Motion Preservation Surgery of the Spine Advanced Techniques and Controversies James J. Yue Associate Professor Department of Orthopaedic Surgery Co-Chief, Division of Spinal Surgery Director, Yale Spine Fellowship Yale University School of Medicine New Haven, Connecticut
Rudolf Bertagnoli Chairman First European Center for Spine Arthroplasty and Associated Nonfusion Technologies (ECSA) Elisabeth Krankenhaus Straubing KKH Bogen, Germany
Paul C. McAfee Chief, Spinal Reconstructive Surgery Orthopaedic Associates St. Joseph’s Hospital Towson, Maryland
Howard S. An The Morton International Professor for Spine Research Department of Orthopaedic Surgery Rush Medical College Chicago, Illinois
1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103–2899
MOTION PRESERVATION SURGERY OF THE SPINE: ADVANCED TECHNIQUES AND CONTRVERSIES
ISBN: 978-1-4160-3994-5
Copyright # 2008 by Saunders, an imprint of Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from thxe publisher. Permissions may be sought directly from Elsevier's Health Sciences Rights Department in Philadelphia, PA, USA: phone: (þ1) 215 239 3804, fax: (þ1) 215 239 3805, 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 Knowledge and best practice in this field are constantly changing. As new research and experience broaden our knowledge, changes in practice, treatment and drug therapy may become necessary or appropriate. Readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of the practitioner, relying on their own experience and knowledge of the patient, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the Editors assumes any liability for any injury and/or damage to persons or property arising out or related to any use of the material contained in this book. The Publisher Library of Congress Cataloging-in-Publication Data Motion preservation surgery of the spine : advanced techniques and controversies / edited by James J. Yue . . . [et al.]. –1st ed. p. ; cm. Includes bibliographical references. ISBN 978-1-4160-3994-5 1. Intervertebral disk–Surgery. 2. Intervertebral disk prostheses. 3. Spine–Surgery. 4. Arthroplasty. I. Yue, James J. [DNLM: 1. Spine–surgery. 2. Arthroplasty, Replacement–methods. 3. Motion. 4. Spinal Diseases–surgery. WE 725 M888 2008] RD771.I6M68 2008 617.50 6–dc22
2007027127
Acquisitions Editor: Kimberly Murphy Developmental Editor: Adrianne Brigido Publishing Services Manager: Joan Sinclair Design Direction: Gene Harris
Printed in China Last digit is the print number: 9 8
7 6 5
4 3 2
1
To my wife and children: Susan, Lauren, and Emily. My three 1 in a millions! —JAMES J. YUE To my mentors, friends, colleagues, and entrepreneurs of these techniques—especially those in the SAS society and those who promote motion preservation. To my wife and children. —RUDOLF BERTAGNOLI I wish to acknowledge the most talented biomechanical engineers who have taught me enough to really make a positive impact in patient care—Manohar Panjabi, Fred Werner, Al Berstein, Helmut Link, and David Paul. I came into contact with each one of these internationally recognized giants at precisely the right time in my career, and I am grateful for their friendship. —PAUL C. MCAFEE
Foreword
The human axial skeleton is composed of 24 mobile segments, with three articulations at each level. Although motion is not identical in degree at each level, it is also not the same in each of the 6 degrees of motion. There are specific range and quality at each anatomical region. Therefore, the preservation of motion is a daunting challenge for the scientists and surgeons. Years of study, testing, and experimentation have gone into enhancing our knowledge of spinal motion and complex technologies are emerging. Biomaterials are evolving to enable properties that are essential for compatibility, safety, and efficacy. Biomechanics are thoroughly tested to mimic physiologic motion as closely as possible, yet being certain durability and wear characteristics are carefully monitored. Animal studies meeting the (high) regulatory standards are done when necessary for tissue compatibility and, in some cases, for efficacy. No animals are like the human so the ultimate trial is pilots (small), then when safe, proceed to larger efficacy randomized control studies. This is difficult as what is needed for regulatory clearance is not necessarily the best control trial. However,
diligent choices are often picked to get the best possible Level I studies and evidence. Minimally invasive and yet maximally beneficial technologies are slowly coming to the fore. The biologics arena is abundant with theory, but with only early proof of concepts. What better motion preservation can there be than tissue engineering, with regenerative technologies, after early, appropriate diagnosis? The editors have brought together between the covers of this first edition of Motion Preservation Surgery of the Spine, a comprehensive textbook covering the bulk of motion technologies. It includes (classic) contributions by several founding scientist surgeons. Even before this first edition has been completed, concepts and ideas for a second edition have been initiated by the editors due to the explosion of innovations in this area. My congratulations to James J. Yue, Rudolf Bertagnoli, Paul C. McAfee, and Howard S. An for this huge task of love and dedication. HANSEN A. YUAN, MD PRESIDENT, SPINE ARTHROPLASTY SOCIETY
v
Preface
Vertebra, from Latin, from vertere, to turn
Some may say that current evolution of the treatment of degenerative disorders of the spine is, in many ways, a revolution. The essence of this apparent revolution is based on many factors including the realization that the pioneering work of Fernstrom and his advocates was not medical heresy but rather medical innovation. The eventual professional and personal persecution of these innovators that ensued was inappropriate and has predictably fueled today’s scientific rapid development of motion-sparing technology. A first-hand account of the medical-socio-legal-political environment by Dr. Alvin McKenzie, a recipient of multi-level Fernstrom ball procedure, is meant to further elucidate the foundation of the recent evolving revolution in the treatment of spondylosis. Whatever procedure a surgeon and his or her patient decide upon to remedy the debilitating effects of spondylosis, the choice should be based on the fundamental principal that the vertebrate spine is a “motion-protective” anatomical structure. An inherent duality exists in this motion-protective function. First, our spine is designed to allow positioning of our cranium and torso in our ever increasing multi-dimensional world, thereby permitting and enhancing our interactive and protective abilities and responsibilities. Coupled to this pro-motion functional adaptation is the intrinsic protection to the neural elements that our vertebrate spinal column provides. Herein lies the duality and balance of our motion-protective spinal column. Our vertebrate spinal column allows us to protect ourselves by allowing us to move and interact with our physical environment but it also protects the neurological elements that give our musculoskeletal and dermatomal systems the ability to perform these functions. Motion Preservation Surgery of the Spine: Advanced Techniques and Controversies was written to provide a general understanding of the basic principles of nonfusion surgery as well as an advanced platform to approach the inevitable revision and/or additional procedures that may become necessary, as may occur with any surgical procedure. Inherent in each chapter are specific case examples that the authors have selflessly provided. In addition, clinical trial data are also provided to allow the reader to gain a sense of the
potential place for a given technology in their practice. As one can ascertain from a review of the author list, we have chosen contributors based on their expertise in a given technology. Often this expertise was obtained from many parts of the globe. This multicultural and multi-national perspective is unique and truly invaluable to the reader given the long-term perspective that countries such as Germany and France have had using nonfusion technology. Shortly after production of Motion Preservation Surgery of the Spine: Advanced Techniques and Controversies began, the Spine Arthroplasty Society requested that the text be included as one of the core teaching texts of their organization. In order to fulfill this honored request, multiple chapters dedicated to the concept of how to study motion-sparing technologies, both in the clinical as well as the laboratory setting, have been included. Future editions, both English and translated, will bring additional materials, learning supplements, and teaching aids. The preparation of this textbook would not have been possible without the inspiration and diligent work of our contributing authors. We are deeply indebted and grateful to them for their tireless patience and determination in completing their chapters. We have also been fortunate to have benefited from the pioneering work of our mentors, including Henry Bohlman, Karin BüttnerJanz, Jürgen Harms, John Kostuik, John P. O’Brien, Robbie Robinson, Arthur Steffe, and Hansen Yuan to name just a few. Our support from our publisher, Elsevier, has been unequalled. Specifically, we would like to thank our Development Editor, Adrianne Brigido. Without her efforts, this book would not have been possible. We also sincerely thank Ms. Kimberly Murphy, Publishing Director, Global Medicine Elsevier, for her invaluable guidance. We believe you will find this textbook informative and a comprehensive platform and foundation for learning about motion-sparing technology of the spine. JAMES J. YUE RUDOLF BERTAGNOLI PAUL C. MCAFEE HOWARD S. AN
vii
Contributors
Jean-Jacques Abitbol, MD Orthopaedic Spine Surgeon, California Spine Group, San Diego, California The CerviCore Cervical Intervertebral Disc Replacement Michael Ahrens, MD University of Luebeck, Neustadt, Germany DASCOR Todd F. Alamin, MD Assistant Professor, Department of Orthopaedic Surgery, Stanford University Medical Center, Stanford, California Invasive Diagnostic Tools
Paul A. Anderson, MD Associate Professor of Orthopaedic Surgery, Department of Orthopaedics, University of Wisconsin, Madison, Madison, Wisconsin Preclinical Evaluation of Dynamic Spinal Stabilization: Animal Models and Basic Scientific Methods S.A. Andrew Scient’x IsoBar TTL Dynamic Rod Stabilization Lucie Aubourg, PhD Clinical Research Manager, LDR Medical, Troyes, France Mobidisc Disc Prosthesis
Todd J. Albert, MD James Edwards Professor and Chairman of Orthopaedic Surgery, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania DISCOVER Artificial Cervical Disc
Stephane Aunoble, MD Unité de Pathologie, Centre Hospitalier Pellegrin, Bordeaux, France Minimally Invasive Posterior Approaches to the Lumbar Spine
Je´rome Allain Professor, University Paris XII, Paris, France Professor, Hôpital Henri Dondor, Créteil, France Mobidisc Disc Prosthesis
Jonathon R. Ball, BMed, BMedSC(Hons), FRACS Neurosurgical Registrar, Royal North Shore Hospital, New South Wales, Australia Cervical Arthroplasty with Myelopathy
Marc Ameil, MD Professor, Department of Orthopaedics, Polyclinique Saint-André, Reims, France Mobidisc Disc Prosthesis Howard S. An The Morton International Professor for Spine Research, Department of Orthopaedic Surgery, Rush Medical College, Chicago, Illinois Animal Models for Human Disc Degeneration; Growth Factors for Intervertebral Disc Regeneration Ravi Ananthan Theken Disc, LLC, Akron, Ohio Theken eDisc: A Second-Generation Lumbar Artificial Disc
Qi-Bin Bao, PhD Pioneer Surgical Technology, Marquette, Michigan Aquarelle Hydrogel Disc Nucleus; NUBAC Intradiscal Arthroplasty Jacques Beaurain, MD Department of Neurosurgery, University Hospital, Dijon, France Mobi-C; Mobidisc Disc Prosthesis Marco Be´rard, MD Alice Hyde Orthopaedic and Sports Medicine Center, Alice Hyde Medical Center, Malone, New York Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis ix
x
Contributors
Ulrich Berelmann, MD M.E. Müller Institute for Surgical Technology and Biomechanics, University of Bern, Bern, Switzerland NuCore Injectable Nucleus: An In Situ Curing Nucleus Replacement Pierre Bernard, MD Centre Aquitain du Dos, Clinique Saint-Martin, Pessac, France Mobi-C Rudolf Bertagnoli, MD Chairman, First European Center for Spine Arthroplasty and Associated Nonfusion Technologies (ECSA), Elisabeth Krankenhaus Straubing, KKH Bogen, Germany Lateral Approaches to the Lumbar Spine: The Anterolateral Transpsoatic Approach; coflex Interspinous Implant for Stabilization of the Lumbar Spine; Hybrid Nonfusion Techniques; Autologous Chondrocyte Disc Transplant: Early Clinical Results; Multilevel Lumbar Disc Arthroplasty; Cervical Arthroplasty Adjacent to Fusion, Multiple-Level Cases, and Hybrid Applications Robert S. Biscup, DO, MS, FAOAO Chairman, Biscup Spine Institute, Fort Lauderdale, Florida Satellite: Spherical Partial Disc Replacement Jason D. Blain Chief Technology Officer, Spinal Elements, Carlsbad, California The Zyre Facet Replacement Device Jon E. Block, PhD President, Jon E. Block, PhD, Inc., San Francisco, California The M6 Artificial Cervical Disc Scott L. Blumenthal, MD Orthopaedic Spine Surgeon, Texas Back Institute, Plano, Texas CHARITÉ Artificial Disc; Simultaneous Lumbar Fusion and Total Disc Replacement Nicholas R. Boeree, BSc, FRCS Orth, FRCS Ed Consultant Orthopaedic Surgeon, Southampton University Hospital Trust, Southampton, United Kingdom Wallis Dynamic Stabilization
Jacob M. Buchowski, MD, MS Assistant Professor of Orthopaedic and Neurologic Surgery, Washington University in St. Louis, St. Louis, Missouri Chief, Degenerative and Minimally Invasive Spine Surgery, Washington University in St. Louis, St. Louis, Missouri Primary Indications and Disc Space Preparation for Cervical Disc Arthroplasty Karin Bu¨ttner-Janz, MD, PhD Professor, Charité Universitätsmedizin, Berlin; Director of Orthopedic Clinic, Vivantes Klinikum im Friedrichshain, Berlin, Germany Classification of Spine Arthroplasty Devices Andrew G. Cappuccino, MD Orthopaedic Spine Surgeon, Buffalo Spine Surgery, Lockport, New York Cervical Disc Replacement Revisions: Clinical and Biomechanical Considerations Allen Carl, MD Professor, Orthopaedic Surgery and Pediatrics, Albany Medical College, Albany, New York Attending Surgeon, Albany Medical Center, Albany, New York Anatomic Facet Replacement System (AFRS) Antonio E. Castellvi, MD Spine Fellowship Director; Spine Surgeon, Florida Orthopaedic Institute, Tampa, Florida Scient’x IsoBar TTL Dynamic Rod Stabilization Joseph C. Cauthen, MD Orthopaedic Spine Surgeon, Neurosurgical and Spine Associates, Gainesville, Florida Repair and Reconstruction of the Annulus Fibrosus with the Inclose Surgical Mesh System Herve´ Chataigner, MD Service de Chirurgie des Scolioses et Orthopedie Infantile, Hôpital St. Jacques, Besançon, France Mobidisc Disc Prosthesis
