restoration of function in the cord after removal of the tumor. Ann Surg 39:934-955, 1904. 28. Cushing H: The special field of neurological surgery. Bull Johns Hopkins Hosp 16:77-87, 1905. 29. Cushing H: The special field of neurological surgery: five years later. Bull Johns Hopkins Hosp 21:325-329, 1910. 30. Dandy WE: Loose cartilage from the intervertebral disc simulating tumour of the spinal cord. Arch Surg 19: 660-672, 1929. 31. Dawbarn RHM: A successful case of spinal resection. New York Med J 49:711-715, 1889. 32. Dwyer AF, Newton NC, Sherwood AA: An anterior approach to scoliosis. A preliminary report. Clin Orthop 62:192-202, 1969. 33. Dwyer AF, Schafer MF: Anterior approach to scoliosis. Results of treatment in fifty-one cases. J Bone Joint Surg 56B:218-224, 1974. 34. Elsberg CA, Beer E: The operability of intramedullary tumors of the spinal cord. A report of two operations with remarks upon the extrusion of intraspinal tumors. Am J Med Sci 142:636-647, 1911.
35. Elsberg CA: Diagnosis and Treatment of Diseases of the Spinal Cord and Its Membranes. Philadelphia, WB Saunders, 1916. 36. Elsberg CA: Experiences in spinal surgery. Surg Gynecol Obstet 16:117-132, 1913. 37. Elsberg CA: The extradural ventral chondromas (enchondroses), their favorite sites, the spinal cord and root symptoms they produce, and their surgical treatment. Bull Neurol Inst NY 1: 350-388, 1931. 38. Eskridge JT, Freeman L: Intradural spinal tumor opposite the body of the fourth dorsal vertebra; complete paralysis of the parts below the lesion; operation; recovery; with ability to walk without assistance within three months. Phil Med J 2:1236-1243, 1898. 39. Goldthwaite JE: The lumbosacral articulation. An explanation of many cases of “lumbago,” “sciatica,” and paraplegia. Boston Med Sci J 164:365-372, 1911. 40. Gower W: A Manual of Diseases of the Nervous System, 2 volumes. London, J & A Churchill, 1886-1888. 41. Gowers WR, Horsley VA: A case of tumour of the spinal cord. Removal; recovery. Med Chir Trans 53:379-428, 1888. 42. Gurunluoglu R, Gurunluoglu A: Paul of Aegina: landmark in surgical progress. World J Surg 27:18-25, 2003. 43. Hadra BE: Wiring the spinous processes in Pott’s disease. Trans Am Orthop Assoc 4:206-210, 1891. 44. Harrington PR, Tullos HS: Reduction of severe spondylolisthesis in children. South Med J 62:1-7, 1969. 45. Harrington PR, Tullos HS: Spondylolisthesis in children. Clin Orthop 79:75-84, 1971. 46. Harrington PR: The history and development of Harrington instrumentation. Clin Orthop Rel Res 93:110-112, 1973. 47. Harrington PR: Treatment of scoliosis. J Bone Joint Surg 44A:591-610, 1962. 48. Hattori S. A new method for cervical laminectomy. Central Jap J Orthop Traum Surg 16:792-794, 1973. 49. Hibbs RA: An operation for progressive spinal deformities. NY Med J 93:1013-1016, 1911. 50. Hodgson AR, Stock FE: Anterior spinal fusion. A preliminary communication on the radical treatment of Pott’s disease and Pott’s paraplegia. Br J Surg 44:266-275, 1956. 51. Holdsworth FW, Hardy A: Early treatment of paraplegia from fractures of the thoraco-lumbar spine. J Bone Joint Surg 35B:540-550, 1953. 52. Holdsworth FW: Fractures, dislocations, and fracture-dislocation of the spine. J Bone Joint Surg 45B:6-20, 1963. 53. Ito H, Asami G: Lumbosacral sympathetic ganglionectomy. Its value as a therapeutic measure for thromboangiitis obliterans. Am J Surg 15:26, 1932. 54. Ito H, Tsuchiya J, Asaini G: A new radical operation for Pott’s disease. J Bone Joint Surg 16:499-515, 1934. 55. Kambin P, Gellman H. Percutaneous lateral discectomy of the lumbar spine. Clin Orthop 174:128-132, 1983. 56. Kaneda K, Abumi K, Fujiya K: Burst fractures with neurologic deficits of the thoraco-lumbar spine. Results of anterior decompression and stabilization with anterior instrumentation. Spine 9:788-795, 1984. 57. Kelly RP, Whitesides TE: Treatment of lumbodorsal fracture-dislocations. Ann Surg 167:705-717, 1968. 58. King D: Internal fixation for lumbosacral fusion. J Bone Joint Surg 30A: 560-565, 1948.
19
20
Chapter 1 ■ History 59. Kleinberg S: The operative treatment of scoliosis. Arch Surg 5: 631-645, 1922. 60. Kocher T: Chirurgische Operationslehre. G Fischer, Jena, 1892. 61. Kocher T: Die Verletzungen der Wirbelsule zugeleich als Beitrag zur Physiologie des menschlichen Ruckenmarks. Mitt Grenzgeb Med Chir 1:415-480, 1896. 62. Kostuik JP: Anterior fixation for fractures of the thoracic and lumbar spine with or without neurologic involvement. Clin Orthop 189:103-115, 1984. 63. Kumar K: Historical perspective: spinal deformity and axial traction. Spine 21:653-655, 1996. 64. Lane WA: Case of spondylolisthesis associated with progressive paraplegia; laminectomy. Lancet 1:991, 1893. 65. Lang JK, Kolenda H: First appearance and sense of the term “spinal column” in ancient Egypt. Historical vignette. J Neurosurg 97(1 Suppl):152-155, 2002. 66. Laquer L: Ueber Compression der Cauda equina. Compressions-Erscheinungen im Gebeite der lumbal und Sacral wurzeln. Eroffnung des Canalis sacralis. Exstirpation eines Lymphangioma cavernosum. Beseitigung fast aller Beschwerden. Neurol Centralbl 10:193-204, 1891. 67. Larson SJ, Holst RA, Hemmy DC, Saiices A Jr: Lateral extracavitary approach to traumatic lesions of the thoracic and lumbar spine. J Neurosurg 45:628-637, 1976. 68. Lister J: Six Papers. Selected by Sir Rickman J. Godley. London, 1912. 69. Love JG: Removal of intervertebral discs without laminectomy. Proceedings of staff meeting. Mayo Clinic 14:800, 1939. 70. Luque ER, Cassis N, Ramirez-Wiella G: Segmental spinal instrumentation in the treatment of fractures of the thoracolumbar spine. Spine 7:256-259, 1982. 71. Luque ER: The anatomic basis and development of segmental spinal instrumentation. Spine 7:256-259, 1982. 72. Luque ER: Interpeduncular segmental fixation. Clin Orthop 203:54-57, 1986. 73. Luque ER: Segmental spinal instrumentation of the lumbar spine. Clin Orthop 203:126-134, 1986. 74. Makins GH, Abbott FC: On acute primary osteomyelitis of the vertebrae. Ann Surg 23:510-539, 1896. 75. Maquet P: Iatrophysics to biomechanics. From Borelli (1608-1679) to Pauwels (1885-1980). J Bone Joint Surg Br 74:335-339, 1992. 76. Marketos SG, Skiadas PK: Galen. A pioneer of spine research. Spine 24:2358-2362, 1999. 77. Markham JW: Surgery of the spinal cord and vertebral column. In Walker AE (ed): A History of Neurological Surgery. New York, Hafner Publishing Company, 1967, pp 370-371. 78. Matas R: Surgical operations fifty years ago. Am J Surg 82:111, 1951. 79. Ménard V: Causes de la paraplegia dans le mal de Pott. Son traitement chirurgical par l’ouverture direct du foyer tuberculeux des vertebras. Rev Orthop 5:47-64, 1894. 80. Meyer HW: The history of the development of the negative differential pressure chamber for thoracic surgery. J Thor Surg 30:114, 1955. 81. Middleton GS, Teacher JH: Injury of the spinal cord due to rupture of an intervertebral disc during muscular effort. Glasgow Med J 76:1-6, 1911.
82. Middleton WEK: A little-known portrait of Giovanni Alfonso Borelli. Med Hist 18:94-95, 1974. 83. Mixter WJ, Barr JS: Rupture of the intervertebral disc with involvement of the spinal canal. N Engl J Med 211: 210-215, 1934. 84. Müller W: Transperitoneale Freilegung der Wirbelsaule bei Tuberkuloser Spondylitis. Deutsche Ztschr Chir 85:128, 1906. 85. Naderi S, Acar F, Arda MN: History of spinal disorders and cerrahiyetülhaniye: a review of a Turkish treatise written by serefeddin Sabuncuoglu in 15th century. J Neurosurg 96:352-356, 2002. 86. Naderi S, Acar F, Mertol T, Arda MN: Functional anatomy of the spine by Avicenna in his eleventh century treatise Al-Qanun fi al-Tibb (The Canons of Medicine). Neurosurgery 52:1449-1453, 2003. 87. Novell JR: From Da Vinci to Harvey: The development of mechanical analogy in medicine from 1500 to1650. J R Soc Med 83:396-398, 1990. 88. Ollier L: Traite des resections et des operations conservatrices qu’on peut pratiquer sur le systeme osseux. Vol 3. G Masson, Paris, 1891, pp 833-857. 89. Oppenheim H, Krause F: Über Einklemung bzw. Strangulationbder Cauda Equina. Dtsch Med Wochenschrift 35: 697-700, 1909. 90. Portal A: Cours d’Anatomie Medicale ou Elements de l’Anatomie de l’Homme. Vol. 1. Baudouin, Paris, 1803. 91. Robinson RA, Smith GW. Anterolateral cervical disc removal and interbody fusion for cervical disc syndrome. Bull John Hopkins Hosp 96:223, 1955. 92. Rogers DL: A case of fractured spine with depression of the spinous processes, and the operation for its removal. Am J Med Sci 16:91-94, 1835. 93. Roy CD: Report of a case of spinal exsection and removal of a tumor from the cord. South Med Rec 20:564-566, 1890. 94. Sachs B, Frankel J: Progressive ankylotic rigidity of the spine. J Nerv Ment Dis 27:1, 1900. 95. Sarpyener MA. Congenital stricture of the spinal canal. J Bone Joint Surg 27: 70-79, 1945. 96. Seddon HJ: Pott’s paraplegia: prognosis and treatment. Br J Surg 22:769, 1935. 97. Smith AG: Account of a case in which portions of three dorsal vertebrae were removed for the relief of paralysis from fracture, with partial success. North Am Med Surg J 8:94-97, 1829. 98. Sorrel-Dejerine Y: Contribution a l’Etude des Paraplegies Pottiques. Paris, 1925. 99. Spink MS, Lewis GL: Albucasis, On Surgery and Instruments. A Definitive Edition of the Arabic text with English translation and Commentary. London, The Wellcome Institute of the History of Medicine, 1973. 100. Taylor AS: Unilateral laminectomy. Ann Surg 51:529-533, 1910. 101. Thompson WAL, Ralston EL: Pseudoarthrosis following spine fusion. J Bone Joint Surg 31A:400-405, 1949. 102. Tyrell F: Compression of the spinal marrow from displacement of the vertebrae, consequent upon injury. Operation of removing the arch and spinous processes of the twelfth dorsal vertebra. Lancet 11:685-688, 1827. 103. Venable CS, Stuck WG, Beach A: The effects on bone of metals: based upon electrolysis. Ann Surg 105:917-938, 1937.
Chapter 1 ■ History 104. Venable CS, Stuck WG: Electrolysis: controlling factor in the use of metals in treating fractures. JAMA 3:349, 1939. 105. Verbiest H. A radicular syndrome from developmental narrowing of the lumbar vertebral canal. J Bone Joint Surg (Br) 36:230-237, 1954. 106. Vincent E: Contribution a la chirurgie rachidienne du drainage vertebral dans le mal de Pott. Rev Chir 1273-1294, 1892. 107. Vincent E: Chirurgie rachidienne et mal de Pott. Rev Chir 18:47-54, 1898. 108. Virchow R: Untersuchungen uber die Entwickelung des Schadelgruendes im gesunden und krankhaften Ziistande, etc. Berlin, G Reimer, 1857. 109. von Eiselsberg AF, Ranzi E: Ueber die chirurgische Behandlung der Him-und Rückenmarkstumoren. Arch Klin Chir 102:309-468, 1913. 110. Watson GL, Paul WE: Contribution to the study of spinal surgery: One successful and one unsuccessful operation
111.
112. 113.
114. 115. 116.
for the removal of the tumor. Boston Med Surg J 153: 114117, 1905. Wilson PD, Straub LR: The use of a metal plate fastened to the spinous processes. American Academy of Orthopaedic Surgeons Instructional Course Lecture, Ann Arbor, MI, 1952. Wiltse LL: History of pedicle screw fixation of the spine. Spine: State of the Art Rev 6:1-110, 1992. Wiltse LL. The history of spinal disorders.Frymoyer JW (ed): The Adult Spine. Principles and Practice. Philadelphia, Lippincott-Raven, 1997, pp 3-40. Yasargil MG. Microsurgical operation of herniated lumbar disc. Adv Neurosurg 4: 81, 1977. Zavaleta MA: Resecion de vertebras. Ann Circ Med Argent 15:497-502, 1892. Zielke K, Pellin B: Neue Instrumente und Implantate zur Erganzung des Harrington Systems. Z Orthop Chir 114:534, 1976.
21
C H A P T E R
(Figure 2.2). Luque reasoned that increasing the number of fixation points along a construct would reduce the force placed upon each individual point and obviate the need for a postoperative cast or brace. Additional beneficial effects of segmental fixation were that it increased the potential corrective power of instrumentation, reduced the potential for construct failure, and resulted in improved fusion rates. The concept of segmental fixation to a contoured rod was widely embraced because it produced greater construct rigidity and allowed for improved control of the sagittal plane. The use of sublaminar wires was adopted by some users of Harrington rod instrumentation. A hybrid form of Paul Harrington’s technique (from Texas) and Eduardo Luque’s technique (from Mexico) was sometimes referred to as the “Tex-Mex” operation. Although the corrective power of sublaminar wires was well-appreciated, many surgeons had reservations in using them because of reports of neurological injury resulting either from direct trauma or from epidural hematoma.32,54 In addition, revision surgery after sublaminar wiring is problematic because scarring may preclude the passage of new wires at the same laminae. In response to these concerns, Drummond18 developed a method for segmental fixation using a button-wire implant passed through the base of the spinous process. This technique does not provide as strong fixation as do sublaminar wires. It avoids, however, passing anything into the spinal canal and thus reduces the risk of direct neurological injury. This compromise of fixation for less risk of neurological injury was seen as a prudent choice by many surgeons operating on healthy, neurologically normal adolescents with idiopathic scoliosis. Nevertheless, some pundits referred to the procedure as the “chicken-Luque” procedure. Increasingly sophisticated multiple hook-rod systems appeared in the 1980s that provided much of the strength of wire fixation but with greater flexibility to address deformities in both the sagittal and the coronal dimensions. The Cotrel Dubousset (CD) system was introduced into the United States in 1986 using a 1⁄4-inch rough-surfaced rod.14 Multiple hooks allowed spinal surgeons to apply compression and distraction over different areas within the same rod. The multiple hook design applied the principles of segmental fixation without the need for sublaminar wires. Significantly, the system provided for a unique mechanism for deformity correction: rod rotation. This proved a powerful force in the correction of scoliosis. Further stability was provided by cross-linking the two parallel rods together. The advantages of the CD system were partially offset, however, by the difficulty of removing the system. The locking mechanism of the hooks was irreversible without destroying the hook or cutting the rod. The Texas Scottish Rite Hospital (TSRH) system was a design advance that addressed the issue of revision surgery. It was similar to the CD system in having multiple hooks and cross links but was designed to allow the removal of the system’s individual components if necessary. Although the features of the TSRH system simplified revision surgery, the top-loading side-tightened system was not universally appreciated. After maturation of the fusion mass, the side-tightened bolts were not always
2
History of Spinal Instrumentation: The Modern Era John K. Houten and Thomas J. Errico The use of internal fixation as a tool for both stabilization and correction of deformity was a major advance in modern spinal surgery. A wide experience in the use of internal fixation in the treatment of the appendicular skeleton was extrapolated to the axial skeleton. This experience has culminated in the wide range of surgical implants currently available to the modern spinal surgeon. A thorough understanding of the evolution of spinal instrumentation should yield a better understanding of both present and future developments.
Dorsal Thoracolumbar Instrumentation In 1975, the Harrington rod represented the “state of the art” in spinal instrumentation. The rod system, originally developed by Paul Harrington for the correction of spinal deformities, was soon used in the treatment of traumatic injuries3,53 (Figure 2.1), degenerative disease,42 and metastatic disease.35,51 The system provided both distraction rods as well as compression rods and hooks. Over the years however, their widespread use led to recognition of their limitations. The use of a distraction system provided excellent correction of coronal plane deformities (scoliosis). Unfortunately, the use of distraction as the sole correction tool resulted in the loss of normal sagittal plane alignment. The loss of normal lumbar lordosis was associated with “flat back syndrome.”12,34 Hook dislodgement and rod breakage also proved to be troublesome complications.22,50 In addition, casting or bracing was generally required in the postoperative period, and this proved to be difficult or impractical in some patients.37 In response to the difficulties encountered with Harrington rods, Eduardo Luque advanced a major concept in the mid-1970s that quietly pushed the future direction of spinal instrumentation: segmental spinal fixation. The issue of bracing was of particular importance to Luque. Practicing in the warm climate of Mexico City, it was difficult for Luque, from a practical standpoint, to use the postoperative casting required in Harrington rod instrumentation. In addition, a large number of his patients, who were from homes of low socioeconomic status, would travel a great distance to seek treatment and would not comply with bracing or would get lost to follow-up. Luque popularized the use of a 3⁄16-inch steel rod secured at each spinal level with sublaminar wires
22
Chapter 2 ■ History of Spinal Instrumentation: The Modern Era
Figure 2.1 (A) AP and (B) lateral radiographs after surgical stabilization of a burst fracture of L3
with Harrington rod internal fixation.
accessible. The following decade saw the introduction of numerous, similar dual-rod systems like Moss-Miami and Isola.12,49 The major variations revolved around the leading and locking mechanisms: side loading, top loading, side tightening, or top tightening. The last decade has seen the introduction of numerous systems that operate with the same design principles, with a shift toward the use of polyaxial screws that make coupling of the fixation points to the rods easier. A major advance that became available with these spinal systems was the exploitation of the pedicle as a site for segmental fixation. This innovation is generally credited to Roy-Camille of Paris. Roy-Camille performed his first operation in 1963 but did not publish the results until 1970.43 Pedicle screws presented many advantages when compared with other tools for spinal fixation. Pedicle screws are biomechanically superior as a point of fixation1 compared with hook- or wire-rod constructs and can be placed into the sacrum, an area to which fixation is otherwise difficult. In addition, they can be placed even after a laminectomy has been performed and can be positioned without entering the spinal canal.17 This advantage allowed for the massive proliferation of spinal instrumentation into the area of degenerative spinal disorders. Prior to the advent of pedicle-screw
instrumentation systems, there had been only sporadic reports of the use of instrumentation for degenerative spinal disorders. The Knodt rod (a small distraction rod system) had been used previously in degenerative disease but was associated with localized loss of lordosis and device dislodgement. In addition, the system needed some lamina for device fixation. Pedicle screw systems, however, can be used after a total laminectomy. Arthur Steffee popularized the use of pedicle screws in the United States in 1984 using a contourable plate. At about the same time, a screw-rod system was in use in Europe, developed by Yves Cotrel of France, that became incorporated into the “Universal” CD system. Controversy soon followed, with both the screw-plate and screw-rod constructs developing a group of proponents.23 Proponents of plates noted that plates were stronger. Most surgeons were ultimately attracted, however, to rods because their use provides greater flexibility, reduces encroachment upon the adjacent facet joints, and leaves more surface area for fusion. The marriage of the long dual-rod constructs to lumbar pedicle screws was an important development that enhanced the surgeon’s ability to accomplish increasingly difficult and complex spinal reconstructions. The use of the polyaxial
23
24
Chapter 2 ■ History of Spinal Instrumentation: The Modern Era
Figure 2.2 (A) AP and (B) lateral radiographs after surgical stabilization of a burst fracture of L3
with segmental sublaminar wire fixation to an angled rod using the technique described by Eduardo Luque.
pedicle screw has further advanced the ease of spinal reconstructions.
Ventral Thoracolumbar Instrumentation Successful use of the Harrington instrumentation kindled interest in developing a ventral system to address neuromuscular scoliosis. Dwyer developed a ventral system for internal fixation using screws connected by a cable.46 Winter attempted a combined ventral and dorsal approach with Harrington and Dwyer instrumentation to treat painful adult idiopathic scoliosis.55 This concept was of particular interest in that these patients were at high risk for pseudoarthrosis and tended to tolerate bracing less well than adolescents.55 The Zielke system, developed in 1975, was the next step in the development of ventral instrumentation. The Zielke device connected transvertebral screws with a threaded rod and nuts and was more rigid than the Dwyer cables. This added both strength and the capacity for incremental correction and derotation, permitting a more powerful correction. The Zielke system produced a
lower pseudoarthrosis rate and somewhat lower recurrence of the flat-back syndrome. In spite of these benefits, the system had many shortcomings. The pseudoarthrosis rate remained high when the system was used as a stand-alone device but was lowered with supplementation of dorsal fixation. This system also suffered from the tendency to shorten the anterior columns and to produce kyphosis. The Dunn device was a ventral implant that consisted of two rods that spanned the distance between two vertebral body bridges: one placed ventrolaterally with a vertebral body staple and the other placed more dorsolaterally with an intervertebral body screw.19 This system was not widely accepted because it was bulky and was associated with vascular complications.25 The ventral Kostuik Harrington instrumentation was an adaptation of short Harrington rods used in conjunction with a pedicle screw developed by Paul Harrington for the use in myelomeningocele. Introduced by John Kostuik in the early 1980s, it was an innovative short-segment ventral fixation device. The screw, when placed ventrolaterally in the vertebral bodies, allows for short-segment ventral correction of the kyphotic deformity associated with burst
Chapter 2 ■ History of Spinal Instrumentation: The Modern Era
fractures. A second neutralization rod was placed parallel to the first rod to enhance stability (Figure 2.3). Over time, cross-fixators were added in an attempt to further enhance stability. Two parallel rods rigidly cross-linked are the biomechanical equivalent of a plate. Most ventral short-
segment constructs subsequently used plates with vertebral body screws. Several other plate designs soon followed that were lower profile. Ryan introduced a plate secured by a rostral and caudal bolt inserted through the vertebral body. The
Figure 2.3 (A) Kostuik-Harrington screws and rods. (B) AP and (C) lateral radiographs after
surgical stabilization of a burst fracture of L4 with Kostuik-Harrington instrumentation.
25
26
Chapter 2 ■ History of Spinal Instrumentation: The Modern Era
single bolt design, however, offered less resistance to rotation than the designs that used two screws or bolts above and below.45 The Yuan I-Plate was an alternative design that consisted of a 3.5-mm stainless steel plate secured with transvertebral screws.56 Black et al.8 published his experience with a low-profile, rectangular, stainless steel plate that had multiple holes that allowed for the placement of three screws at each vertebral level. The Kaneda device represented another stage in development of anterior thoracolumbar instrumentation because it allowed reduction of kyphotic deformities after ventral decompression while providing good strength without incidence of vascular injury.25 The next generation of ventral plates, including the Z-Plate™ (Medtronic/Sofamor Danek, Memphis, TN) and the Anterior Thoracolumbar Locking Plate System (Synthes, Paoli, PA), further improved implant design by providing a lower profile and changing the composition to titanium alloys. In addition, the newer systems allow for both the distraction of kyphotic deformities and the compression of the graft.
Dorsal Cervical Instrumentation The earliest methods used to provide internal fixation for posterior cervical fusions involved the use of spinous process wiring. These techniques, however, are limited in that they often do not provide adequate stiffness or sufficient resistance to rotational movement and extension and cannot be used when the spinous processes have been removed. For internal fixation of C1-2, the Brooks and Gallie techniques use sublaminar wires to compress an autologous bone graft. Although these techniques are reported to be associated with high fusion rates, they have the disadvantages of potentially producing neurological injury from the placement of sublaminar wires and the problem that wires may pull through osteoporotic bone. In addition, there is a small but persistent failure rate associated with the Gallie fusion that may be caused by inadequate immobilization allowing for “grinding down” of the graft. Several instrumentation systems were devised as adjuncts or alternatives to wiring. The Daab plate was a stainless-steel implant shaped like an elongated “H” that could be compressed at either end to fixate it to a spinous process.10,31 This instrumentation represented no significant advantage over the available wiring techniques, and it was probably inferior considering that it typically needing the resection of an intervening spinous process and the associated interspinous ligaments. Halifax clamps are a pair of upgoing and downgoing sublaminar hooks tightened together with a screw that is then secured in position with a locking mechanism (Figure 2.4).15,31 Halifax clamps have the advantage of relatively simple and rapid application. In addition, the area of bone contact is broader than that with wiring and is less likely to pull out of soft bone. They offer C1-2 fixation comparable with that achieved with the Brooks technique.26 Relative disadvantages of the system are that hooks are introduced into the spinal canal, and the implant is relatively “high profile” and has limited application when stabilization is needed over multiple segments.
Figure 2.4 Postoperative lateral cervical radiograph
demonstrating C1-2 internal fixation with Halifax Clamps. In this patient, the posterior clamps were placed to supplement fixation with C1-2 transarticular screws.
In the mid-1980s, Mager introduced transarticular screw placement for internal fixation of C1-2. This is a technically demanding procedure compared with wiring that achieves improved C1-2 stability to flexion-extension and rotation27 than wiring procedures and is associated with the highest published C1-2 fusion rates.28 This technique is not always feasible if there is anatomical variation in the course of the vertebral artery, although there is still benefit in unilateral placement.28 Many practitioners supplement transarticular screws with posterior instrumentation as broken screws have been seen when used as a stand-alone procedure. Lateral mass plate fixation with screws was introduced by Roy-Camille et al.44 This technique of internal fixation is ideal in instances in which the laminae and spinous processes have been removed or fractured. The first technique for screw placement was modified by Magerl and Seeman,38 Heller,29 and An et al.2 The original lateral mass plates were an application of preexisting bone plates with a distance between plate holes of 13mm. The Haid Plate and Synthes reconstruction plates were soon marketed, each offering a choice of two interhole distances. These systems all suffered from insufficient versatility in accommodating the wide variety of interhole differences often needed.13 The AXIS™ system (Medtronic/Sofamor Danek, Memphis, TN) offers plate holes at intervals of 11, 13, and 15mm and a slotted hole design to allow for limitless interhole variations as well as improved ability to contour the plates. The Cervifix™ system (Synthes Spine, Paoli, PA) consists of lateral mass screws that are inserted through clamps along a contourable titanium rod. This provides the flexibility to address variations in anatomy beyond that
Chapter 2 ■ History of Spinal Instrumentation: The Modern Era
offered by the AXIS system. In addition, a hybrid platerod implant is available for occipital screw placement in occipital-cervical fusions. The Starlock™ instrumentation (Synthes Spine, Paoli, PA) is a newer system that allows independent insertion of lateral mass screws, which are then attached to clamps along a contourable titanium rod. Additionally, a tapered rod is available to cross the cervicothoracic junction. Cross-links and laminar hooks are also available. The Summit™ system (Depuy Acromed, Rayham, MA) is another newer design for lateral mass fixation that offers polyaxial screw heads and a top-loading 3-mm titanium rod for increased ease of rod insertion. It has the advantage of approximating the ease of use and flexibility of the lumbar pedicle screw systems presently available.
Ventral Cervical Instrumentation Since the time that the first system was developed by Bohler in the mid 1960s, anterior cervical plating has become a popular means of supplementing a ventral cervical fusion.52 Early in the development of this instrumentation, the potential for screw back-out was recognized as a serious concern for possible complications, including tracheal or esophageal erosion. The first systems widely available were the Caspar™ (Aesculap, San Francisco, CA) and the Orozco (Figure 2.5) (Synthes, Paoli, PA). Both of
Figure 2.5 Postoperative lateral cervical radiograph
demonstrating anterior internal fixation with the Orozco plate.
these systems consisted of simple plates with slots or holes but without any locking devices. Constraint of the screws depended upon obtaining bicortical purchase and “blocking” backout by screw angulation. The rate of screw backout or breakage and graft subsidence was high with the first generation of anterior cervical plates. This led to the development of the Cervical Spine Locking Plate (CSLP) (Synthes, Paoli, PA)41 first introduced in North America in 1991. The CSLP used a titanium expansion screw that secured the screwhead to the plate and, thus, allowed for unicortical purchase without the risk of screw backout. The substitution of titanium for stainless steel allowed for postoperative magnetic resonance imaging. The CSLP reduced the incidence of screw backout48; however, its limitations were a rigid screw trajectory and the fact that the plate was wide and difficult to contour. The Orion™ ventral cervical plate (Medtronic/Sofamor Danek, Memphis, TN) represented the next major product introduction for anterior cervical plating. The plate was manufactured “prelordosed” with a wide variety of screw lengths to allow for unicortical or bicortical purchase. The drill guide was fixed to the plate, providing 15 degrees of rostral and caudal angulation and 6 degrees of medial angulation. Locking screws were added to fix the screws to the plate by overlapping the screw heads. Although the Orion plate saw widespread use and had good reported surgical results,40,47 some surgeons felt that the system was too rigid and shielded the graft from stress, thereby promoting a significant rate of peudoarthrosis.36 Interest in avoiding stress-shielding of a graft while preventing screw backout led to the development of “semiconstrained” plate designs that would allow the graft to share stress with the plate. The A-line™ Cervical Plating System (Surgical Dynamics, Norwalk, CT) and the Codman Anterior Cervical Plate System (Codman & Shurtleff, Raynham, MA) both allow variability in screw direction and are then fixed to the plate with a locking system. Once locked, however, the screws can pivot within the plate hole in response to sufficient force, allowing for slight graft subsidence and maintenance of stress sharing with the graft. The Atlantis™ Plate (Medtronic/Sofamor Danek, Memphis, TN) provides the option of fixed or variable angle screws, making it possible to fashion a “hybrid” construct using both screw types. An additional innovation directed at maintaining load sharing between the graft and plate to create a more optimal fusion environment is the development of slotted plates. Examples of available products applying this design include the ABC™ Plating System (Aesculap, Center Valley, PA), the Premier™ Anterior Cervical Plate System (Medtronic/Sofamor Danek, Memphis, TN), and the DOC™ Ventral Cervical Stabilization System (Depuy Acromed, Rayham, MA). Constructs made with these plates allow the screws to remain in essentially the same position within the vertebra while the screw head moves along a slot in response to graft subsidence. There is some evidence that the ability of the screw to “translate” within the plate is leading to a lower incidence of graft and plate failure in multilevel fusions.21 The success of this technique is fueling sustained interest in these plate designs.
27
28
Chapter 2 ■ History of Spinal Instrumentation: The Modern Era
Ventral fixation of odontoid fractures can be achieved with the placement of one or multiple screws. Although the technique was published in 1971 by Barbour,7 it did not achieve popularity until the late 1980s.9 Controversy developed over whether one or two screw placements is optimal for fixation.20 Several recent papers, however, have not shown improved results with multiple screw placement.4,5 Some surgeons have advocated the application of cannulated screws placed over K-wires in this procedure, citing improved accuracy and the ability to redirect the screw trajectory as technical advantages. Other surgeons, however, prefer the original noncannulated screws, noting the potential risks of K-wire breakage as well as unintended K-wire advancement during screw placement.24
Cage Technology The development of cages to promote interbody fusion traces back to the veterinary work of Bagby in which stainless-steel baskets filled with bone were used to treat Wobbler-neck syndrome in race horses.16 Bagby subsequently pioneered the development of a cage for use in human lumbar interbody fusions.6,33 The implantation of cages as an interbody device through either an anterior or posterior approach has become a widely performed procedure. Titanium-threaded cages include the BAK™ cage (Sulzer Spine-Tech, Minneapolis, MN) and the Ray Threaded Fusion Cage (Surgical Dynamics, Norwalk, CT). Brantigan et al11 introduced cages composed of a radiolucent carbon fiber that allowed for improved postoperative imaging. It is also argued that the carbon fiber material has a modulus closer to that of native bone and, thus, should theoretically be a better fusion substrate than metal.11 Although the initial cage development was done for the cervical spine, the technology was first widely implemented in the lumbar spine. In April 2002, the Food and Drug Administration (FDA) approved the use of the BAK-C™ device (Sulzer Spine-Tech, Minneapolis, MN) for cervical fusion.39 Recent experience with these implants has indicated that fusion rates are achieved comparable with those seen after procedures using uninstrumented allograft.27a To facilitate ventral vertebral reconstruction after anterior and middle column resection, Harms developed a titanium mesh cage that can be packed with bone and is seated into the endplates.29 This implant has found application in cases of vertebral body destruction resulting from metastatic disease, degenerative conditions, and trauma. The Harms cage was considered a valuable innovation even to those surgeons who prefer using struts made of allograft or autograft because a suitable bone graft is sometimes unavailable.
Summary The development of instrumentation for internal fixation of the spine has dramatically improved the ability to successfully provide surgical intervention for a wide variety of spinal disorders. Internal fixation leads to higher fusion rates and provides more powerful means to correct spinal deformities.
In addition, spinal instrumentation allows for reduction or elimination of the need for postoperative external bracing. Over the past 25 years, there has been an amazing increase in the variety of instrumentation available to provide internal spinal fixation. Surgeons are now able to select a specific type of implant that is best suited to address an individual patient’s problem. Improved understanding of biomechanics and clinical experience with today’s instrumentation should promote further advancement in internal fixation and even better patient outcomes in the future.
REFERENCES 1. Abumi K, Panjabi MM, Duranceau J: Biomechanical evaluation of spinal fixation devices. Part III. Stability provided by six spinal fixation devices and interbody bone graft. Spine 14:1249-1255,1989. 2. An HS, Gordin R, Renner K: Anatomic considerations for plate-screw fixation of the cervical spine. Spine 16:S548-551, 1991. 3. Anden U, Lake A, Nordwall A: The role of the anterior longitudinal ligament in Harrington rod fixation of unstable thoracolumbar spinal fractures. Spine 5:23-25, 1980. 4. Apfelbaum R, Lonser R, Veres R, Casey A: Direct anterior screw fixation for recent and remote odontoid fractures. J Neurosurg 93:227-236, 2000. 5. Arand M, Lemke M, Kinzl L, Hartwig E: Incidence of complications of the screw osteosynthesis of odontoid process fractures. Zentralbl Chir 126:610-615, 2001. 6. Bagby GW: Arthrodesis by the distraction-compression method using a stainless steel implant. Orthopedics 11:931-934,1988. 7. Barbour J: Screw fixation in fracture of the odontoid process. South Anst Clin 5:20, 1971. 8. Black RC, Gardner VO, Armstrong GW, et al: A contoured anterior spinal fixation plate. Clin Orthop 227:135-142, 1988. 9. Bome GM, Bedou GL, Pinaudeau M, et al: Odontoid process fracture osteosynthesis with a direct screw fixation technique in nine consecutive cases. J Neurosurg 68:223-226, 1988. 10. Bostman O, Myllynen P, Riska EB: Posterior spinal fusion using internal fixation with the Daab plate. Acta Orthop Scand 55:310-314, 1984. 11. Brantigan JW, Steffee AD, Geiger JM: A carbon fiber implant to aid interbody lumbar fusion. Mechanical testing. Spine 16:S277-282, 1991. 12. Bridwell KH: Spinal instrumentation in the management of adolescent scoliosis. Clin Orthop Feb:64-72, 1997. 13. Cooper PR: The Axis Fixation System for posterior instrumentation of the cervical spine. Neurosurgery 39:612-614,1996. 14. Cotrel Y, Dubousset J: [A new technic for segmental spinal osteosynthesis using the posterior approach]. Rev Chir Orthop Reparatrice Appar Mot 70:489-494, 1984. 15. Cybulski GR, Stone JL, Crowell RM, et al: Use of Halifax interlaminar clamps for posterior C1 -C2 arthrodesis. Neurosurgery 22:429-431, 1988. 16. DeBowes RM, Grant BD, Bagby GW, et al: Cervical vertebral interbody fusion in the horse: a comparative study of bovine xenografts and autografts supported by stainless steel baskets. Am J Vet Res 45:191-199, 1984.
Chapter 2 ■ History of Spinal Instrumentation: The Modern Era 17. Dickman CA, Fessler RG, MacMillan M, Haid RW: Transpedicular screw-rod fixation of the lumbar spine: operative technique and outcome in 104 cases. J Neurosurg 77:860-870, 1992. 18. Drummond D, Guadagni J, Keene JS, et al: Interspinous process segmental spinal instrumentation. J Pediatr Orthop 4:397-404, 1984. 19. Dunn HK: Anterior stabilization of thoracolumbar injuries. Clin Orthop Oct:116-124, 1984. 20. El Saghir H, Bohm H: Anderson type H fracture of the odontoid process: results of anterior screw fixation. J Spinal Disord 13:527-530, 2000. 21. Epstein N: Anterior approaches to cervical spondylosis and ossification of the posterior longitudinal ligament: review of operative technique and assessment of 65 multilevel circumferential procedures. Surg Neurol 55:313-324, 2001. 22. Erwin VVD, Dickson JH, Harrington PR: Clinical review of patients with broken Harrington rods. J Bone Joint Surg Am 62:1302-1307, 1980. 23. Esses ST, Bednar DA: The spinal pedicle screw: techniques and systems. Orthop Rev 18:676-682, 1989. 24. Fuji T, Oda T, Kato Y, et al: Accuracy of atlantoaxial transarticular screw insertion. Spine 25:1760-1764, 2000. 25. Ghanayem AJ, Zdeblick TA: Anterior instrumentation in the management of thoracolumbar burst fractures. Clin Orthop Feb:89-100, 1997. 26. Grob D, Crisco JJ 3rd, Panjabi MM, et al: Biomechanical evaluation of four different posterior atlantoaxial fixation techniques. Spine 17:480-490, 1992. 27. Grob D, Jeanneret B, Aebi M, Markwalder TM: Atlantoaxial fusion with transarticular screw fixation. J Bone Joint Surg Br 73:972-976, 1991. 27a.Hacker RJ, Cauthen JC, Gilbert TJ, Griffith SL: A prospective randomized multicenter clinical evaluation of an anterior cervical fusion cage. Spine 15; 25:2646-2654, 2000. 28. Haid RW, Jr., Subach BR, McLaughlin MR, et al: C1-C2 transarticular screw fixation for atlantoaxial instability: a 6-year experience. Neurosurgery 49:65-68; discussion 69-70, 2001. 29. Heller JG, Carlson GD, Abitbol JJ, Garfin SR: Anatomic comparison of the Roy-Camille and Magerl techniques for screw placement in the lower cervical spine. Spine 16:S552-557, 1991. 30. Heller JG, Zdeblick TA, Kunz DA, et al: Spinal instrumentation for metastatic disease: in vitro biomechanical analysis. J Spinal Disord 6:17-22,1993. 31. Holness RO, Huestis WS, Howes WJ, Langille RA: Posterior stabilization with an interlaminar clamp in cervical injuries: technical note and review of the long term experience with the method. Neurosurgery 14:318-322, 1984. 32. Johnston CE 2nd, Happel LT, Jr., Norris R, et al: Delayed paraplegia complicating sublaminar segmental spinal instrumentation. J Bone Joint Surg Am 68:556-563,1986. 33. Kuslich SD, Ulstrom CL, Griffith SL, et al: The Bagby and Kuslich method of lumbar interbody fusion. History, techniques, and 2-year follow-up results of a United States prospective, multicenter trial. Spine 23:1267-1278; discussion 1279, 1998. 34. Lagrone MO, Bradford DS, Moe JH, et al: Treatment of symptomatic flatback after spinal fusion. J Bone Joint Surg Am 70:569-580, 1988.
35. Livingston KE, Perrin RG: The neurosurgical management of spinal metastases causing cord and cauda equina compression. J Neurosurg 49:839-843, 1978. 36. Lowery GL, McDonough RF: The significance of hardware failure in anterior cervical plate fixation. Patients with 2- to 7-year follow-up. Spine 23:181-186; discussion 186-187, 1998. 37. Luque ER: Segmental spinal instrumentation for correction of scoliosis. Clin Orthop March:192-198, 1982. 38. Magerl F, Seeman P: Stable posterior fusion of the atlas and axis by transarticular screw fixation. In Kehr P, Weidner A (eds): The Cervical Spine. Wien, Springer Verlag, 1987, 322. 39. Matge G: Anterior interbody fusion with the BAK-cage in cervical spondylosis. Acta Neurochir 140:1-8, 1998. 40. Mayr MT, Subach BR, Comey CH, et al: Cervical spinal stenosis: outcome after anterior corpectomy, allograft reconstruction, and instrumentation. J Neurosurg 96:10-16, 2002. 41. Morscher E, Sutter F, Jenny H, Olerud S: [Anterior plating of the cervical spine with the hollow screw-plate system of titanium]. Chirurg 57:702-707, 1986. 42. Morscher E: [Two-stage reposition and fixation of spondyloptosis with Harrington instrumentation and anterior intercorporal spondylodesis (author’s transl)]. Arch Orthop Unfallehir 83:323-334, 1975. 43. Roy-Camille R, Roy-Camille M, Demeulenaere C: [Osteosynthesis of dorsal, lumbar, and lumbosacral spine with metallic plates screwed into vertebral pedicles and articular apophyses]. Presse Med 78:1447-1448, 1970. 44. Roy-Camille R, Saillant G, Mazel C: Internal fixation of the cervical spine by a posterior osteosynthesis with plates and screws. In Sherk H, Dunn E, Eisment F (eds): The Cervical Spine. Philadelphia, JB Lippincott, 1989, pp 390-403. 45. Ryan MD, Taylor TK, Sherwood AA: Bolt-plate fixation for anterior spinal fusion. Clin Orthop Feb:196-202,1986. 46. Schafer MF: Dwyer instrumentation of the spine. Orthop Clin North Am 9:115-122, 1978. 47. Schultz KD, Jr., McLaughlin MR, Haid RW, Jr., et al: Single-stage anterior-posterior decompression and stabilization for complex cervical spine disorders. J Neurosurg 93:214-221, 2000. 48. Spivak JM, Chen D, Kummer FJ: The effect of locking fixation screws on the stability of anterior cervical plating. Spine 24:334-338, 1999. 49. Stambough JL: Posterior instrumentation for thoracolumbar trauma. Clin Orthop Feb:73-88, 1997. 50. Sturz H, Hinterberger J, Matzen K, Plitz W: Damage analysis of the Harrington rod fracture after scoliosis operation. Arch Orthop Trauma Surg 95:113-122, 1979. 51. Sundaresan N, Galicich JH, Lane JM: Harrington rod stabilization for pathological fractures of the spine. J Neurosurg 60:282-286, 1984. 52. Vaccaro AR, Balderston RA: Anterior plate instrumentation for disorders of the subaxial cervical spine. Clin Orthop 112-121, 1997. 53. Wang GJ, Whitehill R, Stamp WG, Rosenberger R: The treatment of fracture dislocations of the thoracolumbar spine with halofemoral traction and Harrington rod instrumentation. Clin Orthop Jul-Aug:168-175, 1979.
29
30
Chapter 2 ■ History of Spinal Instrumentation: The Modern Era 54. Wilber RG, Thompson GH, Shaffer JW, et al: Postoperative neurological deficits in segmental spinal instrumentation: a study using spinal cord monitoring. J Bone Joint Surg Am 66:1178-1187, 1984, 55. Winter RB: Combined Dwyer and Harrington instrumentation and fusion in the treatment of selected
patients with painful adult idiopathic scoliosis. Spine 3:135-141, 1978. 56. Yuan HA, Mann KA, Found EM, et al: Early clinical experience with the Syracuse I-Plate: an anterior spinal fixation device. Spine 13:278-285, 1988.
C H A P T E R
rapidity of onset, and severity of neurologic deficit and associated systemic illness. For instance, pain in conjunction with fever and weight loss, recumbent position, morning stiffness, acute onset, or visceral component allows for initial categorization. The first half of this chapter deals with those disorders that usually present with spinal pain, and the second portion deals with those conditions that present with pain and neurologic deficit. The choice of imaging studies, laboratory tests, and surgical or medical management is summarized in Figures 3.1 and 3.2.
3
Differential Diagnosis of Surgical Disorders of the Spine
Spinal Pain Pain Associated with Fever and Weight Loss
W. Putnam Wolcott, Jacek M. Malik, Christopher I. Shaffrey, Mark E. Shaffrey, and John A. Jane
Patients who present with fever and weight loss associated with spinal pain are at increased risk for an infectious or neoplastic process. Vertebral osteomyelitis, discitis, epidural abscess, and granulomatous processes are the most common infectious conditions affecting the spine. Neoplastic processes, specifically metastatic disease and lymphoma, are often associated with a similar presentation. Neurologic deficits can occur, but are usually present after a period of diagnostic delay that may last weeks to months after the pain and systemic symptoms occur.
Back pain is the most common reason cited for work absence in the United States, affecting approximately 31 million people annually.118,127,135,136 Low back pain (LBP) alone is estimated to cost society between $20 billion and $50 billion a year in the United States. This is a huge financial burden on an already taxed medical system.189 Rapid diagnosis and treatment that result in a rapid return to a premorbid level of activity should greatly reduce costs to the patient and to society. Patients presenting with pain related to the spine do not have a surgical condition in the majority of cases.118,124,189 Only approximately 8% to 10% of those with neurogenic pain have a surgical lesion. These individuals can present with a myriad of symptoms, and the differential diagnosis of spinal disorders is lengthy.189 This chapter presents a logical approach to evaluate the patient who presents with spinal pain or neurologic deficit with a suspected spinal disorder. An algorithm proposed by Borenstein et al. is used as a functional guide and is expanded to include patients who present with a neurologic deficit related to spinal pathology, as well as those who present with pain as their initial complaint.23,24 The following observations are used to classify spinal pathology: (1) the presence or absence of spinal region pain, (2) the characteristics of the pain, (3) the presence or absence of neurologic deficit, (4) the characteristics of the deficit, and (5) the presence of systemic signs and symptoms. Each patient is entered into the algorithm after a thorough history and physical examination. The differential diagnosis process begins with the characterization of back pain, discovery of systemic signs and symptoms, and evaluation for neurologic deficit. With this information, further laboratory and radiologic evaluation can proceed, and ultimately, a diagnosis with appropriate surgical or medical management usually can be achieved. There is some expected overlap between groups in patient presentation; however, the basic framework of the algorithm should help the clinician restrict the differential diagnosis and call attention to uncommon conditions. The causes of spinal pain and neurologic deficit are vast. Therefore an attempt is made to delineate the most common causes by age, location, character, site of pain,
Vertebral Osteomyelitis Patient Population. This disease represents the most common of the pyogenic infections of the axial skeleton.8,91,163 Vertebral osteomyelitis occurs in 2% to 19% of all cases of osteomyelitis, with the largest percentage occurring in the elderly and debilitated.11,39 More than half of patients with vertebral osteomyelitis are older than age 50.163 Often, this diagnosis is initially missed because of its relatively benign and nonspecific presentation. Time from initial presentation to diagnosis ranges from 2 weeks to 5 months, with a mean of 6 to 8 weeks.39,163 Adults tend to demonstrate a more chronic and indolent course with longer delays, whereas the pediatric and immunocompromised groups present with a more acute picture, which results in an earlier diagnosis. Sources of Infection. A definitive source of infection is found in only 40% of cases, with the most common foci being genitourinary, soft tissue, respiratory, and intravenous drug abuse (IVDA).19,51,91 Other less commonly encountered sources include enteric disease, dental extractions, endocarditis, and penetrating trauma. The most common organisms isolated are the grampositive cocci (GPC), which constitute 60% to 70% of all cases of vertebral osteomyelitis1,91,115,163; Staphylococcus aureus is the most prevalent organism, representing 55% to 60% of all positive cultures.51,91,105,163,183 There has been a relative increase (14% to 18%) in other bacteria in more recent series, specifically gram-negative rods (e.g., Escherichia coli, Pseudomonas aeruginosa, and Proteus), which are found predominantly in the parenteral drug abuser or immunocompromised patient.* Signs and Symptoms. The virulence of an organism is related to the timing of diagnosis, which can range from *References 11,51,111,112,153,191,193.
33
34
Section 1 ■ The Fundamentals
Figure 3.1 An algorithm for the evaluation of pain of spinal origin.
weeks to months after initial symptoms. The most common symptom is insidious diffuse back pain, which occurs in approximately 90% of patients.91,115,163 Fever occurs in 40% to 52% of patients. Weight loss, radicular symptoms, myelopathy, spinal deformity, and meningeal irritation are
less common presentations. Neurologic compromise and decreased mobility are late complications. Neurologic deficits are rare in the magnetic resonance imaging (MRI) era; however, neurologic sequelae were present in up to 50% of Hitchon and Yuh’s original series.91
Chapter 3 ■ Differential Diagnosis of Surgical Disorders of the Spine
Figure 3.1 cont’d
Analysis of Findings. Diagnosis is based on pertinent laboratory findings, including an elevated erythrocyte sedimentation rate (ESR), blood and bone cultures, elevated white blood cell (WBC) count, and radiologic studies. The ESR is elevated in approximately 90% of patients. Positive blood cultures and an elevated WBC count are present in only 25% to 42% of adult cases. MRI with superior sensitivity and specificity of 89% to 100% is the gold standard for detection of osteomyelitis.4,42,56,133 The characteristic changes observed on images can take from 2 to 3 weeks to become apparent.4 Bone scans are useful for diagnosis secondary to high sensitivity (90%); however, they can be misleading because other inflammatory processes or neoplasia can mimic an infection on these scans.45,60 Patient Management. Management is based on biopsy results; the preferred method is a percutaneous aspiration and bone biopsy under computed tomography (CT) or fluoroscopic guidance.35 An open biopsy is usually reserved for inadequate tissue acquired via percutaneous biopsy, neurologic compromise, and/or refractory disease.73 Treatment entails bed rest and initiation of broadspectrum antibiotics, followed by definitive antimicrobial therapy based on culture results.11 Ventral decompression
and bone grafting may be required if neurologic sequelae ensue from excessive bone destruction or abscess formation.11,51,91,98,111 Epidural Abscess Patient Population. The incidence of spinal epidural abscess (SEA), including all sources of infection, ranges from 1% to 10%.7,8,11 Adults are primarily affected, with rare occurrences in children. Thoracic involvement is most common, followed closely by lumbar involvement. Cervical SEA occurs in less than 15% of cases43,68,91; dorsal location predominates in approximately two thirds of cases, with an occasional ventral abscess resulting from a direct extension of a located discitis or vertebral osteomyelitis.68 Sources of Infection. Epidural abscesses are usually caused by direct extension of a preexisting osteomyelitis in 20% to 60% of cases,91 direct inoculation from surgical or procedural manipulation, or hematogenous spread from a distant focus. Trauma plays a role in the formation of lateonset abscesses secondary to formation of paraspinal hematomas that act as a physiologic culture medium. The bacteria in SEA are closely related to the organisms found
35
36
Section 1 ■ The Fundamentals
Medical Evaluation of Spinal Pain with Neurologic Deficit
Physical and Neurologic Examination
Cerebral/Brain Stem?
Spinal Cord? Yes
Mental status changes Aphasia Seizures Cranial nerve palsies
Plain X-Ray
CT Scan
Yes Yes
N o
Trauma, Tumor Spondylosis, Stenosis Spondylolisthesis Spinal Dysraphism
N o Tumor, Infarct Hydrocephalus
CT and/or MRI
MRI
Yes
Yes N o
Inflammatory disorder Tumor Vascular etiology
LP Yes
No
Inflammatory disorder MS
Storage disease Metabolic/Nutritional Toxins ALS
No
Dural AV Fistula
Yes
Extra/Intradural Tumors Infarct, Intradural AVM Epidural Abscess, MS Disc Herniation
No
LP
Myelo/CT
Yes
No Yes
Spinal Angiogram
MS, Encephalomyelitis
EMG
Myopathies
?Muscle Bx
Neuropathies
?Nerve Bx
Yes
No Vasculitis Combined Degeneration Metabolic Disorders Paraneoplastic Syndrome
Figure 3.2 An algorithm for the evaluation of spinal pain with neurologic deficit.
in vertebral osteomyelitis, with gram-positive cocci also the most prevalent. S. aureus is isolated in 60% to 65% of the positive cultures, followed by other staphylococcal and streptococcal species.91 Gram-negative rods, although uncommon, occurred with increasing frequency in recent series. Signs and Symptoms. Spine pain is the initial clinical presentation in the majority of cases. There is a higher incidence of fever, leukocytosis, and neurologic compromise than in osteomyelitis7,68,91,191; meningeal signs may be present if the adjacent dura is violated. Misdiagnosis of SEA is common. This entity should be included in the differential diagnosis of any patient who presents with fever and spine pain. Analysis of Findings. Laboratory studies, including ESR and WBC count, are elevated in the majority of patients. MRI is the diagnostic procedure of choice because of its ability to image multiple levels simultaneously.68,109 The T1-weighted images often reveal an isointense or hyperintense extradural lesion, with the T2-weighted images showing a hyperintense collection with greater contrast enhancement.
Patient Management. Without surgical intervention, rapid neurologic deterioration occurs in approximately 10% to 15% of cases, even if the patient is given intravenous antibiotics.91,163 Surgical debridement and decompression, with administration of applicable antibiotics, is appropriate; definitive antibiotic treatment should begin after cultures return. Pediatric Discitis Patient Population. Discitis represents two entirely different entities in the adult and pediatric populations.39,41,50 Classically, discitis is a disease entity of the pediatric population, although it is an uncommon occurrence in the adult, except postoperatively.47 In children discitis has been described as either a chronic inflammatory disorder or a low-grade bacterial infection with a relatively benign course.39,57 Early investigation and detection are important to prevent the progression to a devastating process, such as vertebral osteomyelitis. The incidence of pediatric discitis is unclear, but it has a peak age of occurrence at 6 years of age.191 Sources of Infection. The pediatric population is predisposed to discitis because of the vascular anastomotic
Chapter 3 ■ Differential Diagnosis of Surgical Disorders of the Spine
network around and through the cartilaginous end-plates and discs; vascular obliteration of this surrounding vascular network occurs later in adolescence. The pediatric arterial network functions as a bacterial filter and endpoint. Signs and Symptoms. The clinical presentation of back pain and painful ambulation is related to this entity’s most common location in the lumbar spine. Analysis of Findings. Diagnosis is made via an elevated ESR (in up to 75% of cases), a characteristic MRI or radionuclide bone scan, proper clinical setting, and appropriate age group. MRI, because of its ability to detect hydration changes in the disc, is the most sensitive and specific modality both to detect infection and to follow efficacy of treatment.71,76,109 Gram-positive cocci, specifically staphylococcal and streptococcal species, predominate in positive cultures, occurring in more than 50% of reported cases.50,51,151,191 Nondiagnostic biopsies and blood cultures are found in 20% to 30% of patients in most series. Patient Management. Management consists of bed rest and a brief course of intravenous antibiotics until symptoms attenuate. Then oral antibiotic therapy is instituted. Prognosis is excellent in the uncomplicated case. Adult Discitis Spontaneous discitis is rare in the adult. It is encountered in 2% to 3% of surgical discectomy patients.37,47,73 Clinical presentation reveals back pain at the operated level, usually from 1 to 3 weeks postoperatively. ESR is the most sensitive laboratory test for detection and monitoring of treatment progression. MRI is the investigative procedure of choice, because characteristic changes are visible with MRI well before plain radiographs.4,133 Granulomatous Infections Granulomatous infections include all processes that produce the classic histologic granuloma. These processes include fungal; spirochete; uncommon bacterial organisms such as actinomycosis, nocardia, and brucellosis; and finally, the most common organism, Mycobacterium tuberculosis infections. It is emphasized that patients with other forms of granulomatous disease, such as Wegener’s granulomatosis (a form of vasculitis), are at increased risk of an infectious disorder of the spine.128 Tuberculous Spondylitis Although uncommon in developed countries, tuberculous spondylitis is the most common of the granulomatous infections that affect the axial skeleton.* Patient Population. Although cases of tuberculosis declined in the United States until the mid-1980s, recent epidemiologic data suggest a current resurgence.59,77 The *References 17,19,43,62,77,196.
amount of extrathoracic involvement has increased from less than 8% to more than 18%.17 In developed countries tuberculosis is a disease of the elderly, whereas in underdeveloped countries it predominates in children.77 There are approximately 20,000 cases reported each year in the United States, with the immigrant, low socioeconomic, and immunocompromised populations representing the majority of cases.17 Individuals infected with the human immunodeficiency virus (HIV) are responsible for 87% of matched registries of active tubercular infections in the United States. Skeletal involvement occurs in approximately 1% of all cases. Of those cases, 50% to 60% have an infection of the axial skeleton.77 Sources of Infection. Historically dubbed Pott’s disease, tuberculous spondylitis is usually caused by M. tuberculosis; however, another species of mycobacteria may be the culprit. Tuberculosis spondylitis generally results from hematogenous spread of the pathogen via a pulmonary or genitourinary source. Other routes such as lymphatic dissemination or direct extension from adjacent areas of infection are less frequently encountered. After the initial seeding from a primary focus elsewhere in the body, this granulomatous infection usually progresses via subligamentous spread across the disc space. This occurs along either the posterior or anterior longitudinal ligament, with relative disc-space sparing. Vertebral collapse, spinal deformity, epidural abscess, and subarachnoid seeding after dural erosion are late sequelae. Signs and Symptoms. Clinical presentation involves bone pain over the affected site, most commonly the thoracolumbar spine, in conjunction with fever, malaise, and weight loss. Cervical and sacral involvement are rare, occurring in less than 0.03% of all reported cases.177,196 In the progressive stages of disease kyphosis results from erosive bone destruction. Epidural abscesses are common sequelae, occurring in approximately 50% to 85% of cases.59,62,97 Paraparesis caused by spinal tuberculosis has remained a constant, occurring in 20% of all cases of reported tubercular infections.116 Analysis of Findings. Diagnosis requires evaluation of a urine, sputum, or gastric specimen; a subcutaneous nodule; or bone biopsy to isolate the offending organism. A positive purified protein derivative (PPD) can be helpful, although false negatives can occur in the anergic patient because of age, malnutrition, or immunocompromise. A chest radiograph reveals no evidence of pulmonary disease in 40% to 50% of patients. Plain spine radiographs require 2 months of active tubercular osteomyelitis to become diagnostic.59,77 Radiologic imaging, including MRI and CT/myelogram, is the basis for evaluating treatment progression and surgical planning. MRI is superior to evaluate soft tissue involvement and presence of abscess formation; CT provides better bone detail, especially in the dorsal elements. Patient Management. Treatment modalities are based on positive biopsy and culture, degree of kyphosis, extent of neurologic compromise, and disease refractory to medical management. Neurologic sequelae occur in approximately 10% to 50% of cases of active disease, with 20% of
37
38
Section 1 ■ The Fundamentals
cases of tuberculous spondylitis resulting in paraparesis.116 Ventral decompression with spinal fusion, and debridement with initiation of proper antibiotic therapy, are the accepted modalities of treatment in the case of an evolving deficit. Prognosis, morbidity, and mortality are related to overall age, with children faring better than adults, and to the extent of systemic involvement and preoperative neurologic status.67,85,97,187 Actinomycosis Patient management. Related morphologically to the fungi, Actinomyces israelii is an anaerobic gram-positive bacterium. An infection leads to purulent abscesses, external draining sinuses, and characteristic sulfa granules on microscopic examination. Actinomycosis is found most often in the cervical spine, because the majority of cases involve the mandible and supraclavicular areas with direct extension to the adjacent vertebra.43,91 The actinomycosis infection clinically resembles tuberculous spondylitis, except with less vertebral destruction and disc space narrowing. Signs and Symptoms. Clinical presentation includes neck pain with associated purulent sinus tracts. Diagnosis entails plain radiographs for bone erosion, MRI to evaluate associated abscess, and a biopsy to obtain a culture. Patient Management. After diagnosis, medical treatment with penicillin is the mainstay, with a generally favorable outcome for uncomplicating abscesses.163 Surgical intervention is required for abscess drainage, symptomatic epidural compression of neural structures, or instability that results from bone erosion. Nocardia Sources of Infection. Nocardia asteroides, a grampositive bacterium, usually presents as a systemic illness, but can represent a rare cause of back pain secondary to involvement of the bony spinal column.19,43 Nocardia usually spreads hematogenously from a pulmonary focus to the soft organ systems, but osteomyelitis has also been reported on rare occasions. There is a higher incidence of cervical and thoracic osseous involvement. This is related to the proximity of the pulmonary focus. Spread occurs through venous drainage of the retropharyngeal and mediastinal sites.114 Signs and Symptoms. Initial symptoms are constitutional with eventual radicular pain or myelopathy from extradural compression secondary to abscess formation.114 Patient Management. When there is no evidence of neurologic compromise or abscess formation, the patient is treated with sulfonamide antibiotics. Surgical intervention is warranted in the setting of epidural compression or bony instability. Brucellosis Brucella, a gram-negative rod, is extremely uncommon in developed countries, primarily because of milk pasteurization and uncontaminated food sources.
Source of Infection. A disease that affects farm animals, brucellosis is usually transmitted via ingestion of contaminated food or, reportedly, inhalation. Depending on the magnitude of the inoculum, incubation is a period of days to weeks. Signs and Symptoms. The resultant illness usually presents as a systemic illness with indolent fever, lymphadenopathy, generalized malaise, and occasionally, back pain. Spinal involvement with a lumbar predilection is observed in 54% of cases.132 Of those with spinal involvement, 12% present with some degree of spinal cord impingement.19,43,91 Analysis of Findings. Diagnosis can be made by blood culture, although half are negative; a positive Brucella agglutination test is diagnostic. Plain radiographs are usually nonproductive until late in the disease process. A technetium bone scan or MRI aids in the diagnosis of an active infection. Patient Management. Early institution of treatment with appropriate antibiotics for 6 weeks is the rule for this curable disease. Fungal Infections Fungal infections of the axial skeleton are uncommon, even in endemic areas. Infection occurs by spore inhalation, with resultant pulmonic seeding and systemic spread. Fungal infections in patients with disseminated disease have varying degrees of spinal osseous involvement. This involvement with disseminated fungal infection occurs in 10% to 50% of patients with coccidioidomycosis and blastomycosis infections. There is a much lower incidence of axial involvement for candidiasis or aspergillosis.19,43 Coccidioidomycosis Coccidioidomycosis, endemic to the southwestern United States, has a high rate of spinal involvement and occurs in 20% to 40% of cases of disseminated disease. Radiographs reveal that multiple simultaneous lytic lesions occur in up to 20% of patients with no specific predisposition to vertebral site involvement. Vertebral collapse and neurologic compromise are uncommon.43 Diagnostic plain radiographs, immunodiffusion titers, and biopsy are used to determine treatment with appropriate antifungal pharmacotherapy. Blastomycosis The most virulent fungal agent for osseous involvement, and endemic to the Mississippi River valley, blastomycosis is spread via inhalation, with resultant pulmonary infection and focus. Blastomycosis is hematogenously spread, with a predilection for ventral vertebral involvement. This gives rise to vertebral collapse, joint erosion, and disc invasion. Clinical presentation and bone destruction resemble tuberculous spondylitis. However, blastomycosis more commonly has associated draining sinuses and a greater predisposition to include the dorsal elements.
Chapter 3 ■ Differential Diagnosis of Surgical Disorders of the Spine
Cryptococcus Cryptococcus neoformans is a fungal infection more commonly known for central nervous system involvement in patients with acquired immunodeficiency syndrome (AIDS). It is usually inhaled in an aerosolized form and then spreads hematogenously from a pulmonary location. Osseous involvement occurs in only 10% of cases with the disseminated form.50 Granuloma formation and cellular reaction are minimal. Radiographs reveal dorsal vertebral involvement and disc space sparing. The usual clinical presentation, which is a late sequela of disseminated cryptococcal meningitis, is swelling and pain of the affected vertebral site, as well as decreased spine mobility. Diagnosis is made via latex agglutination test, and positive cerebrospinal fluid (CSF) and blood cultures. Treatment is with amphotericin-B. Disease control, not eradication, is the goal in the immunocompromised host.
age 30 are predisposed toward malignancy.55,126 Malignant lesions of the axial skeleton, whether metastatic or primary, are found more often in ventral locations.20,22,126 Benign processes tend to favor the dorsal elements. The incidence of skeletal metastases outweighs the incidence of primary bone tumors by a margin of 25 to 40:1.12,20,126,140 The overall incidence of primary bone tumors is only 0.4%.20,55,126 Signs and Symptoms. The most frequent clinical presentation for spinal tumors is back pain occurring in up to 85% in the larger series.20–55 The pain associated with neoplastic growth is usually secondary to one of several factors, which include pathologic fracture, tumor vascular engorgement, periosteal stretching, and impingement on nerve roots in the epidural space.97 Metastatic Disease
Candidiasis and Aspergillosis Aspergillus, a mold, and Candida, a yeast, are extremely rare causes of vertebral osteomyelitis, with less than a total of 30 cases of both reported in the literature.49 Both are pathogens of the immunocompromised host. Candidiasis is not as uncommon as other focal or systemic infections found in the intensive care unit (ICU) setting, secondary to the amount of invasive monitoring in critically ill patients.112 Osseous involvement occurs during a prolonged hospitalization and inadequate treatment of a systemic infection. Aspergillus is associated with sinus tract and abscess formation, as well as a radiographic picture similar to tuberculosis. Treatment of both of the aforementioned fungi is with amphotericin-B. Prognosis for aspergillosis spondylitis is poor, even after surgical debridement and drainage.49 Other Infections Wegener’s granulomatosis, syphilis, and parasitic infections, specifically echinococcosis of the spine, are also potential causes of back pain. Usually reported in patients with fulminant systemic infection, these infections must be included as potential sources of vertebral destruction and back pain associated with fever and weight loss. Pain Associated with Recumbency and Night Pain Tumor Nocturnal pain and pain associated with recumbency are hallmarks of destructive lesions of the vertebral column, caused by either a skeletal metastasis or primary bone tumor.* Regrettably, the majority of spinal column tumors are malignant. Patient Population. In the case of skeletal lesions, there are correlations between age, location, incidence, and presentation. Generally, age is directly correlated with the type of lesion. Younger patients tend to have a greater incidence of benign bone tumors, whereas those older than *References 20,26,58,79,118,124,149,178.
Cancer is the second leading cause of death in the United States, with approximately 1.3 million new cases per year.126,140 Metastatic disease in the form of distant foci is evident at autopsy in 40% to 85% of cases of malignancy.140 The spine is the most common site of skeletal metastasis. At least 5% of patients with malignancies suffer from this condition.140 In fact, epidural compression has been reported to be the presenting symptom in approximately 8% to 10% of patients with metastatic disease.22,140 Of these patients nearly all initially complain of back pain, followed by weakness and ataxia. At the time of diagnosis more than 50% will have a paraparesis or bladder/bowel disturbance.22,75,140 The axial skeleton is the leading site of bone metastases that are caused by hematogenous spread through the rich venous network that drains the lungs, pelvis, and thorax. Breast, lung, prostate, and thyroid malignancies account for 50% to 60% of metastatic lesions.22,126,140 Overall, epidural metastases are equally spread throughout the thoracic and lumbosacral spine. The number of symptomatic metastases, however, is highest in the thoracic region. In most series, symptomatic cervical lesions occur in only 6% to 8% of patients.22 Signs and Symptoms. On clinical presentation the patient’s history reveals pain of an insidious, progressive nature, unrelated to mechanical activity unless caused by a pathologic fracture from trauma or an acute compressive axial load. Diffuse pain in the elderly, presumably caused by degenerative causes, can delay diagnosis. The axiom that acute neck or back pain in a patient with a known malignancy is metastatic disease until proven otherwise is a prudent guideline.20 Analysis of Findings. Ultimate diagnosis relies on radiographic studies, including plain radiographs. Bone scans are warranted for suspected occult lesions because approximately 30% to 50% of the trabeculated bone in a vertebral body must be destroyed before it is detected on plain radiography. However, Weinstein’s cervical series detected 99% of metastatic lesions via plain radiographs alone.140 Other radiographic modalities, including MRI and CT/myelography, are helpful in determining the
39
40
Section 1 ■ The Fundamentals
extent of bone destruction and epidural compression, as well as screening for other areas of involvement. In addition to superior sensitivity and specificity, the ability of MRI to scan large areas at once minimizes the chance of missing additional areas of neoplastic involvement. Diagnostic regimens include laboratory studies demonstrating an elevated calcium level, prostate-specific antigen (PSA), or alkaline phosphatase (ALP), as well as pathologic confirmation via biopsy of a primary malignant focus (if present). Metastatic work-up, including both a plain chest radiograph and an enhanced abdominal/chest CT, determines the primary focus in the majority of cases.12,22,140 Patient Management. Treatment options for metastatic disease of the spine include both radiation and surgical intervention. Operative intervention is palliative, with pain control and maintenance of function and stability as the goals. It is usually reserved for neurologic compromise, radiation failure, spinal instability, or uncertain diagnosis. The patient’s preoperative functional status and level of activity directly correlate with the postoperative result.12 Decompression and stabilization via a ventral approach is usually the route of choice; a review of dorsal decompression found an overall unfavorable result, with 12% to 40% below or at baseline preoperative function after surgery.12,140 Rapid progression of neurologic deficit is also a poor prognostic indicator. Patients who suffer progressive neurologic deficits that occur over 24 hours have a 28% to 35% chance of permanent paraplegia, whereas those with slowly evolving deficits regain ambulatory function in approximately 60% to 76% of cases.12,140 Overall, prognosis is directly related to neoplastic type, spinal location, and extent of systemic involvement. Multiple Myeloma Some controversy exists about whether multiple myeloma (MM) is a metastatic lesion or a primary bone malignancy. MM and solitary plasmacytoma account for 45% of all malignant bone tumors.20,22,140,149 The overall incidence ranges from 2 to 5.4 per 100,000, with MM significantly more prevalent than solitary plasmacytoma.20,22 These disorders are the result of abnormal proliferation of plasma cells, which are responsible for immunoglobulin and antibody production and affect the spine in 30% to 50% of reported cases. MM is primarily a disease of the sixth and seventh decades of life and has a predilection for the thoracic spine (50% to 60%), followed by the lumbar spine, and rarely (C)
Osteoid osteoma
Common (10% in spine) Uncommon (40% in spine) Uncommon
10-20 years, M/F ratio 2–4:1 T) Neural arch (C>T, L) Vertebral body (sacrum > vertebrae)
CT: “polka dot” body; MRI: “hot spot” on T1W1 Dense sclerosis, lucent nidua, lesion 50 years of age at presentation Draining vein can extend over multiple segments ~60% spontaneous Type II
Glomus malformation Dorsal aspect cervical cord 90%)1,8,10,14 have been reported with the insertion of Brantigan cages. Brantigan cages filled with autogenous bone, in combination with pedicle screw instrumentation, were compared to those of a control group that had been managed with ethyleneoxide–sterilized allograft bone blocks and pedicle screw fixation.10 Clinical success was derived from the scale of Prolo et al.52 The scores for pain, function, work status, and use of medication were combined. Scores ranging from 4 to 20 were possible. A score of 17 to 20 indicated an excellent result; 13 to 16, a good result; 9 to 12, a fair result; and 4 to 8, a poor result. A clinical success was defined as a 2-year rating of excellent or good. A successful fusion was achieved in 262 (97%) of 271 patients who had been treated with the Brantigan cage compared with 59 (79%) of 75 who had been managed with an allograft. A successful clinical result was achieved in 248 (87%) of 285 patients who had been managed with Brantigan cage and 65 (80%) of 81 who had been managed with an allograft. Brantigan et al.8 examined the efficacy of the carbon cage with pedicle screw fixation. The clinical study consisted of 221 patients at six centers. Fusion success was achieved in 176 (98.9%) of 178 patients. In the management of degenerative disc disease in patients with prior failed discectomy surgery, clinical success was achieved in 79 (86%) of 92 patients, and radiographic arthrodesis in 91 (100%) of 91 patients. Fusion success was not diminished over multiple levels. Harms Cage The vertical upright titanium-mesh cage was developed by Harms et al. in 1991.31 The Harms cage differs from horizontal cages because it needs to be cut for appropriate sizing. The device was created as an anterior load-sharing support for defects created after a ventral corpectomy for treatment of fractures and tumors. Harms used an anterior interbody approach, combined with a posterior approach, in 21 patients with a grade I or II spondylolisthesis. In the 83 patients who had a higher-grade slip (grade III or IV), Harms et al. performed a posterior lumbar interbody arthrodesis and insertion of cage packed with autogenous bone chips. The edges of the cage are designed to dig into the vertebral end plates and prevent dislodgement. To date there have been no widespread multicenter studies.