Iohan Bogorin, MD Service de Chirurgie Orthopédique, du Rachis et de Traumatologie du Sport, Hôpitaux Universitaires de Strasbourg, Strasbourg, France Mobidisc Disc Prosthesis
Boyle C. Cheng, PhD Assistant Professor, University of Pittsburgh, Pittsburgh, Pennsylvania Biomechanics of Nonfusion Devices: Novel Testing Techniques, Standards, and Implications for Future Devices
David S. Bradford, MD Professor, University of California San Francisco Professor and Chair Emeritus, University of California, San Francisco, California History and Evolution of Motion Preservation
Robert J. Chomiak, MD Paradigm Spine, LLC, New York, New York coflex Interspinous Implant for Stabilization of the Lumbar Spine; Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis
Contributors
Christine Coillard, MD Research Centre, Sainte-Justine Hospital, Montreal, Quebec, Canada Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis Christopher Cole Engineer, Theken Disc, LLC, Akron, Ohio Theken eDisc: A Second-Generation Lumbar Artificial Disc Dennis Colleran Vice President, Research and Development, IlluminOss Medical Inc., Tiverton, Rhode Island Innovative Spinal Technologies Dynamic Stabilization Device Domagoj Coric, MD Chief of Neurosurgery, Carolinas Medical Center, Charlotte, North Carolina Carolina Neurosurgery and Spine Associates, Charlotte, North Carolina Cervical Approaches: Anterior and Posterior; NUBAC Intradiscal Arthroplasty G. Bryan Cornwall, PhD, PEng Vice President, Research and Clinical Resources, NuVasive, Inc., San Diego, California The NeoDisc Elastomeric Cervical Total Disc Replacement; Cerpass Cervical Total Disc Replacement Etevaldo Coutinho, MD Spine Surgeon, Santa Rita Hospital, São Paulo, Brazil Lateral Lumbar Total Disc Replacement Andrew H. Cragg, MD Interventional Radiologist, Minnesota Vascular Clinic, Suburban Radiologic Consultants, Edina, Minnesota TranS1 Percutaneous Nucleus Replacement Bryan W. Cunningham, MSc Associate Professor, Department of Orthopaedic Surgery, Johns Hopkins University, Baltimore, Maryland Director, Spinal Research, St. Joseph Medical Center, Towson, Maryland Preclinical Evaluation of Dynamic Spinal Stabilization: Animal Models and Basic Scientific Methods; Cervical Disc Replacement Revisions: Clinical and Biomechanical Considerations; Anatomic Facet Replacement System (AFRS) David Cutter NuVasive Inc., San Diego, California Cerpass Cervical Total Disc Replacement Frank Daday, MBBS, FANZCA Visiting Medical Officer, Allamanda Private Hospital, Gold Coast, Queensland, Australia Overall Revision Strategies: Lumbar
Reginald J. Davis, MD Assistant Professor, Neurosurgery, Johns Hospital Medical Institute; Clinical Instructor, Neurosurgery, University of Maryland Medical Center, Baltimore, Maryland Division Head of Neurosurgery, Greater Baltimore Medical Center, Towson, Maryland PDN-SOLO and HydraFlex Nucleus Replacement; Dynesys Dynamic Stabilization System Rick B. Delamarter, MD Associate Clinical Professor, University of California, Los Angeles, Los Angeles, California; Fellowship Director, The Spine Institute, Santa Monica, California ProDisc-C Total Cervical Disc Replacement; ProDisc-L Total Disc Replacement Joe¨l Dele´crin, MD, PhD Associate Professor, Department of Orthopedic Surgery, Nantes University, Nantes, France Mobidisc Disc Prosthesis Malan DeVilliers, BEng (Mech), MEng, PhD (Eng) Managing Director, Southern Medical (PTY) LTD, Irene, Centurion, South Africa Kineflex Roberto Dı´az, MD Assistant Professor, Department of Neurosurgery, San Ignacio University Hospital, Javeriana School of Medicine, Bogota, Columbia Cervical Disc Replacement Revisions: Clinical Biomechanical Considerations; TranS1 Percutaneous Nucleus Replacement; TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System Juan M. Dipp, MD Chief, Orthopedic and Spine Surgery, Hospital del Prado, Tijuana, Mexico Chief Spine Surgeon, Hospital Angeles, Tijuana, Tijuana, Mexico The PercuDyn System Gary A. Dix, MD, FRCS(C) Medical Director of Spine Services, Anne Arundel Medical Center, Annapolis, Maryland Persistent Pain After Cervical Arthroplasty Thomas B. Ducker, MD, FACS Johns Hopkins Medical School, Baltimore, Maryland Anne Arundel Medical Center, Annapolis, Maryland Persistent Pain After Cervical Arthroplasty Thierry Dufour, MD Neurosurgeon, Centre Hospitalier Regional, Orléans, France Mobi-C; Mobidisc Disc Prosthesis
xi
xii
Contributors
Jacob Einhorn Vice President, Research and Development, Intrinsic Therapeutics, Inc., Woburn, Massachusetts The Intrinsic Therapeutics Barricaid Device Lukas Eisermann, BS Director of Advanced Technology Development, NuVasive, Inc., San Diego, California The NeoDisc Elastomeric Cervical Total Disc Replacement; Cerpass Cervical Total Disc Replacement Thomas J. Errico, MD Chief, Division of Spine Surgery, Department of Orthopaedic Surgery, New York University—Hospital for Joint Diseases; Associate Professor of Orthopaedics and Neurosurgery, New York University School of Medicine, New York, New York The FlexiCore Intervertebral Disc Teddy Fagerstrom, MD Department of Orthopedics, Huddinge University Hospital, Stockholm, Sweden FENIX Facet Resurfacing Implant Daniel R. Fassett, MD, MBA Assistant Professor, Department of Neurosurgery, University of Illinois College of Medicine Peoria, Peoria, Illinois Advanced Spinal Anatomy for Cervical and Lumbar Nonfusion Surgery Jeffrey S. Fischgrund, MD Orthopaedic Spine Surgeon, William Beaumont Hospital, Royal Oak, Michigan The CerviCore Cervical Intervertebal Disc Replacement Ricardo Flores, MD Chief of Neurosurgery, Departments of Neurology and Neurosurgery, Hospital Almater, Mexicali, Mexico The PercuDyn System Jean-Marc Fuentes, MD Hôpital Pellegrin, Bordeaux, France Mobi-C Josue Gabriel, MD Clinical Director, Biodynamics Laboratory, The Ohio State University, Columbus, Ohio Clinical Assistant Professor, Department of Orthopaedic Surgery, The Ohio State University, Columbus, Ohio The Development of a Personalized Hybrid EMG-Assisted/Finite Element Biomechanical Model to Assess Surgical Options Rolando Garcı´a, MD, MPH Spinal Surgeon, Orthopedic Care Center, Aventura, Florida Activ-L Artificial Disc; Dynamic Pedicle-Screw Stabilization with Nucleus Replacement
Fred H. Geisler, MD, PhD Founder, Illinois Neuro-Spine Center at Rush-Copley Medical Center, Aurora, Illinois Statistical Outcome Interpretation of Randomized Clinical Trails; Simultaneous Lumbar Fusion and Total Disc Replacement Ihab Gharzeddine, MD Spine Surgeon, Santa Rita Hospital, São Paulo, Brazil Cervical Disc Replacement Revisions: Clinical and Biomechanical Considerations; Lateral Lumbar Total Disc Replacement; Revision Strategies Following Lumbar Total Disc Replacement Complications Vijay K. Goel, PhD Endowed Chair and McMaster-Gardner Professor of Orthopaedic Bioengineering; Co-Director, Engineering Center for Orthopaedic Research Excellence (E-CORE), Departments of Bioengineering and Orthopaedic Surgery, Colleges of Engineering and Medicine, University of Toledo, Toledo, Ohio Theken eDisc: A Second-Generation Lumbar Artificial Disc; Anatomic Facet Replacement System (AFRS) Jeffrey A. Goldstein, MD Assistant Professor of Orthopaedic Surgery, New York University School of Medicine, New York, New York Director of Spine Service, New York University—Hospital for Joint Diseases, New York, New York Persistent Pain After Lumbar Total Disc Replacement Matthew F. Gornet, MD Staff Physician, The Orthopedic Center of St. Louis, St. Louis, Missouri Maverick Total Disc Replacement Steven L. Griffith, PhD Vice President, Scientific Affairs, Anulex Technologies, Inc., Minnetonka, Minnesota Repair and Reconstruction of the Annulus Fibrosus with the Inclose Surgical Mesh System Geneste Guilhaume Department of Orthopaedic Surgery, Clinique du Parc, Castelnau-le-Lez, France Can Lumbar Disc Replacement Be Used Adjacent to a Scoliotic Deformity? Giancarlo Guizzardi, MD Neurosurgery Unit, Careggi Hospital, Florence, Tuscany, Italy DIAM Spinal Stabilization System
Contributors
Richard D. Guyer, MD Spine Surgeon; Co-Director of the Spine Surgery Fellowship Program, Texas Back Institute, Plano, Texas Socioeconomic Impact of Motion Preservation Technology Nader M. Habela, MD Orthopaedic Associates and Spine Center, St. Joseph Medical Center, Towson, Maryland Indications and Contraindications for Cervical Nonfusion Surgery: Patient Selection Ulrich Reinhard Ha¨hnle, MD Post-Graduate Studies, University of Witwatersrand, Johannesburg, Gauteng, South Africa Orthopedic Surgeon, Nedcare Linksfield Hospital, Johannesburg, Gauteng, South Africa Kineflex Horace Hale CEO, GerraSpine, St. Gallen, Switzerland FENIX Facet Resurfacing Implant Nadim James Hallab, MS, PhD Associate Professor, Department of Orthopedic Surgery, Rush University Medical Center, Chicago, Illinois Material Properties and Wear Analysis David Hannallah, MD, MS Staff Physician, The Cardinal Orthopaedic Institute, Columbus, Ohio Adjacent Segment Degeneration and Adjacent Segment Disease: Cervical and Lumbar Matthew Hannibal, MD Director of Spine Research; Associate Director, San Francisco Orthopedic Research Program, Department of Orthopedic Surgery, San Francisco, California X-STOP Interspinous Process Decompression for Lumbar Spinal Stenosis
xiii
Stephen H. Hochschuler, MD Spine Surgeon, Texas Back Institute, Plano, Texas The Future of Motion Preservation Gordon Neil Holen, DO Spine Surgery, Department of Orthopedic Surgery, Mease Countryside Hospital, Safety Harbor, Florida Total Facet Arthroplasty System (TFAS) Istvan Hovorka, MD Department of Orthopaedics and Sports Traumatology, University of Nice–Archet 2 Hospital, Nice, France Mobi-C Robert W. Hoy, MEng Facet Solutions Inc., Logan, Utah Anatomic Facet Replacement System (AFRS) Kenneth Y. Hsu, MD Co-Medical Director, Department of Orthopedics, St. Mary’s Spine Center, San Francisco, California X-STOP Interspinous Process Decompression for Lumbar Spinal Stenosis Jean Huppert, MD Clinique du Parc, Saint-Priest-en-Jarez, France Mobi-C Cary Idler, MD Spine Fellow, St. Mary’s Medical Center, San Francisco, California X-STOP Interspinous Process Decompression for Lumbar Spinal Stenosis Andre Jackowski, MD, FRCS Department of Spinal Surgery, Royal Orthopaedic Hospital, Birmingham, United Kingdom The NeoDisc Elastomeric Cervical Total Disc Replacement
Alan S. Hilibrand, MD Department of Orthopaedic Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania Adjacent Segment Degeneration and Adjacent Segment Disease: Cervical and Lumbar
Joshua J. Jacobs, MD Associate Dean for Research Development, Department of Orthopaedic Surgery, Rush University Medical Center; Associate Dean for Academic Programs, Department of Orthopaedic Surgery, Rush University Medical Center; Inaugural Crown Family Professor of Orthopaedic Surgery, Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, Illinois Material Properties and Wear Analysis
John A. Hipp, PhD Director, Spine Research Laboratory, Orthopedic Surgery, Baylor College of Medicine, Houston, Texas Chief Scientist, Medical Metrics, Inc., Houston, Texas Quantitative Motion Analysis (QMA) of Motion-Preserving and Fusion Technologies for the Spine
Jorge Jaramillo, MD Department of Orthopaedic Surgery and Rehabilitation, Yale University School of Medicine, New Haven, Connecticut Disc Space Preparation Techniques for Lumbar Disc Arthroplasty
Victor M. Hayes, MD Trinity Spine Center, Tampa, Florida Lumbar Endoscopic Posterolateral (Transforaminal) Approach
xiv
Contributors
Shiveindra B. Jeyamohan, MD Department of Neurosurgery, Thomas Jefferson University, Philadelphia, Pennsylvania Advanced Spinal Anatomy for Cervical and Lumbar Nonfusion Surgery James D. Kang, MD Vice Chairman, Department of Orthopaedic Surgery; Professor, Departments of Orthopaedic and Neurological Surgery; Director, Ferguson Laboratory for Orthopaedic Spine Research, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania Professor, Departments of Orthopaedic Surgery and Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Gene Therapy for Intervertebral Disc Repair and Regeneration Larry T. Khoo, MD Assistant Professor, Department of Surgery, University of California, Los Angeles, Los Angeles, California TranS1 Percutaneous Nucleus Replacement (PNR); TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System Seok Woo Kim, MD, PhD Associate Professor, Department of Orthopaedic Surgery; Chief, Spine Services, Hallym University, Seoul, South Korea Director, International Spine Center, Hangang Sacred Heart Hospital, Hallym University Medical Center, Seoul, South Korea Cervical Disc Replacement Combined with Cervical Laminoplasty Scott H. Kitchel, MD Orthopaedic Physician and Surgeon, Orthopaedic Spine Associates LLC, Eugene, Oregon Cerpass Cervical Total Disc Replacement; The Zyre Facet Replacement Gregory G. Knapik, MS Senior Research Associate Engineer, The Ohio State University Biodynamics Laboratory, Columbus, Ohio The Development of a Personalized Hybrid EMG-Assisted/Finite Element Biomechanical Model to Assess Surgical Options Manoj Krishna, MCh(Orth), FRCS Consultant Spinal Surgeon, University Hospital of North Tees, Stockton-on-Tees, United Kingdom Posterior Lumbar Arthroplasty Greg Lambrecht President and CEO, Intrinsic Therapeutics, Inc., Woburn, Massachusetts The Intrinsic Therapeutics Barricald Device
Carl Lauryssen, MD Director, Research and Education, Olympia Medical Center, Beverly Hills, California The Zyre Facet Replacement Device William Lavelle, MD Fellow, Cleveland Clinic Spine Institute, Cleveland Clinic, Cleveland, Ohio Anatomic Facet Replacement System (AFRS) James P. Lawrence, MD Department of Orthopaedic Surgery and Rehabilitation, Yale University, New Haven, Connecticut Indications and Contraindications for Lumbar Nonfusion Surgery: Patient Selection Jean-Charles Le Huec, MD, PhD Professor and Head, Orthopaedic Department; Chief, Spine Unit; Director, Surgical Research Lab, Bordeaux University Hospital, Bordeaux, France Minimally Invasive Posterior Approaches to the Lumbar Spine; DASCOR Juliano Lhamby, MD Orthopedic Spine Surgeon, Santa Rita Hospital, São Paulo, Brazil Cervical Disc Replacement Revisions: Clinical and Biomechanical Considerations; Lateral Lumbar Total Disc Replacement; Revision Strategies Following Lumbar Total Disc Replacement Complications Gary L. Lowery, MD, PhD Executive Vice President, Research and Technology, Paradigm Spine, LLC, New York, New York coflex Interspinous Implant for Stabilization of the Spine; Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis George Malcolmon Aphatec Spine, Inc., Carlsbad, California The Stabilimax NZ Posterior Lumbar Dynamic Stabilization System Thierry Marnay, MD Department of Orthopaedic Surgery, Clinique du Parc, Castelnau-le-Lez, France Can Lumbar Disc Replacement Be Used Adjacent to a Scoliotic Deformity?
Contributors
William S. Marras, MS, PhD Professor, College of Engineering; Professor, College of Medicine, The Ohio State University Biodynamics Laboratory, Columbus, Ohio The Development of a Personalized Hybrid EMG-Assisted/Finite Element Biomechanical Model to Assess Surgical Options Larry Martin, Jr., MD Resident, Department of Orthopaedic Surgery, Indiana University School of Medicine, Indianapolis, Indiana The Bryan Artificial Disc Joseph M. Marzluff, MD Neurosurgeon, Roper St. Francis Healthcare, Charleston, South Carolina SECURE-C Cervical Artificial Disc Koichi Masuda, MD Professor, Department of Othropedic Surgery and Biochemistry, Rush Medical College at Rush University Medical Center, Chicago, Illinois Animal Models for Human Disc Degeneration; Growth Factors for Intervertebral Disc Regeneration Paul C. McAfee, MD Associate Professor of Orthopedic Surgery and Neurosurgery, Johns Hopkins Hospital, Baltimore, Maryland Chief of Spinal Surgery, St. Joseph’s Hospital, Baltimore, Maryland Indications and Contraindications for Cervical Nonfusion Surgery: Patient Selection; Porous-Coated Motion (PCM) Cervical Arthroplasty; Complications of Anterior Cervical Approaches: Cervical Revision: Approach-Related Considerations; Cervical Disc Replacement Revisions: Clinical and Biomechanical Considerations; Cervical Disc Replacement Combined with Cervical Laminoplasty; Spinal Deformity in Motion-Sparing Technology Jeffrey R. McConnell, MD Clinical Assistant Professor of Surgery, Pennsylvania State University College of Medicine, Hershey, Pennsylvania SECURE-C Cervical Artificial Disc
xv
Lionel N. Metz, MD Medical Student Research Fellow, University of California, San Francisco, School of Medicine, San Francisco, California History and Evolution of Motion Preservation Richard Blondet Meyrat, MD Chief Resident, Department of Neurosurgery, Baylor College of Medicine, Houston, Texas Minimally Invasive Posterior Approaches to the Lumbar Spine Scott Dean Miller, DO Orthopaedic Surgeon, Crystal Clinic Orthopaedic Group, Akron, Ohio Theken eDisc: A Second-Generation Lumbar Artificial Disc Joji Mochida, MD, PhD Professor, Tokai University School of Medicine, Isehara, Kanagawa, Japan Professor and Chairman, Department of Orthopaedic Surgery, Tokai University Hospital, Isehara, Kanagawa, Japan Cell Therapy for Intervertebral Disc Degeneration Richard Navarro Vice President, Research and Development, Theken Disc, LLC, Akron, Ohio Theken eDisc: A Second-Generation Lumbar Artificial Disc Hazem Nicola, MD Orthopedics Spine Surgeon, Department of Biomechanics, Universidad Simón Bolívar, Caracas, Venezuela TranS1 Percutaneous Nucleus Replacement (PNR) Daniel M. Oberer, MD Attending Neurosurgeon, Department of Neurosurgery, Carolinas Medical Center, Charlotte, North Carolina Cervical Approaches: Anterior and Posterior Donna D. Ohnmeiss, MD President, Texas Back Institute Research Foundation, Plano, Texas Socioeconomic Impact of Motion Preservation Technology; Simultaneous Lumbar Fusion and Total Disc Replacement; The Future of Motion Preservation
Alvin H. McKenzie, MD, MChOrth, FRCSC Senior Active Staff, Department of Orthopaedic Surgery, Royal Alexandra Hospital Edmonton, Alberta, Canada The Basis for Motion Preservation Surgery: Lessons Learned from the Past
Carlos E. Oliveira, MD Head of Spine Sector of Orthopedic Department, Hospital de Servidor Publico Estadual, São Paulo, Brazil Anatomic Facet Replacement System (AFRS)
Alan McLeod, PhD Group Director Research and Development: Embroidery Technology, NuVasive (UK) Ltd., Taunton, United Kingdom The NeoDisc Elastomeric Cervical Total Disc Replacement
Douglas G. Orndorff, MD Resident, Department of Orthopaedic Surgery, University of Virginia Medical Center, Charlottesville, Virginia DISCOVER Artificial Cervical Disc
xvi
Contributors
Brett A. Osborn, DO Orthopedic Care Center, Aventura, Florida Dynamic Pedicle-Screw Stabilization with Nucleus Replacement
Kornelis A. Poelstra, MD, PhD Assistant Professor of Orthopaedics, University of Maryland Medical Center, Baltimore, Maryland DISCOVER Artificial Cervical Disc
Corey A. Pacek, MD Resident Physician, Department of Orthopaedic Surgery, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania Resident Physician, Department of Orthopaedic Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Gene Therapy for Intervertebral Disc Repair and Regeneration
Ben B. Pradhan, MD, MSE Director of Clinical Research, The Spine Institute, Santa Monica, California ProDisc-C Total Cervical Disc Replacemet; ProDisc-L Total Disc Replacement
Charles Park, MD Division Chief, Neurosurgery, Harbor Hospital, Baltimore, Maryland Theken eDisc: A Second-Generation Lumbar Artificial Disc Avinash G. Patwardhan, PhD Professor, Department of Orthopaedic Surgery and Rehabilitiation, Loyola University Stritch School of Medicine, Maywood, Illinois Director, Musculoskeletal Biomechanics Laboratory, Edward Hines, Jr. VA Hospital, Hines, Illinois The M6 Artificial Cervical Disc Carlos Fernando Arias Pesa´ntez, MD Neurospine Surgeon, Santa Rita Hospital, São Paulo, Brazil Revision Strategies Following Lumbar Total Disc Replacement Complications Piero Petrini, MD Department of Orthopaedics, City Hospital, Castle, Italy DIAM Spinal Stabilization System Luiz Pimenta, MD, PhD Associate Professor, University of California, San Diego, San Diego, California; Assistant Professor, Department of Neurosurgery, Federal University, São Paulo, Brazil; Assistant Professor, Department of Neurosurgery, Faculdade de Jundiai, São Paulo, Brazil Chief of Spine Surgery, Hospital Santa Rita, São Paulo, Brazil Cervical Disc Replacement Revisions: Clinical and Biomechanical Considerations; Lateral Lumbar Total Disc Replacement; Revision Strategies Following Lumbar Total Disc Replacement Complications; TranS1 Percutaneous Nucleus Replacement (PNR); NUBAC Intradiscal Arthroplasty; TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System Vinod K. Podichetty, MD, MS Director, Division of Research, Cleveland Clinic Florida, Weston, Florida Satellite: Spherical Partial Disc Replacement
Ann Prewett, PhD President, Replication Medical, Inc., Cranbury, New Jersey NeuDisc Artificial Lumbar Nucleus Replacement James P. Price Engineer, Theken Disc, LLC, Akron, Ohio Theken eDisc: A Second-Generation Lumbar Artificial Disc James Robert Rappaport, MD Assistant Clinical Professor, University of Nevada, Reno, Reno, Nevada; Sierra Regional Spine Institute, St. Mary’s Regional Medical Center, Reno, Nevada Kineflex|C Cervical Artificial Disc Christopher Reah, PhD Manager of Embroidery Technology Development, NuVasive (UK) Ltd, Taunton, United Kingdom The NeoDisc Elastomeric Cervical Total Disc Replacement Alejandro A. Reyes-Sa´nchez, MD Professor of Spine Surgery, Facultad de Medicina, Universidad Nacional Autonoma de México, Mexico, Distrito Federal, Head of Spinal Surgery Division, National Institute of Rehabilitation, Mexico, Distrito Federal The M6 Artificial Cervical Disc Souad Rhalmi, MSc Department of Neurosurgery, Research Centre, Sainte-Justine Hospital, Montreal, Quebec, Canada Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis K. Daniel Riew, MD Professor, Washington University School of Medicine, St. Louis, Missouri Mildred B. Simon Distinguished Professor of Orthopaedic Surgery; Professor of Neurological Surgery, Barnes-Jewish Hospital, St. Louis, Missouri Primary Indications and Disc Space Preparation for Cervical Disc Arthroplasty Charles H. Rivard, MD Department of Neurosurgery, Research Centre, Sainte-Justine Hospital, Montreal, Quebec, Canada Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis
Contributors
German Rodrı´guez, MD Attending Physician, Emergency Department, Hospital Del Prado, Tijuana, Mexico The PercuDyn System Thomas F. Roush, MD Orthopaedic Surgeon, Southeastern Spine Institute, Mount Pleasant, South Carolina Simultaneous Lumbar Fusion and Total Disc Replacement Scott A. Rushton, MD Assistant Professor, Department of Orthopedic Surgery, University of Pennsylvania, Philadelphia, Pennsylvania SECURE-C Cervical Artificial Disc Ashish Sahai, MD Clinical Instructor, Standford University Medical Center, Stanford, California Staff Orthopedic Surgeon, VA Palo Alto Health Care System, Palo Alto, California Invasive Diagnostic Tools Samer Saiedy, MD Department of Surgery, St. Joseph Hospital, Baltimore, Maryland Lumbar Anterior Revision: Preoperative Preparation and Approach Considerations Daisuke Sakai, MD, PhD Assistant Professor, Tokai University School of Medicine, Isehara, Kanagawa, Japan Assistant Professor and Attending Surgeon, Department of Orthopaedic Surgery, Tokai University Hospital, Isehara, Kanagawa, Japan Cell Therapy for Intervertebral Disc Degeneration Rick Sasso, MD Associate Professor; Chief of Spine Surgery—Clinical Orthopaedic Surgery, Indiana University School of Medicine, Indianapolis, Indiana Vice-Chairman, Department of Orthopaedic Surgery; Director, St. Vincent Spine Center, St. Vincent Hospital, Indianapolis, Indiana The Bryan Artificial Disc; TranS1 Percutaneous Nucleus Replacement (PNR) Thomas Schaffa, MD General Surgeon, Santa Rita Hospital, São Paulo, Brazil Lateral Lumbar Total Disc Replacement; Revision Strategies Following Lumbar Total Disc Replacement Complications Othmar Schwarzenbach, MD Spital Thun-Simmental AG, Thun, Switzerland NuCore Injectable Nucleus: An In Situ Curing Nucleus Replacement
xvii
Matthew Scott-Young, MBBS, FRACS, FAOrthA Associate Professor, Faculty of Health Sciences and Medicine, Bond University, Gold Coast, Queensland, Australia Visiting Medical Officer, Allamanda Private Hospital, Southport, Queensland, Australia Overall Revision Strategies: Lumbar; Cervical Arthroplasty Adjacent to Fusion, Multiple-Level Cases, and Hybrid Applications Lali H.S Sekhon, MD, PhD, FRACS, FACS Adjunct Associate Professor, Department of Physiology and Cell Biology, University of Nevada School of Medicine, Reno, Nevada Co-Diector, SpineNevada, Reno, Nevada Cervical Arthroplasty with Myelopathy Dilip K. Sengupta, MD Assistant Professor, Department of Orthopedics, Dartmouth College, Hanover, New Hampshire Assistant Professor, Department of Orthopedics, DartmouthHitchcock Medical Center, Lebanon, New Hampshire Dynamic Stabilization System Rajiv K. Sethi, MD Associate Spinal Surgeon, Department of Neurosurgery, Virginia Mason Medical Center, Group Health Spinal Surgery, Department of Neurosurgery, Seattle, Washington History and Evolution of Motion Preservation Farhan N. Siddiqi, MD Trinity Spine Center, Tampa, Florida Lumbar Endoscopic Posterolateral (Transforaminal) Approach Kern Singh, MD Assistant Professor, Department of Orthopedic Surgery, Rush University Medical Center, Chicago, Illinois Animal Modes for Human Disc Degeneration Matthew N. Songer, MD President and CEO, Pioneer Surgical Technology, Marquette, Michigan NUBAC Intradiscal Arthroplasty Gwendolyn A. Sowa, MD, PhD Assistant Professor, Department of Physical Medicine and Rehabilitation; Assistant Professor, Department of Orthopaedic Surgery; Co-Director, Ferguson Laboratory for Orthopaedic Spine Research, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania Assistant Professor, Department of Physical Medicine and Rehabilitation; Assistant Professor, Department of Orthopaedic Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Gene Therapy of Intervertebral Disc Repair and Regeneration
xviii
Contributors
Kristina Spate, MD Department of Vascular Surgery, Yale University School of Medicine, New Haven, Connecticut Management of Complications of the Anterior Exposure of the Lumbar Spine Jean Stecken, MD Neurosurgeon, Neurosurgical Department, Centre Hospitalier Regional, Orléans, France Mobidisc Disc Prosthesis Jean-Paul Steib, MD Professor of Orthopaedic Surgery, Université Louis Pasteur, Faculté de Medicine, Strasbourg, France Surgeon, University Hospital, Department of Orthopaedic Surgery Spine Unit, Strasbourg, France Mobi-C; Mobidisc Disc Prosthesis Jonathan R. Steiber, MD Fellow, Division of Spine Surgery, New York University— Hospital for Joint Diseases, New York, New York The CerviCore Cervical Intervertebral Disc Replacement; The FlexiCore Intervertebral Disc; Persistent Pain After Lumbar Total Disc Replacement Brian J. Sullivan, MD, FACS Director of Brain Services, Anne Arundel Medical Center, Annapolis, Maryland Persistent Pain After Cervical Arthroplasty Bauer E. Sumpio, MD, PhD Professor of Surgery and Radiology, Yale University School of Medicine, New Haven, Connecticut Chief, Vascular Surgery; Program Director, Vascular Surgery Fellowship Training Program, Yale-New Haven Medical Center, New Haven, Connecticut Technique of Anterior Exposure of the Lumbar Spine; Management of Complications of the Anterior Exposure of the Lumbar Spine Andelle L. Teng, MD Fellow, Orthopaedic Spine Service, Orthopaedic Spine Surgery, UCLA Medical Center, Los Angeles, California NFlex Randall Theken, MS Founder and CEO, The Theken Family of Companies, Akron, Ohio Theken eDisc: A Second-Generation Lumbar Artificial Disc Jens Peter Timm Vice President, Research and Development, Aphatec Spine, Inc., Carlsbad, California The Stabilimax NZ Posterior Lumbar Dynamic Stabilization System
P. Justin Tortolani, MD Orthopaedic Spine Surgeon, Orthopaedic Associates, Towson, Maryland Lumbar Anterior Revision: Preoperative Preparation and Approach Considerations Vincent C. Traynelis, MD Professor of Neurosurgery, University of Iowa, Iowa City, Iowa The Prestige Cervical Disc Patrick Tropiano, MD Department of Orthopaedic Surgery, Hôpital CHU Nord, Marseille, France Can Lumbar Disc Replacement Be Used Adjacent to a Scoliotic Deformity? Anthony Tsantrizos, MSc, PhD Disc Dynamics Inc., Eden Prairie, Minnesota DASCOR Alexander W.L. Turner, PhD Research and Testing Associate Manager, NuVasive, Inc., San Diego, California The NeoDisc Elastomeric Cervical Total Disc Replacement; Cerpass Cervical Total Disc Replacement Alexander R. Vaccaro, MD Professor, Department of Orthopaedic Surgery, Rothman Institute, Thomas Jefferson University, Philadelphia, Pennsylvania Advanced Spinal Anatomy for Cervical and Lumbar Nonfusion Surgery Jean-Marc Vital, MD Spinal Disorders Unit, Bordeaux University Hospital, Unité des Pathologies Rachidiennes, Hôpital Pellegrin, Bordeaux, France Mobi-C Archibald von Strempel, MD, DEng Professor, Medizinische Universität Junsbruck, Junsbruck, Austria Chief of the Orthopedic Department, Landeskrankenhaus, Feldkirch, Austria Cosmic: Dynamic Stabilization of the Degenerated Lumbar Spine Corey J. Wallach, MD Spine Fellow, Orthopaedic and Neurosurgery, UCLA Comprehensive Spine Center, Los Angeles, California NFlex Jeffrey C. Wang, MD Chief, Orthopaedic Spine Service; Associate Professor of Orthopaedic and Neurosurgery, UCLA Comprehensive Spine Center, Los Angeles, California NFlex
Contributors
Douglas Wardlaw, MB, ChB, ChM, FRCS(Edinburgh) Honorary Senior Lecturer, University of Aberdeen, Aberdeen, United Kingdom Honorary Professor, The Robert Gordon University, Aberdeen, United Kingdom Consultant Orthopaedic Spinal Surgeon, NHS Grampian, Aberdeen, United Kingdom BioDisc Nucleus Pulposus Replacement Scott A. Webb, DO Surgical Director; Fellowship Director, Florida Spine Institute Clearwater, Florida Spine Surgery; Department of Orthopedic Surgery Mease Countryside Hospital, Safety Harbor, Florida Total Facet Arthroplasty System (TFAS)
xix
Markus Wimmer, PhD Director, Tribology and Human Motions Laboratories, Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, Illinois Material Properties and Wear Analysis Oscar Yeh, PhD Director, Research and Testing, Intrinsic Therapeutics, Inc., Woburn, Massachusetts The Intrinsic Therapeutics Barricaid Device
Ian R. Weinberg, MD University of the Witwatersrand, Johannesburg, Gauteng, South Africa Kineflex
Anthony T. Yeung, MD Voluntary Clinical Instructor, University of California, San Diego, Department of Orthopedic Surgery, La Jolla, California; Desert Institute for Spine Care, Phoenix, Arizona; Executive Director of Intradiscal Therapy Sociey (IITS), Belgium, Wisconsin Lumbar Endoscopic Posterolateral (Transforaminal) Approach; NeuDisc Artificial Lumbar Nucleus Replacement
William C. Welch, MD Chief, Department of Neurological Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania Biomechanics of Nonfusion Devices: Novel Testing Techniques, Standards, and Implications for Future Devices
Christopher A. Yeung, MD Voluntary Clinical Instructor, University of California, San Diego, Department of Orthopedic Surgery, La Jolla, California Lumbar Endoscopic Posterolateral (Transforaminal) Approach
Bradley J. Wessman TranS1, Inc., Wilmington, North Carolina TranS1 Percutaneous Nucleus Replacement (PNR)
Hansen A. Yuan, MD Professor, Department of Orthopaedic and Neurological Surgery, SUNY Upstate Medical University, Syracuse, New York Theken eDisc: A Second-Generation Lumbar Artificial Disc; Aquarelle Hydrogel Disc Nucleus; NUBAC Intradiscal Arthroplasty
Peter G. Whang, MD Assistant Professor, Department of Orthopaedics, Yale University School of Medicine, New Haven, Connecticut Advanced Spinal Anatomy for Cervical Lumbar Nonfusion Surgery Nicholas D. Wharton, MS Senior Engineer, Medical Metrics, Inc., Houston, Texas Quantitative Motion Analysis (QMA) of Motion-Preserving and Fusion Technologies for the Spine Andrew P. White, MD Instructor, Harvard Medical School, Boston, Massachusetts Spinal Surgeon, Department of Orthopaedic Surgery, Beth Israel Deaconess Medical Center, Boston, Massachusetts Adjacent Segment Degeneration and Adjacent Segment Disease: Cervical and Lumbar Thomas Wilson Spine Wave, Inc., Shelton, Connecticut NuCore Injectable Nucleus: An In Situ Curing Nucleus Replacement
James J. Yue, MD Associate Professor, Department of Orthopaedic Surgery; Co-Chief, Division of Spinal Surgery; Director, Yale Spine Fellowship, Yale University School of Medicine, New Haven, Connecticut Indications and Contraindications for Lumbar Nonfusion Surgery: Patient Selection; Disc Space Preparation Techniques for Lumbar Disc Arthroplasty; Activ-L Artificial Disc; NeuDisc Artificial Lumbar Nucleus Replacement; The Stabilimax NZ Posterior Lumbar Dynamic Stabilization System; Can Lumbar Disc Replacement Be Used Adjacent to a Scoliotic Deformity? Considerations for Spinal Arthoplasty in Elderly and Osteoporotic Patients James F. Zucherman, MD Associate Staff Surgeon, St. Mary’s Hospital and Medical Center, San Francisco, California X-STOP Interspinous Process Decompression for Lumbar Spinal Stenosis
CHAPTER
1
The Basis for Motion Preservation Surgery: Lessons Learned from the Past Alvin H. McKenzie
“Structure is determined by need.” — Sir Harry Platt “Structure is determined by function.” — John Hunter “The purpose of man is action.” — Thomas Carlyle1
K E Y l
l
l
l
l
P O I N T S
Mobility and stability are equal partners in the structure of the human spine as determined by our need, function, and purpose of action: to survive and excel. Even with all other properties intact, a spine cannot function at full purpose or in longevity by stability alone—it also requires mobility. The spinal column is articulated, but like its vertebral trabeculae, it has been structured like fractal end-pinned masts or columns that have been reinforced with flying buttresses, spreaders, stays, and shrouds. Its articulating discs permit stable mobility and contribute to the shock absorbency of the structure. The vertebral bodies, not the discs, are the main shock absorbers of the spine: The vertebral trabeculae bend, compress, and rebound; the discs' nucleus pulposus is the hard center of motion and the spherical piston of the segmental shock absorber that indents the adjacent vertebral bodies and stabilizes the loaded spine while the annulus, when taught, allows motion and provides stability to the unloaded spine. When other structures are intact, the ideal motion preservation device should probably adapt to the trabecular nature of the vertebrae; replicate the nucleus pulposus in size, shape, and incompressibility; and be inert, indestructible, and possibly self-lubricating. The main lesson from the past is that established concepts of structure and function may not have always been correct: Acceptance of new ideas occurs when there is need for change, when contingencies prove that the status quo should not prevail, when there is opportunity for new concepts and methods of treatment, and when an innovative choice can be made.