Chapter 41 ■ Lumbar Interbody Cages
Eck et al.,22 in a study of 66 consecutive patients (ages 20 to 81), examined the outcomes of titanium mesh cages implanted into the ventral column. Average follow-up was 2 years (range 24 to 62 months). No cage failure or extrusion was observed. Seventy-eight percent of the cages were judged to be fused by observers examining plain radiographs. The outcome at follow-up for patients with suspected pseudarthrosis was similar to the group of patients with fusions rated as solid.
495
cage (2 cases) Fourteen patients required additional surgery for signs of spinal instability. Brantigan Cage Most complications in the Investigational Device Exemption study were not caused by the cage but rather to the pedicle screw system.11 Seventy-eight (36%) of 219 patients had removal of pedicle screws, 30 (14%) had device-related complications, and 18 (8%) had a revision operation.
Complications of ALIF Penta et al.51 studied 52 patients, using MRI 10 years after an anterior lumbar interbody fusion (ALIF) that had not included the insertion of a fusion cage to determine if the arthrodesis increased the rate of adjacent level disc disease. They found normal adjacent discs associated with a solid fusion to the sacrum in 67% (35) of the patients and concluded that the procedure does not cause adjacent level disc disease. Wetzel studied 12 patients who had failure of a previous posterior lumbar interbody fusion (PLIF). All 12 patients had chronic radiculopathy caused by extensive perineural fibrosis as a result of epidural manipulation that was required during the procedure. Adjacent unfused segments became symptomatic in four cases, which suggested a mechanical problem inherent in the procedure. Glassman26 reported complications were more likely to occur when the cage was inserted through an anterior approach. Titanium cages (trapezoidal) inserted through a dorsal approach retropulsed into the spinal canal, causing neurologic involvement. Subsequently, these patients underwent cage removal and placement of a femoral ring packed with autogenous cancellous bone graft. Ray and BAK Device There have been no reported cases of fracture of titanium cages, whereas a fairly substantial rate of pedicle screw breakage (range, 0.9% to 27%) has been reported.40,41 Kuslich et al.39 reported the reoperation rate for patients who had a fusion cage to be 4% (42 of 947), and reoperation was most common in the first 100 days because of migration of the cage. Dural tears have a reported prevalence of 10.1% in 356 patients, and exclusively occur during the posterior approach.66 Retrograde ejaculation (1.9%), major vessel damage (1.7%), urologic complications (1.4%), and postoperative ileus (2.2%) have been associated with the anterior retroperitoneal approach (in 591 patients).66 Elias et al.23 examined the complication rate in 67 patients who underwent interbody fusion with a threaded interbody cage. A total of 15% (10) were complicated by a dural tear. In three cases bilateral cages could not be placed. Twenty-eight patients (42%) experienced significant low-back pain at 3 months, and in 10 (15%) of these cases it persisted beyond 1 year. In 10 patients postoperative radiculopathy was demonstrated. One patient incurred a permanent motor deficit with sexual dysfunction. Pseudarthrosis was suggested with evidence of motion on flexion and extension films (10 cases), lucencies around the implants (7 cases), and dorsal migration of the
Anterior Open and Anterior Laparoscopic Cage Insertion Regan et al.55,56 looked at 250 consecutive patients who had a laparoscopic arthrodesis at the fourth and fifth lumbar and first sacral vertebrae with insertion of a BAK cage. The control group consisted of 591 patients who had an open anterior arthrodesis with insertion of a BAK device by 42 surgeons at 19 medical centers. Forty percent of patients had a previous laminectomy. The patients who had a laparoscopic procedure at the fourth and fifth lumbar levels had a significantly longer operative time (p < .001), but significantly less blood loss (p = .023) and a significantly shorter hospital stay (p = .003). Six instances of iatrogenic disc herniation as a result of lateral placement of taps, reamers, and cages were noted, leading to nerve root compression in the laparoscopic group. Regan stated that the learning curve was steep and became as safe and effective as the open procedure after the first five to ten procedures55,56 (Figure 41.7). McAfee44 reported one instance of emergent conversion from a laparoscopic to open procedure for a laceration of left common iliac vein (in 100 patients). Other studies45,47 indicate that the most important operative devices in a laparoscopy are the retractors and vascular clamps needed to convert the procedure to a standard open laparotomy. In an effort to avoid vascular complications, some authors49,55 recommend performing the operation in the lateral decubitus position and that the cage be inserted in a transverse direction. This approach can be used from the first to the fifth lumbar levels, but should be avoided at the first sacral level. McAfee et al.49 performed arthrodesis at the first through fifth lumbar levels with the use of a minimally invasive anterior retroperitoneal approach and insertion of a BAK device in a transverse direction. The average hospital stay was 2.9 days. No patient had migration or pseudarthrosis at 24.3 months (range 12 to 40 months) postoperatively. With this technique of lateral insertion, mobilization of the vessels is usually not necessary. Dorsal instrumentation was not used.
Fusion Cages and Osteoinductive Materials Cunningham et al.18 performed an endoscopic study in the thoracic spine in sheep to compare the use of a BAK cage filled with osteogenic protein-1 with a BAK cage filled with autogenous iliac-crest bone graft. Solid fusion was defined as bridging of the trabecular network from end plate to end plate, as seen in the cage on midsagittal microradiographs. The authors evaluated the biomechanic, computed tomography (CT), microradiography, and quantitative histomorphometric measurements.
496
Section 3 ■ Surgical Procedures
Figure 41.7 A 45-year-old man 2 years status post previous anterior cage insertion presents with
worsening back pain and bilateral lower extremity paresthesias that have failed nonoperative intervention. (A) Preoperative anteroposterior radiograph (status post L4-5 cage insertion). (B) Preoperative lateral radiograph. (C) Sagittal MRI with thecal sac displacement. (D) Anteroposterior radiograph following cage removal and femoral allograft insertion with pedicle screw fixation. (E) Lateral postoperative radiograph.
Chapter 41 ■ Lumbar Interbody Cages
Multilevel thoracic decompression was performed in 12 sheep through a thoracoscopic approach. Three noncontiguous sites (the fifth and sixth, seventh and eighth, and ninth and tenth thoracic vertebral levels) were randomly treated with one of five protocols. Group 1 (six specimens) was treated with destabilization only; Group 2, with an empty BAK cage; Group 3, autogenous graft only; Group 4, BAK device with autogenous graft; Group 5, BAK device with osteogenic protein-1. Four months after the operation, groups 3, 4, and 5 had significantly higher levels of segmental stiffness (p < .05) than did the control groups as demonstrated with three of the five testing modalities. Significantly more trabecular bone was present in the treatment groups (p < .05). The authors concluded that the arthrodesis with use of osteogenic protein-1 was biomechanically and histomorphometrically equivalent to that of arthrodesis with use of autogenous iliac-crest graft. Boden6 described his findings in five adult rhesus monkeys that had laparoscopic exposure of the lumbrosacral spine, followed by insertion of a hollow titanium cylindrical cage filled with bone morphogenic protein-2. All five monkeys had solid fusion as determined by manual palpation. CT scans demonstrated fusion with trabecular bone growth through the cage.
Osteoinductive Growth Factors Urist61 was the first to discover the osteoinductive capacity of bone. Demineralized bone matrix, when placed in the muscle pouches of rats, induced ingrowth of connective tissue cells and differentiation of cartilage and bone. Studies and advances in protein isolation technology subsequently yielded evidence of a series of soluble, low molecular weight glycoproteins, identified as BMPs. Because BMPs make up 1% of all bone protein, large volumes of allograft are required to extract small amounts of osteoinductive factors. In the 1980s Wosney65 identified several BMP molecules, of which all but one belong to the TGF-B family of proteins. Several recombinant BMP molecules have been produced as singular molecular species in virtually unlimited quantities using genetically modified cell lines. In 1990 Wang et al.62 described osteoconductivity of rhBMP-2 in a rat ectopic assay system. This activity was also shown for rhBMP-4 and for rhBMP-7 (OP-1). Since 1993 the efficacy of rhBMP-2 has been shown in spinal fusion models in rabbits, dogs, sheep, goats, and rhesus monkeys.5,32 The use of rhBMP-2 in cylindric fusion devices was first studied in animal models. Sandu57 in 1996 reported that successful anterior lumbar fusion had occurred in sheep after implantation of cylindric titanium cages filled with rhBMP-2 soaked collagen sponges. Although only 33% of cages filled with autograft were associated with fusion in this model, all animals with rhBMP-2 filled cages had successful fusion and excellent bony consolidation. In another study cylindric freeze-dried allograft bone dowels were filled with either rhBMP-2 or autogenous bone graft from the iliac crest and implanted into rhesus monkeys.32 At 3 months there was radiographic evidence of fusion in the sites that were implanted with rhBMP-2. By 6 months, there was both radiographic and histologic evidence of solid fusion in all sites implanted with rhBMP-2.
497
The first clinical demonstration of clinical efficacy of rhBMP-2 for spinal fusion was a multicenter trial involving 14 patients.7 All patients had single-level lumbar disease. They were randomized to either tapered Ti fusion cages that contained an rhBMP-2 (1.5mg/ml) soaked bovine collagen sponge or morselized autogenous bone graft from the iliac crest. Of the patients who received rhBMP-2, 10 of 11 achieved fusion by 3 months, and all had fusion by 6 months. Only one of the three controls achieved fusion by 12 months. At this time data-exploring alternative osteoinductive materials are promising but sparse. Although the optimal graft substitute and delivery system for lumbar intervertebral fusion has not been established, the preliminary data from trials incorporating rhBMP-2 has been favorable. Whether to use threaded interbody implants, or traditional femoral ring allografts will depend on further clinical research and outcome analysis.
Summary The key question to answer when considering the effectiveness of cages is to determine whether fusion has taken place. The use of titanium and tantalum cages makes routine assessment of fusion even more difficult because of artifacts on CT and MRI scans.40 The main criteria for determining successful fusion are different for each of the different types of fusion devices. The Brantigan11 cage is radiolucent and allows direct visualization of graft within the cage. The success of the Ray54 cage was evaluated with
Figure 41.8 Lumbar fusion with pedicle screw fixation and
Harm's cage with anterior instrumentation.
498
Section 3 ■ Surgical Procedures
the use of radiographic overlay method of Hutter.34 When superimposed flexion and extensions views were compared, fusion was present when no motion was evident. The BAK device was evaluated on the basis of decreased motion on lateral flexion and extension radiographs, with the use of 3 degrees39 or 5 degrees2 as the essential value (Figure 41.8). Questions relating to the success of an arthrodesis cannot be adequately addressed unless standard criteria are used. Assessment of fusion should be based on increased density of the graft within the disc space, bridging trabecular bone, and continuous intervertebral cortical bridges from the dorsal border of one vertebral body joining the cortex of the caudad vertebral body. Researchers should use conventional criteria for determining fusion, such as bridging trabecular bone in continuity with two adjacent end-plates. Investigators performing prospective, randomized trials should use the same radiographic criteria for both the control group as for the cage group. If CT images do not demonstrate trabecular bone, then the fusion mass is indeterminate. A lack of motion of flexion and extension lateral radiographs should not be equated with successful fusion. The determination of duration and follow-up necessary to adequately identify a fusion mass is, to date, unknown. The rates of fusion after interbody fusion have improved, from 66%58 to 2-year rates of 91%39 with the use of the BAK titanium device, and 96% with the use of the Ray titanium device.54 However, one must keep in mind that longer clinical and radiographic follow-up is necessary before the actual effectiveness (period free of revision) and true fusion rates are known.
9.
10.
11.
12.
13.
14.
15.
16.
17.
REFERENCES 1. Agazzi S, Reverdin A, May D: Posterior lumbar interbody fusion with cages: an independent review of 71 cases. J Neurosurg 91:186-192, 1999. 2. Alpert S: Summary of effectiveness and safety—BAK interbody fusion system—PMA P950002, PMA Document Mail Center (HFZ-401), Center for Disease and Radiological Health. Washington D.C., Food and Drug Administration, Sept 20, 1996. 3. Bagby GW: Arthrodesis by the distraction-compression method using a stainless steel implant. Orthopedics 11: 931-934, 1988. 4. Blumenthal SL, Gill K: Can lumbar spine radiographs accurately determine fusion in postoperative patients? Correlation of routine radiographs with a second surgical look at lumbar fusions. Spine 18:1186-1189, 1993. 5. Boden SD, Horton WC, Martin G: Laparoscopic anterior spinal arthrodesis with rhBMP-2 in a titanium interbody threaded cage. J Spinal Disord 11:95-101, 1998. 6. Boden SD, Schimandle JH, Hutton WC: The use of osteoinductive growth factor for lumbar spine fusion. Spine 20:2633-2644, 1995. 7. Boden SD, Zdeblick TA, Sandu HS, Heim SE: The use of BMP-2 in interbody fusion cages: Definitive evidence of osteoinduction in humans. Trans Am Acad Orthop Surg, 1999. 8. Brantigan J, Steffee A, Lewis M, et al: Lumbar interbody fusion using Brantigan I/F cage for posterior lumbar
18.
19.
20.
21.
22.
23.
24.
interbody fusion and the variable pedicle screw placement system. Spine 25:1437-1446, 2000. Brantigan JW: Pseudoarthrosis rate after allograft posterior lumbar interbody fusion with pedicle screw and plate fixation. Spine 19:1271-1280, 1994. Brantigan JW, McAfee PC, Cunningham BW, et al: Interbody lumbar fusion using a carbon fiber cage implant versus allograft bone. An investigational study in the Spanish goat. Spine 19:1436-1444, 1994. Brantigan JW, Steffee AD, Lewis ML, et al: Lumbar interbody fusion using the Brantigan I/F cage for PLIF and the VSP pedicle screw system: two-year results of a Food and Drug Administration IDE clinical trial. In Husson JL, Le Heuc JC (eds): Intersomatique du Rachis Lumbaire. Montpelier, France, Sauramps Medical, 1996. Brodsky AE, Kovalsky ES, Khalil MA: Correlation of radiologic assessment of lumbar spine fusions with surgical exploration. Spine 16:261-265, 1991. Chen D, Fay LA, Lok J, et al: Increasing neuroforaminal volume by anterior interbody distraction in degenerative lumbar spine. Spine 20:74-79, 1995. Chitnavis B, Barbagallo G, Selway R: Posterior lumbar interbody fusion for revision disc surgery: review of 50 cases in which carbon fiber cages were implanted. J Neurosurg 95:190-195, 2001. Cloward RB: The treatment of ruptured lumbar intervertebral discs by vertebral body fusion. I. Indications, operative technique, after care. J Neurosurg 10:154-168, 1953. Crock HV: Observations on the management of failed spinal operations. J Bone Joint Surg 58B:193-199, 1976. Cunningham BW, Haggerty CJ, McAfee PC: A quantitative densitometric study investigating the stress-shielding effects of interbody spinal fusion devices—emphasis on long term fusions in thoroughbred racehorses. Read at the Annual Meeting of the Orthopedic Research Society, New Orleans, March 9, 1998. Cunningham BW, Kanayama M, Parker LM, et al: Osteogenic protein (rhOP-1) versus autologous fusion in sheep thoracic spine. Spine 24:509-518, 1999. DeBowes RM, Gant BD, Bagby GW, et al: Cervical vertebral interbody fusion in the horse: a comparative study of bovine xenografts and autografts supported by stainless steel baskets. Am J Veter Res 45:191-199, 1984. Diedrich O, Perlick L, Schmitt O, Kraft C: Radiographic characteristics on conventional radiographs after posterior lumbar interbody fusion: comparative study between radiolucent and radiopaque cages. J Spinal Disord 14: 522-532, 2001. Dimar J, Beck D, Glassman S, et al: Posterior lumbar interbody cages do not augment segmental biomechanical stability. Am J Orthop 636-639, August 2001. Eck K, Bridwell K, Ungacta F, et al: Analysis of titanium mesh cages in adults with minimum of two-year follow-up. Spine 18:2407-2415, 2000. Elias W, Simmons N, Kaptain G, et al: Complications of posterior lumbar interbody fusion when using a titanium threaded cage device. J Neurosurg 93:45-52, 2000. Fraser RD: Interbody, posterior, and combined lumbar fusions. Spine 20:167-177, 1995.
Chapter 41 ■ Lumbar Interbody Cages
25. Frymoyer JW, Hanley EN, Howe J, Kuhlman D: A comparison of radiographic findings in fusion and non fusion patients ten or more years following lumbar disc surgery. Spine 4:435-440, 1979. 26. Glassman S, Johnson JB, Raque G, et al: Management of iatrogenic spinal stenosis complicating placement of a fusion cage. A case report. Spine 21:2383-2386, 1996. 27. Glazer PA, Colliou O, Lotz JC, Bradford DS: Biomechanical analysis of lumbrosacral fixation. Spine 21:1211-1222, 1996. 28. Goldstein J, Macenski M, Griffith S, McAfee P: Lumbar sagittal alignment after fusion with a threaded interbody cage. Spine 26:1137-1142, 2001. 29. Grubb SA, Lipscomb HJ: Results of lumbrosacral fusion for degenerative disc disease with and without instrumentation. Two to five year follow-up. Spine 17: 349-355, 1992. 30. Hacker RJ: Comparison of interbody fusion approaches for disabling back pain. Spine 22:660-666, 1997. 31. Harms J: Screw-threaded rod system in spinal fusion surgery. Spine 6:541-575, 1992. 32. Hecht BP, Fischgrund JS, Herkowitz HN: The use of recombinant human bone morphogenetic protein 2 (rhBMP-2) to promote spinal fusion in a non-human primate anterior interbody fusion model. Spine 24:629-636, 1999. 33. Holte DC, O’Brien JP, Renton P: Anterior lumbar fusion using a hybrid interbody graft. European Spine J 3:32-38, 1994. 34. Hutter CG: Posterior intervertebral body fusion. A 25-year study. Clin Orthop 179:86-96, 1983. 35. Kanayama M, Haggerty CJ, Cunningham BW, et al: The biomechanical stability and stress-shielding of lumbar interbody fusion implants: an in-vitro comparative study. Read at the Annual Meeting of the Orthopedic Research Society, New Orleans, March 17, 1998. 36. Klemme W, Owens B, Dhawan A, et al: Lumbar sagittal contour after posterior interbody fusion. Spine 26:534-537, 2001. 37. Kozak JA, Heilman AE, O’Brien JP: Anterior lumbar fusion options. Technique and graft materials. Clin Orthop 300:45-51, 1994. 38. Kumar A, Kozak JA, Doherty BJ, Dickson JH: Interspace distraction and graft subsidence after anterior lumbar interbody fusion with femoral strut allograft. Spine 18:2393-2400, 1993. 39. Kuslich SD, Ulstrom CL, Griffith SL, et al: The Bagby and Kuslich method of lumbar interbody fusion. History, techniques, and 2-year follow-up results of a United States prospective, multicenter trial. Spine 23:1267-1279, 1998. 40. Larsen JM, Rimoldi RL, Capen DA: Assessment of pseudoarthrosis in pedicle screw fixation: a prospective study comparing pain radiographs, flexion/extension radiographs, CT scanning, and bone scintigraphy with operative findings. J Spinal Disord 9:117-120, 1996. 41. Lehman TR, Spratt KF, Tozzi JE: Long-term follow-up of lower lumbar fusion patients. Spine 12:97-104, 1987. 42. Lin PM: Posterior lumbar interbody fusion technique: complications and pitfalls. Clin Orthop 193:90-102, 1985. 43. Lund T, Oxland R, Jost B, et al: Interbody cage stabilization in the lumbar spine. J Bone Joint Surg Br 80:351-359, 1998. 44. McAfee P: Current concepts review: Interbody fusion cages in reconstructive operations on the spine. J Bone Joint Surg Am 81:859-879, 1999.
499
45. McAfee PC: Complications of anterior approaches to the thoracolumbar spine. Emphasis on Kaneda Instrumentation. Clin Orthop 306:110-119, 1994. 46. McAfee PC: Interbody fusion cages in reconstructive operations in the spine. J Bone Joint Surg Am 81:859-889, 1999. 47. McAfee PC, Farey ID, Sutterlin CE: Device-related osteoporosis with spinal instrumentation. Spine 14:919-926, 1989. 48. McAfee P, Lee G, Fedder I, Cunningham B: Anterior BAK instrumentation and fusion: Complete versus partial discectomy. Clin Ortho Rel Research 394:55-63, 2002. 49. McAfee PC, Regan JJ, Geis WP, Fedder IL: Minimally invasive anterior retroperitoneal approach to the lumbar spine. Spine 23:1476-1484, 1998. 50. Otero Vich JM: Anterior cervical interbody fusion with threaded cylindrical bone. J Neurosurg 63:750-753, 1985. 51. Penta M, Sandhu A: Magnetic resonance imaging assessment of disc degeneration 10 years after anterior lumbar interbody fusion. Spine 20:743-747, 1995. 52. Prolo DJ, Oklund SA, Butcher M: Toward uniformity in evaluating results of lumbar spine operations. A paradigm applied to posterior lumbar interbody fusions. Spine 11:601-606, 1986. 53. Ray CD: Threaded fusion cages for lumbar interbody fusions. An economic comparison with 360 degrees fusion. Spine 22:681-685, 1997. 54. Ray CD: Threaded titanium cages for lumbar interbody fusion. Spine 22:667-680, 1997. 55. Regan JJ, McAfee PC, Guyer RD, Aronoff RJ: Laparoscopic fusion of the lumbar spine in a multicenter series of the first 34 consecutive patients. Surg Laparscop Endosc 6:459-468, 1996. 56. Regan JJ, McAfee PC, Mack MJ: Atlas of Endoscopic Spine Surgery. St. Louis, Quality Medical, 1995. 57. Sandu HS, Kabo JM, Turner AS: Augmentation of titanium fusion cages for experimental anterior lumbar fusion. Orthopedic Transactions 21:92, 1997. 58. Stauffer RN, Coventry MB: Anterior interbody lumbar spine fusion. Analysis of Mayo Clinic Series. J Bone Joint Surg 54A:756-768, 1972. 59. Steffee AD, Sitkowski DJ: Posterior lumbar interbody fusion and plates. Clin Orthop 227:99-102, 1988. 60. Togawa D, Bauer T, Brantigan J, Lowery G: Bone graft incorporation in radiographically successful human intervertebral body fusion cages. Spine 26:2744-2750, 2001. 61. Urist MR: Bone formation by autoinduction. Science 150:893-899, 1965. 62. Wang EA, Rosen V, D'Alessandro JS: Recombinant human bone morphogenetic protein induces bone formation. Proc Natl Acad Sci USA 87:2220-2224, 1990. 63. Weiner BK, Fraser RD: Spine update: lumbar interbody cages. Spine 23:634-640, 1998. 64. Wetzel FT, LaRocca H: The failed posterior lumbar interbody fusion. Spine 16:839-845, 1991. 65. Wosney JM: The bone morphogenetic family and osteogenesis. Mol Reprod Dev 32:160-167, 1992. 66. Yuan H, Kuslich SD, Dowdle JA, et al: Prospective multicenter clinical trial of the BAK interbody fusion system. Read at the Annual Meeting of the NASS, New York, Oct. 22, 1997.
C H A P T E R
4 2
Bone Anatomy and Physiology Bone is specialized connective tissue composed of a calcified matrix that contains osteocytes, osteoblasts, and osteoclasts. Osteoblasts synthesize bone matrix, which is composed of hydroxyapatite crystals, organic collagen fibrils, and an amorphous ground substance. The ground substance, composed of glycosaminoglycans, stimulates the calcification of the matrix. Once osteoblasts are surrounded by a completely calcified matrix, they become osteocytes. The secondary structure of bone matures as it is exposed to compressive or tensile forces. Primary bone appears initially in embryonic development, but also occurs in maturing fracture sites. Woven bone is a type of primary bone found at fracture sites during rapid osseous formation and is characterized by a random deposition of fine collagen fibrils that contain a dense population of osteocytes. Because of the disorderly arrangement of these structures, this construct is relatively weak. Primary bone, however, reorganizes to gain a secondary structure more capable of bearing force. Cortical and cancellous bone are both forms of secondary bone and are found in the mature human skeleton. The osteon is the primary structural unit of cortical bone and is characterized by a concentric array of parallel collagen fibers situated around a central or Haversian canal that brings nutrients to compact bone. The Haversian canals communicate with the marrow cavity, the periosteum, and with each other via transverse or oblique Volkmann canals. Cancellous or spongy bone is located deep in cortical bone and contains marrow elements. The beams and struts of cortical bone are covered by a fine endosteum, which is a source of osteoprogenitor cells. Intersecting osseous bars called trabeculae are present in cancellous bone. There are both horizontal and vertical components that are oriented in response to stress and lend strength and support to the bone.58
Dorsal and Lateral Thoracic and Lumbar Fusion George J. Kaptain, Christopher I. Shaffrey, Andrea L. Halliday, Michael Schneier, and John A. Jane
Historic Perspective In 1911 Albee3 and Hibbs40 independently described two methods of arthrodesis to treat patients with thoracolumbar deformities secondary to Pott’s disease. With both techniques greenstick fractures of the spinous process were used to expose fresh cortical surfaces to facilitate fusion. Albee augmented the fusion with tibial autograft. Realizing the potential applications offered by spinal fusion, researchers applied the technique to patients with scoliosis,41 spondylolisthesis,43 and fracture dislocations. MacKenzie-Forbes53 and Hibbs41 described techniques of laminar decortication that extended the fusion bed laterally. In 1924 Hibbs denuded the cartilaginous articulation of the facet joints to enhance fusion in his series of scoliosis patients.41 Howorth,43 McBride,60 and Moe62 further modified facet joint preparations. Hall and Goldstein devised modifications of dorsal element decortication in the treatment of scoliosis.8,31 In 1936 Mathieu and Demirleau57 introduced transverse process fusion as an alternative to dorsal fusions. Adkins2 applied this technique to spondylolisthesis to improve on the results of interlaminar fusions. These techniques were especially valuable in the lumbar spine in cases in which laminectomy precluded dorsal fusion. Watkins86 devised a paramedian approach to the dorsal aspect of the lumbar spine by exposing the transverse processes lateral to the paraspinous musculature. This technique retained the advantage of permitting access to the lateral aspect of the spine without disturbing the spinous process and interspinous ligament. Wiltse and colleagues87,88 were able to gain access to more medial structures through a paramedian approach, splitting the fascial plane between the multifidus and longissimus divisions of the paraspinous musculature, rather than dissecting lateral to the entire muscle group. Recently the application of internal fixation has improved the capability to correct spinal deformity and reduce the incidence of pseudarthrosis after thoracolumbar fusion procedures.33,74,89,94 These devices temporarily immobilize the spine and permit the synthesis of an osseous rather than a fibrous union.
Biology of Bone Grafting Unlike other tissue, bone heals by regeneration with complete restoration of tissue integrity. The components and organization essential to fracture healing are also central to bone synthesis incurred by surgical fusion procedures. A rich vascular network delivers oxygen to the graft site, as well as to a population of inflammatory cells that accumulates in the wound hematoma within hours after surgery.75 Macrophages congregate in the graft bed and remove necrotic debris while secreting growth factors that stimulate the migration of other inflammatory cells.70 Platelets secrete platelet-derived growth factor and other substances that are chemotactic for fibroblasts and other mesenchymal cells necessary for fracture healing.73 Inflammatory cells and platelets that congregate at the surgical or fracture site recruit a population of primitive mesenchymal cells that are stimulated first to proliferate and then to transform to an osseous or chondroid lineage.19 These mesenchymal cells are found in the bone marrow, the endosteum, the cambial or inner layer of the periosteum, and the perivascular spaces.67 Transformation of this pool of undifferentiated cells is induced by bone 500
Chapter 42 ■ Dorsal and Lateral Thoracic and Lumbar Fusion
morphogenic protein (BMP), the final factor essential for bone synthesis.69,90,91 This regenerative process may be divided into three stages: (1) inflammation, (2) repair, and (3) remodeling.75 In the first phase of healing the inflammatory signals summon cells that either participate in the deposition of bone or remodel and ingest necrotic debris. During repair, macrophages remove blood clots, damaged bone matrix, and injured cells. Osteoclasts at the fracture site resorb dead bone and remove mineralized matrix while new blood vessels grow into the graft or fracture site. Cartilaginous matrix or osteoid is deposited by pre-existing osteoblasts, chondroblasts, or cells derived from converted mesenchymal cells. Primary and woven bone are then formed as the osteoid or cartilaginous callus is mineralized. If ossification is incomplete, a fibrous nonunion or pseudarthrosis occurs. As mechanical stress is placed on the construct, remodeling restores the bone to its original shape, structure, and mechanical strength.48 Although both fracture healing and surgical arthrodesis use identical cellular and molecular cascades to effect regeneration, the central challenge of surgical arthrodesis is to incorporate transplanted graft to create a modified osseous structure. Optimum surgical fusion relies not only on the properties of the fusion bed but also on the nature of the graft. Graft material may manifest osteogenic, osteoinductive, or osteoconductive properties together or in isolation. Osteogenesis refers to the ability of the bone graft to transmit viable osteoblasts to the fusion site. Osteogenic grafts are favorable because they supply a population of osteoprogenitor cells independent of the nature of the fusion site. Vascularized autografts contain the highest number of viable osteoprogenitor cells. Autogenous cancellous material also contains large numbers of cells, but few survive transplantation.49 Osteoinduction induces undifferentiated mesenchymal cells to divide and then transform into osseous or chondroid phenotypes.69 BMP, which has been defined and purified by recombinant techniques, is found within the bone matrix and induces the transformation of the undifferentiated cells into those required for fracture healing or arthrodesis.90,91 Osteoconductive grafts supply a scaffolding to support the ingrowth of sprouting capillaries and osteoprogenitor cells from the recipient host bed into the structure of an implant or graft.66 This property is characteristic of autograft, allograft, and bone matrix substitutes. Bone grafts are incorporated by a process known as “creeping substitution” mediated by osteoclasts, in which the resorbed bone graft is replaced with new bone.18 As osteoclasts resorb grafted material, blood vessels invade the implant. The net result is an increase in porosity of the bone graft with a weakening of its initial mechanical strength. As the bulk of the graft is replaced by regenerated bone, the density and strength of the construct increase.66
Graft Selection for Dorsal and Lateral Thoracolumbar Fusions Successful arthrodesis depends on the biologic vitality of both the implant and the fusion bed. An ideal graft contains independently functional vascular channels, a supply
501
of viable osteoprogenitor cells, and a source of BMPs; it is also antigenically compatible to the host tissue.79 An adequately prepared fusion site should offer these elements and, to maximize the area of osseous contact, should also be free of soft-tissue or necrotic debris. Stabilization of the fusion bed also promotes graft incorporation.71 Vascularized autologous bone is the most biologically active graft. Bone implanted with a vascular pedicle is more rapidly incorporated than is nonvascularized material, although the use of this material is rarely practical for most thoracolumbar fusions.72 Cortical and cancellous autograft and allograft are used for the vast majority of thoracolumbar fusions. Cancellous graft differs from cortical bone grafts in the rate and completeness of repair. Autogenous cancellous bone is a better graft material than cortical bone for thoracolumbar fusion applications, because it is more rapidly revascularized and is more osteoinductive.1,80,91 The initial mechanical strength of cortical bone as a grafting material is useful in procedures that require graft rigidity, such as ventral thoracolumbar fusions. Cortical bone grafts, however, become progressively weaker as graft material is resorbed during the incorporation process. Allografts also have osteoinductive and osteoconductive properties, but they are inferior to autogenous substances because they incur immunologic responses from the host.37,72 Allografts are useful in patients with a limited bone supply who require a large quantity of graft material.16 Although autogenous materials are more biologically active than allografts, their procurement can result in a significant number of complications.47,50,81,92 Despite the apparent biologic disparity of these two substances, equivalent clinical outcomes and fusion rates have been reported in a series that compares the use of allograft and autograft in certain patient populations undergoing dorsal thoracolumbar fusions. In a series of pediatric patients with scoliosis, Montgomery et al.63 treated 18 patients with autografts and 12 with allografts. Solid fusion was identified in 94.5% of the autograft group and in 100% of the allograft group. A compilation of the results from this and several other series indicates that similar fusion rates can be achieved through the use of allografts, in conjunction with segmental instrumentation in young patients with thoracolumbar scoliosis.6,12,26,27 The efficacy of allograft in these procedures may be related to the restricted range of motion in the thoracic spine and a conductive environment for fusion in pediatric patients. In contrast, the results of autografting are far superior to those of allografting in adults undergoing lumbar intertransverse fusions. An et al.4 conducted a prospective study that evaluated the efficacy of autografts versus allografts in 20 patients who had undergone instrumented interlaminar and intertransverse fusions. In each patient they applied an autograft on one side and an allograft on the other. They found 80% and 35% fusion rates in the autografted and allografted sides, respectively. Other studies also demonstrate higher pseudarthrosis rates with the use of allograft, compared to autograft in intertransverse and interlaminar fusions of the lumbar spine.21,39,44,64 The greater degree of mobility in the lumbar spine may account for the failure of allograft in these cases.