INTRODUCTION The upright spine has provided mankind with a talisman and occasionally with a hoodoo. The spine's structure and function (mobility, stability, and dynamic equilibrium) have helped engender the world's most successful species, even as its mastlike nature, altered by bows and many articulated segments, has had a bias for breakdown. The upright spine's ingenious form (shape, configuration, balance, articulations, and connectivity) and substance (strength, elasticity, compressibility, and consistency) that constitute its structure seem to have been determined by its possessor's need, function, and purpose of action to survive and gain pre-eminence in a treacherous world. In health, the segmented spine's vital structures and vestments can perform like a multipinned ship's mast with power, agility, endurance, and grace to provide mobility with stability and (usually) freedom from pain. With damage to structure or rigging, it can still function after “bracing the mast” or “reefing the sails” and by careful, energetic sailing until it is “re-stepped by fixing” or fusion (mistakenly thought to produce stability). Since Mixter and Barr,2 there has been an option for the spine's owner to take responsibility for excisional restepping or discectomy. When damage, disorder, or discectomy leaves excessive motion at one of the spine's segments, it often spawns the corrosion of facet arthritis at the same level with instability and breakdown at the next level or at levels beyond. When similar misadventure leaves loss of motion at one of the segments, 3
4
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
the stiffness may stay the course in the short term but might ultimately forge the same superjacent and subjacent degenerative cascade that dogs instability or “successful” spinal fusion. The essential lesson to be learned from the past is that the spine cannot function at full purpose or in longevity without the essential duality of “stability and motion.” At the end of the 20th century, this concept of duality had been ignored and even suppressed. To paraphrase Shakespeare, “Whether ’tis nobler in the mind to suffer the slings and arrows of [spinal treatment learned from the past] or—by opposing end them: that is the question.”3 The answer may lie in those unheeded or rejected lessons from the past that deserve a second look.
and balance. It is articulated with flexible, firm discs and compressible and expandable bodies internally supplied with strong constantly remodeling trabeculae that bend or bow to transmit multidirectional forces along their axes and margins and give shelter to hemopoietic cells. Without this amazing spine, prehensile, bipedal self-sustaining human being with controlled positioning of limbs and body that can deliver power and agility where it is needed for flight, fight, work, or play with efficiency, endurance, and grace might not be the sceptered chieftain of the world but some extinct species of cephalopod or quadrapus. When it ails, we should aim not to diminish it or restrain it but to restore it. The Vertebral Body
THE SPINAL COLUMN We could first re-examine the structure, function, and needs of the spine, possibly the most complex organ we have. Pinned at the pelvis and sacrum like a fractal two-footed, keel-stepped, and single deck–stepped mobile mast, the conjoined spine then rises to the heavens with its spreaders, where its fractal, diverging condyles cup and double-pin the cranium: the receptor, modulator, transponder, and control center that contains and supersedes the lookout. The spine's dimensions accord remarkably well with a fir main mast of a ship (height ¼ beam 12 5; diameter ¼ 1 inch per 3 feet of height; critical load varies with the mast diameter by the power of three so that doubling the diameter would allow 8 times the load).4 Critical load, determined by the Euler formula, also varies inversely with the square of the mast height, so that halving the height by bowing or other means could quadruple the maximum load. It has been suggested that bowing of a spine could increase its critical load from 2 N/mm2 to 4 N/mm2 or from about 300 psi to 600 psi. With the weight of the cranium and the effects of spine's many bony spreaders and tensioned muscular shrouds and ligamentous stays, the spine develops its three mobile curves or three bows that decrease mast height and, like other bows, store energy to initiate motion, aid balance, and prolong endurance.5 The cervical spine remains mobile, stable, and capable of transport and relay of vital signals and elements. It provides suspension, motor power, circulation, and innervations for the body's mobile arms that aid balance but are primarily deployed for their myriad prehensile tasks. The thoracic spine also transports and transmits and is stabilized by the arrangement of its braced and cross-braced ribs and muscles that can be enhanced by generating the thoracic pressure cage.5 While giving lodging to the body's wind-catcher and circulatory center, the thorax provides platforms and leverage to the upper limbs. Below stands the lumbar spine supported by its own investments and its muscular shrouds that deploy to the thorax, lower limbs, and abdominal wall and join with the sacral-pelvic complex to form the hold or hull. Reinforced by the abdominal pressure cage, the trunk is capable of stable mobility and of housing a share of the body's filters, transporters, transmitters, energy converters, and magazines. The spine's bipedal and prehensile appendages synchronize with the spine for all manner and direction of motion and speed of propulsion or for feats of agility and strength. The healthy spine is expandable, contractile, resilient, rotatable, inclinable, flexible, lungeable, reboundable, deflectable, and stackable for stability
The vertebral body is composed primarily of cancellous bone contained anteriorly and laterally by a thin mantle of cortical bone. At the back, it assumes a thicker cortex to support the pedicles and other posterior vertebral elements, and above and below, it has thicker peripheral vertebral plates that give rise to fibers of the annulus fibrosis. Posterocentrally in the nuclear recess lies the cribriform cartilage plate that accommodates a nucleus pulposus,6 and is composed of chondral cartilage of notochord origin with no direct blood supply and no continuous underlying osseous plate. Instead, there are Y- and inverted Y-shaped, vertically directed, branching, or fractal7 trabecular columns that blend into the cribriform plate on one side while fibers from the nucleus and the inner lamellae of the disc penetrate it on the other (Fig. 1–1). The trabecular columns form primary and secondary compression trabeculae: slender, compound, intermediate length masts that can bend and rebound not unlike a fresh wishbone. They fuse where they meet the transecting spreader-like tension trabeculae that prevent vertical buckling, and increase critical load. Paracentrally and posteriorly about the nuclear recess, the primary compression trabeculae are numerous and well developed, and in the central nuclear recess, they are less dense and weaker.8 They
n F I G U R E 1–1. Fractal design of trabeculae in upper vertebral body at junction of cribriform plate and peripheral vertebral plate.
CHAPTER 1
The Basis for Motion Preservation Surgery: Lessons Learned from the Past
5
density in response to local stresses (Wolff's law).11 The vertebral body column is also braced by its flying buttresses, the pedicles, laminae, and appendages. The side aisle of the buttresses, roofed with ligamentum flavum, becomes the halyard sleeve that guards the spinal cord, ganglia, and nerve roots; the articular facets permit guided and constrained spinal motion, and afford each segment two of its three points of stability; the spinous and transverse processes act as mast spreaders for muscular and ligamentous shrouds and stays that augment spinal shape, stability, and motion. Possibly inspired by the fractal structure of the vertebral trabecular mesh that gives strength, compressibility, and rebound to the vertebral bodies, industry has created an airless metal tire, the Michelin Tweel (Fig. 1–4).12 The Intervertebral Disc
increase their moment of inertia and critical load by expanding axially like barrel staves en route to the nuclear recess opposite (Fig. 1–2).9 Secondary compression trabeculae from the bony vertebral plates run obliquely to the adjacent vertebral body walls. Primary tension trabeculae, which are more numerous close to the vertebral plates, run horizontal to the planes of the end plates, slightly bowed and at right angles to the primary compression trabeculae. Secondary tension trabeculae extend into the pedicles. Young's Modulus (the modulus of elasticity that corresponds to the critical load of the vertebral body) lies between that of wood and magnesium or aluminum.4 Because trabecular bowing stores energy, and because the average healthy vertebral body can be compressed by one tenth of its height without trabecular fracture, a vertebra may be so compressed and rebound to its normal height (Fig. 1–3).10 The vertebral body's strength varies with trabecular
The intervertebral discs, firmly bound to adjacent vertebral bodies, share their stresses and responsibility for controlled mobility through their annulus fibrosis and a nucleus pulposus, which also contribute to two (cephalad and caudal) cribriform end plates. Like the vertebral body, it derives blood from the anterior and lateral vertebral plexus that will drain to the Batson/azygos system of veins.13 The annulus fibrosis is composed of laminated fibrocartilage with obliquely directed fibers that are anchored to the adjacent vertebral plates centrally and peripherally by extensions of trabecular fibers and peripherally by the periosteum and the overlying anterior and posterior longitudinal ligaments. Successive laminar layers of the annulus run obliquely in alternate directions and, therefore, become tight and rigid under tension of the normally positioned healthy nucleus pulposus.6 The annulus may bulge when the nucleus pulposus, under vertical loading, indents the cribriform plates and causes the adjacent vertebral bodies to approach one another by up to 4 mm at each level. The nucleus pulposus is gelatinous and composed of “viscous proteoglycans imbedded with loose fibrous strands arranged in a felted mesh of undulating bundles that contain a profusion of fusiform cells resembling reticulocytes and vacuolar darkly nucleated chondrocytes.”6 The nucleus pulposus is hygroscopic and may
n F I G U R E 1–3. Discography with compression and rebound of vertebral bodies (Roaf).
n
n F I G U R E 1–2. Primary (central) compression trabeculae bowing between cribriform plates: Secondary (peripheral) compression trabeculae running to vertebral walls. Both reinforced by primary horizontally bowed primary tension trabeculae.
F I G U R E 1–4. Michelin’s fractal airless metal tire.
6
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
change internal tension by an ability to imbibe fluid.14 In health, the nucleus pulposus lives and breathes by its contact with the cribriform cartilage plate. It exchanges metabolites and fluid in response to pressure gradients and osmalarity between the nucleus and the adjacent vertebral bodies. The healthy, in place, and intact nucleus pulposus, encased by multiple layers of annulus fibrosus, is not unlike the core of a classic golf ball (a round bag of latex compressed by layer after layer of elastic bands) that behaves like a hard sphere.15 As Robert Roaf and others10,16 have demonstrated, noncritical vertical loading of healthy vertebrae with an intact annulus and healthy nucleus pulposus and cribriform plate will cause the vertebral plates to spring inward while the nucleus pulposus retains its spherical shape (Fig. 1–5). As the rigid nucleus indents the adjacent vertebral bodies, it creates stability of the segment by friction of the nucleus with the indented body as the disc space narrows and only then will the annulus bulge. If the loading is relieved, the rebound of the vertebral body will return the nucleus pulposus to its place, tighten the annular fibers, restore shape of the annulus, and maintain stability. When the critical load is exceeded, the vertebral body will fracture, the nucleus pulposus will remain intact, and the segment can become unstable.10 Notwithstanding any previous concepts or teachings about the roles of the vertebrae and the discs, the vertebrae, not the discs, are the main shock absorbers of the spine. The annulus fibrosis is the check-ligament that permits limited universal intervertebral mobility. With normal loading, the healthy nucleus pulposus is transposed by vertebral body pressure to tighten the annulus and produce ligamentous intervertebral stability, and with heavy loading, the rigid, spherical nucleus obtains stable seating in the nuclear recesses, where it acts as a piston to depress the cribriform plates, bend the trabeculae. and produce contact stability. With return of normal loading, the vertebral body rebounds, the nucleus resumes its position, and the annulus tightens.
and breakdown of nuclear tissue and biochemical changes in the disc. Interdiscal biologic irritants such as proteoglycans, prostaglandin E2, inflammatory cytokines (tissue necrosis factor-a and interleukins) that can be activated by matrix metalloproteinases (extracellular kinases) create a medium with a low pH that invites inflammatory cells into the disc.18,19 The granulation tissue forms scar tissue that promotes shrinkage of the nucleus and buckling of the annulus. With migration and dislocation of scar tissue and more breakdown of the fibrocartilage of the internal annulus, external annular buckling can lead to ruptures and possibly protrusions of degenerate discs (Fig. 1–6).20 The breakdown in the nucleus pulposus and annulus will ultimately cause alteration of the stresses affecting the adjacent vertebral bodies and their vertebral trabeculations. An irregular trabecular system may ultimately evolve (and observations suggest that initial changes occur in the horizontal trabeculae near the vertebral body plates8). Annular ruptures with or without protrusions may permit disc fluid to leak from the disc to generate backache, headache, or radicular pain without sensory or motor loss. Anterior ruptures of necrotic discs with low pH may cause abdominal pain to mimic abdominal catastrophes or pain of gynecologic origin.21 Commonly, the degenerative disc becomes unstable and causes back pain of facet joint origin that can be accompanied by disc protrusions with the potential for neurologic impairment.22 Efforts by the patient to control pain from instability by hyperlordosing the spine and locking the facets can lead to posterior element hypertrophy with canal and foramenal stenosis.23,24 The patient may or may not proceed from a state of good disc health to invalidism or follow any of a host of scenarios that may be slow, intermittent, or rapid. The degenerative disc may never become surgical by current criteria but can produce states of impairment
Pathologic Changes in the Vertebral Body: Disc Complex
The post-traumatic, degenerative changes due to aging in the vertebral body, the intervertebral disc, and the nucleus pulposus probably begin with infraction of the cribriform cartilage plate17 and nutrition failure of the nucleus pulposus succeeded by apoptosis
Spreading inward (late 30s) Radial tears
Angular deformity of annular rings (30s) Clefting (40s) Progressive softening (from age 30)
n F I G U R E 1–5. The vertebral body is compressed; the nucleus maintains its shape (Roaf).
n F I G U R E 1–6. Fahrni’s chronology of degenerative changes of the intervertebral disc.
CHAPTER 1
The Basis for Motion Preservation Surgery: Lessons Learned from the Past
that erode health, employability, well-being, social and economic equanimity, and happy pursuits. Motion preservation with stability and freedom from pain within reasonable parameters of safety should be one of the great medical missions of physicians and spinal surgeons of this new century, whatever the severity or stage of the spinal disorder.
M
REQUIREMENTS FOR THE IDEAL MOTION PRESERVATION DEVICE Prerequisites An effective motion preservation device (prosthesis) for the spine should expand the indications for spinal surgery. A cavalier attitude toward reconstructive spinal surgery is not being advocated, but there is a need of more and improved treatment choices for those who may fail to meet current guidelines for current treatment modalities but have unresolved chronic spinal pain of disc origin that impairs health and imposes significant social and economic hardship. A prosthesis may be used to maintain or restore discogenic instability and improve pain relief after discectomy or at sites of previous discectomy or to “top off” degenerative discs above or below levels requiring fusion for other reasons.25 Patients selected for arthroplasty should be those with proven discogenic pain or instability, or both, with or without disc protrusion without advanced degenerative arthritis or structural annular or bony faults at the level of potential arthroplasty. Also, these patients have failed to respond to sustained comprehensive conservative treatment and have lost their ability to maintain well-being, productivity, strength, and agility. Other prerequisites would accord with the nature and extent of spinal pathology demonstrated by routine standing radiographs, magnetic resonance imaging (MRI) and discography, and observance of all the usual preoperative medical precautions. The Replacement Device or Prosthesis An intervertebral prosthesis should ideally restore the deficiencies of the motion segment it is to occupy: It should restore to the segment its mobility, stability, and freedom from pain and enable it to function in longevity with power, agility, and endurance. Biomechanics The prosthesis should maintain intersegmental stability and permit appropriate rotation about the X, Y, and Z axes (i.e., flexion Xþ and extension X- 10 degrees at L1–2 to 15 degrees at L5-S1; left rotation Yþ and right rotations Y- 2 degrees average, and lateral bending right Zþ and lateral bending left Z- 6 degrees average), as well as optimum translation along the X, Y, and Z planes (i.e., yaw left Xþ or yaw right X- 2 mm, elevate (ascend) Yþ or sink (descend) Y- 2 mm, or shift fore Zþ or aft Z- 2 mm) with coupling (the phenomenon of consistent association of one motion (rotation or translation) about or along one axis or plane with another motion about or along another axis or plane (Fig. 1–7).8 The prosthesis should permit those excursions required for normal spinal function but should also preserve segmental stability by limiting motion in accordance with the foregoing limits.
7
Coupling n
F I G U R E 1–7. Coupling: the phenomenon of consistent
association of rotation or translation about or along one axis or plane with another motion about or along a second axis or plane. (White AA, Panjabi MM: Clinical Biomechanics of the Spine, 2nd ed. Philadelphia, JB Lippincott Company, 1990.)
Integrity of the prosthesis and the segment in which the prosthesis resides will depend on the following factors: l Comprehensive surgical management of all significant spinal pathology, l Preparation of an adequate approach to the intervertebral space that is capable of or may be modified to become capable of providing safe entry for and biomechanically sound containment of an appropriate prosthesis, l Presence of biomechanically sound opposing vertebral segments with reasonably compatible trabecular systems within the involved vertebral bodies, l Subjacent and superjacent segmental health and reasonably normal annular, ligamentous, and neuromuscular controls, and l Availability of a comprehensive selection of suitable, inert, indestructible, properly sized, self-lubricating, or self-incorporating prostheses. Lessons from the Past: The Logic of the Times (1950s and 1970s) By mid-20th century, spinal problems were well recognized: Discectomy was frequently followed by disc space collapse, loss of mobility, recurrent protrusions, canal and foramenal stenosis, scarring with root entrapment, instability, and facet arthritis. l Decompression, although seldom advocated or performed, could lead to recurrent stenosis, instability, perineural fibrosis, dural tears, failure of pain relief, and neurologic sequelae. l Fusion was often attended by perioperative morbidity, pseudarthroses, stenosis, sacroiliitis, fixation failure, and adjacent level instability. Spinal fusion, with its shortcomings, became the gold standard of care for the failed spine. l Stabilization of the spine after discectomy by Lucite pegs (Gardner, 1950s) and methyl acrylic (Cleveland, 1955; Hamby, 1957) had met with cool or no enthusiasm.26 l
Arthroplasty of the spine was considered to be a nasty phrase, as per the history of the specialists demonstrates: l Paul Harmon, from 1959 to 1961, used vitallium balls through an anterior approach to stabilize vertebral segments to assist fusion. On nine out of 13 occasions, he found that they could work well as stand-alone stabilizers (the first disc arthroplast27) but had his California license suspended and spent 2 years in South America before his restoration.
8 l
l
l
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
Ulf Fernstrom, from 1962 to 1972, favored steel ball arthroplasy of the spine over Hirsch's spinal wires to prevent and treat spinal instability. In 1966, he reported treating 105 patients by steel ball arthroplasty from a posterior approach with few complications, safe fixation, slow subsidence, retained stability, absence of spurring, and assisted mobility.28 By 1972, he reported 195 total patients treated at 262 lumbar levels by posterior approach and 13 patients treated by anterior cervical approach, with excellent results in 65% of patients. Although 85% of his patients had been on total disability pension before surgery, 85% had returned to work after his surgery.29 It was said by his detractors, “15% of his patients never worked again!” For his efforts, he was removed from his seat at the University of Udevalla to the village of Hudiksvaal, north of the Arctic Circle, with suppression of further opportunity to publish. Reitz and Joubert, (1964) had carried out steel ball, hemispherical, and Silastic top arthroplasties at 32 cervical and nine lumbar levels in South Africa before having their surgery suspended.30 Al McKenzie, in 1971, reported short-term results of 40 steel ball arthroplasties carried out during 1969 and 1970.31 Twenty-five years later, in 1995, the author was finally able to have published his report of 10- to 20-year (17-year average) results of re-examining 67 of 103 patients treated at 155 levels, whose age at surgery averaged 44 years.32 Although obliged to suspend the procedures in 1974, the author reported excellent and good long-term results for 83% of patients in the discectomy and prosthesis group and 75% excellent and good results for the spondylosis and prosthesis group, with 95% of all patients having returned to work. One prosthesis became displaced and was exchanged, whereas another was removed for discitis. Disc space preservation had been excellent and good in 55%, fair in 28%, and poor in 17% (Fig. 1–8).
n F I G U R E 1–9. Need good exposure and thorough discectomy with accurate sizing, angulation, distraction, occasional intrusion on pedicle, protection of cauda and nerve, and accurate centering in nuclear recess.
What About the Intervertebral Metal Ball Arthroplasty? An inert, indestructible, properly sized, and correctly seated sphere appears to be an acceptable motion preservation device for use between vertebral bodies. In the lumbar spine, it can usually be inserted safely from a posterior approach (Fig. 1–9) and in the cervical spine from an anterior approach (Fig. 1–10). By its firmness, a carefully sized metal sphere can seat securely in a clean nuclear recess to restore intervertebral height, motion, and stability. When thus installed, it simulates the nuclear fulcrum to permit coupling in all planes and axes. Like the nucleus pulposus, the hardest part of the vertebral body/annulus/nucleus complex, it functions as the piston of the shock-absorbing mechanism of the intervertebral segment and achieves coaptation with the trabecular pattern already developed in the subjacent and superjacent vertebral bodies that
n n
F I G U R E 1–8. Fair preservation of disc space with minimal
enthesopathy at target level: good preservation at superjacent level after 34 years.
F I G U R E 1–10. Anterior approach in cervical spine (rarely in lumbar spine). Should restore optimum lordosis. After larger prosthesis, this man returned to underwater research on Great Barrier Reef for the Australian government.
CHAPTER 1
The Basis for Motion Preservation Surgery: Lessons Learned from the Past
n
F I G U R E 1–11. This patient has been able to pursue golf, weather permitting every snow-free day since he underwent surgery 34 years ago.
9
superstitions, and beliefs that regulated members for inclusion, hunting, defense, war, domesticity, agriculture, and industry. Surviving societies had strong establishments that could select the leader and control the loyal followers. Challenges to the order brought fight or flight, and harmony might bring a place in the choir.36 Order bred habit, tradition, and sometimes prejudices, entrenched leadership, entrenched succession, and the status quo. Except in war, chaos, or catastrophe, there would usually be “no need” for change. Dror37 wrote that either continuity of a status quo or acceptance of a new idea hung on the weighting of need, contingencies, opportunity, and choice. New ideas, unless conceived by the leaders, would fail unless there was a survival threatening need for it; unless failure to accept might result in a contingency that could prejudice their wisdom or authority; and unless there was opportunity for them to accept the “new” without loss of honor and especially when the choice for change could be made by the leaders. Tribal man may still be alive and well. If we are to see spinal arthroplasty supercede discectomy and spinal fusion as a favored option, we may need to persuade the sages that a need for change has occurred; that the contingencies might reflect unfavorably on those who embrace the status quo; that opportunities for change are at hand; and that worthy choices can be made (Fig. 1–12). Quo Vadis?
accord with Wolffe's law. An appropriately sized prosthesis of spherical shape is disinclined to initiate bone resorption, so that any subsidence of a sphere occurs only slowly while preventing instability at the level of placement and diminishing degenerative disc changes at levels above and below (Fig. 1-11). If all of the foregoing were true, what happened? What logic shelved the procedure for more than 30 years? Speculation runs from inspired personal or political considerations to preservation of national image or the mislogia that Socrates33 defined as the hatred of new ideas. The simple answer may be that spinal fusion had been designated as the gold standard of care in the belief that once it was perfected, all our spinal troubles would be over. In spite of an historical profusion of spinal tools, instruments (screws, hooks, rods, plates, and cages), and surgical approaches and the martialing of biologic agents, the comfort of the biopsychosocialists, and the outlawing of tobacco, the need for a gold standard remained unrequited. Contingencies continued to plague success. Surgeons once dedicated to the pursuit of better spinal fusions still looked for opportunities to deliver better spinal care. So far, the choice of arthroplasty has mostly been limited to those achieved by an anterior approach, with some of those surgeries producing new kinds of morbidity or being succeeded by prosthetic loosening.
Those who elect to have change in the management of spinal disorders, who would propose motion preservation in the spine, must know their customer, the spinal column. They must understand its needs; anticipate the contingencies; determine where, when, and why the opportunities for motion preservation should be exercised; and what the choice of prosthesis should be. As one of Fernström's patients demonstrates (Fig. 1–13), long-term success is possible in spinal surgery.
The Need for Change Review of man's past history, errors, and oversights will reveal that mankind has developed a genetic resistance to new ideas and to change.34 When tribal man bonded for protection, his little band grew into or was absorbed into a society.35 Societies spawned self-appointed coteries (condo police) who supplied rules, myths,
n
F I G U R E 1–12. Large prostheses below appear to have preserved all levels of this lady’s spine.
10
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
n
F I G U R E 1–13. One of Fernstrom’s patients in his 81st year, 38 years after disc replacement, clearing snow from his walk.
REFERENCES 1. Carlyle T: On Heroes and Hero Worship and the Heroic in History. Project Guttenburg, 1841. 2. Mixter WJ, Barr JS: Rupture of the intervertebral disc with involvement of the spinal canal. New Eng J Med 211:210–215, 1934. 3. Shakespeare W: Hamlet. Second quarto, London, 1604. Printer, James Roberts for Nicolas Lang. 4. www.classicmarine.co.uk/Articles/masts.htm. Accessed December 5, 2006. 5. Gordon JE: Structures, or Why Things Don't Fall Down. London, Pelican Books, 1978. 6. Parke WW: Applied Anatomy of the Spine, Arthrology. In Rothman RH and Simeon FA (eds.): The Spine, 2nd ed. Philadelphia, WB Saunders, 1982, pp. 27–35. 7. Richardson MJ, Gillespy T III: Fractal Analysis of Trabecular Bone. University of Washington, Department of Radiology. www.rad. washington.edu/exhibits/fractal.html. Accessed December 10, 2006. 8. White AA, Panjabi MM: Clinical Biomechanics of the Spine, 2nd ed. Philadelphia, JB Lippincott Company, 1990. 9. Van Nostrand's Scientific Encyclopedia, 4th ed. Princeton, D. Van Nostrand Company, Inc., 1968, pp. 388–389. 10. Roaf R: A study of the mechanics of spine injuries. J Bone Joint Surg 42B:810–823, 1960. 11. Wolff J: Das Gesetz der Transformation de Knochen. Berlin, Hirschwald, 1892. (English translation: Maquet P, Furlong R: The law of bone remodeling. Berlin, Springer-Verlag, 1986.)
12. www.autoblog.com/2006/05/16/video-michelin-tweel-in-motion; www. michelin.com. Accessed December, 2006. 13. Piersol GA: Piersol's Human Anatomy. Philadelphia, JB Lippincott Company, 1930. 14. Charnley J: The imbibition of fluid as a cause of herniation of the nucleus pulposus. Lancet 1:124–127, 1952. 15. Armstrong JR: Lumbar Disc Lesions. London and Edinburgh, E & S Livingston, Ltd., 1952. 16. Lindahl O: Mechanical properties of dried defatted spongy bone. Acta Orthop Scand 47:11–19, 1976. 17. Schmorl G, Junghanns H: The Human Spine in Health and Disease, 2nd ed. New York, Grune & Stratton, 1971. 18. Herkowitz HN, Dvorak J, Bel G, et al: The Lumbar Spine, 3rd ed. Philadelphia, Lippincott Williams & Wilkins, 2004. 19. Mitchell PEG, Hendry NGC, Billewicz WZ: The chemical background of intervertebral disc prolapse, J Bone Joint Surg 43B:141–151, 1961. 20. Fahrni H: Age changes in lumbar intervertebral discs. Can J Surg 13:65–71, 1970. 21. Fernstrom U: Intervertebral disc degeneration with abdominal pain. Acta Chir Scand 113(6):436–437, 1957. 22. White AH: Lumbar Spine Surgery. Toronto, CV Mosby, 1987. 23. Kirkaldy-Willis WH: Lumbar spinal stenosis. Clin Orthop 99:30–50, 1974. 24. Kirkaldy-Willis WH: Spinal stenosis. Clin Orthop 115:2–3, 1976. 25. Steffe A: Personal communication, October 23, 2005. 26. Garcia R: Challenges to Spine Surgery Meeting. Puerto Rico, January 2002. 27. Harmon P: Personal communication, 1969. 28. Fernstrom U: Arthroplasty with intercorporal endoprosthesis in herniated disc and in painful disc. Acta Chir Scand 355 (suppl):154–159, 1966. 29. Fernstrom, U: Personal Communication. 30. Reitz H, Joubert MJ: Replacement of cervical intervertebral disc with a metal prosthesis. S Afr Med J 38:881–884, 1964. 31. McKenzie AH: Steel ball arthroplasty of lumbar discs. J Bone Joint Surg Br 54:266, 1972. 32. McKenzie AH: Fernstrom intervertebral disc arthroplasty: Long term evaluation. Orthopaedics Intl 3:313–324, 1995. 33. Plato: Dialogues of Plato. U.S.A., Pocket Books Inc., 1950. 34. Wells HG: The Outline of History. Doubleday Company, Garden City, NY, 1940. 35. Toynbee A: A Study of History. Oxford, UK, Oxford Press, 1972. 36. Keegan J: A History of Warfare. Vintage Books, Random House, Inc., 1993. 37. Dror Y: Submission to the United Nations Secretariat. New York, July 11, 1995.
CHAPTER
2
History and Evolution of Motion Preservation Rajiv K. Sethi, Lionel N. Metz, and David S. Bradford
K E Y l
l
l
l
l
P O I N T S
This chapter focuses on various spirited and innovative approaches to motion preservation at the intervertebral disc. A device must be durable, biocompatible, and address the complex functions of the disc that provides both shock absorption and intervertebral motion. In many disc replacement designs, articular motion preservation is addressed at the expense of shock absorption. We have classified the described devices by their common design principles. The long-term outcomes of currently used disc replacements will influence future designs.
INTRODUCTION The intervertebral disc is an integral part of the triarticular vertebral motion segment. This joint lies between adjacent vertebral bodies and allows flexion and extension in the sagittal and coronal planes, rotation in the horizontal plane, and limits translation in the horizontal plane between two vertebral bodies. The disc has a viscoelastic nucleus pulposus that absorbs and redistributes axial load to the surrounding annulus fibrosus. Its elastic bandlike annulus fibrosus stores and returns energy from axial loads and constrains the mobility of the motion segment. The relative importance of these complex functions varies at different locations—cervical, thoracic, or lumbar—in the spine. When considering the challenge of disc replacement, it is important to determine which of these functions and material properties are important to implement. This discussion reviews the evolution of a variety of conceptual designs in disc arthroplasty in the past several decades, with particular attention to the following core questions: Is it feasible to implant the device? Does the device offer a mechanical advantage over fusion? Will the device offer motion preservation that results in an outcome preferable to fusion and maintain the mechanics of the remaining healthy segments? How long will such a device function and will the failure rate be acceptable? And if there is a failure, can salvage be undertaken with little risk to the patient? Will the device function properly long enough to merit implantation? These key questions
go into the design of the implant. Only long-term clinical studies will provide the definitive answers. The precise events of intervertebral disc degeneration have been widely studied but not completely understood. However, it is known that when disc tissues become damaged or degenerated, pathologic joint mechanics as well as pathologic changes in innervation and nociception cause back pain, which is usually exacerbated by movement.1 When this pain is persistent despite conservative treatment, operative management can be considered. Operative intervention for back pain such as a spinal fusion is itself fraught with complications and often has poor outcomes.2 One core principle of the orthopaedic approach to degenerative joint disease is that pathologic motion causes pain, and if that motion can be arrested, the patient will benefit. This concept of joint fusion guided the orthopaedic treatment of arthritis in the knee and hip for many years and continues to be the mainstay of spine surgery. The evolution toward arthroplasty to treat painful knees and hips has raised the important question of whether degenerative joint disease of the spine might be better treated by arthroplasty than by arthrodesis. Considering that fusion of a single lumbar motion segment for degenerative disease of the spine may result in pain relief in some patients, artificial disc replacement technology should at least replicate that or do better. Total joint replacement for end-stage arthritis of the hip and knee has revolutionized the field of orthopaedic surgery. Both primary total hip and knee replacement have resulted in high rates of patient satisfaction, and surgeons and patients have become accustomed to excellent long-term results. Surgeons hope to achieve similar benefits with disc arthroplasty. In broad terms, the intervertebral disc serves two distinct roles: (1) facilitation of limited motion between two vertebral bodies, and (2) absorption of shock through viscoelastic properties of the nucleus pulposus/annulus fibrosis construct.3 Thus, the application of arthroplasty to the degenerated intervertebral disc is more complex than that of the knee or hip. The efficacy of disc replacement is not only reliant on a mechanically sound replacement of the disc but also on the assurance that the disc is the primary source of pain. This would imply that the facet joints are not a contributory factor to the back pain. The innervation 11
12
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
of the disc is also poorly understood, and therefore, disc replacement may not truly relieve the patient's symptoms.4,5 Early arthroplasty techniques attempted to use the abovementioned conceptual tenets and hypotheses. In 1955, David Cleveland published the first paper on disc arthroplasty utilizing methyl-acrylic in 14 patients. At 5 months after the operation, 100% of these patients had resolution of their back pain. To aid in the stabilization of the lumbar disc space, Paul Harmon devised a Vitallium sphere that he implanted through an anterior retroperitoneal approach in 13 patients from 1959 to 1961. In 1964, Ulf Fernstrom published his results of 105 lumbar cases in which he posteriorly implanted a stainless steel ball bearing device. The implant group had a 12% rate of low back pain following implantation versus 60% in the control group. RATIONALE FOR DISC ARTHROPLASTY Arthrodesis at a single vertebral level is not associated with the same degree of functional loss as knee or hip arthrodesis.6 Even patients with multiple fused levels adjust quite well to fusion and enjoy significantly improved quality of life at long-term followup.7 Most surgeons would agree that simple motion preservation is not a good enough reason to pursue disc replacement. Although the loss of mobility in a single fused segment is small, it has a potentially large effect on the adjacent segments that compensate for the functional loss. This compensatory mechanism is implicated in the high rate of adjacent segment degeneration seen after short fusions.8–10 Long-term studies confirming that disc arthroplasty is associated with a lower incidence of adjacent segment degeneration are needed to verify the value of this intervention.