502
Section 3 ■ Surgical Procedures
Indications Spinal fusion is used to maximize the weight-bearing capacity of the spine while limiting pathologic motion across joint complexes. The objectives of this procedure include the relief of pain, preservation of neurologic function, and maintenance of spinal alignment. Currently, dorsal and dorsolateral fusion techniques are used to manage spinal instability or deformity that result from congenital or idiopathic scoliosis, isthmic spondylolisthesis, spinal fractures, primary and metastatic tumors, and spinal infections, as well as an assortment of degenerative conditions. Absolute indications for fusion include conditions that compromise the structural integrity of the spine that could result in neural injury (Table 42.1). Spinal fusion has also been shown to arrest the progression of deformity in congenital and idiopathic scoliosis24 and in patients with spondylolisthesis. In his review of the treatment of spondylolisthesis Hanley concluded that fusion is indicated in patients who suffer from a documented progression of spondylolisthesis, in children who present with a symptomatic grade III or IV slip, or in those with a low-grade slip if symptoms persist despite conservative therapy.34 Degenerative abnormalities of the lumbar spine may result in symptoms from either compression of neural structures or from abnormal mechanical stress placed on the facet-intervertebral disc joint complex. Alterations in the mechanical properties of the spine are multifactorial but probably begin with desiccation of the intervertebral disc. Although disc desiccation occurs with aging, in certain persons this process is followed by bulging of the annulus fibrosus, hypertrophy of facet joints and the ligamentum flavum, and the eventual development of spinal stenosis. Lumbar stenosis is characterized by constriction of the vertebral canal, resulting in compression of neural elements, which causes neurogenic claudication, low back pain, radicular symptoms, or even incontinence (rarely). Although these spondylotic changes may restrict the mobility of the lumbar spine, most symptoms result from compression of the neural elements and are relieved by spinal decompression. The benefit of adding a fusion after decompressions is therefore controversial. Spondylotic changes associated with lumbar stenosis, on occasion, result in unstable conditions or deformities, such as degenerative spondylolisthesis or scoliosis; these changes may further restrict the area of the vertebral canal. The use of spinal arthrodesis for either the treatment of degenerative disc disease or lumbar disc herniation continues to be an area of debate, and no definitive benefit has
been proven.76,84 However, fusion has been shown to be effective in treating instability (spondylolisthesis) or deformity (scoliosis) resulting from degenerative disc disease. Lumbar stenosis in association with degenerative spondylolisthesis has been studied prospectively by Herkowitz and Kurz,38 who randomized 50 patients according to decompression alone versus decompression in conjunction with a noninstrumented fusion. They found good to excellent outcomes in 24 out of 25 patients treated with decompression and fusion, whereas only 11 out of 25 patients treated with decompression alone experienced good or excellent results. Criteria for instability may be evaluated radiographically by the presence of traction spurs55 or by abnormal angular or translational motion on lateral flexion-extension radiographs. Reported results of fusion for patients who present with localized back pain with these radiographic abnormalities has varied widely with success rates ranging from 30% to 90%.34 Intraoperative assessment of stability is another factor that influences the decision to augment lumbar decompression with fusion. Fox et al.28 demonstrated that patients who undergo multiple-level decompressions are more likely to suffer from postoperative instability and slip progression. The lateral extent of the decompression, which is essential for the relief of radicular symptoms, also impacts on postoperative stability. Lombardi et al.52 compared three groups of patients. The first group had total laminectomies with preservation of the facet joints. The second group had laminectomies, as well as bilateral facetectomies. The third group had laminectomies, partial facetectomies, and intertransverse fusions. Worse outcomes and a higher incidence of slip progression were noted in the group that received facetectomy without fusion.52 Internal fixation should also be considered for patients who are candidates for spine fusion. Before the advent of fixation systems, failure rates of noninstrumented fusions for scoliosis ranged from 20% to 40%, with a 35% incidence of loss in correction. Adding segmental fixation has decreased the failure rate to 10%.85 Instrumentation has also improved fusion rates and clinical outcomes in patients with degenerative disease of the lumbar spine.30,85,93 Zdeblick described a series of 124 patients who underwent lumbar or lumbosacral fusion. He noted a fusion rate of 65% in patients who were treated without instrumentation. In the remaining two patient groups semirigid instrumentation resulted in a 77% fusion rate, whereas rigid instrumentation resulted in successful arthrodesis in 95% of cases. Clinical outcomes paralleled the rate of successful fusion.94
TABLE 42.1
Indications for Spinal Fusion Recommended
Relative
Rare
Two-third column injury Trauma Tumor Infection Iatrogenic instability Isthmic instability
Degenerative spondylolisthesis Abnormal movement on dynamic films with appropriate pain or neurologic deficit
Routine discectomy Stable spinal stenosis
Modified from Sonntag VKH, Marciano FF: Spine 20:138S-142S, 1995, with permission.
Chapter 42 ■ Dorsal and Lateral Thoracic and Lumbar Fusion
Surgical Technique Positioning After the patient is intubated, compressive devices are placed on the lower extremities. Bladder catheterization should be considered for lengthy procedures or when significant blood loss is expected. All dorsal approaches to the lumbar spine are performed with the patient in the prone position, with limited hip flexion for lumbosacral fusion and moderate hip flexion for thoracic fusions. With procedures involving the lower thoracic and lumbar spine, the patient’s arms are abducted and the elbows are flexed. The upper extremities are adequately padded, particularly at the elbow to avoid ulnar nerve injury; the arm is not abducted beyond 80 degrees to avoid brachial plexus injury. With dissections involving the upper or midthoracic spine, or in patients with arthritis of the shoulder or elbow, the arms should be padded and fixed in the anatomic position. With concerns about arm position affecting the performance of radiographs, a preoperative study should be performed, and positioning should be adjusted before beginning surgery. Prevention of epidermal, appendicular, and ocular compressive injuries should be ensured during patient positioning by proper padding over all pressure points. Before transfer, all electrocardiogram leads and wires should be removed from the chest and abdomen to avoid skin breakdown. Chest rolls or frames are used so that the abdomen hangs freely to avoid excessive intraabdominal pressure. The anterior superior iliac spine (ASIS) and knees are padded, and a pillow is placed under the legs so that the toes hang freely. The dorsalis pedis pulse is palpated to guard against femoral artery compression. Urinary catheters should be free of tension, and if chest rolls are used, the breasts should be displaced medially. Finally, goggles should be placed over the eyes to prevent pressure injury to the globes.
503
Soft-Tissue Approaches Median Approach to the Thoracic and Lumbar Spine The median approach to the thoracolumbar spine permits bilateral exposure of the dorsal elements and vertebral canal through one fascial incision. The spinous processes are palpated, and a longitudinal incision is directed through the epidermis, dermis, and superficial fascia to expose the deep thoracolumbar fascia. Self-retaining retractors are then placed, and the fascia is incised with either a monopolar cautery or a scalpel. Once the tips of the spinous processes are visualized, a periosteum or Cobb elevator is used to maintain lateral traction on the paraspinous musculature to facilitate visualization of muscular insertions into the spinous process and medial lamina. Optimal hemostasis is maintained only if these structures are sharply severed directly at their insertion, using either a second Cobb elevator, electrocautery, or curved Mayo scissors. The detachment of these structures facilitates a subperiosteal dissection of the soft tissue from the spinous process, lamina, and the medial portion of the facet joint. The structure of the thoracic vertebrae affords an exposure of the transverse process at this point. Gauze sponges are inserted into the dissection plane, using periosteum elevators to improve exposure and hemostasis. After this is completed, self-retaining retractors can be advanced deep into the superficial fascia. The insertions of the rotares muscle group can be separated at the point at which they insert into the caudal tip of each lamina. The transverse process of the lumbar vertebrae is located in a deeper sagittal plane relative to the facet joint and pars interarticularis. The terminal branches of the segmental lumbar artery are located rostral, caudal, and lateral to the facet joints (Figures 42.1 and 42.2). Identification and coagulation of these vessels during the course of dissection will reduce bleeding. Arterial bleeding from these
Figure 42.1 Lateral view of the lumbar segmental artery as branches course proximal to the
neural foramina before becoming apparent dorsal to the intertransverse ligament.
504
Section 3 ■ Surgical Procedures
Figure 42.2 Dorsal oblique view of the course of the lumbar segmental artery as it pierces the
intertransverse ligament.
vessels should be controlled with bipolar electrocautery to avoid transmission of thermal damage to nerve roots exiting the nearby neural foramina. Fracture of the transverse process should be avoided while performing a periosteal dissection lateral to the zygapophyseal joints. Exposure of the ventral face of the transverse process is hazardous because this area houses both the anterior transverse artery and the exiting nerve root. Advancement of self-retaining retractors permits visualization of the dorsal aspect of the dorsal elements. Care should be taken to periodically release self-retaining retractors so soft-tissue trauma is reduced. Exposure of the L5 transverse process and the ala of the sacrum is complicated by multifidus and sacrospinalis muscle origins that are firmly attached at the sacral ala and dorsal sacrum. The lateral aspect of the superior facet of the sacrum and sacral ala should be periosteally exposed and decorticated to the depth of the L5 transverse process for lumbosacral fusions. Large, crank-type retractors that can produce excessive force on paraspinal musculature should be avoided when attempting to expose transverse processes. Hand-held retractors provide sufficient exposure of the facet joints and transverse processes to permit decortication and pedicle screw placement, while limiting the possibility of muscle necrosis. Paramedian Approach to the Lumbar Spine A median incision is made, as described previously, and the thoracolumbar fascia is widely exposed.88 A longitudinal incision is made in the deep fascia (bilaterally, if necessary) two finger widths lateral to the spinous process. The fascial plane that separates the multifidus from the longissimus muscles is defined, and a finger is inserted into this avascular plane until a facet joint is palpated. Self-retaining retractors are then placed deep into the lumbodorsal fascia, and a subperiosteal dissection of the musculature is performed over the transverse process, facet joint, and lateral lamina (if indicated). Knowledge of the anatomy of the terminal branches of the segmental artery is necessary to control arterial bleed-
ing. Exposure at the L5-S1 level may be complicated by the laterally projecting fibers of the multifidus as they sweep toward their sacral insertion. These muscles should be detached subperiosteally from their attachments on the sacral ala to permit decortication for lumbosacral fusion. Bone Dissection Interlaminar and Intertransverse Fusion Any remaining soft tissue is removed, and the dorsal aspect of the lamina or transverse process is decorticated with rongeurs, gouges, or osteotomes; the bone chips are set aside to be used with other graft materials. Decortication of the lateral aspect of the superior facet, the region of the mamillary and accessory processes, and the transverse processes should proceed medially to laterally. This decortication should not extend further than the dorsal surface of the transverse process. The intertransversarii muscles and the intertransverse ligament should not be detached to avoid damage to the contents of the neural foramen. In lumbosacral fusions it is essential to decorticate the superior articular facet of the sacrum, as well as the ala of the sacrum to prevent pseudarthrosis. To avoid postoperative degenerative change or spondylolisthesis, the facet joint immediately above or below the fusion mass must not be violated. Onlay graft material is then placed over the decorticated area (Figures 42.3 and 42.4). Facet Joint Preparation In the thoracic spine en bloc removal of the superior facet can be accomplished with a straight osteotome, using Hall’s method (Figure 42.5).8 An initial cut is made at the level of the caudal border of the transverse process and is extended medially and caudally. The cartilage and synovium of the joint are removed to expose the underlying cancellous bone. Graft is inserted into this space to facilitate fusion. Moe’s method of facet stripping uses a curette or gouge to first reflect the osteocartilaginous rind of the superior facet joint.62 This flap is left connected to the caudal border of
Chapter 42 ■ Dorsal and Lateral Thoracic and Lumbar Fusion
505
Figure 42.3 Dorsal view of an interlaminar fusion demonstrating
the decortication of the lamina and facet joint surfaces with the placement of bone graft contralaterally. Note that the facet joint of the rostral border of the fusion site should not be disturbed to guard against postoperative instability.
Figure 42.5 Dorsal view demonstrating a method of facet joint
preparation in the thoracic spine.
the rostral transverse process and is reflected caudally. The same procedure is used to denude the caudal facet of the motion segment. This flap is reflected rostrally and left attached to the rostral margin of the transverse process. Soft tissue or cartilage is removed from the fusion site and grafting material is added to the site. In the lumbar spine osteocartilaginous plates may be removed using osteotomes. Initial cuts should be oriented parallel to the axis of the joint space. Osteocartilaginous plates of bone are removed from the superior and inferior joint surface. These pieces are morcellized or cut into strips and may be used as graft material. The importance of complete soft-tissue removal in the areas lateral to the facet joint and overlying transverse process during intertransverse fusion cannot be overemphasized.
Complications and Avoidance Figure 42.4 Dorsal view of a dorsolateral fusion after total
laminectomy. Fusion bed involves the dorsal aspect of the transverse process, the facet joint, and the pars interarticularis.
Positioning Hemorrhage Intraoperative venous hemorrhage can be greatly reduced if the abdomen is properly placed during positioning.
506
Section 3 ■ Surgical Procedures
Hemodynamic pressure in the epidural and paravertebral venous plexus is greatly increased by the functional obstruction of the caval system through an increase in intraabdominal pressure.7 Venous hypertension is reduced if the abdomen hangs freely during the procedure. The knee-chest, kneeling, and prone positions on a spinal frame maintain abdominal decompression while permitting access to the dorsal aspect of the spine. Loss of Lumbar Lordosis Loss of lumbar lordosis (“flat back” syndrome) has been associated with pain and gait abnormalities in patients who have undergone thoracolumbar fusion for scoliosis.36 Tan et al.82 compared the degree of lumbar lordosis in patients placed in various positions for dorsal thoracolumbar surgery and found that the closest approximation to normal lumbar lordosis is obtained by placing the patient in the prone position with extended lower extremities. Compression and Traction Injuries Neural, ocular, and epidermal structures may be damaged by compression and traction during positioning. These injuries are minimized by proper padding of pressure points, such as the knees and the region of the lateral femoral cutaneous nerves, and by avoiding stretching of the brachial plexus by positioning the arms at an 80-degree angle or less to the body. Surgical Pseudarthrosis The success of fusion is determined clinically by amelioration of symptoms and, anatomically, by evidence of bridging trabecular bone on radiographs. Pseudarthrosis after spinal fusion is defined by the incomplete development of a rigid osseous construct, thus permitting motion across a joint complex. The diagnosis of pseudarthrosis is determined with plain radiographs, flexion and extension views, plain tomography, and computed tomography (CT). These images may demonstrate the absence of osseous material at the graft site or movement at the joint with flexion-extension images. These studies are not as sensitive or as specific for determining the status of a fused segment when compared with intraoperative observations during a “second look” surgery.9,14,45 The difficulty in radiographically characterizing a fusion mass is further complicated in the thoracic spine in which motion is limited, and in instrumented fusions in which artifact obscures imaging. Failure to achieve a solid fusion has been associated with increased morbidity in a variety of pathologic conditions. Pseudarthrosis is known to allow the progression of deformity in scoliosis,24,32,59,62,63 as well as a recurrence of symptoms in spondylolisthesis.5,35 The significance of pseudarthrosis as a complication of surgery, however, is not always straightforward. DePalma and Rothman22 and Frymoyer et al.29 have all demonstrated that solid arthrodesis does not correlate with patient satisfaction, alleviation of pain, or disability. The treatment of pseudarthrosis depends on the recognition of its varied clinical presentation. Pseudarthrosis after spinal surgery for scoliosis or spondylolisthesis may
present with a progression of deformity, radiologic instability, pain, myelopathy, or radiculopathy. Reexploration may be required to establish a definitive diagnosis and repair of pseudarthrosis. The diagnosis and treatment of pseudarthrosis in patients with persistent low back pain or neurologic symptoms after lumbosacral fusion is a more difficult issue. The “failed back” syndrome includes individuals with persistent symptoms after surgical treatment of a variety of conditions, such as lumbar stenosis, lumbar disc herniation, spondylolisthesis, or degenerative disc disease. In the absence of infection, hardware failure, or new neurologic deficit, it is essential to avoid reexploration until a sufficient interval has elapsed to allow for maturation of the fusion mass. At the conclusion of this interval radiologic evaluation may or may not demonstrate a pseudarthrosis. Other sources of persistent pain or neurologic deficit include disc herniation, painful degenerative changes at adjacent levels, and residual spinal or foraminal stenosis. Magnetic resonance imaging (MRI), CT, and myelography are helpful to delineate the presence of these conditions. Reexploration of the surgical site should proceed only in patients who do not respond to conservative therapy. Treatment of nonunion should begin with an identification of factors that inhibit bone healing. The distribution of forces within the graft site influences the effectiveness of healing. In patients with scoliosis, pseudarthrosis varies with the degree of correction, as well as with the expanse of curvature.59 Pseudarthrosis is also more commonly a problem in patients with spondylolisthesis, especially in those individuals with a higher grade of subluxation.11 In both scoliosis and spondylolisthesis normal spinal biomechanics are altered so that large tensile forces are imposed on the graft site, limiting the capacity of fusion. Motion across the fusion site is also associated with pseudarthrosis. The pseudarthrosis rate increases with an increasing number of spinal segments because of a longer fusion mass and more total motion across the area.22,74 Spinal instrumentation reduces pseudarthrosis rates by limiting motion across the area.74,89,93,94 Schwab and colleagues85 reported significantly improved union rates in patients with lumbar spondylosis. In cases of fusion for degenerative disc disease, noninstrumented fusions have pseudarthrosis rates ranging from 20% to 40%, whereas the addition of pedicle screw fixation decreases the rate to 5%. Although it is generally accepted that instrumentation facilitates fusions, some authors report no difference in rates of arthrodesis using internal fixation.61 The decision to use instrumentation must be tempered with the understanding that intraoperative device-related complication rates have been reported to be approximately 5% to 10%, although the majority of these occurrences do not result in permanent sequelae.30 Additionally, although hardware improves fusion rates in most instances, up to 20% of instrumented patients require reoperation for the removal of hardware.10,74 These rates, however, are comparable to the incidence of revision surgery for the repair of pseudarthrosis in noninstrumented fusions.30 Methods of pseudarthrosis repair in patients with scoliosis are related to the degree of deformity. A one-
Chapter 42 ■ Dorsal and Lateral Thoracic and Lumbar Fusion
stage repair of pseudarthrosis is possible if the patient has maintained an appropriate spinal contour. Under these conditions the fusion mass is exposed, the instrumentation is removed, the bone is decorticated, and large quantities of autologous cancellous bone are applied. Internal fixation is then reapplied and compressive forces are placed across the area of nonunion. A two-stage procedure is advocated in persons who develop significant scoliosis as a result of pseudarthrosis. In these patients the fusion mass is exposed, the hardware is removed, and multiple osteotomies are performed in an effort to realign the spine. The deformity is reduced during a course of halofemoral traction after which the patient undergoes bone grafting and reinstrumentation.20,78 Ventral discectomy and bone grafting followed by dorsal instrumentation and fusion have also been recommended. Repair of lumbosacral dorsal and lateral fusions can be approached by a re-exploration of the fusion mass, with decortication and application of additional bone graft. Fusion rates in patients undergoing pseudarthrosis repair are significantly improved by the application of dorsal segmental instrumentation.74,94 Interbody fusion has also been advocated as an alternative in patients with pseudarthrosis and persistent low back pain.77 Although successful fusion has been reported in 81% of patients, the approach is not familiar to many spine surgeons and is associated with complications, including hemorrhage resulting from great vessel injury, unintentional sympathectomy, and retrograde ejaculation in males.83 The general medical condition of the patient also influences fusion and fracture healing. Smoking has been associated with an increased incidence of pseudarthrosis.15,17 Partial oxygen arterial pressures were significantly reduced in the population of smokers, reinforcing the concept that oxygen delivery is an essential component in bone deposition.15 Osseous healing is also facilitated by immunologically mediated messages between cells. If possible, glucocorticoid administration should be tapered before surgery to increase the possibility of engraftment. Nutritional status is also an essential preoperative consideration in patients undergoing elective arthrodesis; malnourished patients may benefit from dietary supplementation before their surgery.12,49 Hemorrhage Significant blood loss from thoracolumbar fusions may be the result of perforation of the terminal branches of lumbar and thoracic segmental arteries.54 These vessels are encountered while exposing the transverse processes in the vicinity of the facet joint and the proximal portion of the transverse process. They lie deep in the intermediate fascia of the paraspinal musculature; the rostral and caudal articular arteries lie superior and inferior to the facet joint, the communicating artery travels lateral to the joint on the dorsal surface of the transverse process, and the interarticular artery courses medially (Figures 42.1 and 42.2). Identification and coagulation of these vessels before dissection will avoid hemorrhage and excessive use of cautery. Venous bleeding can be the result of increased intraabdominal pressure with a resultant increase in flow through Batson’s plexus. Bleeding can be avoided by careful posi-
507
tioning, as outlined in the previous section. Venous bleeding from the decorticated segment of bone at the graft site may also occur. Although some degree of hemorrhage is preferable to a dry fusion bed, excessive intraoperative bleeding is avoidable if decortication follows all other portions of the procedure. Autologous blood banking should be offered as an alternative to patients when operative blood loss may require blood transfusion. This option is essential in scoliosis surgery in which exposures are expansive and blood loss may be significant. Oga et al.65 outlined an algorithm using a cell saver and autologous blood banking. In their system harvested autologous blood and cell savers were used in patients in whom operative blood loss was expected to exceed 1000ml. The cell saver alone was used in procedures in which blood loss of less than 1000ml was expected. Blood loss in their series ranged from 2000 to 3000ml. Homologous transfusion was required in only 10% of their patients.65 Infection Infection rates in spinal surgery vary with the complexity and duration of the procedure. Whereas 2% to 4% of noninstrumented fusions become complicated by infection, procedures that involve the placement of internal fixators result in rates between 6% and 11%.56 Staphylococcus aureus is the most common pathogen isolated from deep wound infections. Streptococcal species and a variety of gram-negative enteric organisms have also been identified as pathogens. Seeding may occur intraoperatively by direct inoculation of the organism; therefore adherence to strict sterile technique minimizes risk. Familiarity with decortication, harvesting, and instrumentation techniques shortens operative time and therefore minimizes the possibility of wound colonization. The use of allograft and other implants has also been associated with an increased risk of infection. The possibility that these materials will transmit microbes is minimized if sterile packaging is opened at the last possible moment to prevent contamination.63 Hematogenous seeding of wound infections is also possible during and after surgery. The recognition and treatment of remote infections minimize the possibility of colonization from this source. Urinary catheters should be removed soon after completion of the procedure to prevent the possibility of urinary tract infections. Patient mobilization, pain control, and spirometry should be encouraged to prevent atelectasis and pneumonia. Perioperative antibiotics and intraoperative wound irrigation are effective adjuncts that should minimize infection rates.42 A first- or second-generation cephalosporin sufficiently covers the most common pathogens encountered in this population of patients. Preservation of the viability of tissues within the wound is of paramount importance; self-retaining retractors should be repositioned often to reduce tissue damage, and monopolar electrocautery should be used sparingly. If placed, drains should be directed away from the perineum and should be removed by the second postoperative day.56 Re-exploration of wound infections is indicated not only to obtain diagnosis but also to débride infected and necrotic material. A complete exposure of the fusion mass
508
Section 3 ■ Surgical Procedures
should include the removal of unincorporated bone graft and necrotic material. After the wound is thoroughly irrigated, washed bone graft material may be replaced. Levi et al.51 have demonstrated that deep, instrumented wound infections should be treated without the removal of hardware. They were able to eradicate 16 of 17 deep wound infections by using a combination of débridement and systemic and local antibiotic administration.51 Postfusion Spondylosis Interlaminar fusion has been associated with vertebral canal stenosis, as well as with acquired pars defects. Maturation of the fusion mass may result in the inward protrusion of the lamina to obstruct the vertebral canal. These complications usually occur at the cranial edge of the fusion site and may be associated with hypertrophy of the ligamentum flavum.13,46,54 Postfusion stenosis is accentuated in persons with associated disc bulges or congenital stenosis; this complication may be avoided through careful patient selection. Alternative treatments that overcome this phenomenon include intertransverse fusion, as well as a procedure, as described by DiPierro et al., to decompress nerve roots and stabilize the lumbar spine in patients with lumbar stenosis, and involve unilateral decompressive hemilaminectomy (Figure 42.6).25 The contralateral lamina and facet are decorticated and packed with autograft. The impact of laminar hypertrophy in dorsal fusions may be minimized by this unilateral decompression. General Medical Complications Lumbar fusion is commonly used in patients with degenerative spinal conditions. The utility of spinal arthrodesis
in several forms of degenerative disease, however, is speculative. In an effort to clarify the risk imposed by spinal fusion in comparison with spine surgery without fusion, Deyo et al.23 compared the cost, complications, and reoperation rates in elderly patients on Medicare. Patient selection was based on the diagnosis as defined by International Classification of Diseases, Ninth Revision Clinical Modification. These investigators found that mortality, the rate of nursing home hospitalization, and the duration of hospitalization were significantly increased in patients who required a concomitant fusion for a given diagnosis. Outcome, as measured by the incidence of reoperation 4-year follow-up, was no different in both groups.23 In a separate review of 1000 patients who underwent spinal fusion, Prothero et al.68 found that urinary retention and paralytic ileus were the most common medical complications. Deep venous thrombosis occurred in 4.2% of patients. The risk of acquiring thrombotic complications, however, was significantly reduced by a factor of 2 when the duration of postoperative bed rest was decreased.68
Conclusion Dorsal and dorsolateral spine fusion are currently used for a variety of traumatic, neoplastic, congenital, and degenerative conditions. The development of segmental instrumentation systems has permitted correction of spinal deformity and reduced pseudarthrosis rates at the price of a slightly increased operative complication rate. Spinal instrumentation will fail with time unless adequate bony fusion occurs. Meticulous preparation of the fusion bed and adequate quantities of bone graft are key factors in obtaining a solid fusion. Complications are reduced by a thorough understanding of operative anatomy and proper preoperative, intraoperative, and postoperative care.
REFERENCES
Figure 42.6 Three-dimensional reconstruction of fine-slice
computed tomography data showing a unilateral dorsal decompression with contralateral interlaminar fusion. (From DiPierro CG, Helm GA, Shaffrey CI, et al: J Neurosurg 84: 166-173, 1996, with permission.)
1. Abbot LC, Schottstaedt ER, Saunders JB, Bost FC: The evaluation of cortical and cancellous bone as grafting material: a clinical and experimental study. J Bone Joint Surg 29:381-414, 1947. 2. Adkins EWO: Lumbo-sacral arthrodesis after laminectomy. J Bone Joint Surg 37B:208-233, 1955. 3. Albee FH: Transplantation of portions of the tibia into the spine for Pott’s disease. JAMA 57:885-886, 1911. 4. An HS, Lynch K, Toth J: Prospective comparison of autograft vs. allograft of adult posterolateral lumbar spine fusion: differences among freeze dried, frozen, mixed grafts. J Spine Disord 8:131-135, 1995. 5. Apel DM, Lorenz MA, Zindrick MR: Symptomatic spondylolisthesis in adults: four decades later. Spine 14: 345-348, 1989. 6. Aurori BF, Weierman RJ, Lowell HA, et al: Pseudarthrosis after spinal fusion for scoliosis. A comparison of autogenic and allogeneic bone grafts. Clin Orthop 199: 153-158, 1985.
Chapter 42 ■ Dorsal and Lateral Thoracic and Lumbar Fusion 7. Batson OV: The function of the vertebral veins and their role in the spread of metastases. Ann Surg 112:138-149, 1940. 8. Bernardi R, Mueller W: Thoracic fusion techniques using bone. In Menezes AH, Sonntag VKH (eds): Principles of Spine Surgery. New York, McGraw-Hill, 1996, pp 1173-1183. 9. Blumenthal SL, Gill K: Can lumbar spine radiographs accurately determine fusion in postoperative patients? Correlation of routine radiographs with a surgical second look at lumbar fusion. Spine 18:1186-1189, 1993. 10. Blumenthal SL, Gill K: Complications of the Wiltse pedicle screw fixation system. Spine 18:1867-1871, 1993. 11. Boxall D, Bradford DS, Winter RB: Management of severe spondylolisthesis in children and adolescents. J Bone Joint Surg 61A:479-486, 1979. 12. Bridwell KH, O’Brien MF, Lenke LG, et al: Posterior spinal fusion supplemented with only allograft bone in paralytic scoliosis. Does it work? Spine 19:2658-2666, 1994. 13. Brodsky AE: Post-laminectomy and post-fusion stenosis of the lumbar spine. Clin Orthop 115:130-139, 1976. 14. Brodsky AE, Kovalsky ES, Khalil MA: Correlation of radiologic assessment of lumbar spine fusions with surgical exploration. Spine 16:S261-S265, 1991. 15. Brown CW, Orme TJ, Richardson HD: The rate of pseudarthrosis (surgical nonunion) in patients who are smokers and patients who are nonsmokers: a comparison study. Spine 11:942-943, 1986. 16. Butterman GR, Glazer PA, Bradford DS: The use of allografts in the spine. Clin Orthop 324:75-85, 1996. 17. Carpenter CT, Dietz JW, Leung KYK, et al: Repair of a pseudarthrosis of the lumbar spine. J Bone Joint Surg 78A:712-720, 1996. 18. Chase CW, Herndon CH: The fate of autogenous and homogenous bone grafts. A historical review. J Bone Joint Surg 37A:809-841, 1955. 19. Cook SD, Rueger DC: Osteogenic protein. I. Biology and applications. Clin Orthop 324:29-38, 1996. 20. Cummine JL, Lonstein JE, Moe JH: Reconstructive surgery in the adult for failed scoliosis fusion. J Bone Joint Surg 61A:1151-1161, 1979. 21. Dawson EG, Lotysch M, Urist MR: Intertransverse process lumbar arthrodesis with autogenous bone grafts. Clin Orthop 154:90-96, 1981. 22. DePalma AF, Rothman RH: The nature of pseudarthrosis. Clin Orthop 59:113-118, 1968. 23. Deyo RA, Ciol MA, Cherkin DC, et al: Lumbar spine fusion. A cohort study of complications, re-operations, and resource use in the Medicare population. Spine 18: 1463-1470, 1993. 24. Dickson JH, Erwin WD, Rossi D: Harrington instrumentation and arthrodesis for idiopathic scoliosis: a twenty-one year follow up. J Bone Joint Surg 72A: 678-683, 1990. 25. DiPierro CG, Helm GA, Shaffrey CI, et al: Treatment of lumbar stenosis by extensive unilateral decompression and contralateral autologous bone fusion: operative technique and results. J Neurosurg 84:166-173, 1996. 26. Dodd CAF, Fergusson CM, Freedman L, et al: Allograft versus autograft bone in scoliosis surgery. J Bone Joint Surg 70B:431-434, 1988.
27. Fabry G: Allograft versus autograft in idiopathic scoliosis surgery: a multivariate statistical analysis. J Pediatr Orthop 11:465-468, 1991. 28. Fox MW, Onofrio BM, Hanssen AD: Clinical outcomes and radiological instability following decompressive lumbar laminectomy for degenerative spinal stenosis: a comparison of patients undergoing concomitant arthrodesis versus decompression alone. J Neurosurg 85:793-802, 1996. 29. Frymoyer JW, Hanley EN, Howe J, et al: A comparison of radiographic findings in fusion and nonfusion patients ten or more years following lumbar surgery. Spine 4:435-440, 1979. 30. Garfin SR: Summation. Spine 19:2300S-2305S, 1994. 31. Goldstein LA: Results in the treatment of scoliosis with turnbuckle plaster cast correction and fusion. J Bone Joint Surg 41A:321-335, 1959. 32. Graham JJ: Pseudarthrosis in scoliosis: routine exploration of forty-five operative cases. J Bone Joint Surg 50A:850, 1968. 33. Grubb SA, Lipscomb HJ: Results of lumbosacral fusion for degenerative disc disease with and without instrumentation. Two and five year follow up. Spine 17: 349-355, 1992. 34. Hanley EN: The indications for spinal fusion with and without instrumentation. Spine 20:143S-153S, 1995. 35. Hanley EN, Levy JA: Surgical treatment of isthmic lumbosacral spondylolisthesis. Analysis of variables influencing results. Spine 14:48-50, 1989. 36. Hasday CA, Passoff TL, Perry JP: Gait abnormalities arising from iatrogenic loss of lumbar lordosis secondary to Harrington rod instrumentation in lumbar fractures. Spine 8:501-511, 1983. 37. Heiple KG, Chase SW, Herndon CH: A comparative study of the healing process following different types of bone transplantation. J Bone Joint Surg 45A:1593-1616, 1963. 38. Herkowitz HN, Kurz LT: Degenerative lumbar spondylolisthesis with spinal stenosis. A prospective study comparing decompression with decompression and intertransverse process arthrodesis. J Bone Joint Surg 73A:802-808, 1991. 39. Herron LD, Newman MH: The failure of ethylene oxide gas-sterilized freeze-dried bone graft for thoracic and lumbar spine fusion. Spine 14:496-500, 1989. 40. Hibbs RH: An operation for progressive spinal deformities. NY J Med 93:1013-1016, 1911. 41. Hibbs RA: A report of fifty-nine cases of scoliosis treated by the fusion operation. J Bone Joint Surg 6A:3-37, 1924. 42. Horwitz NH, Curtin JA: Prophylactic antibiotics and wound infection following lumbar disc surgery: a retrospective study. J Neurosurg 43:727-731, 1975. 43. Howorth MB: Evolution of spinal fusion. Ann Surg 117:278-289, 1943. 44. Jorgenson SS, Lowe TG, France J, Sabin J: A prospective analysis of autograft versus allograft in posterolateral lumbar fusion in the same patient. A minimum of 1 year follow up in 144 patients. Spine 19:2048-2053, 1994 45. Kant AP, Daum WJ, Dean MS: Evaluation of lumbar spine fusion. Plain radiographs versus direct surgical exploration and observation. Spine 20:2313-2317, 1995.
509
510
Section 3 ■ Surgical Procedures
46. Kestler OC: Overgrowth (hypertrophy) of lumbosacral grafts causing complete block. Bull Hosp Jt Dis Orthop Inst 27:51-57, 1966. 47. Kurz LT, Garfin SR, Booth RE: Harvesting autogenous iliac crest bone grafts: a review of complications and techniques. Spine 14:1324-1331, 1989. 48. Kushner A: Evaluation of Wolff’s law in bone formation. J Bone Joint Surg 22A:589-596, 1940. 49. Lane JM, Muschler GF: Principles of bone fusion. In Rothman RH, Simeone FA (eds): The Spine, ed 3. Philadelphia, WB Saunders, 1992, pp 1739-1755. 50. Laurie SWS, Kaban LB, Mulliken JB, Murray JE: Donor site morbidity after harvesting rib and iliac bone. Plast Reconstr Surg 73:933-938, 1984. 51. Levi ADO, Dickman CA, Sonntag VKH: The management of postoperative infections after spinal instrumentation. J Neurosurg 86:975-980, 1997. 52. Lombardi JS, Wiltse LL, Reynolds J, et al: Treatment of degenerative spondylolisthesis. Spine 10:821-827, 1985. 53. MacKenzie-Forbes A: Technique of an operation for spinal fusion as practiced in Montreal. J Orthop Surg 2B:509-514, 1920. 54. MacNab I: The blood supply of the lumbar spine and its application to the technique of intertransverse lumbar fusion. J Bone Joint Surg 53B:628-638, 1971. 55. MacNab I: The traction spur. An indicator of segmental instability. J Bone Joint Surg 53A:663-670, 1971. 56. Massie JB, Heller JG, Abitbol JJ, et al: Postoperative posterior spinal wound infections. Clin Orthop 284:99-108, 1992. 57. Mathieu P, Demirleau P: Traitement chirurgical du spondylolisthesis douloureaux. Rev Orthop 23:352, 1936. 58. Matthews JL: Bone structure and ultrastructure.. In Urist MR (ed): Fundamental and Clinical Bone Physiology. Philadelphia, Lippincott-Raven, 1980, pp 4-44. 59. May VR, Mauck WR: Exploration of the spine for pseudarthrosis following spinal fusion in the treatment of scoliosis. Clin Orthop 53:115-122, 1967. 60. McBride ED: A mortised transfacet bone block for lumbosacral fusion. J Bone Joint Surg 31A:385-393, 1949. 61. McGuire RA, Amundson GM: The use of primary internal fixation in spondylolisthesis. Spine 18:1662-1672, 1993. 62. Moe JH: A critical analysis of methods of fusion for scoliosis. An evaluation in two hundred and sixty six patients. J Bone Joint Surg 40A:529-554, 1958. 63. Montgomery DM, Aronson DD, Lee CI, Lamont RL: Posterior spine fusion: allograft versus autograft bone. J Spinal Disord 3:370-375, 1990. 64. Nugent JP, Dawson EG: Intertransverse process lumbar arthrodesis with allogenic fresh-frozen bone graft. Clin Orthop 287:107-111, 1991. 65. Oga M, Ikuta H, Sugioka Y: The use of autologous blood in the surgical treatment of spinal disorders. Spine 17: 1381-1385, 1992. 66. Prolo DJ: Morphology and metabolism of fusion of the lumbar spine. In Youmans JR (ed): Neurological Surgery, ed 4. Philadelphia, WB Saunders, 1996, pp 2449-2460. 67. Prolo DJ, Rodrigo JJ: Contemporary bone graft physiology and surgery. Clin Orthop 200:322-342, 1985.