INTERVERTEBRAL DISC REPLACEMENT COMPREHENSIVE CLASSIFICATION SYSTEM Ideas regarding disc replacement are not always the result of methodic review of previous patent designs and literature. Thus, newer patents often appear to be less sophisticated brethren of established designs. Therefore, rather than listing the patented inventions in chronologic order, the following discussion of disc replacement design describes the designs as derivations of prototypic designs. To this end we have attempted to classify these devices by the central mechanical principles. The 14 classifications used are as follows: ball bearing, ball and socket, fixed dome, articulating plates, hard plates/hard core, hard plates/soft core, screw-in dowel, spring and piston, complex mechanical/vertebral body replacement, fluidfilled bag, simple elastomer/polymer prostheses, spiral nucleoplasty, in situ polymerization, and biologic. Ball Bearing The Fernstrom Ball, shown in Figure 2–1, is often credited with being one of the first substantial attempts at intervertebral disc arthroplasty.12 The device is the prototypic simple ball bearing design. The ball is placed into the disc space after discectomy, allowing for movement in several planes. This device has an extremely low surface contact area on initial implantation, which leads to a predictable subsistence through the cancellous bone beneath. This subsistence was severe in many cases, often leading to pain, disc height loss, and eventually loss of motion and many times fusion at the implanted level. Subsequently, the use of the Fernstrom ball has been discontinued in the United States. Ultimately, the indications for use of the device have evolved to encompass fusion rather than motion.
BIOMECHANICAL CHALLENGES The vertebral column is a polyarticular structure that functions under the influence of a delicate balance of forces. Generally, the anterior structures, the vertebral bodies and discs, support 90% of the axial load, whereas the posterior elements, primarily the facet joints, support the remainder of the load in the lumbar spine. It is paramount that any disc replacement maintain the delicate load distribution and emulate the mechanics of the invertebral disc (IVD) as closely as possible to avoid facet arthrosis or adjacent segment degeneration. Disc arthroplasty can potentially avoid the compensatory motion in adjacent segments that occurs after short fusions, thus preventing adjacent segment disease. However, at present, there are no disc arthroplasty devices in common use that protect the spine by absorbing axial loads in addition to preserving articular motion. Motion preservation in the posterior elements of the spine is arguably a simpler task than for the intervertebral disc. Dynamic stabilization systems such as Dynesys, as well as facet joint arthroplasty are in development. These systems are covered in later chapters of this volume. The remainder of this chapter focuses on various spirited and innovative approaches to motion preservation at the intervertebral disc. We have made extensive use of the thorough chronologic cataloguing of spine motion preservation patents performed by Szpalski et al in their 2002 review11; thus, we would like to make a special acknowledgement of their efforts.
n F I G U R E 2–1. Fernstrom ball. http://jborden.org/etc/ C2056846532/E20060723070512/index.html. Accessed January 23, 2007.
CHAPTER 2
History and Evolution of Motion Preservation
13
that incorporated an elastic annulus around his ball joint.16 Xavier's design in 1998 added criss-crossed wires that would aid the stability of the construct.17 Fixed Dome The Sabitzer and Fuss device (Fig. 2–4) is a prototype for the fixed dome concept.18 In this device, the disc is replaced by a rigid dome that is firmly affixed to the inferior vertebral body and freely articulates with the inferior end plate of the superior vertebral body. The latter has a natural concavity that offers some restraint of movement as well as load repartition.11 Kehr designed a cervical implant in which there is a mounting bracket resting on the inferior vertebral body with an arm that anchors into the side of the vertebral body.19 A domed articulating prosthesis attaches to the bracket and articulates with the upper cartilaginous end plate. Articulating Plates
The idea of using a ball-and-socket design is no doubt inspired by its successful application total hip arthroplasty.13 This association is clearly demonstrated by Ibo's ball-and-socket implant (Fig. 2–2).14 In contrast, the Graham prosthesis (Fig. 2–3) involves a cylindrical anchor that resides in each of two adjacent vertebral bodies and a ball-and-socket joint that lies in line with the axis of load.15 The device also offers some compressive cushioning mediated by spacers around the ball joint. In 1994, Mazda put forth a similar device
The articulating plate designs use hard materials with low friction coefficients to resurface the end plates, restore disc space height, and distribute the surface stress.11 The concept involves one concave surface and one convex surface, resulting in a wide range of rotational and translational motions—up to 6 degrees of freedom like the native disc. These designs are derivations of the simple ball-andsocket designs used in hip arthroplasty. Although not the first example of this principle, Salib and Pettine's design (Fig. 2–5) had a small articulating surface and limited distribution of surface stresses.20 Shinn and Tate describe a very similar design for their device.21 Evolutions of this concept involve larger curved articulating surfaces, such as the device by Yuan (Fig. 2–6).22 Shelokov patented a similar device that uses two dome-shaped condyles, similar to the design of the leading total knee replacements.23 Lesoin et al24 designed an articulating plates device that screws into the vertebral bodies, much like the Salib and Pettine design. Cauthen designed a slit-and-rib11 implant (Fig. 2–7) that constrains rotational motion in the coronal and horizontal planes and allows for sagittal rocking chair–type motion and possibly some
n F I G U R E 2–3. The Graham ball-and-socket device is designed with a ball-and-socket joint in-line with the cylindrical anchors, which are seated in the adjacent vertebral bodies.
n F I G U R E 2–4. The Sabitzer and Fuss device involves a domed prosthesis that articulates with the upper end plate and anchors to the lower vertebral body.
n
F I G U R E 2–2. Ibo ball-and-socket prosthesis.
Ball-and-Socket
14
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
n F I G U R E 2–5. Salib and Pettine incorporate the ball-and-socket type of articulation into two plates, one concave and one convex.
anterior or posterior translation.25 It would be easy to imagine a tremendous amount of stress at the base of the thin rib during side bending. Hard Plates/Hard Core The previously discussed articulating plate designs have the limitation of high friction in cases in which metal on metal articulation is used. This predisposes them to problems with metallic wear debris, which may build up exponentially once wear is initiated. Both of these problems were addressed by the addition of a polymer core. In these designs, the core is either floating, unconstrained, between two concave end plates, such as in the Schellnack-Büttner Janz (SB) CHARITÉ26 (Fig. 2–8), first implanted in 1984 by Karin Büttner Janz, or it is anchored to the flat surface of one endplate while it articulates to the concave surface of the other end plate, as in the semiconstrained ProDisc design,27 first implanted in 1990 by Thierry Marnay. Both of these designs are discussed in detail in later chapters. Most designs use a polyethylene or other plastic type of core articulating with one or both metal end plates. In 1974, Hoffman-Daimler invented
n F I G U R E 2–6. Yuan incorporated the articulating plate design into implants with large articulating surfaces.
n F I G U R E 2–7. Cauthen’s slit-and-rib design uses two articulating components, both convex at the articulating surfaces. It constrains rotational motion in the coronal and horizontal planes but allows for rocking motion in the sagittal plane.
such a device that consisted of two metal plates with a complex plastic core.28 The device was never implanted into humans. In 1978, Weber patented a design slightly different from the prototype design, using two concave polyethylene surfaces and an ovoid ceramic core for use in the lumbar and cervical spine.29,30 Various materials have been used for the core, such as metal,31 fluoroplastic,32 and ceramic. Ojima used hydroxyapatite-coated plates to allow for better bony incorporation of the implant,33
´ Artificial Disc (DePuy Spine, Inc., n F I G U R E 2–8. CHARITE Raynham, MA). (Photograph Courtesy of DePuy Spine, Inc.)
CHAPTER 2
n F I G U R E 2–9. Erickson and Griffith device used a polymer core to decrease friction with the articulating superior end plate.
whereas others, Bryan and Kunzler,34 used plate extensions that could be screwed into the anterior aspect of the superior and inferior vertebral bodies. Variations on the hard plate/hard core theme involved a core secured to one of the plates. Gordon's35 variation used a hemispheric bearing anchored to the inferior plate but articulating with the superior plate, similar to the ProDisc. Erickson and Griffith's design (Fig. 2–9) used a concave core. These designs offer rotational motion as well as some translational motion, depending on the difference between the radii of the convex and concave surfaces.36 The Bullivant design appears to be very similar to the ProDisc II design, shown in Figure 2–10, with the core attached to the superior metal end plate instead of the inferior one. Hard Plates/Soft Core Few of the above-mentioned designs adequately address axial shock absorption or try to re-create the indispensable viscoelastic properties of the IVD.37 This is the limitation of the rigid materials used in the core of these designs. Several designs have used the
History and Evolution of Motion Preservation
15
concept of an articulating core and attempted shock absorption by using a compressible elastomer as the core material. Khvisuyk's38 design is composed of a silicon cushion between two metallic plates. The Downey prosthesis39 was similar and had two large bolts to screw it into the superior and inferior end plates. Another example is the AcroFlex, by DePuy Spine (Raynham, MA), shown in Figure 2–11. It uses a silicone elastomer core bonded to two titanium plates.40 It is designed for implantation in the lumbar spine. Other designs involving a soft core secured to both plates were brought forth by Harms et al41 and Oka et al.47 Harms described a silicone rubber core that was sandwiched between two hard plastic plates. Oka's similar invention was composed of a composite body made of a polyvinyl alcohol hydrogel flanked by ceramic or metallic end plates. Zdeblick and McKay's43 design called for rigid end plates separated by a core with properties similar to that of the nucleus pulposus. Several similar designs followed by Graf.44–46 Baumgartner and Takeda47,48 used projecting rims on the inside of the end plate around the elastomeric core to constrain some of the translation and rotational motion as well as limit the extent of compression.11 Some designs used capsule-like elastomer cores between hard plates with or without spikes.49–51 Bainville et al52 recognized the risk of core migration after several million cycles called for an antiexpulsion device around the core that would still allow for compression of the soft core. Viart and Marin's device (Fig. 2–12)53 may have prevented core migration by using a doughnut-shaped core, but without a fulcrum in the center of the core, the device is likely to have limited sagittal and coronal rotational motion. Graf envisioned a device that allowed for motion through asymmetric compression of the cushion, while providing shock absorption (Fig. 2–13). His blueprint used standard concave end plates and a hydrophilic gel core. The mobile articulation hard plate/soft core designs have the additional consideration of an increased friction coefficient of many elastomeric materials compared with harder polymers. The nonarticulating soft core designs, such as the Graf device, undergo a tremendous amount of shear stress when resisting translation,54 which is a concern for device longevity.
n
F I G U R E 2–10. ProDisc II Lumbar Total Disc Replacement, courtesy of Synthes, Inc (West Chester, PA). The device has two metallic end plates separated by a polyethylene core that locks into the lower end plate.
n
F I G U R E 2–11. The second generation AcroFlex-100 consists of an HP-100 silicone elastomer core bonded to two titanium end plates (DePuy Spine, Inc., Raynham, MA). (Courtesy of Spineuniverse.com.)
P A R T
16
I
Introduction to Motion Preservation Surgery of the Spine
n
F I G U R E 2–12. Viart and Marin’s device uses a doughnutshaped soft core designed to prevent core migration relative to the two hard plates.
n
F I G U R E 2–14. Bryan and Kunzler’s cylindric implant is screwed into place and secured by anchoring brackets. The mobile elements, or “resilient bodies,” reside in the center of the implant.
Screw-In Dowel The screw-in prostheses strongly resemble the BAK interbody fusion implants55 manufactured by Zimmer. They are shortgrooved cylinders bisected along the long axis to create hollow half cylinders. The functional components reside inside the half cylinders created by the bisection. One theoretic advantage of these devices is the ease of implantation. Bryan et al designed several of these devices; an example is shown in Figure 2–14, with varying motion-sparing or cushioning inner components.56, 57 He refers to the inner components simply as “resilient bodies,” which reflects the difficulty in predicting the ideal materials for this application. Also in 2000, Cauthen58 designed an implant using two threaded, solid half cylinders separated by a ball bearing that was articulated with concavities in a flat surface (Fig. 2–15). The design was intended for use in the cervical spine. Mehdizadeh59 used the concept of two threaded half cylinders but linked them with small springs to provide elasticity to the construct.
in IVD replacement. Patil's spring box (Fig. 2–16) is designated as our prototype for the spring-and-piston design.60 It contained overlapping half boxes linked with several interposed springs. Dumas et al also describe a spring-linked plate design.61 Beer and Beer62 patented a device very similar to the Patil device that includes screw plates for fixation to the anterior surface of the superior and inferior vertebral bodies. Butterman's complex design included pistons as shock absorbers.63 The Pisharodi expandable device, shown in Figure 2–17, encapsulated the springs in a hollow bag and would be collapsible to a fraction of its expanded size to facilitate implantation.64 The spring implant created by Hedman
Spring and Piston The function of the healthy disc in absorbing axial shock37 has led many inventors to investigate the role of small springs and pistons
C3
C2
C1
C2
C3
n
F I G U R E 2–13. Graf’s device constrained the soft core with interlocking teeth on the two hard plates, preventing core migration and presumably ascribing motion to differential compression rather than articulation.
n
F I G U R E 2–15. Cauthen’s design incorporates the ease of a screw-in device with the simplicity of a ball-and-socket type of articulation.
CHAPTER 2
History and Evolution of Motion Preservation
17
n
n
et al of the Kostuik team65 was implanted in large animal models, which is a more advanced stage of development than most of the previously described devices. The device involved two titanium plates, hinged posteriorly, and two springs to create an elastic construct (Fig. 2–18). The springs in this construct were said to have withstood more than 100 million cycles without failure and were implanted with some success in sheep models.10 Overall, the success of hard plate/hard core prostheses has led to a decrease in the development of spring-containing designs.
at either end. Many other devices combine a multitude of mechanical principles and properties to re-create the disc's native properties and functions.
F I G U R E 2–16. Patil’s cervical interlocking plates are separated by springs, which impart a cushioning effect to the device.
F I G U R E 2–18. Kostuik’s implant relied on two springs for compressive shock absorption and was wedge shaped to maintain lumbar lordosis.
Fluid-Filled Bag Typically, these devices consist of a soft fillable bag with a specific shape and material properties that is filled with fluid of liquid or gaseous form. In 1975, Froning67 designed a discoid bladder
Complex Mechanical/Vertebral Body Replacement There have been many proposed motion preservation devices for use in the lumbar spine that use complex designs that do not fit neatly into one category. These complex mechanical devices often have the burden of dozens of moving parts that may cause potential licensing companies to shy away. Simple devices often win out because production, quality control, and difficulty of implantation hinder the adaptation of complicated or clumsy devices. One clever device by Main et al (Fig. 2–19) used an expandable system capable of replacing the disc as well as a substantial amount of the adjacent vertebral bodies.66 It incorporates an “elastomeric suspension medium” around each of the device's suspension plates
n
F I G U R E 2–17. Pisharodi’s expandable prosthesis uses springs to cushion axial loads and, more importantly, can be implanted in a collapsed state.
n
F I G U R E 2–19. Main’s device replaced the vertebral bodies as well as the disc. It has a mobile element in the center of its length that may allow for some degree of normal biomechanics at that level.
18
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
that was to be inserted into the disc space in deflated form and then filled with fluid once positioned in place. The disc bladder design has been one of the most popular forms of disc prosthesis because of its inherent shock absorption and small size before implantation. Several others made variations on the size, shape, material, and filling fluids used in these devices.68–78 Simple Elastomer/Polymer Prostheses The observation that some elastomers have similar viscoelastic properties to the IVD79 led to the trial of several of these materials as simple disc replacements. Many of these designs were very crude, simply using a polymer cast in the shape of the nucleus. The prototype design by Nachemson, in the 1950s, sought to replace the degenerated nucleus pulposus with a resilient material, such as silicone.80 Others also used silicone, but various similar materials have been used to achieve the desired amount of compressibility and friction.81–83 A major distinction of these devices is that no artificial end plate is used as an articulating surface, so the native cartilaginous or bony end plates form the articular surfaces. There are many other notable designs.84,85 Although the physical properties of native nucleus pulposus (NP) and these materials may be similar, the remodeling potential of the NP allows these material properties to be maintained for several decades, whereas even the most resilient elastomer is likely to fall prey to wear over this time period. Kotani's woven disc prosthesis,86 shown in Figure 2–20, might be less susceptible to shear forces but may be
prone to a substantial foreign body reaction from microfiber degradation.87 Spiral Nucleoplasty Most of these devices are similar to the above-mentioned elastomer prostheses. The novelty in this design is that the elastomer can be implanted through a much smaller annulotomy. This is one of the few device types that can be inserted through a minimally invasive posterior approach after removal of the nucleus. In 1975, Stubstad patented his coiled elastomer implant (Fig. 2–21),88 spawning several similar designs, collectively referred to as spiral nucleoplasty devices.89 In Situ Polymerization Hydrogels and other biocompatible polymers have been investigated. This approach has appeal because it can potentially be performed through a posterior minimally invasive approach. Despite these benefits, the design lacks the ability to resist degradation. Challenges arising from polymer resorption and degradation, and precipitation of a foreign body reaction may hinder the efficacy of such systems. Biologic Cells have the remarkable ability to produce impressively resilient tissues capable of being strong yet flexible. Both biologic tissues and synthetic materials with these properties are subject to wear and eventually failure. However, only living tissues have the unique ability to adapt and remodel, properties that allow tissues to retain or even increase their resiliency. Unfortunately, these properties can also lead to pathologic remodeling to the point where the native function of the tissue is lost as in degenerative disc disease.90,91 Nonetheless, it is quite reasonable to assume that motion preservation of the spine would be best accomplished by a faithful restoration of the living tissues. Thus, many inventors have sought to find appropriate cells, carriers, and implantation techniques to biologically regenerate the intervertebral disc.92,93 Most of these designs seek to restore the desiccated nucleus pulposus in degenerative disc disease. Two cellular candidates have prevailed in these designs: (1) chondrocytes, which produce a large
n
F I G U R E 2–20. Kotani used advanced techniques in polymer design to weave a three-dimensional disk-shaped structure out of ultrahigh-molecular-weight polyethylene fibers. Spray coated with ceramic granules, the implant was designed to incorporate into the surrounding structures.86
n
F I G U R E 2–21. This spiraled variant of Stubstadt’s elastomer implants might have been implanted minimally invasively.