68. Prothero SR, Parkes JC, Stinchfield FE: Complications after low-back fusion in 1000 patients: a comparison of two series one decade apart. Clin Orthop 306:5-11, 1994. 69. Reddi AH, Weintroub S, Muthukumaran N: Biologic principles of bone induction. Orthop Clin North Am 18:207-212, 1987. 70. Rifas L, Shen V, Mitchell K, Peck WA: Macrophagederived growth factor for osteoblast-like cells and chondrocytes. Proc Natl Acad Sci USA 81:4558-4562, 1984. 71. Sauer HD, Schoettle H: The stability of osteosynthesis bridging defects. Arch Orthop Trauma Surg 95:27-30, 1979. 72. Schaffer JW, Field GA, Goldberg VM: Fate of vascularized and nonvascularized autografts. Clin Orthop 197:32-43, 1985. 73. Scher CD, Shepard RC, Antoniades HN, Stiles CD: Platelet-derived growth factor and the regulation of the mammalian fibroblast cell cycle. Biochim Biophys Acta 560:217-241, 1979. 74. Schwab FJ, Nazarian DG, Mahmud F, Michelsen CB: Effects of spinal instrumentation on fusion of the lumbosacral spine. Spine 20:2023-2028, 1995. 75. Simmons DJ: Fracture healing perspectives. Clin Orthop 200:100-112, 1985. 76. Sonntag VKH, Marciano FF: Is fusion indicated for lumbar spinal disorders? Spine 20:138S-142S, 1995. 77. Stauffer RN, Coventry MD: Anterior interbody lumbar spine fusion. J Bone Joint Surg 54A:756-768, 1972. 78. Steinmann JC, Herkowitz HN: Pseudarthrosis of the spine. Clin Orthop 284:80-90, 1992. 79. Stevenson S, Emery SE, Goldberg VM: Factors affecting bone graft incorporation. Clin Orthop 323: 66-74, 1996. 80. Stringa G: Studies of the vascularization of bone grafts. J Bone Joint Surg 39B:395-420, 1957. 81. Summers BN, Einstein SM: Donor site pain from the ilium. J Bone Joint Surg 71B:677-680, 1989. 82. Tan SB, Kozak JA, Dickson JH, Nalty TJ: Effect of operative position on sagittal alignment of the lumbar spine. Spine 19:314-318, 1994. 83. Tiusanen H, Seitsalo S, Osterman K, Soini J: Anterior interbody lumbar fusion in severe low back pain. Clin Orthop 324:153-163, 1996. 84. Turner JA, Ersek M, Herron L, et al: Patient outcomes after lumbar spine fusion. JAMA 268:907-911, 1992. 85. Vaccaro AR, Garfin SR: Internal fixation (pedicle screw fixation) for fusions of the lumbar spine. Spine 20: 157S-165S, 1995. 86. Watkins MB: Posterolateral fusion of the lumbar and lumbosacral spine. J Bone Joint Surg 35A:1014, 1953. 87. Wiltse LL, Spencer CW: New uses and refinements of the paraspinal approach to the lumbar spine. Spine 13:696-706, 1988. 88. Wiltse LL, Bateman JG, Hutchinson RH: The paraspinal sacrospinalis splitting approach to the lumbar spine. J Bone Joint Surg 50A:919-926, 1968. 89. Wood GW, Boyd RJ, Carothers TA, et al: The effect of pedicle screw/plate fixation on lumbar/lumbosacral
Chapter 42 ■ Dorsal and Lateral Thoracic and Lumbar Fusion autogenous bone graft fusions in patients with degenerative disc disease. Spine 20:819-830, 1995. 90. Urist MR: Bone: formation and autoinduction. Science 150:893-899, 1965. 91. Urist MR, Hay PH, Dubuc F, Buring K: Osteogenic competence. Clin Orthop 64:194-220, 1969. 92. Younger EM, Chapman MW: Morbidity at bone graft donor sites. J Orthop Trauma 3:192-195, 1989.
93. Yuan HA, Garfin SR, Dickman CA, Mardjetko SM: A historical study of pedicle screw fixation in thoracic, lumbar and sacral spinal fusions. Spine 19:2279S-2296S, 1994. 94. Zdeblick TA: A prospective, randomized study of lumbar fusion: preliminary results. Spine 18:983-991, 1993.
511
3.3 TRAUMA SURGERY C H A P T E R
4 3
injury (including associated injuries), the patient characteristics (e.g., age, medical risk factors, bone quality, desire/ability to tolerate use of a halo orthosis), and the physician’s experience. Although much less common, penetrating trauma to the occipitocervical junction presents unique issues that relate to the specific location and trauma modality (bullet, knife, etc.). This class of injuries will not be specifically discussed in this chapter but is addressed in the chapter on Penetrating Spinal Cord Injuries. Although most of the principles from blunt trauma are applicable to penetrating trauma, it is important to point out some important differences. Compared with blunt trauma, penetrating trauma typically results in less ligamentous injury and therefore for a similar fracture may be more stable. However, penetrating trauma more commonly result in trauma to vascular or other important regional structures.
Trauma Surgery: Occipitocervical Junction Curtis A. Dickman, Tanvir F. Choudhri, and Jürgen Harms The occipitocervical junction includes the skull base at the foramen magnum, C1, C2, and the associated ligamentous, neural, and vascular structures. Because of the important neural and vascular structures in the region, occipitocervical junction injuries have the potential to cause significant neurologic morbidity or mortality. Therefore, careful recognition, diagnosis, and management of these injuries are essential. The association with high-impact trauma and occipitocervical junction injuries is well recognized. However, the potential for injuries even with relatively minor trauma should be remembered, especially with abnormal bone (e.g., osteoporosis) and/or ligaments (e.g., rheumatoid arthritis). Occipitocervical junction injuries can be classified in several ways (Box 43.1). One useful system describes occipitocervical junction injuries as isolated ligamentous injuries, isolated fractures, or mixed ligamentous and bony injuries. Occipitocervical junction trauma can also be described by the site and/or level(s) of injury. At most sites, classification systems have been developed for specific injury patterns (e.g., C2 odontoid fractures). Finally, occipitocervical junction injuries can be described based on their stability. Stability is generally determined with clinical and radiographic assessment, sometimes using dynamic flexion/extension radiographs. A stable injury does not demonstrate significant radiographic deformity, pain, or neurological dysfunction with normal physiological loads and movement. An example of a stable injury would be an isolated C2 spinous process fracture, which meets the above criteria. Some injuries are clearly unstable, such as occipitocervical dislocations. Other injuries may initially appear stable but have a reasonable chance of developing delayed instability with time, gravity, movement, and/or relaxation of paraspinal muscle spasm. This category reflects the reality that clinical and radiographic assessment of long-term stability may be indeterminate. The above classification systems are helpful in injury assessment and planning management. However, the management of a patient with occipitocervical junction trauma is best determined by considering the nature of the
General Principles The initial management of occipitocervical junction injuries is focused on basic trauma management principles including establishment and maintenance of airway, breathing, and circulation; careful immobilization and transportation; and recognition and management of any associated injuries. These principles have evolved over time and have been published in numerous settings.28,29 Occipitocervical junction injuries are frequently recognized on routine cervical spine imaging. However, these injuries may be difficult to detect on initial diagnostic studies. Clinical suspicion based on history and physical exam can aid recognition. Routine radiographs and clinical assessment are often inadequate to fully characterize the injury, and more specialized imaging is usually indicated. Coronal, curved coronal, sagittal, and/or 3-D CT reconstruction views can be extremely helpful in characterizing the presence and nature of injury. MRI imaging may be difficult or impossible to obtain acutely but can often provide essential information on spinal canal compromise and may suggest the presence and degree of ligamentous injury. Dynamic imaging with plain x-rays, CT, and/or MRI can be valuable in assessing stability but should be performed carefully. Occasionally, stability is checked with real-time fluoroscopy during careful flexion and extension controlled by a qualified examiner. For example, fluoroscopic flexion/extension imaging may be helpful when there is urgent need to assess the stability of the cervical spine in an unresponsive patient. Once occipitocervical junction injuries are diagnosed, management decisions are based on several factors, including the extent and stability of injury, the presence or progression of neurological deficits, and patient-specific factors that influence the risks with different treatments. Nonoperative management typically includes some type of rigid (halo) or semirigid (collar) orthosis. Operative management is generally indicated for injuries that are unstable, have significant potential for delayed instability, have 512
Chapter 43 ■ Trauma Surgery: Occipitocervical Junction
513
BOX
43.1 Occipitocervical junction injury classification systems
A. Location of bone or ligamentous injury Pure ligamentous injuries
Occipitoatlantal dislocations Transverse ligament injuries Rotatory C1-2 dislocations Isolated fractures
Occipital condyle fractures C1 (lateral mass, ring) C2 (odontoid, body, hangman, posterior element) Mixed ligamentous and bony injuries B. Site/Level of Injury
Occipital bone (C0) C0-1 ligaments C1 C1-2 ligaments C2
(e.g., condyle fracture) (e.g., occipitoatlantal dislocation) (e.g., lateral mass, ring fractures) (e.g., transverse ligament injuries) (odontoid, body, hangman, posterior element fractures)
C. Degree of stability
Stable Low probability of delayed instability High probability of delayed instability Unstable
progressive neurological deficits, and/or cause significant deficits or symptoms that are not controlled with nonoperative measures. Operative planning may include obtaining additional imaging (e.g., dedicated studies for image-guidance), ensuring the availability of appropriate instrumentation, and arranging neurophysiologic monitoring where appropriate.
Diagnosis and Management Occipitocervical Dislocations Occipitocervical dislocations are relatively uncommon ligamentous injuries that usually result from hyperflexion and distraction during high-impact blunt trauma.1,36 These injuries are highly unstable, frequently fatal, and usually result in significant neurologic injury from stretching, compression, and/or distortion of the spinal cord, brainstem, and cranial nerves.8 In addition, significant morbidity and mortality can result from vertebral artery injury. Recognition and rapid management of these injuries may limit further injury, but even with appropriate care, neurologic deficits can progress. Although initially felt to be rare, several series of trauma fatalities have revealed an incidence between 8% and 19%.1 Lateral cervical spine radiographs may recognize occipitocervical dislocations (sensitivity 0.57), especially in severe injuries. However, these injuries can be difficult to diagnose with plain radiographs alone, especially with lesssevere dislocations. In addition, the frequent presence of coexisting significant head trauma can delay recognition of spinal injury. Diagnostic clues include prevertebral softtissue swelling, an increase in the dens-basion distance, and separation of the occipital condyles and C1 lateral masses (Figure 43.1). CT imaging with reconstruction
Figure 43.1 Lateral cervical x-rays demonstrating
occipitocervical dislocation. The craniovertebral instability is apparent in the two images. (From Dickman CA, Spetzler RA, Sonntag, VKH [eds]: Surgery of the Craniovertebral Junction. New York, Thieme, 1998.)
views (sensitivity 0.84) usually provides a better assessment of fractures and alignment than plain x-rays. The presence of subarachnoid hemorrhage supports but does not confirm the diagnosis. MRI imaging can be helpful for diagnosis (sensitivity 0.86), to assess the extent of spinal cord compression and injury, and to demonstrate compressive hematoma lesions.29 Based on the injury pattern, occipitoatlantal dislocations have been classified by Traynelis44 into four types: type I (anterior), type II (longitudinal), type III (posterior), and “other” (complex). A number of diagnostic radiographic criteria have been described that assess the relationship between the skull base and cervical spine (Figure 43.2). Although developed for lateral plain x-rays, these criteria can also be used on sagittal-reconstruction CT views provided there are no significant artifactual distortions. Wackenheim’s clival line extends along the posterior surface of the clivus and should be tangential to the tip of the dens.45 Ventral or dorsal translation of the skull in relation to the dens will shift the clival line to either intersect or run posterior to the dens, respectively. Power’s ratio is based on the relationship of the B-C line (from the basion to the C1 posterior arch) and the O-A line (between the opisthion and the C1 anterior arch).39 Normal BC/OA ratios average 0.77, while pathologic ratios (greater than 1.0) typically represent occipitocervical dislocations. However, false-negatives can occur with longitudinal or posterior dislocations. The Wholey dens-basion technique assesses the distance from the basion to the dens tip.46 Although variability is common, in adults the average distance is about 9 mm and pathological distances are greater than 15mm.12 Dublin’s method, the least reliable method, measures the distance from the mandible (posterior ramus) to the anterior part of C1 (normally 2 to 5mm) and C2 (normally 9 to 12mm).15 Initial management of these injuries focuses on immobilization, almost always with a halo orthosis. Cervical collars are potentially dangerous because they may produce distraction and thereby promote further injury. Similarly, traction can cause neurologic worsening (2 of 21 patients) and should be avoided or used with extreme caution.28,29 Nonoperative management does not provide definitive treatment of these injuries because of the significant liga-
514
Section 3 ■ Surgical Procedures
B B A
A
O C
C D
Figure 43.2 Four radiographic methods for assessing occipitocervical dislocation:
(A) Wackenheim’s clival line; (B) Power’s ratio (BC/OA); (C) Wholey dens-basion technique; and (D) Dublin’s method. See text for details. (From Barrow Neurologic Institute, with permission.)
mentous disruption that cannot be expected to heal even with prolonged rigid (halo) external immobilization (11 of 40 patients had a nonunion and/or neurologic deterioration).29 Operative stabilization consists of an occipitocervical arthrodesis with rigid internal fixation (discussed below and in the chapter on High Cervical and Occipitocervical Plate, Rod, Wire, and Bone Techniques). Decompression and restoration of alignment may also be necessary to maximize neurologic recovery.
IA
IIA
IB
IIB
Transverse Ligament Injuries Isolated traumatic transverse ligament injuries are unstable injuries that can result in significant upper cervical spinal cord injury either during the initial trauma or afterwards. These injuries are more common in hyperflexion injuries. Because transverse ligament injuries may be difficult to recognize on initial (neutral) plain x-rays, an elevated index of suspicion is required in some settings—for example, high-impact trauma. Transverse ligament injuries are suggested or diagnosed with radiographic imaging. A widened atlantodental interval (ADI) on flexion lateral cervical radiographs (greater than 3mm in adults, greater than 5mm in children) suggests transverse ligament insufficiency. Thin-cut CT imaging with reconstruction views may suggest the diagnosis by demonstrating a C1 lateral mass avulsion fracture at the ligamentous insertion. Thin-cut MRI with attention to the transverse ligament when using gradient
Figure 43.3 Classification of transverse ligament injuries. Type I
injuries are disruptions of the transverse ligament in its midportion (IA) or periosteal insertion laterally (IB). Type II injuries lead to transverse ligament insufficiency through fractures that disconnect the C1 lateral mass tubercle (insertion of the transverse ligament) via a comminuted fracture (IIA) or an avulsion fracture (IIB). (From Barrow Neurologic Institute, with permission.)
echo sequences can directly demonstrate a transverse ligament injury.13 If the diagnosis is uncertain, dynamic (flexion/extension) imaging is appropriate for cooperative patients. Based on CT and MRI traumatic transverse ligament injuries can be classified into two categories (Figure 43.3). Type I injuries involve disruptions of the midportion
Chapter 43 ■ Trauma Surgery: Occipitocervical Junction
(IA) or periosteal insertion laterally (IB). Type II injuries involve fractures that disconnect the C1 lateral mass tubercle for insertion of the transverse ligament via a comminuted fracture (IIA) or an avulsion fracture (IIB).11 The management of transverse ligament injuries is based on the type of injury. Type I injuries are pure ligamentous injuries that cannot be expected to heal with nonoperative external fixation. Therefore, operative stabilization with a posterior C1-2 arthrodesis and fixation is indicated. The surgical options include C1-2 posterior wiring, C1-2 Halifax clamps, C1-2 transarticular screws, and/or C1-2 segmental screw fixation (see later section and Chapter 111). Type II injuries have a much higher chance of healing with halo immobilization (up to 74%).11 If a nonunion is still present after a prolonged period of immobilization (greater than 3 months) then operative stabilization is generally appropriate. Rotatory C1-2 Subluxations Rotatory C1-2 subluxations are ligamentous injuries that are more common in children and adolescents. These injuries typically present with neck pain and a fixed, rotated “cocked-robin” head position. Open mouth x-rays may demonstrate an asymmetry of the C1 and C2 lateral masses. CT imaging can confirm the rotatory subluxation diagnosis and demonstrate coexisting fractures. C1-2 axial rotation greater than 47 degrees confirms the diagnosis. Three-view CT imaging (15 degrees to left, neutral, and 15 degrees to right) can also be helpful in establishing the diagnosis.16,17 MRI imaging may detect a coexistent transverse ligament injury. The treatment of C1-2 rotatory subluxations is generally nonoperative. Axial traction with a halter device or Gardner-Wells tongs can usually achieve reduction of the injury. Prolonged traction and/or the use of muscle relaxants may be needed. Periodic imaging may help to assess progress but clinical improvement in the alignment and symptoms often provides confirmation of a successful reduction. Operative reduction and fixation are reserved for irreducible injuries, recurrent subluxations, and transverse ligament injuries. Occipital Condyle Fractures Occipital condyle fractures generally occur with axial trauma and are almost always unilateral (greater than 90%). These injuries are classified into three types according to Anderson and Montesano.3 Type I injuries are comminuted fractures that result from axial trauma. Type II fractures are extensions of linear basilar skull fractures. Type III injuries, the most common, are avulsion fractures of the condyle that can result from a variety of mechanisms. The incidence of occipital condyle fractures has been estimated to be between 1% and 3% of blunt craniocervical trauma cases.33 Although plain radiographs (usually open mouth x-rays) may occasionally identify the injury, they have an unacceptably low sensitivity (estimated 3.2%) and should not be relied on when the diagnosis is suspected. CT imaging with reconstruction views provides the best assessment of fracture pattern and alignment.33, 37
515
Occipital condyle fractures are generally stable and therefore are typically managed with an external nonrigid orthosis (collar) until the fracture heals (often 12 weeks). Type III fractures are felt to be more prone to instability and, when significant displacement or clinical concern exists, halo immobilization may be appropriate. Operative stabilization with an occipitocervical fusion is generally reserved for situations when there are associated cervical fractures or significant ligamentous injuries. C1 Fractures Isolated C1 fractures comprise approximately 5% of cervical spine fractures. These injuries occur with axial trauma with or without lateral bending.27 Open mouth x-rays may suggest the injury, but CT imaging with reconstruction views provides the best assessment of fracture pattern and alignment. Fractures can include almost any part of the ring or lateral masses of C1. Aside from unilateral lateral mass fractures, the fractures usually occur at multiple sites (Figure 43.4). Jefferson fractures are four-part fractures with bilateral ventral and dorsal ring fractures. The assessment of these injuries is focused on evaluating the integrity of the transverse ligament and on recognizing any additional fractures. The management of C1 fractures is based on the integrity of the transverse ligament that can be assessed indirectly with several radiographic criteria such as a widened atlantodental interval (ADI, greater than 3mm) and increased spread of the lateral masses of C1 over C2 (greater than 6.9mm, Rule of Spence)42 or directly through high-resolution MRI (Figure 43.5). If the transverse ligament is intact, isolated C1 fractures are generally stable and can be treated with an external orthosis (e.g., SOMI) primarily for symptom control until the fracture heals. With transverse ligament insufficiency, operative stabilization is indicated by using a C1-2 fusion technique such as dorsal C1-2 wiring techniques, C1-2 transarticular screws, C1 lateral mass-C2 pars/pedicle screws, or ventral C1-2 screw fixation (see chapter on High Cervical and Occipitocervical Plate, Rod, Wire, and Bone Techniques). The surgical choice is primarily based on patient anatomy and fracture pattern as well as surgeon experience/preference. Postoperatively, most operations employing rigid internal fixation can be managed with a non-rigid external orthosis (e.g., collar or SOMI) but C1-2 dorsal wiring without additional instrumentation generally warrants the use of a halo.41 C2 Fractures C2 fractures comprise about 20% of all cervical spine fractures and are classified as either odontoid, body, or other fractures (e.g., hangman, laminar, or spinous process). Odontoid Fractures C2 odontoid fractures can occur from a number of mechanisms but most often are caused by hyperextension injuries. Although lateral cervical spine x-rays may demonstrate some fractures, especially those with displacement,
516
Section 3 ■ Surgical Procedures
Figure 43.4 C1 lateral mass fracture. Axial CT images (A-B) and coronal (C) and sagittal (D) CT reconstruction views of right C1 lateral mass fracture from high-speed MVA. The fracture healed with 3 months’ external immobilization.
this technique can easily miss fractures, especially those with degenerative changes or minimal displacement. Open-mouth radiographs are very helpful to diagnose most odontoid fractures, but these also may be inconclusive. Thin-cut CT images with sagittal and coronal view reconstruction views are the best way to diagnose and characterize odontoid fractures as well as to find associated fractures and plan treatment.23, 26 Anderson and D’Alonzo classified odontoid fractures into three types based on the location of the fracture line through the odontoid tip (type I), odontoid base (type II), or C2 body (type III)2 (Figure 43.6). Type I fractures are essentially avulsion fractures of the odontoid tip and are rare, generally stable, and usually managed with an external semirigid (collar) or rigid (halo) orthosis. Type II fractures are the most common type of odontoid fracture. These fractures are unstable and prone to nonunion because they occur in an area of relatively reduced osseous vascularity. Therefore, rigid halo immobilization or surgical stabilization is often necessary. Hadley et al. described Type IIA fractures that are comminuted fractures at the base of the dens with associated free fragments.25 These fractures are considered particularly unstable and surgical stabilization is advisable, usually with a posterior C1-2
fusion. Type III fractures involve the vertebral body and are discussed below. C2 Body Fractures The C2 body can be defined as the C2 bone mass caudal to the dens and ventral to the pars interarticularis bilaterally. Benzel et al.6 have classified C2 body fractures based on the orientation of the fracture line: coronal, sagittal, or transverse (a.k.a. horizontal rostral). The transverse type of C2 body fracture is a more appropriate description of type III odontoid fractures. The coronal and sagittal types represent “vertical” fractures. Of the vertical fractures, the coronal type was much more common (4:1) and resulted from multiple (four) different mechanisms. Sagittal type C2 body fractures were caused by axial loading trauma. Figure 43.7 shows an example of C2 body fracture. Although standard cervical radiographs will often recognize the fracture, the injury is best characterized with high-resolution CT scanning with multiplanar reconstruction views. It is important to look for radiographic evidence of involvement of the foramen transversarium and clinical signs of vertebral artery injury. If there is a significant degree of suspicion, an assessment of the vertebral
Chapter 43 ■ Trauma Surgery: Occipitocervical Junction
517
Figure 43.5 Axial MRI images demonstrating an intact (A) and ruptured (B) transverse ligament. (From Dickman CA, Spetzler RA, Sonntag, VKH [eds]: Surgery of the Craniovertebral Junction. New York, Thieme, 1998.)
Odontoid Type I
Odontoid Type II
Odontoid Type III
Figure 43.6 C2 odontoid fractures as described by Anderson
and D’Alonzo. Type II fractures are better described as C2 body fractures as discussed later. (From Barrow Neurologic Institute, with permission.)
artery with CT, MR, or transfemoral catheter angiography should be obtained. The stability of C2 body fractures can be assessed either with fracture characteristics (displacement, etc.) or with careful dynamic (flexion/extension) imaging when stability appears likely. A majority of C2 body fractures can be managed nonoperatively. Depending on the alignment, degree of displacement, and fracture location either a collar or halo may be advisable. Occasionally, surgical intervention with a posterior C1-2 fusion is indicated, particularly for highly unstable fractures and in patients prone to nonunion. Other C2 Fractures Traumatic spondylolisthesis fractures of the axis (a.k.a. hangman fractures) are characterized by bilateral fractures through the C2 pars/pedicle (Figure 43.8). Although these fractures may be unstable, they do not generally cause sig-
nificant compromise of the spinal canal or neurologic injury. Effendi and colleagues18 have classified these injuries into three groups based on mechanism. Type I fractures occur with axial loading and hyperextension. Type II fractures are hyperextension injuries with rebound hyperflexion. Type III fractures are primarily flexion injuries with rebound extension. Levine and Edwards34 modified the system by adding a Type IIA, which represents flexion-distraction injuries. Type I and II injuries are generally stable and can usually be managed in a collar. With significant displacement (greater than 4 to 6 mm) halo immobilization may be advisable. Type IIA injuries are more likely to be unstable, especially with displacement greater than 4 to 6mm or angulation more than 11 degrees. If one or both of these findings are present, surgical stabilization may be necessary. Type III injuries are unstable and typically require surgical stabilization. Isolated C2 laminar or spinous process fractures are stable and therefore are usually managed with an orthosis (e.g., collar). Combination Occipitocervical Junction Injuries Combination occipitocervical junction fractures involve bony and ligamentous injuries of the foramen magnum (e.g., occipital condyles), C1, and/or C2. These injuries are usually unstable, occur with high-impact trauma, and frequently result in death or major neurologic injury. Management of these injuries is similar to that of occipitocervical dislocations. Initial management involves airway management, craniovertebral immobilization, and medical stabilization. Patients who are medically stable are consid-
518
Section 3 ■ Surgical Procedures
Figure 43.7 C2 body fracture. An 80-year-old woman presented with neck pain after a fall.
A lateral cervical radiograph (A) suggests C1-2 instability from a C2 fracture. A sagittal CT reconstruction image (B) suggests a C2 body fracture with a transverse fracture line (also described as type III odontoid fracture) and a vertical (coronal) fracture line in addition. Axial CT images (C-D) confirm the C2 vertical fracture component.
ered for more prolonged stabilization with rigid external halo immobilization and/or surgical stabilization. For incomplete spinal cord injuries, decompression of any compressive bony or hematoma lesions may also be necessary and are performed when the patient is medically stable. With complete spinal cord injuries, the timing of surgical stabilization and/or decompression is less urgent. Combined C1-2 fractures occur with axial trauma with or without lateral bending. Although plain x-rays may indicate a combined fracture, a CT with multiplanar reconstruction views is usually necessary to fully characterize the fractures and alignment and to plan treatment.
Compared with isolated C1 and C2 fractures, combined C1-2 fractures are typically associated with a higher rate of instability, nonunion and neurologic injury. Treatment of these injuries is based on the degree and location of bony and ligamentous injuries. Because of the instability, rigid external (halo) and/or internal fixation are usually required. Standard surgical procedures (e.g., posterior C1-2 interspinous fusion) may not be possible because of the extensive fractures. Advances in instrumentation and surgical technique have allowed the development and increased use of newer types of surgical stabilization such as C1-2 transarticular screws or C1-2 segmental fixation.30,40
Chapter 43 ■ Trauma Surgery: Occipitocervical Junction
519
Figure 43.8 C2 hangman fracture. A 21-year-old woman presented with neck pain after a motor
vehicle accident. Initial studies with a lateral cervical radiograph (A), sagittal CT reconstruction image (B), and axial CT image (C) demonstrate the fracture through the C2 pars/pedicle with moderate displacement. The fracture healed with 3 months’ external immobilization as evidenced by the delayed sagittal CT reconstruction (D) and axial CT (E) images.
Surgical Procedures General Principles Preoperative There are multiple indications for surgical intervention with occipitocervical trauma. Decompression may be necessary to relieve compromise of the spinal canal or neural
foramina from bone or soft tissue (e.g., hematoma) lesions. Internal stabilization may be necessary to treat acute or impending instability, to promote fracture healing, and to improve and/or correct alignment. Preoperative care is focused on optimizing medical stability; obtaining the necessary imaging to assess the injury location, alignment, and stability; and determining the
520
Section 3 ■ Surgical Procedures
nature and timing of any needed intervention. The timing of surgery is based on the patient’s medical stability, the degree of spinal compression, the presence or progression of neurologic deficits, and the availability of optimal operating room equipment and personnel. When appropriate likelihood of benefit exists, incomplete spinal cord injuries with compressive lesions warrant surgical intervention as soon as possible. This is particularly true when there are progressive neurologic deficits. However, it is important to note that neurologic deterioration may be related to the natural history of the neurologic injury and/or medical deterioration (such as hypoxia, hypotension, and/or fever) that would not necessarily be assisted with surgical intervention. Rather, it is possible that the patient would have a better chance of tolerating, and hopefully benefiting from, the procedure by delaying surgery until the medical issues have been optimized. Where possible, early surgical intervention is desirable to promote early mobilization and transfer to rehabilitation. Preoperative planning includes selection of a primary surgical plan, as well as backup plans, which may become necessary. When needed, specialized equipment (imageguidance, instrumentation, etc.) and/or neurophysiologic monitoring should be reserved or arranged in advance. When possible, preoperative studies should be loaded onto image-guidance equipment (if used) in advance to permit preoperative surgical planning. Intraoperative The intraoperative setup and positioning are directed by the nature of the injury and surgical approach. In general, occipitocervical junction trauma procedures use a midline dorsal approach in the prone position with cranial fixation or a high ventral cervical approach in the supine position. Transoral, transfacial, and far lateral skull base approaches are not commonly used in the trauma setting. When there is sufficient neurologic function and degree of potential new or exacerbated neurologic injury, spinal monitoring (sensory and/or motor evoked potentials) may prove useful for determining whether the final surgical positioning is satisfactory. Exposure of the occipitocervical junction for trauma may require special considerations. For example, throughout the case, careful attention is advised to maintain appropriate alignment and minimize/avoid pressure on unstable or compressed neurologic structures. Traumatic injuries to the subcutaneous and paraspinal soft tissues can distort and obscure anatomic landmarks. Additional exposure (length of incision, number of levels, etc.) may aid recognition and management of the abnormal anatomy because of increased exposure of adjacent normal anatomy. Decompression and stabilization are two primary objectives of surgery. Decompression of neurologic structures may be accomplished by correcting alignment or removing compressive bone, ligaments, or other spaceoccupying lesions such as hematomas. The goals of stabilization are to achieve stability and, where appropriate and possible, to maintain/improve alignment, maximize neurologic function, and improve symptoms. Achieving bony
fusion is the best way to achieve long-term stability. At surgery, the standard principles of arthrodesis should be followed with careful attention to the exposure and preparation of bone fusion surfaces and the choice of structural or morselized bone graft material. For the majority of cases, internal fixation with instrumentation is utilized to maintain alignment and promote osseous union. Nonrigid external orthoses (e.g., collar, SOMI) do not provide substantial immobilization of the occipitocervical junction. Therefore, instrumentation should be optimized with some (or all) of the following strategies: including all segments involved in the construct, using larger diameter or length screws as possible, and achieving bicortical purchase where advisable. Occipitocervical junction trauma operations may require special closure considerations. For example, these cases may have an increased risk of postoperative infection as the incision typically extends beyond the suboccipital hairline. Copious irrigation is therefore advised before closure. Leaving one or more Jackson Pratt or Hemovac drains may reduce the incidence of postoperative hematoma or seroma collections. However, these drains should be used cautiously if the dura was compromised from the trauma or during the procedure. In the event of dural compromise, primary closure and/or augmentation (patch, fibrin glue product, etc.) and extra attention to fascial closure are often used. A running locked suture may be used for the skin closure. For cases with significant dural compromise that is not possible to repair, several days of postoperative spinal drainage via local (through or near incision) or distal (typically lumbar) placement of an intrathecal catheter may reduce chances of a postoperative spinal fluid collection or leak. Postoperative Postoperatively, a rigid external orthosis (halo) is used if instrumentation is not used or if concern exists regarding the instrumentation or bone quality. Otherwise, some type of nonrigid orthosis is advisable in most cases. Because standard cervical collars do not immobilize the occipitocervical junction well, special orthoses are often used (e.g., SOMI braces). Although not officially studied or approved for cervical spine application, use of an external spinal stimulator may improve the fusion rate. Stimulators may be used as a primary adjunct in patients prone to nonunion (e.g., smokers) or in an attempt to salvage a nonunion. Patients with spinal cord injury require special attention to nutrition, skin care, pulmonary toilet, DVT prophylaxis, and often psychiatric support. Patients with poor nutrition are prone to wound healing problems and sutures may need to be left for a prolonged period (2 to 3 weeks or more). Postoperative imaging with plain x-rays and/or CT imaging is generally obtained when possible to assess final anatomy and alignment, extent of decompression, and position of instrumentation. Interval imaging is followed as needed to assess bony fusion. When sufficient stability is achieved either from the internal fixation and/or bone fusion, the orthosis can be weaned. Dynamic imaging with
Chapter 43 ■ Trauma Surgery: Occipitocervical Junction
Occipitocervical Junction Fusions Occipitocervical junction fusions are performed through a posterior midline approach. Ventral occipitocervical junction fusions may be technically possible, but the transoral approach for trauma is prone to infection and may be difficult to expose because of the altered anatomy and difficult to close because of the instrumentation. Finally, the ventral approach is not ideal for placement of instrumentation because the surgeon is limited in the extent of rostrocaudal exposure. At surgery, careful transfer to the prone position is required. The patient’s head is fixed with a Mayfield head clamp unless the patient is already in a halo ring/vest. In this case, it is possible to turn the patient in the halo ring/vest. After locking the halo ring to the operating table with a Mayfield halo adapter, the posterior part of the halo vest and connecting bars are disassembled to permit adequate exposure. As much as possible, the head should be positioned in an appropriate alignment such that the patient will naturally look forward (i.e., avoid hyperflexed or hyperextended positioning to maximize patient visualization and comfort). The iliac crest region is prepped to harvest bone graft. The exposure should extend from the inion down to C3 at least, with the ability to continue further caudally as necessary. Decompression of the foramen magnum should be performed if necessary, but the ability to achieve a midline fusion and take advantage of the thicker midline bone is limited by an extensive midline suboccipital decompression. Structural unicortical strips are harvested from the iliac crest along with cancellous bone. Local autograft from the cranium or posterior spinal elements is significantly less effective in achieving fusion. Allograft is least likely to achieve fusion and generally should not be relied on. Instrumentation options include inverted U rods with wiring, inverted-Y plate/screw constructions, and specialized cranial plate attachments for polyaxial cervical screws. 5,31 The midline bone is thickest and allows placement of longer screws with better purchase. The construct should be extended to at least C2 and sometimes lower to achieve optimal fixation. However, advances in instrumentation have made the longer constructs to the lower cervical spine or cervicothoracic junction uncommon unless additional subaxial cervical spine injuries exist. Dorsal C0-1 transarticular screw fixation has recently been described by Grob24 and by Gonzales et al.22 The utility of this procedure is still evolving. The instrumentation options and techniques are discussed further in Chapter 111. Postoperatively, a collar or SOMI brace is used until bony fusion occurs (usually 12 weeks). Halo immobilization is used when the bone quality or fixation is suboptimal.
approach. C1-2 fusion requires sacrifice of the movement at C1-2 (primarily rotation); therefore, for appropriate fractures with an intact transverse ligament, odontoid screw fixation may be preferable. At surgery, the positioning is similar to that used in occipitocervical junction fusions. However, if transarticular screw placement is planned, then the head should be flexed as possible to facilitate screw placement. The exposure should extend from the foramen magnum through C3. If the dorsal elements of C1 and C2 are intact and do not need to be decompressed, then structural autograft from the iliac crest is harvested for placement between or along the posterior elements of C1 and C2. Careful exposure, preparation, and decortication of the fusion surfaces are important to maximize the chances of achieving fusion. The caudal edge of C1 is a common site for nonunion and deserves special attention. Instrumentation options include C1-2 wiring alone or with additional screw instrumentation, C1-2 Halifax clamp fixation, C1-2 transarticular screw fixation, and C1-2 segmental screw fixation.30,40 The wiring options include the Brooks, Gallie, and Sonntag interspinous fusion operations.* The relative advantages and disadvantages of the various options are listed in Box 43.2. Postoperatively, a collar or SOMI brace is used until bony fusion occurs (usually 12 weeks). Halo immobilization is used when the bone quality or fixation is suboptimal. Odontoid Screw Fixation Ventral odontoid screw fixation is appropriate for many unstable C2 odontoid fractures that require operative fixation. The main advantages of odontoid screw fixation are the preservation of C1-2 mobility and the relatively *References 7,9,14,20,21,35,38.
43.2 BOX
flexion/extension views can provide an assessment of stability and bony fusion.
Advantages and disadvantages of dorsal C1-2 fusion operation
C1-2 Wiring (Brooks, gallie, sonntag)
Advantages: Familiar technique, avoids screw Disadvantages: Least rigid, requires more external fixation, higher nonunion rate C1-2 Transarticular screw fixation
Advantages: Most rigid Disadvantages: Potential for vertebral artery injury C1-2 Segmental fixation
Advantages: Familiar technique, avoids screw, very rigid Disadvantages: Venous plexus bleeding, potential vertebral artery injury C1-2 Sublaminar hooks (Halifax clamps)
Dorsal C1-2 Fixation Dorsal C1-2 fusions are indicated for unstable C1 and/or C2 fractures and are performed through a dorsal midline
521
Advantages: Avoids screw placement risks Disadvantages: Less rigid than screws, weak in extension, may narrow canal
522
Section 3 ■ Surgical Procedures
short and well-tolerated nature of the procedure. However, the procedure is not possible for many patients and fractures because of anatomic limitations. For example, patients with short necks, barrel chests, inability to tolerate cervical extension, insufficient transverse ligaments, oblique fracture lines and/or significantly comminuted fractures are poor candidates for this procedure. For these patients, a posterior C1-2 fusion is typically chosen. At surgery, the patient is positioned supine with the head extended, usually in a fixed position with a Mayfield head holder. Biplanar fluoroscopy is generally used. Using a lower cervical incision about at C5-6, a standard high ventral cervical approach is followed to the C2-3 region. A variety of standard or specialized retractors can be used to maintain exposure. Using a Kerrison rongeur or high-speed drill, a midline trough is made in the anterior-superior C3 vertebral body. Next, by using the trough, a power drill with a 2mm bit is used to drill a pilot hole from the anterior inferior border of C2, across the fracture line, and to the tip of the odontoid process. The appropriate length screw is determined by preoperative x-ray or CT measurements, intraoperative fluoroscopy, and/or by measuring the length of drill bit. If reduction of the fracture is needed, then a lag screw of an appropriately shorter length should be selected. Even without significant fracture displacement, lag screws can promote fusion by providing a compressive force across the fracture line. The screw is threaded into the pilot hole in the same trajectory. One or two screws may be placed,
but one is typically used because the outcomes appear similar.32,43 Several odontoid screw systems exist with specialized instrumentation and screws.4 One system includes cannulated screws that can be placed over a threaded drill bit (Figure 43.9).10 See Chapter 111 for additional details on the instrumentation. Patients are managed with a postoperative external orthosis (collar) until the fracture heals (usually 10 to 12 weeks). Success rates are high, with fusion rates between 81% and 96%.32,43 In the event of a nonunion, a posterior C1-2 technique can be used. Ventral C1-2 Fixation Ventral C1-2 transarticular fixation can be used if an odontoid screw fixation is not successful during an anterior approach. In addition, the approach may be used if a posterior approach is not feasible for some reason or a posterior C1-2 fusion has failed. The technique involves bilateral screws through the lateral vertebral body of C2 into the lateral mass of C1. Careful preoperative assessment of the course of the vertebral artery is essential to determine whether the procedure is feasible. At surgery, the positioning, approach, and exposure are similar to odontoid screw placement. An entry point is marked with a pilot hole at the groove between the C2 body and superior articular facet. This point is just medial to the vertebral artery. The screw trajectory is about 20 degrees lateral and rostral as needed to
Figure 43.9 C2 odontoid screw placement with a cannulated screw system. Initially a K-wire drill bit is placed (A). Next, the screw is carefully threaded over the drill bit under fluoroscopic guidance (B-C). (From Dickman CA, Spetzler RA, Sonntag, VKH [eds]: Surgery of the Craniovertebral Junction. New York, Thieme, 1998.) Continued
Chapter 43 ■ Trauma Surgery: Occipitocervical Junction
engage the C1 lateral mass securely. The screws are placed with fluoroscopic guidance (ideally biplanar) (Figure 43.10). Although placing onlay bone graft may be possible, the technique does not allow direct placement of bone between C1 and C2 and aims to have fusion occur at the articulation between the C1 and C2 lateral masses. Therefore the C1-2 facet should be scraped with a small curette as possible to promote arthrodesis. Postoperatively, a collar or SOMI brace is used until bony fusion occurs (usually 12 weeks). Halo immobilization is used when the bone quality or fixation is suboptimal. Ventral C2-3 Fixation Ventral C2-3 discectomy and fusion are used for some traumatic C2 hangman fractures that demonstrate suffi-
523
cient instability to warrant operative fixation.19 By using a high cervical incision, a standard high anterior cervical approach and limited C2-3 discectomy is performed. If there is canal compromise from osteophytes or a herniated disc, then a more extensive dorsal osteophytectomy and/or discectomy is performed. The alignment, if abnormal, is optimized as much as possible. After preparing the end plates, a C2-3 arthrodesis is performed with structural iliac crest autograft or allograft. Then C2-3 ventral cervical plating is performed (see the chapter on High Cervical and Occipitocervical Plate, Rod, Wire, and Bone Techniques and the chapter on Ventral Subaxial Cervical Fixation Techniques). Extra attention is required for the C2 screws because of the unique anatomy. A narrow lowprofile plate is preferred to facilitate placement. The patient is managed with a postoperative external orthosis (collar) for 6 or more weeks depending on the degree of preoperative instability. Dorsal C2-3 fixation does not adequately treat these fractures unless direct dorsal C2 screw placement across the fracture is used. C1-3 dorsal segmental instrumentation with intervening screw or sublaminar wiring at C2 is another alternative procedure.
Summary
Figure 43.9 cont’d
Figure 43.10 Anterior C1-2 transarticular instrumentation.
(From Barrow Neurological Institute, with permission.)
Occipitocervical junction trauma can result in a variety of injury patterns involving the regional bony, ligamentous, neurologic, and vascular structures. Because of the vital nature of these threatened structures, accurate diagnosis and careful management are required. In particular, careful attention is directed to achieving and maintaining an appropriate alignment from the onset of trauma. After initial airway management and medical stabilization, relevant diagnostic imaging should be obtained. Although many of the injuries are able to be recognized on plain radiographs, high-resolution CT scanning with multiplanar reconstruction views generally provides the most useful information. MR imaging may be difficult to obtain but is usually best to assess any spinal canal compromise and the integrity of important ligaments. Flexion/extension imaging is most useful for cooperative patients who do not have significant spinal canal compromise. The primary focus of the imaging is to identify and characterize injuries and to guide management. If instability is documented or presumed based on imaging, then some combination of external and/or internal stabilization is necessary to protect neurologic function and permit mobilization. If malalignment is present, correction with an orthosis, traction, and/or operation is considered depending on the degree of deformity and its relation to current or potential neurologic injury. When needed, traction should be performed cautiously and only with a solid understanding of the injury, as distraction can exacerbate certain injuries (e.g., occipitoatlantal dissociation). Operative procedures generally require rigid intraoperative fixation via a halo ring/adaptor or Mayfield head clamp. Surgical intervention is focused on decompressing significant compressive lesions (bone, hematoma, etc.), restoring alignment, and achieving stabilization with arthrodesis and usually internal fixation. Advances in instrumentation and surgical technique (e.g., image guid-
524
Section 3 ■ Surgical Procedures
ance, surgical innovation) have led to the development of better, stronger internal fixation constructs that can spare motion (e.g., odontoid screw fixation), reduce the number of levels to be fused, and avoid or minimize use of uncomfortable orthoses such as halos, which have inherent risks themselves (pulmonary compromise, skull pin site complications, etc.). These instrumentation techniques are discussed further in the chapter on High Cervical and Occipitocervical Plate, Rod, Wire, and Bone Techniques. Achieving bony fusion is an important goal of stabilization, and careful attention to technique is required. Although many trauma patients are good fusion candidates (young, healthy patients), liberal use of autograft is advised in most cases, because many patients may be or become critically ill and malnourished because of spinal or systemic injuries. Furthermore, nonunions can be difficult to manage and may require more substantial operative intervention. Overall, the treatment of occipitocervical injuries must be individualized on the basis of patient/injury characteristics and surgeon experience.
13.
14.
15.
16.
17.
18.
REFERENCES 1. Alker AJ, Oh YS, Leslie EV: High cervical spine and craniocervical junction injuries in fatal traffic accidents: a radiological study. Ortho Clin North Am 9:1003-1010, 1978. 2. Anderson LD, D’Alonzo RT: Fractures of the odontoid process of the axis. J Bone Joint Surg Am 56:1663-74, 1974. 3. Anderson PA, Montesano PX: Morphology and treatment of occipital condyle fractures. Spine 13:731-6, 1988. 4. Apfelbaum RI, Lonser RR, Veres R, Casey A: Direct anterior screw fixation for recent and remote odontoid fractures. J Neurosurg 93(2 Suppl):227-36, 2000. 5. Apostolides PJ, Dickman CA, Golfinos JG, Papadopoulos SM, Sonntag VK: Threaded Steinmann pin fusion of the craniovertebral junction. Spine 21:1630-7, 1996. 6. Benzel EC, Hart BL, Ball PA, Baldwin NG, Orrison WW, Espinosa M: Fractures of the C-2 vertebral body. J Neurosurg 81:206-12, 1994. 7. Brooks AL, Jenkins EB: Atlanto-axial arthrodesis by the wedge compression method. J Bone Joint Surg Am 60: 279-284, 1978. 8. Bucholz RW, Burkhead WF: The pathological anatomy of fatal atlanto-occipital dislocations. J Bone Joint Surg Am 61A:248-250, 1979. 9. Dickman CA, Crawford NR, Paramore CG: Biomechanical characteristics of C1-2 cable fixations. J Neurosurg 85: 316-22, 1996. 10. Dickman CA, Foley KT, Sonntag VK, Smith MM: Cannulated screws for odontoid screw fixation and atlantoaxial transarticular screw fixation. Technical note. J Neurosurg 83:1095-100, 1995. 11. Dickman CA, Greene KA, Sonntag VKH: Injuries involving the transverse atlantal ligament: classification and treatment guidelines based upon experience with 39 injuries. Neurosurgery 38:44-50, 1996. 12. Dickman CA, Greene KA, Sonntag VKH: Traumatic injuries of the craniovertebral junction. In Dickman CA, Spetzler RA, Sonntag, VKH (eds): Surgery of the
19.
20.
21. 22.
23.
24. 25.
26.
27.
28.
29.
30.
Craniovertebral Junction. New York, Thieme, 1998, 175-196. Dickman CA, Mamourian A, Sonntag VK, Drayer BP: Magnetic resonance imaging of the transverse atlantal ligament for the evaluation of atlantoaxial instability. J Neurosurg 75:221-7, 1991. Dickman CA, Sonntag VK, Papadopoulos SM, Hadley MN: The interspinous method of posterior atlantoaxial arthrodesis. J Neurosurg 74:190-8, 1991. Dublin AB, Marks WM, Weinstock D, Newton TH: Traumatic dislocation of the atlanto-occipital articulation (AOA) with short-term survival: With a radiographic method of measuring the AOA. J Neurosurg 52:541-546, 1980. Dvorak J, Hayek J, Zehnder R: CT-functional diagnostics of the rotatory instability of the upper cervical spine. Part 2. An evaluation on healthy adults and patients with suspected instability. Spine 12:726-31, 1987. Dvorak J, Panjabi MM, Hayek J: Diagnosis of hyper- and hypomotility of the upper cervical spine using functional computerized tomography. Orthopade 16:13-9, 1987. Effendi B, Roy D, Cornish B, Dussault RG, Laurin CA: Fractures of the ring of the axis. A classification based on the analysis of 131 cases. J Bone Joint Surg Br 63:319-27, 1981. Francis WR, Fielding JW, Hawkins RJ, Pepin J, Hensinger R: Traumatic spondylolisthesis of the axis. J Bone Joint Surg Br 63:313-8, 1981. Gallie WE: Skeletal traction in treatment of fractures and dislocations of the cervical spine. Ann Surg 106:770-776, 1937. Gallie WE: Fractures and dislocations of the cervical spine. Am J Surg 46:495-499, 1939. Gonzalez LF, Crawford NR, Chamberlain RH, Perez Garza LE, Preul MC, Sonntag VK, Dickman CA: Craniovertebral junction fixation with transarticular screws: biomechanical analysis of a novel technique. J Neurosurg 98:202-9, 2003. Greene KA, Dickman CA, Marciano FF, Drabier JB, Hadley MN, Sonntag VK: Acute axis fractures. Analysis of management and outcome in 340 consecutive cases. Spine 22:1843-52, 1997. Grob D: Transarticular screw fixation for atlanto-occipital dislocation. Spine 26:703-7, 2001. Hadley MN, Browner CM, Liu SS, Sonntag VK: New subtype of acute odontoid fractures (type IIA). Neurosurgery 22:67-71, 1988. Hadley MN, Dickman CA, Browner CM, Sonntag VK: Acute axis fractures: a review of 229 cases. J Neurosurg 71:642-7, 1989. Hadley MN, Dickman CA, Browner CM, Sonntag VK: Acute traumatic atlas fractures: management and long-term outcome. Neurosurgery 23:31-5, 1988. Hadley MN, Walters BC, Grabb PA, Oyesiku NM, Przybylski GJ, Resnick DK, Ryken TC, Mielke DH: Guidelines for the management of acute cervical spine and spinal cord injuries. Clin Neurosurg 49:407-98, 2002. Hadley MN, Walters BC, Grabb PA, Oyesiku NM, Przybylski GJ, Resnick DK, Ryken TC, Mielke DH: Management of acute central cervical spinal cord injuries. Neurosurgery 50:S166-72, 2002. Harms J, Melcher RP: Posterior C1-C2 fusion with polyaxial screw and rod fixation. Spine 26:2467-71, 2001.
Chapter 43 ■ Trauma Surgery: Occipitocervical Junction
31. Hurlbert RJ, Crawford NR, Choi WG, Dickman CA: A biomechanical evaluation of occipitocervical instrumentation: screw compared with wire fixation. J Neurosurg 90:S84-90, 1999. 32. Jenkins JD, Coric D, Branch CL Jr: A clinical comparison of one- and two-screw odontoid fixation. J Neurosurg 89:366-70, 1998. 33. Leone A, Cerase A, Colosimo C, et al.: Occipital condylar fractures: A review. Radiology 216:635-44, 2000. 34. Levine AM, Edwards CC: The management of traumatic spondylolisthesis of the axis. J Bone Joint Surg Am 67: 217-26, 1985. 35. Melcher RP, Puttlitz CM, Kleinstueck FS, Lotz, JC, Harms J, Bradford DS: Biomechanical testing of posterior atlantoaxial fixation techniques. Spine 27:2435-2440, 2002. 36. Muhonen MG, Menezes AH: Weaver syndrome and instability of the upper cervical spine. J Pediatr 116:596-9, 1990. 37. Mody BS, Morris EW: Fracture of the occipital condyle: case report and review of the world literature. Injury 23:350-2, 1992. 38. Naderi S, Crawford NR, Song GS, Sonntag VK, Dickman CA: Biomechanical comparison of C1-C2 posterior fixations. Cable, graft, and screw combinations. Spine 23:1946-55, 1998.
525
39. Powers B, Miller MD, Kramer RS, Martinez S, Gehweiler JA Jr: Traumatic anterior atlanto-occipital dislocation. Neurosurgery 4:12-17, 1979. 40. Resnick DK, Benzel EC: C1-C2 pedicle screw fixation with rigid cantilever beam construct: case report and technical note. Neurosurgery 50:426-8, 2002. 41. Sonntag VK, Hadley MN, Dickman CA, Browner CM: Atlas fractures: treatment and long-term results. Acta Neurochir Suppl (Wien) 43:63-8, 1988. 42. Spence KF Jr, Decker S, Sell KW: Bursting atlantal fracture associated with rupture of the transverse ligament. J Bone Joint Surg Am 52:543-9, 1970. 43. Subach BR, Morone MA, Haid RW Jr, et al.: Management of acute odontoid fractures with single-screw anterior fixation. Neurosurgery 45:812-9,1999. 44. Traynelis VC, Marano GD, Dunker RO, Kaufman HH: Traumatic atlanto-occipital dislocation: Case Report. J Neurosurg 65:863-870, 1986. 45. Wackenheim A: Roentgen Diagnosis of the Craniovertebral Region. New York, Springer-Verlag, 1974. 46. Wholey MH, Bruwer AJ, Baker HL: The lateral roentgenogram of the neck (with comments on the atlanto-odontoid-basion relationship). Radiology 71: 350-356, 1958.
C H A P T E R
4 4
lation of characteristic neurologic findings determined by the anatomic location of an injury. Examples include Brown-Sequard syndrome, central cord syndrome,13 anterior cord syndrome, and posterior cord syndrome (Table 44.1). A rectal examination is essential (particularly in the neurologically injured patient in order to document any degree of sacral sparing) and should be accompanied by bulbocavernosus reflex testing to assess for spinal shock. Spinal shock is the transient loss of all motor, sensory, and reflex function distal to the level of an acute spinal cord injury. The classification of a neurologic deficit as complete or incomplete cannot be determined until spinal shock has resolved. The radiographic evaluation often begins with the ATLS screening series that includes a cross-table lateral view of the cervical spine from the occiput to C7/T1. Care should be taken that the lower part of the cervical spine is completely visualized; superimposition of the shoulders may be overcome with caudally directed manual traction on the patient’s arms. Experience at multiple centers has demonstrated that most missed cervical fractures and subluxations are those present at the lower aspect of the cervical spine.8,14 A swimmer’s view often proves useful for complete visualization of the cervicothoracic junction.4 An openmouth odontoid view, an anteroposterior (AP) and lateral plain radiograph of the entire spine should be obtained if a fracture is found resulting from the frequent occurrence of noncontiguous spinal injuries. Radiographic findings suggestive of cervical instability are summarized in Table 44.2. Segmental injuries are common and the presence of injury at one level should prompt a careful search for subtle injuries elsewhere in the spine.14 Computed tomography (CT) should be routinely used to provide a more accurate delineation of osseous injuries.1,2 Sagittally reconstructed images are helpful in illustrating the sagittal alignment of the spine as well as injuries at the cervicothoracic junction. It is often helpful in demonstrating those fracture lines passing in the plane of the transaxial CT cuts. Magnetic resonance imaging (MRI) is used to further evaluate the nature and extent of neural and connective soft tissue injury. As such, MRI may be used to identify intracanilicular associated disc herniations, spinal cord contusions and often ligamentous disruption, and occult fractures.5,6 Flexion and extension dynamic radiography is frequently used in the awake neurologically intact patient with isolated neck pain and normal plain radiographs. These films are often repeated in patients with persistent neck pain to minimize masked instability secondary to acute muscle spasm.
Trauma Surgery: Cervical Spine Scott D. Daffner, Alexander R. Vaccaro, Charles Stillerman, and L. Erik Westerlund
Injury to the cervical spine should be suspected in any patient complaining of neck pain after trauma. Initial management of the multiply injured patient will be dictated by established advanced trauma life support (ATLS) protocols, with priority directed to management of airway, breathing, and circulatory compromise. The “chin lift and jaw thrust” method of securing an airway may decrease the space available for the spinal cord (beyond that seen with nasal or oral intubation) and should be avoided in the patient with a known or suspected cervical spine injury. Spinal precautions (to include cervical spine immobilization) should be maintained throughout the early stages of evaluation and resuscitation of the multi-trauma patient.3 The most common causes of injury to the neck are motor vehicle accidents (MVAs), diving into shallow water, and sporting-related activities. A thorough history of a given accident may further influence clinical suspicion for the presence of a cervical spine injury. Did the patient strike his or her head? Was there evidence of cranial impact to the windshield from inside the vehicle? Was the patient ejected? Was there any indication of weakness or paralysis noted at the accident scene? Was the patient neurologically intact at the scene with later deterioration in neurologic function? Information gathered through such questioning will guide clinical suspicion for neck injury and may provide important prognostic information when neurologic compromise is present. Obtaining information regarding prior history of injury, underlying preexisting cervical spine disease, or systemic conditions (such as ankylosing spondylitis) is important as well. The physical examination of the patient with known or suspected cervical spine injury begins at the patient’s head and progresses distally. It is complete only after a thorough evaluation of the entire musculoskeletal system has been performed.5 Abrasions or lacerations about the scalp, face, or neck provide mechanistic clues, alerting the examining physician to the potential for underlying spine trauma. The posterior cervical spine should be palpated carefully to evaluate for focal tenderness, step-off, or hematoma. Range of motion should be prohibited until the radiographic evaluation of the neck has been completed. All voluntary motion of the arms, hands, fingers, legs, feet, and toes should be observed, graded, and recorded, along with any noted sensory or deep tendon reflex compromise. Incomplete spinal cord lesions are described by a constel-
Soft-Tissue Neck Injuries Isolated soft tissue injury is a common occurrence that has been variably described as whiplash, cervical sprain, cervical strain, acceleration injury, and hyperextension injury.8,19,21,24,25 Each of these is nearly always the result of an excessive acceleration force acting to violently extend the neck beyond normal restraints. The overwhelming majority of these injuries occur as the result of MVAs.23 Symptoms may include nonfocal neck pain with or without accompanying radicular symptoms, isolated cervical 526
Chapter 44 ■ Trauma Surgery: Cervical Spine
527
TABLE 44.1
Incomplete Spinal Cord Injury Syndromes
Central cord syndrome
The central cord syndrome is the most commonly encountered of all incomplete spinal cord injuries. Its presence is characterized by upper extremity motor weakness with relative sparing of the lower extremities. Expected neurologic recovery is fair to poor.
Anterior cord syndrome
Anterior cord syndrome results from damage to the interior two - thirds of the spinal cord with sparing of the posterior third. There is loss of motor function and pain and temperature sensation. There is preservation of vibration and position sense. Potential for recovery is variable.
Posterior cord syndrome
The posterior cord syndrome is the least common. Injury to the posterior columns results in loss of vibration and position sense. There may be sparing of crude touch. Potential for functional recovery is fair.
Brown-Sequard syndrome
Brown-Sequard syndrome is an uncommon injury pattern that is secondary to injury to half of the spinal cord. This is characterized by ipsilateral motor weakness and loss of proprioception, and contralateral loss of light touch, pain, and temperature sensation. Prognosis for ambulation is excellent in this setting.
TABLE 44.2
Radiographic Findings Suggestive of Cervical Instability Direct Evidence of Instability
Indirect Evidence of Instability
Angulation > 11 degrees between adjacent segments16 Anteroposterior translation > 3.5 mm16 Segmental spinous process widening on lateral view18
Increased retropharyngeal soft-tissue margin5 Avulsion fractures at or near spinal ligament insertions Minimal compression fractures of the anterior vertebral bodies7,10,11,15 Nondisplaced fracture lines through the posterior elements or vertebral body9
Facet joint widening17 Malalignment of spinous processes on AP view Rotation of facets on lateral view12 Lateral tilt of vertebral body on AP view12
radiculopathy, cervical myelopathy, and various related lumbar pain syndromes. Closed head injuries may be associated with these injuries. Intracranial manifestations include chronic headache, concussion, extra axial/intracranial bleeding, and sympathetic dysfunction. Additionally, psychiatric changes, including sleep disturbance, depression, mood changes or frank personality changes, may occur.19 The most common radiographic finding is the loss of normal cervical lordosis as seen on a lateral plain radi-
ograph.18 Delayed flexion and extension radiographs are again obtained 2 weeks after resolution of acute muscle spasm to evaluate for evidence of potential destabilizing soft tissue disruption.20 Bone scan has a limited role in screening for occult fractures in selected patients with atypical chronic pain.19 If positive, a CT scan may then be performed for further evaluation. Early intervention and treatment is based on the presenting injury subtype, including its pathomechanics,
528
Section 3 ■ Surgical Procedures
severity, and the overall medical status of the patient.22 In the setting of a whiplash-type injury, initial use of a soft collar will improve comfort in many patients, although use should be limited to a 2- to 4-week period to minimize dependence, muscle atrophy, and decreased neck range of motion.23 Isometric exercises and gentle, supervised range of motion should be initiated as soon as symptoms permit (or within 2 weeks of injury). The regimen should be performed several times a day and should include neck flexion/extension, rotation, and lateral flexion. Enlisting the assistance of a physical therapist may be beneficial, particularly in the early phases of recovery.
Transient Quadraplegia A neurapraxia-type injury to the cervical spinal cord resulting in transient quadriplegics is most commonly seen in athletes participating in contact sports. The incidence among collegiate football players is 7.3 per 10,000 athletes. Plain radiographs are normal in this setting. The mechanism of injury is most often axial compression combined with hyperflexion or hyperextension. Sensory and motor neurologic deficits are bilateral and usually persist from several minutes to 48 hours following trauma. There is an association with developmental cervical stenosis, although effective guidelines for identification of predisposed athletes have been difficult to establish. Efforts to establish sensitive and specific screening methods to reliably identify “at-risk” athletes are under way.
Injuries to the Occipitocervical Articulation Injuries to the occipitocervical junction are being recognized with increasing frequency while patient mortality rates are declining. Improved outcomes are likely a direct benefit of present trauma protocols that begin at the scene of an injury, supporting those who would not have previously survived. Heightened suspicion and early detection (with current imaging techniques) have further contributed to the above-noted trends. The occipital condyles are paired, semi-lunar-shaped projections from the inferior aspect of the occiput that articulate with the atlantal lateral masses.26 This articulation bears lit-
tle intrinsic osseous stability, depending instead on the external and internal craniocervical ligaments for constraint. The internal craniocervical ligaments (tectorial membrane, cruciate ligament, and paired alar and apical ligaments) provide the predominance of the intrinsic occipito-atlantal stability. Injury to the craniocervical junction commonly occurs by three primary forces: distraction, compression, and lateral rotation.39 Injuries may be mild and stable or life threatening (with complete osteoligamentous disruption).33 Occipital Condyle Fractures Occipital condyle fractures are most often identified incidentally on head CT in the unconscious patient, though awake patients with complaints of deep suboccipital pain or occipital headache should be suspected of having sustained an injury to the occipital cervical junction.26,37 The neurologic examination in survivors is often normal, although mild cord injury and cranial nerve injury has been reported. Classification of occipital condyle fractures is based on CT morphology (Figure 44.1).26,38 A type I fracture is a comminuted fracture of the condyle resulting from impaction of the condyle by the lateral mass of C1. The mechanism is often a direct blow to the head. A type II injury is characterized by the presence of a related basilar skull fracture. Type III injuries are avulsion fractures occurring at the attachment site of the alar ligaments. They may be bilateral in up to 50% of cases and, in this circumstance, are associated with an atlanto-occipital dislocation. Treatment of stable type I and II injuries is cervical immobilization in a hard collar, cervicothoracic brace, or halovest for 6 to 8 weeks. Type II fractures demonstrating separation of the occipital condyle from the occiput may have inadequate lateral column support, thus requiring 8 to 12 weeks of halovest immobilization. Instability is commonly noted in type III injuries and is demonstrated by occipito-atlantal anteriorposterior displacement, longitudinal diastasis, or joint incongruity. Injuries identified as unstable are best managed with a posterior occipital-cervical arthrodesis. Occipitocervical Dislocation/Dissociation Until recently, few cases of patients surviving this entity had been reported.28-32,40 Occipitocervical dislocation or dissociation often results from high-energy trauma, is
Figure 44.1 Occipital condyle fracture classification. A type I fracture is a comminuted fracture
of the condyle resulting from the impaction of the condyle by the lateral mass of C1. The mechanism is often a direct blow to the head. A type II injury is characterized by the presence of a related basilar skull fracture. A type III injury is an avulsion-type fracture occurring at the attachment site of the alar ligaments.
Chapter 44 ■ Trauma Surgery: Cervical Spine
529
Figure 44.2 A lateral plain radiograph revealing longitudinal
diastasis of the occipital-C1 articulation.
highly unstable, and is frequently fatal (Figure 44.2). Highresolution CT (with or without MRI) is often required to evaluate these injuries because they may be difficult to appreciate on plain radiographs unless significant displacement is present (Figure 44.3A). Occipitocervical instability (subluxation and dislocation) is classified according to the direction of displacement of the occiput.27,34,36 Type I injuries are anterior subluxations of the occipital condyle relative to the atlantal lateral masses. These represent the most commonly observed injury pattern. Type II injuries are vertical displacements of the occipital condyles greater than 2mm beyond normal. C1-2 distraction injuries are included in this category. Type III injuries are posterior occipital dislocations and are exceedingly rare. In evaluating these injuries, more than 2mm of subluxation at the atlanto-occipital articulation indicates a functional loss of integrity of the major occipital-cervical stabilizers such as the alar ligaments and the tectorial membrane.33,39 Treatment of occipitocervical instability is through closed or open reduction and surgical stabilization.35 Traction is to be avoided in these injuries (Figure 44.3B,C).
Injuries to the First Cervical Vertebra Traumatic Transverse Atlantal Ligament Insufficiency Insufficiency or disruption of the transverse atlantal ligament (TAL) may occur following a violent flexion force to the upper cervical spine. Associated head injuries are common. Although survival after acute traumatic rupture had previously been thought unusual, it is now being reported
Figure 44.3 A, A sagittal MRI revealing longitudinal diastasis of
the occipital C1 and C1-2 articulation. B, Lateral x-ray film of type IIB occipital-atlantal-axial dislocation. Note that in addition to the longitudinal distraction of the occiput relative to the atlas, a distractive injury also exists at the atlantoaxial segment. Continued
with increasing frequency.43,52,67 Findings range from normal to transient quadriparesis. Permanent quadriparesis is rare given the fatal sequelae that typically follow complete injury at this level.46,52,60 Associated clinical signs include cardiac and respiratory changes secondary to brain stem compromise, or dizziness, syncope, and/or blurred vision as a result of vertebral artery disruption. Symptoms may be exacerbated by neck flexion. A lateral plain radiograph
530
Section 3 ■ Surgical Procedures
often demonstrates abnormal translation (greater than 5mm) at the atlanto-dens interval (ADI). Nonoperative treatment strategies have generally failed to provide satisfactory results, and the treatment of choice in most patients is a C1/C2 arthrodesis.
Acute disruption of the TAL may also be noted in association with a Jefferson-type burst fracture of C1.56,59,65 Treatment in this circumstance should consist of cervical immobilization for 10 to 12 weeks awaiting union of the C1 arch. Persistent instability after completion of cervical immobilization may then be addressed with a C1/C2 fusion.47,56 Traumatic Rotatory Subluxation Acute trauma is an unusual cause of acute C1-2 rotatory subluxation. The clinical presentation of C1-2 rotatory subluxation is the complaint of neck pain with findings of torticollis, and it is more commonly seen in children than adults. Four types of C1-2 rotatory injuries have been described (Figures 44-4 and 44-5).45 Type I injuries involve poor rotational changes without associated subluxation. In the type II pattern there are 3 to 5mm of displacement of C1 on C2 (with one lateral mass acting as a pivot while the other rotates anteriorly). Type III injuries have more than 5mm of forward displacement of both lateral masses. Type II and III injuries are both associated with TAL incompetence, and neurologic involvement is common. Treatment is with halo or Gardner-Wells traction-reduction, followed by external immobilization for 2 to 3 months. Delayed instability is managed with a posterior stabilization procedure. Fixed or irreducible deformities as well as delayed presentation of this condition are again best managed with surgical stabilization. Fractures of the First Cervical Vertebra
Figure 44.3 cont’d C, Postoperative lateral x-ray film demonstrates
the screw-cable-rib construct used to stabilize this occipitalatlantal-axial instability. Posterior C1-2 transarticular screw fixation was used to provide rigid fixation across the atlantoaxial level, thereby blocking rotational movement at this level. Multiple titanium cables were also placed to achieve occiput to C2 fixation. Rib was used because it conforms to the occipital-atlantal-axial contour. (B and C, from Stillerman CB, Ranjan SR, Weiss MH: Cervical spine injuries: diagnosis and management. In Wilkins RH, Rengachary SS. Neurosurgery, ed 2, vol II. New York, McGraw-Hill Medical Publishing, 1995, p 2883.)
Fractures of C1 occur either as an isolated injury or often in combination with a fracture to the C2 vertebrae. The most common associated cervical spine injuries are a type II odontoid fracture and spondylolisthetic fracture of C2.44,53-56,63 Fractures of C1 comprise up to 10% of all spine injuries and are thus encountered with relative frequency.55 Neurologic injury is unusual.41,42,54,63 Fractures of C1 are classified generally into three categories. This classification scheme has proven useful in determining treatment options, expected clinical course, and prognosis (Figure 44.6). Type I fractures are limited to involvement of the dorsal arch, are often bilateral, and will typically occur at the junction of the lateral masses and posterior arch. This is the most common pattern of C1
Figure 44.4 Classification of rotatory subluxation. Type I, simple rotatory displacement without
anterior shift. The odontoid acts as a pivot point. Type II, rotatory displacement with anterior displacement of 3 to 5mm. The lateral articular process is the pivot point. Type III, rotatory displacement with anterior displacement of more than 5mm. Type IV, rotatory displacement with posterior translation. (From Fielding JW, Hawkins RJ Jr.: Atlanto-axial rotatory fixation [fixed rotatory subluxation of the atlanto-axial joint]. J Bone Joint Surg 59A:37-44, 1977.)
Chapter 44 ■ Trauma Surgery: Cervical Spine
fracture and likely occurs secondary to hyperextension in conjunction with an axial load. A type II atlas fracture is a unilateral lateral mass injury that occurs as the result of an asymmetrically applied axial load. Intraarticular extension is not common but is reported.50 A type III (or Jefferson) fracture is a burst-type fracture that involves three or more fractures through the anterior and posterior aspects of the C1 ring. The mechanism of this second most common pattern is that of a pure axially applied load. Plain radiographs are useful in the evaluation of these injuries and will often demonstrate widening of the retropharyngeal soft tissue shadow from C1 to C3 (though these changes may take 6 or more hours to develop).61,66 The open-mouth odontoid view will show lateral displacement of the lateral masses in a Jefferson-type fracture and may appear normal with the more common Type I posterior arch fracture. If total combined lateral displacement of the C1 lateral masses over C2 is greater than 6.9mm,42,63,66 the transverse atlantal ligament has been disrupted, resulting in an unstable injury.48,53,56,65 Type II fractures appear radiographically as unilateral displacement of the affected lateral mass on an open-mouth odon-
toid radiograph. Improvements in technique and image quality have made CT in the plane of the C1 ring helpful in fully defining these injuries. The most important factor governing treatment and outcome is the occurrence of other simultaneous occurring injuries.44,53,56 Treatment of isolated C1 fractures has traditionally been nonoperative, although some European centers have reported the successful surgical reduction and stabilization of markedly displaced Jefferson burst fractures. Results with nonoperative treatment have been good,50,64 although mild neck pain is a chronic sequelae in up to 80% of these patients.54 There has been no reported correlation between fracture union/nonunion and functional outcome.*
Fractures of the Second Cervical Vertebra Fractures of the Odontoid Process Fractures of the odontoid process of the axis are relatively common among injuries of the upper cervical spine, although the exact prevalence is not well established. Odontoid fractures in young adults are most often secondary to high-energy trauma, such as MVAs or violent blows to the head (57% to 81%).69,72,76,83 Those sustained by the elderly or very young are more commonly due to lowerenergy falls.74,86,89 As with other upper cervical injuries, the importance of clinical suspicion is critical to early recognition, since several studies have reported a high incidence of missed injuries, especially in patients with depressed mental status. The degree of neurologic involvement is widely variable; however, the majority of patients have a normal neurologic examination. Odontoid fractures are best visualized on lateral and open-mouth AP plain radiographs, as well as on reformatted sagittal CT images (because routine axial imaging may miss the fracture). The most widely adopted classification system is that proposed by Anderson and D’Alonzo69 following their experience with 60 patients with odontoid fractures treated over an 8-year period. This classification identifies three fracture types based on the anatomic location of the fracture line (Figure 44.7). Type I fractures are the least common and are described as an oblique fracture involving the
Figure 44.5 A three-dimensional CT scan revealing a traumatic
rotatory dislocation of C1-2.