CHAPTER 2
amount of aggrecan, the substance that imbibes water in the nucleus and imparts viscoelastic properties to the disc, and (2) mesenchymal stem cells, which are capable of both proliferating and differentiating into nucleus pulposus–like cells. These technologies are areas of active research and are covered in more detail in later chapters. LESSONS LEARNED Several lessons about spinal implants have been learned. The contact surface between the implant and the vertebral end plate must be maximized in order to prevent subsistence of the device.11 This concept was demonstrated by the Fernstrom ball implants, which have a very small end plate contact surface. A larger contact surface allows for wider load repartition and less stress on the end plates; however, this factor mandates that disc arthroplasty with rigid materials will likely always require an anterior approach. The complex role of the disc, providing both viscoelastic load distribution and articular motion, creates a difficult design challenge. Most designs used today, namely the CHARITÉ III and ProDisc II, have largely abandoned viscoelastic considerations and focused on motion preservation. Other designs using the viscoelastic properties of silicone, or elastic properties of springs or pistons, or both, put less importance on motion preservation, particularly in the transverse plane. Seldom has a device aimed to accurately reproduce both roles, and none has been very successful in completing this task. Indeed, experience has shown that materials best suited for millions of cycles of articular motion are hard and have low friction coefficients. These materials, such as metals, polyethylene, and ceramics, inherently lack additional viscoelastic properties. CONCLUSION Each of the designs presented attempts to emulate one or more of the essential biomechanical properties of the intervertebral disc. Long-term clinical outcomes of the few disc replacements that have been evaluated in clinical trials continue to be measured; these outcomes will guide the next generation of disc arthroplasty designs. Surgeons must be increasingly circumspect when considering performing new procedures whose long-term efficacy is not established. The appropriately high treatment expectations of our patients will be properly addressed by prudent clinical judgment in selecting the best, as opposed to the newest, intervention. REFERENCES 1. Vernon-Roberts B, Pirie CJ: Degenerative changes in the intervertebral discs of the lumbar spine and their sequelae. Rheumatology 16:13–19, 1977. 2. Ibrahim T, Tleyjeh IM, Gabbar O: Surgical versus non-surgical treatment of chronic low back pain: a meta-analysis of randomised trials. Int Orthop Nov. 21, 2006, epub. 3. Lotz JC, Hsieh AH, Walsh AL, et al: Mechanobiology of the intervertebral disc. Biochem Soc Trans 30:853–858, 2002. 4. Aoki Y, Akeda K, An H, et al: Nerve fiber ingrowth into scar tissue formed following nucleus pulposus extrusion in the rabbit anularpuncture disc degeneration model: effects of depth of puncture. Spine 31:E774–E780, 2006. 5. McCarthy PW: Innervation of lumbar intervertebral disks—a review. J Peripher Nerv Syst 3:233–242, 1998.
History and Evolution of Motion Preservation
19
6. Chow DH, Luk KD, Evens JH, Leong JC: Effects of short anterior lumbar interbody fusion on biomechanics of neighboring unfused segments. Spine 21:549–555, 1996. 7. Padua R, Padua S, Aulisa L, et al: Patient outcomes after Harrington instrumentation for idiopathic scoliosis: a 15- to 28-year evaluation. Spine 26:1268–1273, 2001. 8. Ghiselli G, Chen J, Kaou M, et al: Adjacent segment degeneration in the lumbar spine. J Bone Joint Surg 86:1497–1503, 2004. 9. Eck JC, Humphreys SC, Hodges SD: Adjacent-segment degeneration after lumbar fusion: a review of clinical, biomechanical, and radiologic studies. Am J Orthop 28:336–340, 1999. 10. Kostuik JP: Intervertebral disc replacement: Experimental study. Clin Orthop Rel Res 337:27–41, 1997. 11. Szpalski M, Gunzburg R, Mayer M: Spine arthroplasty: a historical review. Eur Spine J 11:65–84, 2002. 12. Fernstrom U: Arthroplasty with intercorporal endoprothesis in herniated disc and in painful disc. Acta Chir Scand Suppl 357:154–159, 1966. 13. Charnley J: The long-term results of low-friction arthroplasty of the hip performed as a primary intervention. J Bone Joint Surg Br 54:61–76, 1972. 14. Ibo I, Pierotto E, inventors: Prosthesis of the cervical intervertebralis disk. US patent 5,755,796, May 26, 1998. 15. Graham DV, inventor: Artificial disk. US patent 5,246,458, September 21, 1993. 16. Mazda, inventor: Prothèse discale intervertébrale, French Patent 2694882, 1994. 17. Xavier R, Xavier S, Xavier S, inventors: Vertebral body prosthesis. US patent 6,063,121, May 16, 2000. 18. Sabitzer and Fuss, Austrian Patent 405237B, 1997. 19. Kehr, European Patent 069926, 1996. 20. Salib RM, Pettine KA, inventors: Intervertebral disk arthroplasty. US patent 5,258,031, November 2, 1993. 21. Shinn GL, Tate JD, inventors: Artificial intervertebral disk prosthesis. US patent 5,683,465, November 4, 1997. 22. Yuan HA, inventor: Low wear artificial spinal disc. US patent 5,676,701, October 14, 1997. 23. Shelokov AP, inventor: Articulating spinal disc prosthesis. US patent 6,039,763, March 21, 2000. 24. Lesoin et al, French Patent 2718635, 1995. 25. Cauthen, JC, inventor: Articulating spinal implant. US patent 6,179,874, January 30, 2001. 26. Link HD: History, design and biomechanics of the LINK SB Charité artificial disc. Eur Spine J 11:98–105, 2002. 27. Rousseau MA, Bradford DS, Bertagnoli R, et al: Disc arthroplasty design influences intervertebral kinematics and facet forces. Spine J 6:258–266, 2006. 28. Hoffman-Daimler, German Patent 2263842, 1974. 29. Weber, Swiss Patent 624673, 1978. 30. Weber, Swiss Patent 640131, 1979. 31. Keller A, inventor: Surgical instrument set. US patent 4,997,432, March 5, 1991. 32. Savchenko et al, Russian Patent 2140229, 1999. 33. Ojima, European Patent 0317972, 1989. 34. Bryan V, Kunzler A, inventor: Human spinal disc prosthesis. US patent 5,865,846, February 2, 1999. 35. Gordon J, inventor: Intervertebral disc replacement prothesis. US Patent 20050234553, October 20, 2005. 36. Huang RC, Giardi FP, Commisa FP Jr, Wright TM: The implications of constraint in lumbar total disc replacement. J Spinal Discord Tech 16:412–417, 2003. 37. Vuono-Hawkins M, Langiana NA, Parsons JR, et al: Materials and design concepts for an intervertebral disc spacer. II. Multidurometer composite design. J Appl Biomater 6:117–123, 1995. 38. Khvisuyk, USSR Patent 895433, 1982. 39. Downey EL, inventor: Vertebra prosthesis. US patent 5,147,404, September 15, 1992.
20
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
40. Fraser RD, Ross ER, Lowery GL, et al: AcroFlex design and results. Spine J 4:245–251, 2004. 41. Harms et al, German Patent 3911610, 1990. 42. Oka M, Gen S, Ikada Y, Okimatsu H, inventors: Artificial intervertebral disc. US patent 5,458,643, October 17, 1995. 43. Zdeblick and McKay, World Patent 0074606, 2000. 44. Graf, French Patent 2772594, 1999. 45. Graf, French Patent 2775891, 1999. 46. Graf, French Patent 2801782, 1999. 47. Baumgartner, Europe Patent 0566810, 1992. 48. Takeda M, inventor: Display apparatus. US patent 5,270,697, December 14, 1993. 49. Grammon and Gauchet, French Patent 2723841, 1996. 50. Yves-Alain, inventor: Elastic disc prosthesis. US patent 5,676,702, October 14, 1997. 51. Bisserie M, inventor: Intervertebral disk prosthesis. US patent 5,702,450, December 30, 1997. 52. Bainville D, Laval F, Roy-Camille R, et al, inventors: Intervertebral disk prosthesis. US patent 5,674,294, October 7, 1997. 53. Viart and Marin, French Patent 2895985, 2001. 54. Bao QB, Yuan HA: Prosthetic disc replacement: the future. Clin Orthop Rel Res 394:139–145, 2002. 55. Nibu K, Panjabi MM, Oxland T, Cholewicki J: Multidirectional stabilizing potential of BAK interbody spinal fusion system for anterior surgery. J Spinal Disord 10:357–362, 1997. 56. Bryan V, Kunzler A, inventor: Human spinal disc prosthesis. US patent 5,865,846, February 2, 1999. 57. Bryan et al, World Patent 0013620, 2000. 58. Cauthen JC, inventor: Articulating spinal implant. US patent 6,019,792, February 1, 2000. 59. Mehdizadeh HM, inventor: Disc replacement prosthesis. US patent 6,231,609, May 15, 2001. 60. Patil AA, inventor: Artificial intervertebral disc. US patent 4,309,777, January 12, 1982. 61. Dumas et al, French Patent 2734148, 1996. 62. Beer JC, Beer JM, inventors: Synthetic intervertebral disc. US patent 5,458,642, October 17, 1995. 63. Butterman GR, inventor: Intervertebral prosthetic device. US patent 5,827,328, October 27, 1998. 64. Pisharodi M, inventor: Artificial spinal prosthesis. US patent 5,123,926, June 23, 1992. 65. Hedman TP, Kostuik JP, Fernie GR, Maki B, inventors: Artificial spinal disc. US patent 4,759,769, July 26, 1988. 66. Main JA, Wells ME, Keller TS, inventors: Vertebral prosthesis. US patent 4,932,975, June 12, 1990. 67. Froning EC, inventor: Intervertebral disc prosthesis and instruments for locating same. US patent 3,875,595, April 8, 1975. 68. Edeland HG: Suggestions for a total elasto-dynamic intervertebral disc prosthesis. Biomater Med Devices Artif Organs 9:65–72, 1981. 69. Monson GL, inventor: Synthetic intervertebral disc prosthesis. US patent 4,863,477, September 5, 1989.
70. Frey O, Koch R, Planck HMF, inventors: Joint endoprosthesis. US patent 4,932,969, June 12, 1990. 71. Strong MD, inventor: Method of determining the quality of a crimped electrical connection. US patent 5,197,186, March 30, 1993. 72. Baumgartner W, inventor: Intervertebral prosthesis. US patent 5,171,280, December 15, 1992. 73. Fuhrmann G, Gross U, Kaden B, et al, inventors: Intervertebral disk endoprosthesis. US patent 5,002,576, March 26, 1991. 74. Monteiro, Spanish Patent 2094077, 1997. 75. Krapiva PI, inventor: Disc replacement method and apparatus. US patent 5,645,597, July 8, 1997. 76. Gauchet, World Patent 0035387, 2000. 77. Minda and Schmidt, European Patent 1157675, 2001. 78. Weber and Da Silva, World Patent 0190786, 2001. 79. Kennedy JP: From thermoplastic elastomers to designed biomaterials. J Polymer Sci Part A: Polymer Chemistry 43:2951–2963, 2005. 80. Carl A, Ledet E, Yuan H, Sharan A: New developments in nucleus pulposus replacement technology. Spine J 4:325–329, 2004. 81. Schneider PG, Oyen R: Surgical replacement of the intervertebral disc. First communication: replacement of lumbar discs with siliconrubber. Theoretical and experimental investigations (author's transl). Z Orthop Ihre Grenzgeb 112:1078–1086, 1974. 82. Schneider PG, Oyen R: Proceedings: Disk displacement. Experimental studies—clinical consequences. Z Orthop Ihre Grenzgeb 112:791–792, 1974. 83. Kuntz JD, inventor: Intervertebral disc prosthesis. US patent 4,349,921, September 21, 1982. 84. Marcolongo and Lowman, World Patent 0132100, 2001. 85. Banks, World Patent 0112107, 2001. 86. Kotani Y, Abumi K, Shikinami Y, et al: Artificial intervertebral disc replacement using bioactive three-dimensional fabric: design, development, and preliminary animal study. Spine 27:929–935, 2002. 87. Willert HG: Reactions of the articular capsule to wear products of artificial joint prostheses. J Biomed Mater Res 11:157–164, 1977. 88. Stubstad JA, Urbaniak JR, Kahn P, inventors: Prosthesis for spinal repair. US patent 3,867,728, February 25, 1975. 89. Studer and Schärer, European Patent 1157876, 2001. 90. Duance VC, Crean JK, Sims TJ, et al: Changes in collagen crosslinking in degenerative disc disease and scoliosis. Spine 23:2545–2551, 1998. 91. Crean JK, et al: Matrix metalloproteinases in the human intervertebral disc: role in disc degeneration and scoliosis. Spine 22:2877–2884, 1997. 92. Crevensten G, Walsh AJ, Ananthakrishnan E, et al: Intervertebral disc cell therapy for regeneration: mesenchymal stem cell implantation in rat intervertebral discs. Ann Biomed Eng 32:430–434, 2004. 93. Lotz JC, Kim AJ: Disc regeneration: why, when, and how. Neurosurg Clin N Am 16:657–663, vii, 2005.
CHAPTER
3
Classification of Spine Arthroplasty Devices Karin Bu¨ttner-Janz
K E Y l
l
l
l
l
P O I N T S
Numerous devices for arthroplasty of the lumbar and cervical spine have been developed notably in the past 10 years, which are classified in this chapter according to main categories and subcategories. A definite classification of the devices following clinical indications is not possible yet because of missing data and clinical experience. However, the present knowledge about biomechanics and derived thereof of clinical effectiveness of devices is being critically carried together as a fundament for a classification based on indications. The five main categories of the classification are theoretically aimed at uniform topographic regions of the lumbar and cervical spine. The three categories for dorsal implantation so far pertain only to the lumbar spine, and for the cervical spine, nucleus replacement is used only to a small degree. Because clinical long-term objectives preferentially stand in the foreground, the main categories and subcategories of spine arthroplasty devices are listed according to the materials and material combinations, implant designs, biomechanical principles, and kinds of application with respect to the devices. By using this classification, all presently available and probably all future devices can be classified, although it will be need to be determined whether a sixth main category for the replacement of the complete functional spinal unit will have to be added.