531
*References 49,51,54,57,58,62.
Figure 44.6 Classification of fractures of C1. Type I fractures are limited to involvement of the
posterior arch, are often bilateral, and will typically occur at the junction of the lateral masses and posterior arch. A type II atlas fracture is usually a unilateral injury defined by involvement of the lateral mass (with fracture lines passing both anteriorly and posteriorly) as the result of an asymmetrically applied axial load. A type III (or Jefferson) fracture is a burst-type fracture that involves three or more fractures through the anterior and posterior aspect of the C1 ring.
532
Section 3 ■ Surgical Procedures
superior tip of the dens. Type II odontoid fractures occur at the junction of the base of the dens and the body of the axis. This is the most common of the three types and the most controversial when discussing treatment.72,75 Type II fractures have the highest rate of nonunion when treated nonoperatively, especially in the elderly. In type III fractures, the fracture line occurs in the body of the axis (primarily involving cancellous bone). Isolated type I odontoid fractures are considered stable (unless they are associated with instability involving the occipital-cervical junction) and may be treated with a Philadelphia collar or similar orthosis. Type III fractures are often successfully managed with collar or halo immobilization. Type II fractures, however, lack both periosteum and cancellous bone at the fracture site, increasing the propensity for non-union.68 Fractures that are significantly displaced may be realigned with traction reduction and immobilized with a halo vest until definitive treatment measures are selected. Factors considered to be associated with non-healing of Type II fractures include the degree of displacement, angulation, age of patient, loss of fracture reduction, and medical co-morbidities. Surgical stabilization, when chosen, may proceed through an anterior or posterior approach, depending on patient variables and fracture subtype.70,77
quently a result of an MVA, is similar in terms of location to the originally described hangman’s fracture, but from a mechanistic standpoint bears little resemblance to the fracture subtype characteristic of judicial hanging.79,80 The majority of these injuries are the result of motor-vehicle trauma and are infrequently associated with injury to the spinal cord (6.5%).73 The basic mechanism of injury is that of rapid deceleration with forward flexion, axial impaction of the head on the windshield, followed by hyperextension after impact.79,80 Each of the three primary fracture types77-79 (Figure 44.8) are characterized further by variations of this mechanism. Dynamic radiography may be required to differentiate injury types. Type I fractures occur through the neural arch in the region just posterior to the vertebral body. There is less than 3mm of translation and no angulation at the fracture site. This fracture subtype is the result of hyperextension and axial load. These may be treated with immobilization in a cervical orthosis for 3 months. Type II fractures are divided further into type II and type IIA injuries. Type II fractures have greater than 3mm of displacement and significant angulation. The mechanism of injury is a combined force comprised of hyperex-
Traumatic Spondylolisthesis of the Axis Traumatic spondylolisthesis of the axis is a bipedicular fracture of the second vertebrae that has been of interest for decades given its unique distinction as the “hangman’s fracture.71,82,84,85,88 The lesion encountered today, fre-
Type I
Type I Type II
Type II
Type II
Type III
Figure 44.8 Classification of traumatic spondylolisthesis of the Figure 44.7 Classification of odontoid fractures. Type I fractures
are the least common and are described as an oblique fracture involving the superior tip of the dens. Type II odontoid fractures occur at the junction of the base of the dens and the body of the axis. This is the most common type. Type III odontoid fractures are characterized by the fracture line passing through the cancellous bone of the vertebral body.
axis. Type I fractures occur through the neural arch in the region just posterior to the vertebral body. There is less than 3mm of translation and no angulation at the fracture site. Type II fractures have greater than 3mm of displacement and significant angulation. Type III fractures describe a type I (pars) fracture with an associated bilateral facet dislocation at C2-3. The critical feature is the classic presence of a freefloating posterior arch of C2.
533
Chapter 44 ■ Trauma Surgery: Cervical Spine
tension and axial loading (extension immediately followed by flexion). Fracture reduction may be achieved with skeletal traction in extension with immediate or delayed conversion to halo-vest immobilization. Surgical stabilization is infrequently necessary, although several surgical options exist in patients in whom a reduction cannot be maintained or in those unable to tolerate prolonged halo traction or halo-vest immobilization. In reducible fractures a primary pedicle screw osteosynthesis of C2 is possible. In fractures that are not anatomically reducible, or in cases of displaced nonunion, an anterior C2-3 arthrodesis is a viable treatment option (Figure 44.9). Type IIA fractures are distinguished by an oblique fracture line often running from posterior-superior to anterior-inferior along the length of the pars. The mechanism is a flexion-distraction force. This fracture subtype is seen less than 10% of hangman’s fractures. Reduction is by extension and slight axial loading; axial traction will accentuate the deformity. Reduction should be followed by immobilization in a halo-vest for 3 months. Type III fractures describe a type I (pars) fracture with an associated bilateral facet dislocation at C2-3. The critical feature is the classic presence of a free-floating posterior arch of C2. These are unstable and irreducible by closed means, requiring surgical intervention. An additional group of injuries may also be described as traumatic spondylolisthesis of C2-3 with either bilateral
laminar fractures (Type IV) or bilateral facet fractures of the inferior articular processes of C2 (Type V). The mechanism of both types is flexion or shear, producing a highly unstable pattern.78-81,87
Injuries to the Lower Cervical Spine The C3 through C7 vertebrae are similar in anatomy, biomechanics, and generally incur fracture patterns that are similar. However, the C7 vertebrae is exposed to greater axial compression and flexion load because of its location at the junction of the cervical and thoracic spine. Closed indirect injuries to the head and neck often produce patterns of injury that are therefore characteristic to the lower cervical vertebral column. The most severe neurologic sequelae arise as a result of a translational deformity, establishing ligamentous integrity as critically important to stability and treatment. Both two- and three-column models are presently used in discussing traumatic pathoanatomy of the lower cervical spine. The three-column model was originally described in 1984 with specific reference to thoracolumbar injuries93 but has since been modified to address cervical spine stability. It may be of greater utility to discuss the cervical spine as a two-column entity composed of an anterior and posterior column.90,96,97,102 In the two-column model, the anterior spine consists of the posterior longitudinal ligament (PLL) and all remaining ventral structures, while the posterior column is comprised of all structures dorsal to the PLL (Table 44.3). The anterior and posterior columns are then reciprocally affected by flexion and extension moments.96 A mechanistic classification of subaxial cervical spine injuries was described by Allen et al.90 in 1982. This classification divides mid- and low-cervical fractures into six groups based on force vector (initial dominant force) and subsequent incremental tissue failure (based on the attitude of the spine at failure). Abnormal relationships between adjacent vertebrae imply ligamentous failure, suggesting a shear force mechanism (since ligaments do not fail in compression). The three most common injury groups are compressive flexion, compressive extension, and distractive flexion. Vertical compression injuries occur with intermediate frequency, whereas distractive extension and lateral flexion injuries occur the least.90 The presence of neurologic injury has not been strongly associated with any individual group within the classification, although it is
TABLE 44.3
The Two-Column Model of the Lower Cervical Spine
Figure 44.9 A lateral plain radiograph following an anterior
C2-3 fusion for late instability of a type II hangman’s fracture.
Anterior Column Components
Posterior Colum Components
Anterior longitudinal ligament Intervetebral disc & anulus fibrosus Vertebral body Posterior longitudinal ligament
Pedicles and posterior vertebral arch Posterior interspinous ligament complex
534
Section 3 ■ Surgical Procedures
related to progressive osteoligamentous disruption or the severity of injury within a particular subgrouping. Injuries as identified on plain radiographs should undergo further evaluation with CT scanning and possibly MRI. Assessment of plain radiographic, CT, and MRI findings assist in the evaluation of spinal stability. Compressive Flexion Compressive flexion injuries are caused by a ventral and axially directed load of increasing intensity. Compressive fractures without facet fracture or subluxation are usually stable injuries. Higher stages of injury involve increased ventral osseous and dorsal ligamentous injury and may be unstable (Figures 44-10 and 44-11). Treatment is tailored accordingly, although a frequent complication is late instability with conservative management.* Surgical intervention often involves a cervical corpectomy and instrumented fusion with a structural graft. Adjunctive dorsal stabilization may be necessary in highly unstable advanced-stage lesions. Compressive Extension Compressive extension injuries result in a spectrum of pathology ranging from unilateral vertebral arch fractures to bilateral laminar fractures and finally to vertebral arch fractures with full ventral displacement of the vertebral body (Figure 44.12). Management is tailored based on injury severity and instability. An initial dorsal reduction *References 91,92,94,95,99-101,103.
Stage 1
and stabilization procedure is often required, followed by adjunctive ventral stabilization if necessary. Distractive Flexion Distractive flexion injuries are also known as the flexion dislocation injuries. There is typically little osseous injury except for minor compression failure of the caudal vertebral segment. However, there is severe ligamentous damage involving the dorsal facet capsule complex, ligamentum flavum, and interspinous ligaments, and (depending on the presence of a unilateral or bilateral dislocation) injury to the posterior longitudinal ligament and intervertebral disc (Figures 44-13 and 44-14). A significant number of patients with this injury have also had an associated closed head injury.98 Radiographic changes may be minimal in the early stages (flexion sprain) of this injury subtype. MRI is often useful to delineate the full extent of soft tissue disruption (including injury to the disc), although obtaining this study in an awake, alert, and cooperative patient should not delay traction reduction when plain radiographs demonstrate a translational displacement. Some physicians recommend a prereduction MRI prior to closed or open reduction of this injury subtype. All injuries in this family should be considered at risk for further displacement, making surgical stabilization the primary mode of treatment. Following a successful closed reduction, MRI should be obtained to evaluate for the presence of a herniated disc (Figure 44.15). If present, a ventral decompression and stabilization is the preferred surgical approach. If a closed reduction is not feasible, the surgical approach is predicated on the presence of an
Stage 2
Stage 4
Stage 3
Stage 5
Figure 44.10 Compression flexion injury. Stage 1: Blunting and rounding-off of anterosuperior
vertebral margin. Stage 2: Loss of anterior vertebral height with anteroinferior beaking. Stage 3: Fracture line extending from anterior surface of vertebral body extending obliquely through the subchondral plate (fractured beak). Stage 4: Less than 3mm of the posteroinferior vertebral margin into the neural canal. Stage 5: Greater than 3mm of displacement of the posterior aspect of the vertebral body with complete disruption of the posterior ligamentous complex. The vertebral arch is intact. (From Rizzolo SJ, Cotler JM: Unstable cervical spine injuries:specific treatment approaches. J Am Acad Orthop Surg 1:57-66, 1993.)
535
Chapter 44 ■ Trauma Surgery: Cervical Spine
Figure 44.11 (A) A sagittal CT reconstruction revealing an advanced stage flexion compression
cervical spinal injury. (B) A plain radiograph following an anterioposterior cervical decompression and stabilization procedure.
Stage 1a
Stage 2
Stage 1b
Stage 3
Stage 1c
Stage 4
Stage 5
Figure 44.12 Compression extension injury. Stage 1: Unilateral vertebral arch fracture through
the articular process (stage 1a), the pedicle (stage 1b), or lamina (stage 1c), either with or without a rotary spondylolisthesis of the centrum. Stage 2: Bilaminar fracture at one or more levels. Stage 3: Bilateral fractures of the vertebral arch with partial-width anterior vertebral body displacement. State 4: Partial-width anterior vertebral body displacement. Stage 5: Complete anterior vertebral body displacement. (From Rizzolo SJ, Cotler JM: Unstable cervical spine injuries: specific treatment approaches. J Am Acad Orthop Surg 1:57-66, 1993.)
536
Section 3 ■ Surgical Procedures
Stage 1
Stage 2
Stage 4 Stage 3
Figure 44.13 Distractive flexion. Stage 1: Flexion sprain injury with facet subluxation in flexion
and divergence of spinous processes. There may be some blunting of the anterosuperior vertebral margin (similar to stage 1 compression-flexion injury). Stage 2: Unilateral facet dislocation with or without rotary spondylolisthesis. Stage 3: Bilateral facet dislocation with up to 50% vertebral body displacement. Stage 4: Completely unstable motion segment with full-width vertebral body displacement. (From Rizzolo SJ, Cotler JM: Unstable cervical spine injuries: specific treatment approaches. J Am Acad Orthop Surg 1:57-66, 1993.)
extruded disc fragment. If present, a ventral decompression is required with or without an attempted ventral open reduction followed by a stabilization procedure. In the absence of an extruded disc fragment, a dorsal open reduction and stabilization procedure is often sufficient to obtain adequate spinal stability. Vertical Compression A vertical compression fracture is described as a cervical burst fracture caused by an axial-loading mechanism. Osseous failure is considered more significant by far than injury to the ligamentous structures in this injury (Figure 44.16). Treatment with halo immobilization is usually sufficient, although injuries at the cervical thoracic junction (C7) have a tendency to settle into kyphosis, which may require surgical intervention.95 Ventral surgical decompression and stabilization is often necessary in patients with an incomplete neurologic deficit.
may occur in a ventral-to-dorsal direction through the vertebral body. Less severe injuries may have little displacement, making radiographic detection difficult.90,99 The presence of a ventral avulsion fracture resulting from an avulsion of the anterior longitudinal ligament may provide a clue to this injury type. This injury is especially unstable in a patient with ankylosing spondylitis or diffuse idiopathic skeletal hyperostosis (DISH) where there exist two rigid moment arms joined at an unstable junction. Distraction extension injuries are commonly associated with neurologic impairment. Patients frequently present with neurologic evidence of a central cord syndrome with significant weakness involving the upper extremities and relative sparing of the lower extremities. Spontaneous recovery is common.90 Most distractive extension injuries without disc space disruption are stable and may be treated nonoperatively with late flexion-extension radiographs to confirm stability. Unstable injuries benefit from a ventral reconstructive procedure with ventral plate fixation acting as a ventral tension band.
Distractive Extension Distractive extension injuries are usually caused by forces acting to place the anterior elements under tension (Figures 44-17 to 44-20). Ventral disc space widening is the characteristic radiographic finding, although failure
Lateral Flexion Lateral flexion injuries are secondary to asymmetric compressive loading resulting in unilateral vertebral body compression failure and ipsilateral posterior arch fracture
537
Chapter 44 ■ Trauma Surgery: Cervical Spine
Figure 44.15 A sagittal MRI revealing significant cord edema
and hemorrhage at the level of a C6-7 bilateral facet dislocation (type IV distractive flexion injury). Note the soft tissue density behind the body of C6, which may represent an extruded disc fragment.
Stage 1
Stage 2
Figure 44.14 (A) A lateral plain radiograph revealing evidence
of a C4-5 unilateral facet dislocation (type II distractive flexion injury). Note the 25% anterior subluxation of C4 on C5. (B) A transaxial CT scan revealing a left-sided unilateral facet dislocation. Note that the left C4 inferior articular process is anterior to the left C5 superior articular process.
Stage 3
Figure 44.16 Vertical compression injury. Stage 1: Central
“cupping fracture” of the superior or inferior vertebral end plate. Stage 2: Fracture of both superior and inferior end plates. Stage 3: Displacement and fragmentation of the vertebral body. (From Rizzolo SJ, Cotler JM: Unstable cervical spine injuries: specific treatment approaches. J Am Acad Orthop Surg 1:57-66, 1993.)
538
Section 3 ■ Surgical Procedures
(Figure 44.21). As noted previously, this is the least common pattern of lower cervical spine disruption and is often stable, requiring cervical immobilization for 6 to 12 weeks.
Summary Stage 1
Stage 2
Figure 44.17 Distractive extension injury. Stage 1: Failure of
anterior ligamentous complex, which may present as a widening of the disc space or a nondeforming transverse fracture through the centrum. Stage 2: Injury may be identified radiographically by an anterior marginal avulsion fracture of the centrum. Posterior ligamentous disruption may be identified by posterior displacement of the superior vertebra. (From Rizzolo SJ, Cotler JM: Unstable cervical spine injuries: specific treatment approaches. J Am Acad Orthop Surg 1:57-66, 1993.)
Over half of the reported 50,000 annual new spinal cord injuries seen in the United States each year occur in the cervical spine. Of these, 11,000 have some degree of permanent deficit. Further advancement and implementation of advanced trauma life support (ATLS) protocols may be expected to continue producing increases in the incidence of multi-trauma survivors with proportional increases in the incidence of cervical spine trauma presenting in our emergency departments. These trends and statistics underscore the socioeconomic importance of cervical spine trauma, and they add greater emphasis to the critical nature of early injury recognition, evaluation, and proper treatment.
Figure 44.18 (A) A sagittal MRI revealing a distraction extension injury at the C4-5 level with
retrolisthesis of C4 on C5. (B) A plain lateral radiograph following an anterior tension band (instrumented fusion) reconstruction of this injury.
Chapter 44 ■ Trauma Surgery: Cervical Spine
Figure 44.19 (A) A lateral plain x-ray revealing a high-grade distraction extension cervical injury
at the C4-5 level. (B) A lateral plain x-ray following a posterior-anterior reconstruction procedure to obtain adequate spinal stability.
539
540
Section 3 ■ Surgical Procedures
Figure 44.20 (A) A sagittal MRI revealing a high-grade distraction extension injury at the C7-T1
level. (B) The patient underwent a posterior open reduction and stabilization procedure using a cervicothoracic plate rod implant to obtain adequate spinal stability.
Stage 1
Stage 2
Figure 44.21 Lateral flexion. Stage 1: Asymmetric compression fracture of the centrum with
associated ipsilateral vertebral arch fracture seen on lateral radiograph; no displacement is noted on the anteroposterior (AP) view. Stage 2: Displacement is evident on the AP as well as the lateral radiograph. There may also be tension failure of the contralateral ligaments and facet joint. (From Rizzolo SJ, Cotler JM: Unstable cervical spine injuries: specific treatment approaches. J Am Acad Orthop Surg 1:57-66, 1993.)
Chapter 44 ■ Trauma Surgery: Cervical Spine
REFERENCES 1. Alker G: Computed Tomography With and Without Myelography. Philadelphia, JB Lippincott, 1989. 2. Allen BL, Ferguson RL, Lehman TR, Obrien RP: A mechanistic classification of closed indirect fractures and dislocations of the lower cervical spine. Spine 7:1-27, 1982. 3. American College of Surgeons: Advanced Trauma Life Support Manual. Chicago, American College of Surgeons, 1984. 4. Clark CR, Ingram CM, El Khoury GY, Ehara S: Radiographic evaluation of cervical spine injuries. Spine 13:742-747, 1988. 5. Dunn EJ, Blazer S: Soft tissue injuries to the lower cervical spine. Instr Course Lect 36:499-512, 1987. 6. Goldberg AL, Rothfus WE, Deeb ZL: The impact of magnetic resonance on the diagnostic evaluation of acute cervicothoracic spinal trauma. Skeletal Radiol 17:89-95, 1988. 7. Herkowitz HN, Rothman RH: Subacute instability of the cervical spine. Spine 9:348-357, 1984. 8. Kirshenbaum KJ, Nadimpalli SR, Fantus R, Cavallino RP: Unsuspected cervical spine fractures associated with significant head trauma: the role of CT. J Emerg Med 8:183-198, 1990. 9. Lee C, Kim KS, Rogers LF: Sagittal fractures of the cervical vertebral body. AJR 139:55-60, 1982. 10. Mazur JM, Stauffer ES: Unrecognized spinal instability associated with seemingly simple cervical compression fractures. Spine 8:687-692, 1983. 11. Mori S, Nobuhiro O, Ojima T, et al: Observation of “tear drop” fracture-dislocation of the cervical spine by CT. J Jpn Orthop Assoc 57:373-378, 1983. 12. Scher AT: Unilateral locked facet in cervical spine injuries. AJR 129:45-48, 1977. 13. Schneider RC, Thompson JM, Bebin J. The syndrome of acute central cervical spinal cord injury. J Neuro Neurosurg Psych 21:216-227, 1958. 14. Vaccaro AR, An HS, Lin S, et al: Non-contiguous injuries to the spine. J Spine Disord 5:320-329, 1992. 15. Webb JH: Hidden flexion injury of the cervical spine. J Bone Joint Surg 55B:322-327, 1976. 16. White AA, Punjabi MM: Update on the evaluation of instability of the lower cervical spine. Instr Course Lect 36:499-520, 1987. 17. Woodring JH, Goldstein SJ: Fractures of the articular process of the spine. Am J Radiol 139:341-344, 1982. Soft-Tissue Neck Injuries 18. Daffner RH: Evaluation of cervical cerebral injuries. Semin Roentgenol 27:239-253, 1992. 19. Gay Jr, Abbott KH: Common whiplash injuries of the neck. JAMA 152:1968-2704, 1953. 20. Herkowitz HN, Rothman RH: Subacute instability of the cervical spine. Spine 9:348-357, 1984. 21. Hohl M: Soft tissue injuries of the neck in automobile accidents: factors influencing prognosis. J Bone Joint Surg (Am) 56:1675-1682, 1974. 22. Joliffe VM: Soft tissue injury of the cervical spine: consider the nature of the accident. BMJ 307:439-440, 1993. 23. LaRocca HL. Cervical spine syndrome: diagnosis, treatment and long term outcome. In Frymoyer JW,
541
Ducker TB, Hadler NM, et al. (eds): The Adult Spine. New York, Raven Press, 1991, pp 1051-1062. 24. McNab I: Acceleration injuries of the cervical spine. J Bone Joint Surg (Am) 46:1797-1799, 1964. 25. Wickstrom JK, Martinez JL, Rodriguez RPO, Haines DM: Hyperextension and hyperflexion injuries to the head and neck of primates. In Neckache and Backache, Springfield Press, 1970, pp 108-117. Injuries to the Occipitocervical Articulation 26. Anderson PA, Montesano PX: Morphology and treatment of occipital condyle fractures. Spine 13:731-736, 1988. 27. Anderson PA, Montesano PX: Traumatic Injuries of the Occipito-Cervical Articulation. In Camins, O’Leary (eds): Disorders of the Cervical Spine. Baltimore, Williams & Wilkins, 1992. 28. DeBeer JDV, Thomas M, Walters J, Anderson P. Traumatic atlanto-axial subluxation. J Bone Joint Surg (Br) 70B:652-655, 1988. 29. Dibenedetto T, Lee CK: Traumatic atlanto-occipital instability: a case report with follow-up and a new diagnostic technique. Spine 15:595-597, 1990. 30. Eismont FJ, Bohlman HH: Posterior atlanto-occipital dislocation with fractures of the atlas an odontoid process: report of a case with survival. J Bone Joint Surg (Am) 60:397-399, 1978. 31. Evarts CM: Traumatic occipito-atlantal dislocation: report of a case with survival. J Bone Joint Surg (Am) 52:16531660, 1970. 32. Gabrielsen TO, Maxwell JA: Traumatic atlanto-occipital dislocation. AJR 97:624-629, 1966. 33. Harris JR Jr, Carson GC, Wagner LK, Kerr N: Radiographic diagnosis of traumatic occipitovertebral dissociation: comparison of three methods of detecting occipitovertebral relationships on lateral radiographs of supine subjects. AJR. 162:887-892, 1994. 34. Jevitch V: Traumatic lateral atlanto-occipital dislocation with spontaneous bony fusion: a case report. J Spinal Disord 4:251-263, 1991. 35. Koop SE, Winter RB, Lonstein JE: The surgical treatment of instability of the upper part of the spine in children and adolescents. J Bone Joint Surg (Am) 66:403-411, 1984. 36. Montane I, Eismont FJ, Green BA: Traumatic occiptoatlantal dislocation. Spine 16:112-116, 1991. 37. Stroobants J, Fidlers L, Storms JL, et al: High cervical pain and impairment of skull mobility as the only symptoms of an occipital condyle fracture: a case report. J Neurosurg 81:137-138, 1994. 38. Wackenheim A: Roentgen Diagnosis of the Craniovertebral Region. Berlin, Springer-Verlag, 1974. 39. White AA, Panjabi MM: Kinematics of the spine. In Clinical Biomechanics of the Spine (ed 2). Philadelphia, JB Lippincott, 1990, pp 92-97. 40. Zigler JE, Waters RL, Nelson RW, et al: Occipitocervico-thoracic spine fusion in a patient with occipitocervical dislocation and survival. Spine 11:645-646, 1986. Injuries to the First Cervical Vertebra 41. Alker G, Oh Y, Leslie E, et al: Post mortem radiology of head and neck injuries in fatal traffic accidents. Radiology 114: 611-617, 1975.
542
Section 3 ■ Surgical Procedures
42. Bucholz R, Burkhead W: The pathological anatomy of fatal atlanto-occipital dislocations. J Bone Joint Surg (Am) 61:248-250, 1979. 43. Dunbar HS, Ray BS: Chronic atlanto-axial dislocations with late neurologic manifestations. Surg Gyncol Obstet 113:747, 1961. 44. Esses S, Langer F, Gross A: Fracture of the atlas associated with fracture of the odontoid process. Injury 12:310-312, 1981. 45. Fielding JW, Hawkins RJ, Hensinger RN, et al: Atlantoaxial rotary deformities. Orthop Clin North Am 9: 955-967, 1978. 46. Gonzalez TA, Vance M, Helper M: Legal Medicine: Pathology and Toxicology. New York, Appleton Century, 1940, p 312,. 47. Hamilton AR. Injuries of the atlanto-axial joint. J Bone Joint Surg (Br) 33:434-435, 1951. 48. Hatchette S. Isolated fracture of the atlas. Radiology 36:233-235, 1941. 49. Highland T, Salciccioli G: Is immobilization adequate treatment of unstable burst fractures of the atlas? A case report with long-term follow-up evaluation. Clin Orthop 201:196-200, 1985. 50. Hinchey J, Bickel W: Fractures of the atlas: review and presentation of data on eight cases. Ann Surg 121: 826-832, 1945. 51. Jefferson G: Remarks on fractures of the first cervical vertebra. Br Med J 2:153-157, 1927. 52. Krantz P: Isolated disruption of the transverse ligament of the atlas: an injury easily overlooked at post-mortem examination. Injury 12:168-170, 1980. 53. Levine A: Avulsion of the transverse ligament associated with a fracture of the atlas: a case report. Orthopaedics 6:1467-1471, 1983. 54. Levine A, Edwards C: Fractures of the atlas. J Bone Joint Surg (Am) 73:680-691, 1991. 55. Levine A, Edwards C: Treatment of injuries in the C1/C2 complex. Orthop Clin North Am 17:31-44, 1986. 56. Lipson SJ: Fractures of the atlas associated with fractures of the odontoid process and transverse ligament ruptures. J Bone Joint Surg (Am) 59:940-943, 1977. 57. Marlin A, Williams G, Lee J: Jefferson fractures in children. J Neurosurg 58:277-279, 1983. 58. Milward F: An unusual case of fractures of the atlas. BMJ 1:458, 1933. 59. O’Brien JJ, Butterfield WL, Gossling HR: Jefferson fracture with disruption of the transverse ligament. Clin Orthop 126:135-138, 1977. 60. Pennecot GF, Leonard P, DesGachons SP, et al: Traumatic ligamentios instability of the cervical spine in children. J Pediatr Orthop 4:339-345, 1984. 61. Penning L: Paravertebral hematoma in cervical spine injury: incidence and etiologic significance. AJR 136: 553-561, 1981. 62. Plaut H: Fracture of the atlas resulting from automobile accidents: a survey of the literature and a report of 6 cases. AJR 40:867-890, 1938. 63. Segal L, Grimm J, Stauffer E: Non-union of fractures of the atlas. J Bone Joint Surg (Am) 69:1423-1434, 1987. 64. Sherk H, Nicholson J: Fractures of the atlas. J Bone Joint Surg (Am) 52:1017-1024, 1970.
65. Spence KF, Decker S, Sell KW: Bursting atlantal fracture associated with rupture of the transverse ligament. J Bone Joint Surg (Am) 52:543-549, 1970. 66. Templeton P, Youhg J, Mirvis S, et al: The value of retropharyngeal soft tissue measurements in trauma of the adult cervical spine: cervical spine soft tissue measurements. Skel Radiol 16:98-104, 1987. 67. Wigren A, Anici F: Traumatic atlanto-axial dislocation without neurologic disorder, J Bone Joint Surg (Am) 55:642, 1973. Fractures of the Second Cervical Vertebrae 68. Amling M, Hahn M, Wening VJ, et al: The microarchitecture of the axis as the predisposing factor for fracture of the base of the odontoid process. J Bone Joint Surg (Am) 76:1840-1846, 1994. 69. Anderson LD, D’Alonzo RT: Fractures of the odontoid process of the axis. J Bone Joint Surg (Am) 56:1663-1674, 1974. 70. Bohler J: Anterior stabilization for acute fractures and nonunions of the dens. J Bone Joint Surg (Am) 64:18-27, 1982. 71. Bucholz RW: Unstable hangman’s fractures. Clin Orthop 154:119-124, 1981. 72. Clark CR, White AA III: Fractures of the dens: a multicenter study. J Bone Joint Surg (Am) 67:1340-1348, 1985. 73. Francis WR, Fielding JW, Hawkins RJ, et al: Traumatic spondylolisthesis of the axis. J Bone Joint Surg (Br) 63:313-318, 1981. 74. Hanigan WC, Powell FC, Elwood PW, Henderson JP: Odontoid fractures in elderly patients. J Neurosurg 78: 32-35, 1993. 75. Holsbeeck EV, Stoffelen D, Fabry G: Fractures of the odontoid process: conservative and operative treatment, prognostic factors. Acta Orthop Belg 59:17-21, 1993. 76. Husby J, Sorensen KH: Fractures of the odontoid process of the axis. Acta Orthop Scand 45:182-192, 1974. 77. Jeanneret B, Vernet O, Frei S, Magerl F: Atlantoaxial mobility after screw fixation of the odontoid: a computed tomographic study. J Spinal Disord 4:203-211, 1991. 78. Levine AM, Edward CC: The management of traumatic spondylolisthesis of the axis. J Bone Joint Surg (Am) 67:217-226, 1985. 79. Levine AM, Edwards CC: Treatment of injuries in the C1/C2 complex. Orthop Clin North Am 17:31-44, 1986. 80. Levine AM, Edwards CC: Traumatic lesions of the occipitoatlantoaxial complex. Clin Orthop 239:53-68, 1989. 81. Levine Am, Rhyne Al: Traumatic spondylolisthesis of the axis. Semin Spine Surg 3:47-60, 1991. 82. Mollan RAB: Hangman’s fracture: injury. Br J Accident Surg 14(3):265-267, 1982. 83. Osgood RB, Lund CC: Fractures of the odontoid process. N Engl J Med 198:61-72, 1928. 84. Pepin JW, Hawkins RJ: Traumatic spondylolisthesis of the axis: hangman’s fracture. Clin Orthop 157:133-138, 1981. 85. Schneider RC, Livingston KE, Cave AJE, et al: “Hangman’s fracture” of the cervical spine. J Neurosurg 22:141-154, 1965. 86. Seimon LP: Fracture of the odontoid process in young children. J Bone Joint Surg (Am) 59:943-947, 1977.
Chapter 44 ■ Trauma Surgery: Cervical Spine
87. Starr JK, Eismont FJ: Atypical hangman’s fracture. Spine 18:1954-1957, 1993. 88. Termansen NB: Hangman’s fracture. Acta Orthop Scand 45:529-539, 1974. 89. Wisoff HS: Fracture of the dens in the aged. Surg Neurol 22:547-555, 1984. Injuries to the Lower Cervical Spine 90. Allen BL, Ferguson RL, Lehman TR, Obrien RP: A mechanistic classification of closed, indirect fractures and dislocations of the lower cervical spine. Spine 7:1-27, 1982. 91. Capen DA, Nelson RW, Zigler J, et al: Surgical stabilization of the cervical spine: a comparative analysis of anterior and posterior spine fusions. Paraplegia 25: 111-119, 1987. 92. Capen DA, Zigler J, Garland DE: Surgical stabilization in cervical spine trauma. Contemp Orthop 14:25-32, 1987. 93. Denis F: Spinal instability as defined by the three column spine concept in acute spinal trauma. Clin Orthop 189: 65-76, 1983. 94. Evans DK: Dislocations of the cervicothoracic junction. J Bone Joint Surg (Br). 65:124-127, 1983. 95. Grady MS, Howard MR, Jane JA, Persing JA: Use of the Philadelphia collar as an alternative to the halo vest in patients with C2-C3 fractures. Neurosurgery 18:151-156, 1985.
96. Harris JH: Radiographic evaluation of spinal trauma. Orthop Clin North Am 17:75-86, 1986. 97. Panjabi MM, White AA III, Johnson RM: Cervical spine mechanics as a function of ligament transection. J Bone Joint Surg (Am) 57:582, 1975. 98. Rorabeck CH, Rock MG, Hawkins AJ, Bourne RB: Unilateral facet dislocation of the cervical spine: an analysis of the results of treatment in 26 patients. Spine 12:23-27, 1987. 99. Stauffer ES: Management of spine fractures C3-C7. Orthop Clin North Am 17:45-53, 1986. 100. Stauffer ES, Kelly EG: fracture dislocations of the cervical spine: instability and recurrent deformity following treatment by anterior interbody fusion. J Bone Joint Surg (Am) 59:45-48, 1977. 101. Van Peteghem PK, Schweigel JF: The fractured cervical spine rendered unstable by anterior cervical fusion. J Trauma 19:110-114, 1979. 102. White AA III, Southwick WO, Panjabi MM: Clinical stability in the lower cervical spine: a review of past and current concepts. Spine 1:15-27, 1976. 103. Zigler J, Rockowitz N, Capen D, et al: Posterior cervical fusion with local anesthesia: the awake patient as the ultimate spinal cord monitor. Spine 12:206-208, 1987.
543
C H A P T E R
The thoracic spine typically consists of 12 vertebrae (T112) and is the longest segment of the spinal column. However, in this chapter the thoracic spine is considered from level one to ten (T1-10). It is the thoracolumbar junction from thoracic vertebrae 11 to the second lumbar vertebrae (T11-L2). The primary spinal curves are present at birth, are maintained through life, and are relatively rigid or stiff. The secondary curves are more flexible and are the result of development or adaptation. The cervical lordosis is created first, at approximately 3 to 9 months of age, when the infant begins supporting his or her head and sitting upright. The lumbar lordosis develops later (between 12 and 18 months), as the child begins to ambulate and assumes and upright posture.41 A thorough knowledge of the thoracic spine and thoracolumbar junction anatomy facilitates a greater understanding of the biomechanical, radiographic, and surgical techniques that are used to surgically treat spinal fractures.
4 5
Trauma Surgery: Thoracic and Thoracolumbar Spine James S. Harrop, Alexander R. Vaccaro, Kevin T. Foley, and Iain Kalfas
In the United States approximately 160,000 patients a year incur traumatic spinal column injuries, with 10% to 30% of them having a concurrent spinal cord injury.31,51 Cervical and lumbar (L3-5) spine fractures comprise the majority of these vertebral column injuries. However, between 15% and 20% of traumatic fractures occur at the thoracolumbar junction (T11-L2), whereas 9% to 16% occur in the thoracic spine (T1-10).29,39 Thoracic and thoracolumbar fractures are associated with a high incidence of neurologic injuries because of the significant kinetic energy required to create these lesions and the relatively small size of the spinal canal in the thoracic region.51 These patients are typically younger males who were involved in high-speed motor vehicle accidents. The transfer of axial loads through the transition from the rigid thoracic kyphosis to the mobile lumbar region results in a high incidence of thoracolumbar junction fractures. Trauma to the thoracic spine and thoracolumbar junction, as a result of its anatomy, and biomechanical characteristics, is efficiently classified according to its radiographic presentation, biomechanical deficiencies, and clinical presentation of the patient. A thorough understanding of the principles of trauma care and an awareness of the unique biomechanic properties of the thoracic spine afford the treating physician a knowledge base that facilitates the effective management of these patients, both nonoperatively and operatively.
Thoracic Spine The thoracic spine differs from the cervical and lumbosacral spines because of its articulation with the rib cage (T1-12) through an extensive ligamentous support network, its coronal facet joint orientation, and its small spinal canal to neural element ratio (Figure 45.2). There is significant variability in what is considered the “normal” sagittal curvature of the thoracic spine. This value has been reported to be between 20 and 45 degrees,34,85,97 with each individual vertebral body contributing approximately 3.8 to 3.9 degrees of kyphosis through its wedgedshaped angulation.85 This variability is further influenced by age (increases with aging) and gender; females have a greater degree of kyphosis than males.34,97 There is also a significant degree of variability on a segmental basis, particularly at the transitional regions with the lordotic cervical and lumbar spine.11,13,94 The apex of thoracic kyphosis is at the seventh thoracic vertebrae. The thoracic spine typically has a mild, right-sided lateral curvature.41,97 The causes of the rightsided lateral curve is debated but is believed to be either the result of hand dominance (right-hand majority) or created by pulsations of the thoracic aorta.41 The thoracic vertebrae increase in size (height, width, and depth) as the spine is descended.7,11,85,105,108 These bodies differ from the other vertebral bodies in that the ventral portion is shorter than the dorsal portion, which contributes to the thoracic kyphosis. In the transverse plane the thoracic vertebrae have a triangular configuration and appear heart shaped. The diameter of the thoracolumbar pedicles also increases in size as the spine is descended and have an oval configuration, with a greater height (mean 15 to 16mm) than width (mean 8 to 9mm).105 The thoracic pedicles are situated toward the rostral portion of the vertebral body in close proximity to the superior disc space (Figure 45.3). The pedicle location on the vertebral body progressively migrates as the spine is descended in a caudal direction. In the rostral thoracic spine, the pedicles have a greater angle of insertion into the vertebral body (approximately 15 degrees for T1).
Anatomy The vertebral column provides humans with the ability to maintain an upright posture, protects the neural and visceral organs (i.e., heart, lungs, abdominal contents), and aids with motility. It consists of 29 vertebrae arranged in four major curves, two primary curves (thoracic and sacral), and two compensatory or secondary curves (cervical and lumbar).41 The vertebral column also provides a protective environment for the spinal cord and neural elements. The vertebral body, pedicles, and dorsal elements surround the spinal cord, allowing the spinal nerves to exit through the paired neural foramina The laminae are formed as a dorsal-medial extension of the pedicles and fuse in the midline to create the spinous processes (Figure 45.1). 544
545
Chapter 45 ■ Trauma Surgery: Thoracic and Thoracolumbar Spine
This lessens to approximately 10 degrees at T3-11, and finally to 5 degrees at T12.7,70,97 The medial pedicle cortical wall is approximately two to three times thicker than the lateral wall.61 Clinically, this is manifested by a greater incidence of lateral compared to medial breeches during the insertion of thoracic pedicle screws. This protects the medially located neural elements. The thoracic transverse processes project laterally from the dorsal articular pillars and decrease in length caudally.29 However, unlike the lumbar spine the relationship between the transverse process and the midpoint of the pedicle is not as clearly defined. McCormick et al.70 showed that there is a significant degree in the variability of the transverse process to pedicle relationship. The mid-
Vertebral body
Pedicle
Laminae
Facet joints
point of the T1 transverse process is approximately 5mm rostral to the center of the pedicle, whereas at T12 the transverse process to pedicle relationship changes to approximately 6mm caudal.70 This distance difference is greater than 1cm between T1-12, and is approximately 0 at the T6-7 level.70 The thoracic spine facet articulations are considered apophyseal joints and are composed of a ligamentous capsule with a synovial lining. The ligament is thicker than their cervical counterpart’s lining. The joints are located at the rostral and caudal borders of the lamina and situated medial to the transverse processes (Figure 45.3). The inferior facet’s ventral surface articulates with the superior facet’s dorsal surface. Thoracic facet joints are oriented in a coronal plane and therefore limit the degree of flexion and extension of the thoracic spine.28 The ribs articulate with the thoracic vertebrae at two locations. The rib tubercle articulates with the transverse process of the vertebral body at the costotransverse articulation, except at T1, T11, and T12. This articulation is supported with a large superior costotransverse ligament, which connects the rostral rib segment to the caudal transverse process (Figure 45.2). The second rib articulation is the rib head with the same numbered vertebral body and the rostral disc space through the two costal demifacets (T2-10). The strong ligamentous structures that compose the costovertebral joint make the thoracic disc the strongest of all the vertebral discs.1a The superior demifacet (rostral on the vertebral body and caudal to the rib) is located over the pedicle, such that the sixth rib articulates with the fifth and sixth vertebral bodies and overlies the sixth vertebral pedicle. Because of the rostral location of the thoracic pedicle on the vertebral body, the sixth rib overlies the T5-6 disc space. Understanding the anatomic
Spinous process
Figure 45.1 Axial CT image of L2 vertebral body identifying the
dorsal elements.
Costovertebral articulation
Disc space
Pedicle
Transverse process
Figure 45.2 Axial CT image of T6 vertebral body identifying the
Figure 45.3 Sagittal CT reformatted image of T2-3 vertebral
relationship between the vertebral body and the rib head’s articulations.
bodies illustrating the relationship of the pedicle to the intervertebral disc space in the thoracic spine.
546
Section 3 ■ Surgical Procedures
relationship of the rib head to the pedicle allows the surgeon to remove the rib-head and identify the neurovascular bundle, along with the neural foramina and thecal sac at that level. In the thoracic spine the laminae are short and broad, which prevents hyperextension.35 The volume of the spinal canal varies throughout the vertebral column, and is the narrowest in the midthoracic region (T3-9).7,85,95 The transverse spinal canal diameter decreases from T1-3 and then increases caudally into the lumbar region. The anteroposterior diameter, however, is more varied.85,95 Therefore, in the thoracic region a minor degree of canal encroachment may compromise the narrow canal and may result in neurologic compromise.9,10 The thoracic spinal cord has the most tenuous blood supply.85 The small canal size, limited blood supply, and high degree of energy required to create a thoracic fracture results in a 90% incidence of neurologic deficit in patients who sustain a thoracic fracture.74 Thoracolumbar Junction The transition from a relatively rigid thoracic kyphosis to a mobile lumbar spine occurs at the thoracolumbar junction. This transition generally occurs at T11-12, although in elderly female patients the thoracolumbar inflexion point migrates caudally, as a result of their increased degree of thoracic kyphosis.11,94,97 The lowest thoracic ribs (T11 and T12) afford less stability at the thoracolumbar junction region compared to the rostral thoracic region, because they do not connect to the sternum and are free floating. Only a single rib articulation is present on the eleventh and twelfth vertebral bodies, and there are no accessory ligamentous attachments, such as the rib’s tubercle to the vertebral body via the costotransverse ligament, or the ligamentous attachment to the transverse process.41 Conversely, the surrounding thoracolumbar ligaments, such as the interspinous and thoracolumbar fascia, which provide a significant amount of stability, are the strongest caudally.105 The thoracolumbar junction facet joints are again of the apophyseal type and are composed of a ligamentous capsule with a synovial lining. As mentioned previously, the joints of the midthoracic region are oriented in the coronal plane, limiting flexion and extension while providing substantial resistance to anteroposterior translation.29 In the lumbosacral region the facet joints are oriented in a more sagittal alignment, which increases the degree of potential flexion and extension at the expense of limiting lateral bending and rotation. Therefore in the region of the thoracolumbar junction there is a gradual change in facet orientation to a more oblique orientation in the upper lumbar spine. Depending on the spatial orientation of the spinal column (i.e., flexion or extension), the facet joints may support a third of the axial load. These joints, however, provide substantial support and resistance to approximately 35% to 45% of the torsional and shear forces experienced in this region.27,39 During infancy the spinal cord terminates at the end of the vertebral column or lumbosacral junction. However, the location of the terminal end of the spinal cord or conus medullaris varies as the infant develops.68 In neonates the spinal cord terminates between the first and third lumbar
vertebrae, whereas in adults it is positioned between the twelfth thoracic vertebrae and the second lumbar vertebrae.68
Imaging Plain radiographic studies are often difficult to interpret because of surrounding soft tissue and bony overlap. Patients with spinal fractures often sustain concurrent lifethreatening injuries and may be obtunded or sedated and require emergent resuscitation. Therefore it is not uncommon in these clinically unstable trauma patients for fractures to not be identified early in the resuscitative period. It has been reported that between 5% and 15% of multisystem trauma patients have occult fractures not diagnosed on their initial evaluation.17,30,63 Although thoracic vertebral fractures are only a minor proportion of traumatic fractures, they are extremely difficult to visualize compared to other vertebral or appendicular fractures. Approximately 20% to 50% of superior thoracic spinal fractures are not diagnosed by admission plain radiographs.29,99,103 Therefore, all suspected spinal trauma admissions should be immobilized until a thorough and detailed spinal evaluation can be performed. If appropriate stabilization precautions are not taken in this patient population, unforeseen neurologic compromise may result.40 Initial imaging studies performed on all thoracolumbar trauma patients should consist of standard anteroposterior (AP) and lateral plain radiographs. These initial radiographs should be reviewed in an organized and systematic manner. On the AP view the pedicles, vertebral bodies, disc spaces and spinous processes, and soft tissue relationships should be examined in detail. The vertebral body heights, disc space relations, vertebral body alignment, and paraspinal swelling should be carefully assessed on the lateral view. These plain radiographs are particularly useful to assess overall sagittal and coronal balance. If a deformity exists, a useful radiographic measurement of the degree of deformity is the Cobb measurement, which is the subtended angle measured between a perpendicular line drawn from the superior end-plate of the vertebral body above the injured body and the inferior end-plate one level below the injured vertebral body (see Figure 45.20). This method of measuring spinal sagittal angulation has been shown to have the highest degree of intraobserver and interobserver reliability.62 In the presence of a vertebral injury the entire spine should be imaged in an orthogonal manner, because of the high incidence (5% to 20%) of noncontiguous spinal fractures.16,87,92 The rostral thoracic spine can be difficult to visualize on lateral plain radiographs because of the patient’s shoulders and body habitus, and a swimmer’s view may allow better visualization of the cervicothoracic junction down to the third thoracic vertebral body.22 Radiographically, a typical superior end-plate thoracic fracture manifests visually as loss of vertebrae height, with or without malalignment, a widened paraspinal line, and possibly, a widened mediastinum.103 Because of difficulties in imaging the upper thoracic region, a high index of suspicion is required by the physician to avoid missing
Chapter 45 ■ Trauma Surgery: Thoracic and Thoracolumbar Spine
injuries at this level. The physician perhaps should have a low threshold for ordering supplemental imaging modalities to assist in the diagnosis, such as computed tomography and magnetic resonance imaging. Computed tomography (CT) provides a fast and accurate imaging assessment of the vertebral column. This imaging modality has been very useful in the classification and understanding of these fractures.26,49,69 CT is more sensitive in detecting fractures than plain radiographs in the thoracic spine. It is also useful in defining the threedimensional anatomy of complex fractures through reformation in the sagittal and coronal planes. CT imaging permits a more detailed and precise assessment of fracture configuration by its ability to remove extraneous artifacts, such as the ribs and soft tissue structures. This facilitates the quantification of the degree of spinal canal encroachment by retropulsed middle column bone and the presence of a nondisplaced laminae fracture on transaxial imaging, all not possible with standard plain radiography (Figure 45.4). Serial CT has confirmed the spontaneous remodeling and bone reabsorption of retropulsed bone fragments in the spinal canal at the time of long-term follow-up of traumatic burst fractures.8,23,60 CT image reconstruction is also invaluable at the cervicothoracic junction because of the overlay of the scapula, shoulders, and surrounding tissues. In the obtunded patient this technique has been reported to identify more than 10% of fractures not visualized on plain radiographs.55 CT, however, has a limited capacity to visualize disc herniations, epidural or subdural hematomas, ligamentous disruption, or spinal cord parenchymal changes.33 Magnetic resonance imaging (MRI) has further improved the ability to visualize and comprehend the pathoanatomy of the soft-tissue, ligamentous, intervertebral disc and the neural element disruption that occurs
547
after spinal injury. Unfortunately MRI is not always available because of its expense, takes a longer time to implement, and cannot be performed on patients with specific ferromagnetic implants. Today it has supplanted CT myelography as the imaging tool of choice of the neuroaxis, because it is faster, noninvasive, and allows improved visualization of the spinal cord parenchyma.57 MRI provides the physician the ability to identify edema and/or hemorrhage of the spinal cord57 (Figure 45.5). These images have been correlated with neurologic outcomes, where the presence of hemorrhage in the spinal cord parenchyma is associated with minimal neurologic recovery.93 MRI evaluation is especially useful at the thoracolumbar junction because of the variable location of the cauda equina and conus medullaris in the adult population at this level. A neurologic examination can be difficult to interpret at the conus/cauda equina transition level, as a result of the presence of lumbar spinal nerve sparing, the presence of concurrent injuries, sedation, in-dwelling catheters, and delayed reflex recovery (Figure 45.6). Accurate neural visualization may help in clarifying the pathoanatomy in this clinical situation.
Biomechanics The thoracic spine assumes a relatively rigid kyphotic posture between the flexible lordotic cervical and lumbar segments, with the thoracolumbar junction providing the transition between the thoracic and the lumbar spine. The spinal column provides humans with the ability to maintain an upright position by balancing all applied forces that affect the spine while keeping the head centered over the pelvis. The vertebral body is the primary load-bearing structure of the spine, with the intervertebral disc transferring all forces applied to the adjacent
Figure 45.4 T12 burst fracture sustained after an MVA. (A) Note the high definition of the posterior displaced fragment (arrow), along with the vertebral body sagittal fracture (B, arrow) and the associated laminae fracture.
548
Section 3 ■ Surgical Procedures
vertebral bodies.39,86 The annulus fibrosus of the intervertebral disc supports a significant portion of all applied axial and lateral loads and resists tension and shearing.67 The spinal ligamentous structures are essential in maintaining overall sagittal balance. The posterior longitudinal
A
Figure 45.5 T2-weighted MRI of cervical spine of C5 ASIA A
spinal cord injured patient after an automobile crash. The images illustrate severe intraparenchymal edema, along with hemorrhage in the spinal cord.
S
ligament (PLL) is a relatively weak ligament that provides some restriction to hyperflexion, along with the ligamentum flavum. The thick anterior longitudinal ligament (ALL) functions in resisting spinal hyperextension and distraction.82 This thick ligament has fatigue loading values that are approximately double any other spinal ligaments,67,82,86 and its strength increases caudally from C3 to the sacrum.82 The intrinsic strength of the spinal ligaments is only an isolated factor in the overall stability of the spinal column. The lever arm by which these ligaments act on the spine also significantly affects the overall stability of the vertebral column (Figure 45.7). For example, the facet capsules are very strong ligaments and act with a short lever arm by their relationship to the instantaneous axis of rotation (IAR), whereas the intraspinous ligaments are relatively weak but act with a great lever arm because of their increased distance from the IAR. Based on the ligaments’ relative strengths, it would seem that the intraspinous ligaments are not important, but both ligaments significantly affect the strength and structure of the spine. The thoracic spine differs from the remainder of the spinal column because it is supported and maintains articulations with the ribs (Figure 45.2). The intact rib cage increases the axial load-resisting capacity of the thoracic spine by a magnitude of four. The rib cage and facet articulations limit rotation, and therefore most thoracic spine fractures occur from a flexion or axial compression force vector.47 The majority of stability in flexion is provided by the costovertebral articulations.42 A significant factor in the degree and extent of fracture character is the rate of force impact loading.102 The IAR is defined as the axis where all forces and moments are balanced on a specific motion segment.43 This axis is a geometric concept and does not apply to a specific anatomic location.44 However, in the normal thoracic and thoracolumbar spine the IAR is located in the ventral vertebral body. The IAR is analogous to a fulcrum, where during flexion the structures anterior to this axis will come under compression, whereas structures dorsal to this axis are exposed to tensile forces. The bending moment or arm is defined as the product of the lever arm (distance at right angle to the point of force application) and the magnitude
A
Figure 45.7 The lever arm of the ligaments to the instantaneous Figure 45.6 T12 burst fracture with retropulsion of fragment
into canal. Note that the fragment is compressing the conus medullaris and there is resulting spinal cord edema. Clinically the patient has a severe lower-extremity paresis and loss of bowel and bladder function.
axis of rotation (IAR) greatly influences the stability of the spine. The weaker interspinous ligaments work at the greatest distance from the IAR and therefore provide significant resistance to gravitational influences. ALL, Anterior longitudinal ligament; Cap, capsular ligaments; IS, interspinous ligaments; PLL, posterior longitudinal ligament.
Chapter 45 ■ Trauma Surgery: Thoracic and Thoracolumbar Spine
of force transmitted on the object. Therefore forces or loads that are applied to the IAR at a greater distance are at a biomechanic advantage as a result of their greater moment arm. Gravitational forces exert a significant axial load on the vertebral column in the standing adult human. The center of gravity of the body, located where all forces are counterbalanced such that there is no net movement when positioned through this point, is approximately 4cm anterior to the first sacral vertebra.105 This results in a ventral bending (angular) vector acting on the spinal column. This bending force draws or attracts the ventral spinal column closer to the center of gravity such that a lower energy state may be achieved by the paraspinous musculature. The dorsal ligamentous complex and dorsal paraspinal musculature, acting as a dorsal tension band, counteracts these forces, such that the net sum of the vectors acting on the spine equals zero. Therefore the dorsal ligamentous, osseous, and musculature components are essential for overall support of the spine to prevent a change in the spine’s sagittal alignment. Trauma resulting in disruption of the spinal ligaments or osseous structures may change the net vector sum from zero, resulting in the potential for spinal imbalance. These new vectors acting on the spine, if not corrected, may result in a gradual spinal deformity associated with or without pain or neurologic deterioration. Whiteside used the analogy of a construction crane to illustrate this mechanical principle (Figure 45.8).106 In this analogy the weight to be lifted is anterior to the crane where the boom (anterior vertebral column) is under compressive forces and the guide wires (posterior columns) are under tension. Failure of
549
either supporting structure independent of the other will result in mechanical failure or collapse. The thoracic and thoracolumbar vertebrae are at an increased risk for developing compression fractures after trauma as a consequence of axial loads resulting from the natural kyphotic curvature of the thoracic spine.45 The kyphotic posture results in the placement of axial forces on the ventral portion of the vertebral body. An axial load causes all points that are ventral to the IAR of the spine to come closer together while simultaneously all points that are dorsal are spread apart. Therefore if the strength of the ventral vertebral body is exceeded, a fracture of the vertebral body occurs, resulting in a vertebral compression fracture (VCF). The traumatic forces may also exceed the strength of the dorsal vertebral body and ligamentous elements, resulting in disruption of the dorsal tension band. The destruction of the ventral vertebral stabilizing elements (vertebral body, disc, ligaments, annulus) causes the IAR to migrate dorsally to the region with intact supporting structures.45,46 The dorsal migration of the IAR causes the previous mechanical advantage of a longer level arm from which the dorsal ligaments and muscles acted to be shortened. This migration of the IAR simultaneously increases the distance of the center of gravity to the IAR, thereby placing further distraction on the dorsal spinal column and compression on the ventral spinal column (Figure 45.7).45,46 The shortened dorsal lever arm and potentially disrupted dorsal tension band, along with a greater ventral bending moment arm, creates a viscous cycle that places the vertebral bodies at a greater risk for further compression fractures and subsequent deformity progression.
Figure 45.8 (A and B) Whiteside’s analog of the spine illustrates the delicate equilibrium of the
anterior compression vectors against the posterior tension vectors. (Copyright Cleveland Clinic Foundation, 2003.)
550
Section 3 ■ Surgical Procedures
Concurrent Injuries Thoracic Spine Fractures The osseous structures, ligaments, rib cage, and the anatomy of the thoracic spine’s structural integrity provide more resilience to sustaining fractures then the remaining vertebral column. Therefore when a fracture occurs over this region, the physician must be aware that a high degree of energy was required to produce this lesion. These forces on impact are dispersed and transmitted to the soft tissue and viscous elements contained within and around the thoracic cavity, resulting in a high incidence of concurrent injuries. The incidence of concurrent injuries is reported to be greater than 80%, and involves the thorax, appendicular skeleton, and abdominal region.84,92,103 These high-energy impacts also affect remote areas from the trauma, such as the cranial vault. Petitjean et al.84 reported a 65% incidence of head injuries after highvelocity impacts, which resulted in incomplete thoracic spinal cord injury with 12% of these injuries classified as severe (Glasgow Coma Scale [GCS] score less than 8).84 The thoracic cavity consists of the heart, lung, trachea, esophagus, and great vessels. Any of these structures may be injured as a result of blunt trauma and manifest as hemodynamic instability. Tearing or rupture of the aorta has been associated with thoracic vertebral fractures.101 Hemodynamic instability may result from injury to any vascular structure, or even from blood loss secondary to a thoracic vertebral fracture.24,104 Hemothorax has also been reported to occur between 24% and 32% of patients with thoracic fractures.2,37 Pulmonary injuries have been reported in 85% of patients and typically consist of pulmonary contusions.84 Infrequently, perforation of the esophagus and tracheal injuries have also been associated with thoracic fractures.20,79,100 The mechanism of perforation is believed to be caused by ischemia of the esophageal tissue after becoming devascularized as a result of a deceleration-traction injury.20,100 Thoracolumbar Junction Fractures The thoracolumbar region is more vulnerable to concurrent injuries than the thoracic region because it is not provided the protection of the thoracic rib cage. Petitjean84 reported a 71% incidence of associated blunt abdominal injuries after thoracolumbar fractures. Typically these consist of hollow viscous injuries, such as intestinal perforations, mesenteric avulsions, or solid organ injuries.4,84,88 The most common mechanism of abdominal injuries is distraction or seat-belt injuries.4,52,84,88 Blunt abdominal aortic dissections are associated with distraction-rotational injuries of the thoracolumbar region.52 These aortic injuries can range from an intimal tear to a fullthickness laceration. CT provides accurate imaging of this injury in the stable and asymptomatic patient. A large degree of energy is involved in this distractiontype mechanism, accounting for the large number of associated injuries. Multiple-level thoracic and lumbar fractures are also associated with a high incidence of abdominal injuries.52,88 Axial load injuries, particularly in patients that have jumped or fallen and landed on their feet, may manifest as
both thoracolumbar fractures and also calcaneal fractures. Isolated transverse process fractures of the lumber spine should not be overlooked as a minor injury. Miller reported a 48% incidence of concurrent abdominal injuries associated with transverse process fractures.76 Therefore a physician treating vertebral column injuries must not only be aware of the presence of spinal fractures, but also the possibility of concurrent, nonspinal, soft tissue and bony injuries.
Classification Fracture classification schemes provide physicians with the ability to organize and treat fractures through predetermined protocols to maximize patient outcomes. A classification scheme must therefore be simple to understand, easy to apply, and have consistent outcomes with the application of the same treatment algorithm. Unfortunately, the thoracic spine does not have a uniformly agreed upon classification scheme, and the systems available for the thoracolumbar junction injuries are inconsistently used because of their complexity, difficulty in application, and clinical irrelevance.* Nicholl83 developed the first detailed thoracic and thoracolumbar spinal fracture classification scheme and attempted to define unstable versus stable fractures after trauma in a series of flexion and flexion rotation injuries. This classification system was originally intended to guide the treatment and work status of injured miners. Nichol emphasized the importance of the dorsal interspinous ligament in spinal stability.83 Later, Holdsworth,50 after clinical failures in immobilizing pure flexion fractures per the recommendations of Watson, Jones, and Davis, further studied the importance of the spinal ligamentous complexes after thoracolumbar junction injuries. He classified fractures, according to their mechanism of injury, into four main types: flexion, flexion and rotation, extension, and compression. Holdsworth50 further classified these fractures as unstable if the posterior ligamentous complex, consisting of the intervertebral disc, spinous ligaments, facet capsules, and the ligamentum flavum, was disrupted. He noted that in compression, flexion, and extension injuries, the dorsal ligamentous complex is typically not ruptured and these fractures were therefore considered stable. However, he reported that flexion and rotational injuries were at a much greater risk for disruption of the dorsal ligamentous complex and subsequent instability. Rennie and Mitchell89 added a fifth category of thoracolumbar junction fractures consisting of flexion distraction fractures or seat-belt injuries, based on the description and reporting of Chance.18 Kelly and Whiteside59 reported that without dislocation of the dorsal elements of the spinal column neurologic injuries rarely occur. They classified fractures based on structural criteria and considered the spine to consist of not one, but rather two separate supportive columns (Figure 45.9). The primary weight-bearing ventral column is composed of the vertebral bodies, and a second structural column consists of the posterior neural arches and *References 25,26,32,50,59,65,66,72,83.
551
Chapter 45 ■ Trauma Surgery: Thoracic and Thoracolumbar Spine
Anterior column
Anterior vertebral body column
Posterior column
Posterior facet complex columns
Figure 45.9 Axial CT of lumbar vertebral body illustrating Kelly
and Whiteside’s two-column theory of spine stability, which classifies the spine into two equal columns of support consisting of equal anterior and posterior columns.
ligaments. The structural classification scheme provided surgeons the ability to predict the degree of instability of the spine based on the degree of resulting structural damage after trauma. Based on this assessment, treatments were devised to enhance neurologic and spinal stability. Later, Louis65 further modified this structural classification scheme by proposing a third column. Louis’ threecolumn concept consisted of one ventral column, the vertebral bodies, and two dorsal columns involving each facet articulation (Figure 45.10). Technological advancements in the form of superior imaging studies have allowed a greater understanding of the pathoanatomy of spinal trauma. Several biomechanic studies have documented that an isolated rupture of the posterior spinal ligamentous complex is insufficient to create instability.5,6,85,91 However, if the PLL, along with the posterior annulus fibrosis, is also disrupted, then the vertebral column will become unstable, particularly in flexion. Denis used the enhanced imaging techniques CT, along with in vitro biomechanic data to further modify the spinal column theories into a different three-column classification scheme (Figure 45.11). In this classification the ventral column consists of the ventral longitudinal ligament (VLL), the anterior annulus fibrosis, and the anterior half of the vertebral bodies. The middle column consists of the PLL, the dorsal annulus fibrosis and the dorsal half of the vertebral bodies. Lastly, the posterior column, analogous to what Holdworth defined as the dorsal ligamentous complex, consists of the bony neural arch, posterior spinous ligaments, and ligamentum flavum, as well as the facet joints. According to the Denis classification scheme, rupture of the dorsal ligamentous complex only creates instability if and when there is concurrent disruption of at least the PLL and dorsal annulus.
Figure 45.10 Axial CT of lumbar vertebral body illustrating
Louis’ three-column figure theory of spine structure. This classification system divides the spine into an anterior column consisting of the vertebral body and two equal posterior columns consisting of the facet complexes.
Anterior column
Middle column
Posterior column
Figure 45.11 Axial CT of lumbar vertebral body illustrating
Denis’ three-column classification scheme of spine stability. This classification system divides the spine into an anterior column consisting not only of the anterior half of the vertebral body, but also the anterior longitudinal ligament. The middle column consists of the posterior half of the vertebral body and the posterior longitudinal ligament. The posterior column consists of the facets, laminae, and ligamentous complex.
Denis defined failure of the anterior column alone under compression (compression fracture) with an intact posterior column as a stable fracture (Figure 45.12). Burst fractures were defined as being generated through an axial compressive load, and involved failure of the anterior and
552
Section 3 ■ Surgical Procedures
A
R
Figure 45.13 An 18-year-old male who sustained a 25-foot fall.
Axial CT image of L2 burst fracture detailing the canal encroachment of the retropulsed fracture (arrow). Note the disruption of the posterior vertebral wall and the splaying of the pedicles. Figure 45.12 L1 vertebral compression fracture (arrow). Note the posterior vertebral body’s height is maintained, and in the Denis classification only the anterior column is disrupted.
middle columns (Figure 45.13). Severe tensile injuries resulted in seat-belt fracture or flexion distraction injuries, which involve a disruption of the posterior and middle columns with an intact anterior column that serves as a fulcrum or hinge (Figure 45.14). The last category in Denis’ scheme is fracture dislocations, which are defined as a mechanical failure of all three columns, which makes them extremely unstable injuries (Figure 45.15). Denis organized instability into three categories: mechanical, neurologic or both mechanical and neurologic. Mechanical instability may result in a late kyphotic deformity. For example, a seat belt or severe compression fracture with compromise to the dorsal ligamentous complex may result in the spinal column falling into kyphosis, rotating around the intact middle column because of the deficient dorsal tension band. External immobilization, and when appropriate, operative reduction and stabilization, may prevent this deformity progression. Neurologic instability may occur after a severe burst fracture in which the middle column has ruptured under axial loads. This disruption and retropulsion of bone fragments into the spinal canal predisposes these patients to an increased risk for neurologic injury, especially with increased spinal motion. Denis reported that 20% of patients with severe burst fractures and dorsal ligamentous injuries that were treated nonoperatively with external immobilization, developed a subsequent neurologic deficit. Neurologic and mechanical instability may develop after a burst fracture or fracture/dislocation, with or without an initial neurologic deficit. These are very unstable
Figure 45.14 L2 Chance or seatbelt fracture. Note this flexion
distraction injury splits the pedicle and facets (arrows), leaving the anterior column intact.
Chapter 45 ■ Trauma Surgery: Thoracic and Thoracolumbar Spine
553
T8 T7
Figure 45.15 A 38-year-old male driven over by a truck presented with thoracic T7-8 dislocation
and complete loss of motor and sensation below the injury. The spine was fractured and dislocated, therefore separating all three columns of stability as detailed in the Denis classification.
injuries and according to Denis’ analysis require decompression and internal stabilization. Magerl developed a comprehensive classification scheme based on pathomorphologic criteria. This system is modeled after the AO long bone fracture classification (Miller) and consists of a grid with three major fracture types, subdivided into three more groups, which are further subgrouped into three more categories.75 The main categories include: (A) compression injuries of the vertebral bodies, (B) distraction injuries that affect the anterior and posterior elements, and (C) axial torque or multidirectional injuries with translation that also affects the anterior and posterior elements. The severity of the fractures is inherent with this scheme as one progresses from type A to C. Although this classification system is extremely organized and comprehensive, it is underused clinically because of its complexity. McCormack72 created a classification system based on a load-sharing principle that uses a graded point system based on the integrity of the vertebral bodies or anterior and middle column. Points are based on the amount of vertebral body comminution, spread of fragments at the fracture site, and the amount of corrected traumatic kyphosis (Figure 45.16). Traumas with the highest degree of energy cause bone fragments to be greatly dispersed such that they are displaced further and not in continuity with the other fragments. This classification can be applied preoperatively to quantitatively estimate the extent of disruption of the anterior and middle columns. The patients with increased point values (greater than six) have a large void or gap, resulting in the least supportive anterior and middle columns. The fractures with the greatest dispersion of fragments and least bone contact create the highest degree of cantilever bending loads on the pedicle screw implants and predisposes the posterior instrumentation to failure.73,107 This classification scheme assists the surgeon in deciding if ventral spinal support is necessary after dorsal instrumentation, based on the
premise that inadequate anterior column support will result in excessive loads being transferred to the dorsal elements (and instrumentation), thus increasing the risk for failure.
Treatment Options and Strategies There is a significant amount of controversy regarding not only the need for surgical versus nonsurgical management of thoracolumbar injuries, but also the approach and type of surgery if operative intervention is chosen. No definitive treatment algorithm has been universally accepted for this spinal disorder, despite the numerous classification systems that exist. Stability of the vertebral column over the thoracic and thoracolumbar region, like the remainder of the spine, is dependent on the integrity of the osseous and ligamentous components. Once these structures are disrupted the stability of the vertebral column can become compromised, resulting in an unstable spine. One difficulty in treating these fractures is that the definition of instability of the spine is difficult to assess, based on clinical and radiographic findings. White and Panjabi have the most detailed description of instability: “the loss of the ability of the spine under physiological loads to maintain relationships between vertebrae in such a way that there is either damage or subsequent irritation to the spinal cord or nerve roots, and in addition, there is development of incapacitating deformity or pain due to structural changes.”105 However, even this definition leaves a large degree of ambiguity because of the large spectrum of spinal disorders. Treatment algorithms are based on maximizing neurologic recovery and prevention of neurologic decline and deformities, while maximizing pain relief. This is accomplished by identifying fractures that are “unstable” such that their structural support can be augmented to maximize clinical outcomes. Each patient and management strategy must be individualized depending on the type of
554
Section 3 ■ Surgical Procedures
Figure 45.16 McCormick grading scheme or load-sharing classification. Communition of
fragments based on a sagittal CT reformat as either (A1) little (60%), three points. Apposition of fragments based on an axial CT image as either (B1) minimal one point, (B2) spread defined as greater than 2mm in less than 50% of the body two points, or (B3) wide, 2mm in greater than 50% of the body three points.
Chapter 45 ■ Trauma Surgery: Thoracic and Thoracolumbar Spine
Figure 45.16 cont’d Kyphosis correction on plain radiographs
(C1) 10 degrees, three points. (Copyright Cleveland Clinic Foundation, 2003.)
fracture, location, other traumatic comorbidities, neurologic outcomes, and medical comorbidities. Nonoperative Strategies The majority of fractures in the thoracic and thoracolumbar regions consist of compression, stable burst, and isolated dorsal column fractures. These fractures are generally classified as stable because there is little risk of neurologic compromise from osseous compression or deformity progression. The acquired stability of the preserved ligaments and osseous structures usually prevent acute instability; however, a low potential for chronic or glacial instability still remains. Glacial instability usually presents as mechanical pain, but could also present as a neurologic deficit. VCFs, as defined by the Denis’ three-column scheme, consist of a loss of anterior column or ventral vertebral body height as a result of axial compression (Figure 45.12). These fractures are considered stable anatomically, if the dorsal ligamentous complex, along with the dorsal verte-
555
bral body, is not disrupted. The intact dorsal ligaments and dorsal elements provide a significant tension band. Further stability is also provided by the remaining intact anterior and middle columns. Neurologic function should not be impaired as a result of these fractures, because the dorsal cortex of the vertebral body is not violated and because there is no encroachment of fracture fragments on or into the spinal canal. In approximately a third of patients these vertebral compression fractures unfortunately are persistently painful and can require bracing, prolonged bed rest, and narcotic therapy. Such VCFs may be satisfactorily treated nonoperatively with a hyperextension brace or corsette. This bracing technique maintains alignment through three-point bending principles and results in axial loads being displaced through the dorsal vertebral body and fracture. The redirection of axial loads through the dorsal portion of the fracture may impede the degree of anterior column subsidence and therefore minimize the extent of eventual thoracic kyphosis. In the elderly, symptomatic compression fractures associated with severe pain and functional morbidity, which have not responded to a minimum of 6 weeks of conservative management, have been treated with polymethylmethacrylate (PMMA) augmentation, either via vertebroplasty or via a cavitation and endplate elevation procedure (Figure 45.17). These techniques have been associated with a significant improvement in patient function and pain relief.3,64 Burst fractures, as defined by Denis, are vertebral body fractures involving the anterior and middle columns, such that the ALL and vertebral body, including the dorsal vertebral body cortex, are disrupted (Figures 45.4, 45.6, 45.11, and 45.13). A burst fracture in a neurologically intact patient without posterior ligamentous or dorsal element fractures is usually considered a stable injury. James et al.54 confirmed this clinically in his reported series of patients with burst-type fractures with an intact dorsal bony architecture, all of whom became stable with a bracing regimen. Thoracic burst fractures (T1-10) comprise a minor subset of burst injuries, representing approximately 5% to 10% of the total number of burst fractures. These burst fractures are inherently more stable because of the presence of the costovertebral ligamentous complex, along with the support of the rib cage.15 Therefore, like VCFs, conservative therapy has been the mainstay of treatment through bracing and postural reduction.5,6,36,81 The incidence and degree of kyphosis and neurologic deterioration after a thoracic and thoracolumbar burst fracture are not known. In the neurologically intact patient with a mild kyphotic deformity (