INTRODUCTION The constantly increasing number of spine arthroplasty devices allows for no complete overview of implants and materials that are already being clinically applied for the preservation of spinal motion. In this chapter, the majority of the presently known devices are dwelled on as much such as to use and explain their major characteristics for the classification of the respective implant or material groups. For this chapter, we requested that 60 companies complete a questionnaire. Twenty-one of these companies responded. The author's personal experience with motion preservation devices over 25 years also aided in the evaluation of implants and materials as well as coherences. DEFINITION The term spine arthroplasty refers to a replacement in the region of the spine, by which a motion is achieved (“arthro” ¼ joint,
“plasty” ¼ replacement). The common criteria of all devices for spine arthroplasty are their respective implant design and material characteristic for a repair, preservation, or improvement of function of the cervical or lumbar functional spinal unit (FSU), including motion. The term implant is used for a replacement with a specific design, and the term material is used for substances including their characteristics. Device is the generic term for both. FSU is used synonymously for spinal motion segment. ANATOMY AND BIOMECHANICS The intervertebral disc and the facet joints of the same motion segment together generate a three-joint complex. This functional unit of a spinal motion segment and its dependence on the biomechanical loads differs from the biomechanics of the hip or knee joint. The intervertebral discs increase in height from cranial to caudal and are mostly higher ventrally than dorsally in the cervical and lumbar spine, thus resulting in a ventrally open intervertebral angle. During the load transfer within an intact intervertebral disc, the nucleus pulposus with its high internal pressure protects the annulus fibrosus against overloading. With its internal pressure, the nucleus pulposus on the one hand maintains the distance between the vertebral bodies, and on the other hand, it firmly holds them together because the pressure acting on the annulus fibrosus strains the lamellae of the fibrous ring going from vertebra to vertebra.1 The nucleus pulposus thus secures the intervertebral distance and the height of the foramina intervertebralia, as well as part of the stability of the FSU. The annulus fibrosus functions by limiting all segmental movement, including rotation; it also modulates range of motion. The annulus fibrosus thus has a strong influence on the quality of the segmental motion and the stability of the spinal motion segment. The facet joints and the ligaments of the spine (particularly the ventral and dorsal longitudinal ligament and the muscles) also have an influence on the stability. During movement of adjacent vertebrae, the intact intervertebral disc adjusts to the change of the distance between vertebral bodies by mass movements within the nucleus pulposus, resulting in the intervertebral inclinations. Depending on the age of the 21
22
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
person, the disposition for degeneration, the spinal topography, the functional loads, and other factors, the range of motion of different functional spinal units of the spine differ in magnitude. There are, however, mean values for the segmental extension, flexion, lateral bending, and axial rotation for the cervical as well as the lumbar spine.2,3 These mean values for extension, flexion, lateral bending, and axial rotation vary up to 2.5-fold in the caudal motion segments of the cervical spine and up to 8.5-fold in the caudal motion segments of the lumbar spine. The sagittal position of the facets differs for the cervical spine compared with the lumbar spine and has an effect on the extent of mobility, especially axial rotation. In the lumbar spine, the mean facet joint angle in the transversal plane increases from 25 to 53 degrees from L1-L2 to L5-S13 as evidence for the action of facet joints in resisting shear forces. At the same time, the position of the center of rotation changes. Whereas the facet joints are real joints, the intervertebral discs do not have the typical anatomic structures of a joint despite their mobility. Phylogenetically, only a limited morphologic adaptation to the upright human gait took place. This can be seen in physiologic changes to the aging intervertebral discs and the high rate of morbidity of the cervical and lumbar spine worldwide. The closed functional interaction of an intervertebral disc with the facet joints of the same motion segment can be seen in corresponding degenerative changes of the facet joints as a result of degeneration of an intervertebral disc with the decreased height of the intervertebral space. Conversely, a resection of dorsally stabilizing spinal structures, including parts of the facet joints, leads to higher loads and in the long term degeneration of the intervertebral disc at the same level. The physiologic curve of the spine from cervical to lumbosacral can produce damping, or shock absorption, for the spine during load transfer. However, in many cases with degenerative changes this function no longer exists because of the erect position of the lumbar or cervical spine. This reactive adaptation leads to the question of the importance of a damping function of the human spine. Relief of the strain of the facet joints by virtue of a craniocaudal damping caused by the intervertebral disc is not given and not necessary when the facet joints are in a vertical position, as is the case in the lumbar spine. Comparative consideration of the construction of intervertebral discs of quadrupeds also shows characteristics of damping, although there is no vertical spinal load.4 Anatomically and phylogenetically, there is no relevance to the shock absorption function for a load relieving of facet joints of same level or of neighboring intervertebral discs and facet joints given. Therefore the integrated damping function of an intervertebral disc is rather to be seen as a prerequisite for intervertebral mobility. However, vibration loads may cause a malfunction in the region of the intervertebral disc in humans.5 With respect to all 23 intervertebral discs of the human body, the loss of one shock absorber should not affect the vibration transferred into adjacent levels. Putz6 writes that intervertebral discs absorb shock only to a limited degree. BACKGROUND OF MOTION PRESERVATION DEVICES Motion preservation devices can directly influence the biomechanical function of the motion segment depending on their design,
material, the region of the spine intended for the implantation, and the place of implantation into the FSU. The biomechanical functioning of the spinal motion segment may, in turn, have shortor long-term effects on the patient's symptoms and clinical progress. Depending on the effect on the FSU, the devices can also influence adjacent and other distant segments with respect to their own mechanics or material characteristics.7 Owing to the complex structure of the FSU, the high functional loads, the different degrees of degenerative changes with different, large pathologic significance, and numerous individual influences, motion preservation devices have been developed and implanted in patients for many years. The ideal case should lead to a normalization of the complete FSU, with positive effects on the adjacent segments. The complexity of the spinal motion segment, and the not finalized trial and probation of spinal devices can be seen not only in the different approaches within the FSU but also in the different methods for the treatment of degenerative spinal disorders. On the one hand, the still limited knowledge about the lumbar and cervical FSU in connection with motion preserving devices and, on the other hand, the lack of data on long-term outcomes after the insertion of the different implants and materials are reflected in this relationship. OVERVIEW OF MOTION PRESERVATION DEVICES At present, there is no precise delineation of the groups of motion preservation devices according to anatomic, biomechanical, and clinical criteria. The objective of this chapter is to present a classification of known devices and those assessed as feasible, preferentially those that have already been used in humans. The devices are classified according to l The spinal region being treated; l The morphologic structure within the spine that is being replaced; l The material or materials from which the device is made; l The type of mechanical construction, including the primary and secondary mechanism of fixation; l The biomechanical effect of the device within the FSU; l The approach to the spine by which the device is implanted; and l The clinical indication for which the device is used. Previously, motion preservation devices for the spine were categorized according to the type of replacement, such as l Total disc replacement (TDR) of the cervical and lumbar spine; l Nucleus replacement; l Posterior dynamic stabilization; l Facet joint replacement. The classification of spine arthroplasty devices developed in this chapter is primarily categorized by l The topographic structure of the FSU to be replaced; l The extent of degenerative changes; l Biomechanical principles; and l The design and material composition of the device. As far as is possible, specific subcategories will be developed for the main categories in the classification to increase the understanding of the devices and materials.
CHAPTER 3
Total Disc Replacement This category refers to artificial intervertebral discs for the cervical and lumbar spine, which are intended to replace the complete function of a normal intervertebral disc, that is, take over the function of the nucleus pulposus and the annulus fibrosus. The cartilaginous plates in the region of the end plates of the vertebral bodies, also belonging to the intervertebral disc, are replaced by the prosthetic plates of a TDR. Motion preservation devices must meet the following requirements: 1. Physiologic mobility, in combination with physiologic translation; 2. Re-creation and preservation of the distance within the intervertebral space, as in the adjacent intervertebral disc; 3. Stabilization, including prevention of micro-instabilities; 4. Damping of load transfer, so there is no hindrance of the three major tasks. Physiologic mobility is different in its range of motion in the lumbar and cervical spine with respect to the mean extension, flexion, lateral bending, and axial rotation. During extremes of movement, there should be no nonphysiologic apposition of the adjacent vertebral bodies. The translation is greater in the cervical spine than in the lumbar spine. Intervertebral height should be restored without hyperdistraction, and orientated at the height of the adjacent intervertebral space. The facet joints as part of the “three-joint complex”, have to be considered in all requirements of total disc replacements, including the physiologic range of motion and avoidance of hypermobilities and microinstabilities. The increased load on the lumbar spine increases the risk that the initial height of the compressible elastic implants will degenerate, resulting in a decrease of the intervertebral height. In this situation, segmental hypermobility and destabilization may occur. Under the main category of TDR, there are three subcategories: functional three-component prostheses, functional twocomponent prostheses, and functional one-component prostheses (Table 3–1). Functional Three-Component Prostheses
Functional three-component prostheses for total disc replacement have two articulation surfaces as their main characteristic. The CHARITÉ Artificial Disc (DePuy Spine, Inc., Raynham, MA) is the first prosthesis with the aim of the total replacement of the intervertebral disc. The device was developed in 1982 in Berlin/Germany.8 The CHARITÉ prosthesis has a cranial and caudal symmetric articulation surface composed of the sliding core of a ball-and-socket type and two symmetric articulation surfaces of the prosthetic plates fixed to the two vertebral bodies. The prosthesis is specifically designed for implantation into the lumbar TABLE 3–1.
Classification of Spine Arthroplasty Devices
spine. The three components of the prosthesis with metal-polyethylene articulating surfaces allow a translation within the motion segment that is relatively well adapted to physiologic conditions. The prosthesis is constrained during extension and flexion, and during lateral bending, because the metal and polyethylene surfaces may come into contact. During axial rotation, it is unconstrained. The motion amplitudes in all planes are not in the physiologic extent, so that after a balanced implantation of the prosthesis, hypermobility with a higher stress on adjacent spinal structures, including the facet joints, may result, especially during axial rotation. There are indications that the facet joints will undergo postoperative degeneration at same level over the long term.9 The very limited elastic characteristics of the polyethylene result in a minimal damping function of the prosthesis. In summary, the CHARITÉ prosthesis completely replaces the cartilaginous plates and the nucleus pulposus. However, the functions of the annulus fibrosus of a natural intervertebral disc are not reproduced by the device (Fig. 3–1). A modified CHARITÉ disc is in development with lateral anchoring spikes on the prosthetic plates. Other functional three-component prostheses such as the Mobidisc and Mobidisc C (LDR, Troyes, France), the Kineflex and Kineflex|C (Spinal Motion, Inc., Mountainview, CA), the Activ-L (Aesulap Spine, Tuttlingen, Germany), the Dynardi (Zimmer Spine, Minneapolis, MN), the Secure-C (Globus Medical, Inc., Audubon, PA), and the Baguera (International Center Cointrin, Geneva, Switzerland) also function according to the ball-and-socket principle and also partly replace a normal intervertebral disc, with numerous differences in design, in the multiplicity of components, in the material, the saving of the sliding core, and in mechanism for fixation of the prosthetic plates to the vertebral bodies. In some devices a metal-to-metal contact of prosthetic components may occur at maximal ranges of motion. The Kineflex and the Kineflex|C have metal-metal bearing surfaces, with the mobile core being held by a retention ring of the prosthetic plates. All components are made of cobalt chrome. There is one core height for the lumbar spine and one core height for the cervical spine, but there are many different end plates. Lumbar lordosis is achieved by angled inferior plates (Figs. 3–2 and 3–3). In the case of the Dynardi (Dynamic Artificial Disc), the osseous end plates of the vertebral bodies have to be adjusted to the partially convex shape of the prosthetic. The two symmetric sliding surfaces of the prosthesis consist of ball-and-socket articulation
Total Disc Replacement
A. Functional three-component prostheses B. Functional two-component prostheses C. Functional one-component prostheses
23
n
F I G U R E 3–1. CHARITE´ Artificial Disc.
24
P A R T
n
I
Introduction to Motion Preservation Surgery of the Spine
F I G U R E 3–2.
Kineflex|C. n
F I G U R E 3–5.
Baguera L (with two versions of the sliding core).
and the lower prosthetic plate as well as from the design of the sliding core itself. The prosthetic plates are composed of a TiAlV alloy (Figs. 3–5 and 3–6). Functional Two-Component Prostheses
n
F I G U R E 3–3.
Kineflex.
surfaces. A central pin within the plates and corresponding holes in the sliding core prevent luxation of the insert. The kinetic center is dorsally displaced. The first implantation of this prosthesis was performed in 2006 (Fig. 3–4). The Secure-C prosthesis differs from the biarticular ball-andsocket type; it allows a greater extension and flexion compared with lateral bending, because along with a cranial ball-and-socket sliding surface, a second sliding surface caudally is present in the shape of a cylindrical segment. The sliding core of the Mobidisc shows, added to the ball-andsocket surface, an additional plane sliding surface that includes a limitation of axial rotation. However, it is not suited for a limitation of rotation of the prosthesis because the ball-and-socket surface allows an unlimited axial rotation. The same basic biomechanical principle also applies to the Baguera, where a damping is integrated. It results from a cavity between the PE-sliding core
n
F I G U R E 3–4.
Dynardi.
Functional two-component prostheses have one articulation surface as their major characteristic. The first functional two-component prosthesis with the aim of a total replacement of the intervertebral disc is the ProDisc-L (Synthes, West Chester, PA) for the lumbar spine. The first model of this device was developed in the '80s after the CHARITÉ prosthesis. Approximately 15 years later, after modifications of the prosthesis, the ProDisc-C (Synthes, West Chester, PA) for the cervical spine was developed and the first implantation was performed in 2002. Both prostheses have a ball-and-socket articulation surface in the combination polyethylene-metal and a keel anchor of the plates (Figs. 3–7 and 3–8). The major biomechanical difference between subcategory A prostheses and subcategory B prostheses is the constant center of rotation of subcategory B prostheses with its differing biomechanics, including segmental translation and load on facet joints. Only the results of longterm clinical studies can clarify whether the biomechanical differences will result in different treatment outcomes. Additional characteristics of the individual prostheses, such as differing radii of the ball-andsocket articulation surfaces or the position of the kinetic center of the prostheses, will also produce physiologic effects. Other subcategory B prostheses are the Maverick (Medtronic Sofamor Danek, Inc., Memphis, TN), FlexiCore (SpineCore Inc.,
n
F I G U R E 3–6. Baguera C.
CHAPTER 3
Classification of Spine Arthroplasty Devices
n
n
F I G U R E 3–7.
ProDisc-L.
Summit, NJ), Cervicore (Stryker Spine, Summit, NJ), PCM (Cervitech, Inc., Rockaway, NJ), DISCOVER (DePuy Spine, Raynham, MA), Prestige (Medtronic Sofamor Danek, Memphis, TN) and Discocerv (Scient'x USA, Inc., Maitland, FL). The Maverick device with the first metal-metal sliding partners in a TDR has a dorsal center of rotation, adapted to lumbar kinetics. The first two-component TDR with an integrated (stiff) limitation of rotation resulting from an internal stop of metallic prosthetic components is the FlexiCore. This prosthesis for the lumbar spine has no translation. The Prestige, which is used in the cervical spine, has an articulation principle that differs from the ball-and-socket principle. It has a ventrally extended socket to achieve a greater translation. The CerviCore, being a metal-metal prosthesis, completely differs from the partly spherical shape of the articulation partners, with the aim of
n
F I G U R E 3–8.
ProDisc-C.
25
F I G U R E 3–9. Discocerv.
approximating physiologic mobility. In addition to the sliding partner polyethylene-metal and metal-metal for the cervical spine, there are also ceramic-ceramic articulation surfaces. An example of a device with this type of articulation is the Discocerv, which was first implanted in 1999. The upper convex ceramic surface consists of alumina, and the lower cup consists of zirconia (Fig. 3–9). In comparison with the ball-and-socket prostheses, different radii lead to different material loads in the finite element model.10 For example, the DISCOVER device presents material innovations in TDRs with highly cross-linked ultra high-molecular-weight polyethylene (UHMWPE) for the core and the bioactively coated TiAlV plates that result in improved postoperative radiographic diagnostics (Fig. 3–10). More material improvements as well as new materials, material combinations, and coatings, including poly-ether-ether-ketone (PEEK) and carbon, are being developed. The biomechanics of presently known prostheses of subcategory B enables a partial replacement of a natural intervertebral disc, depending on different adaptations of the prosthesis for the simulation of the physiologic conditions. The development of modular prostheses started with the porous coated motion (PCM) disc. This prosthetic system includes a constrained version that eliminates translation. All sliding surfaces of the prostheses of subcategories A and B are meant to function according to the low-friction principle of the joint components. The ball-and-socket type shows a long-term consistency when an adequate material combination and correct prosthetic implantation are used. This can be seen from the follow-up examinations for the CHARITÉ prosthesis.11,12 Likewise, the tribological
n
F I G U R E 3–10. DISCOVER.
26
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
n
F I G U R E 3–11. M6.
pairing Metasul (metal-metal) in total hip replacement has proven successful over the long term.13 Compared with the sliding surfaces in total hip replacements, the Maverick and Kineflex have spiralshaped grooves on the convex surfaces of their sliding partners. In summary, the development of functional three- and twocomponent TDRs is not complete because despite a sufficient assortment of components and with the correct selection of components, exact implantation of prostheses with the existing prosthetic models, and depending on its design, a functional overload, such as hypermobility within the motion segment, may result in nonphysiologic stress on the facet joints. As a result, present long-term comparison studies of patients with fusion surgeries and patients with implanted prostheses are not definitely suited to prove the qualitative difference between these two treatment alternatives. It cannot even be ruled out that patients with implanted TDRs of present designs may profit less in the long term than patients with fusion surgeries. The time has come to develop and implant TDRs that have a natural mobility to mimic the intervertebral disc, so that the spinal motion segment will be physiologically replaced in the area of the intervertebral disc. Functional One-Component Prostheses
In this subcategory of prostheses, the device allows physiologic mobility including the damping function, that is, a total disc replacement. These are compact prostheses, so there are no articulation surfaces. There is compressible material between two prosthetic plates. Polycarbonate polyurethane elastomere is used preferentially. The functions result from the material characteristics. If the material combination of each compact prosthesis is the same throughout, differing mean segmental ranges of motion for each direction
n
F I G U R E 3–12.
Physio-L.
as well as between cervical and lumbar spine cannot be achieved. Therefore, physiologic motion resulting from the material characteristics is not possible. A forerunner of this type of prosthesis, with a low number of implantations in the 1980s, was the AcroFlex (DePuy Spine Inc., Raynham, MA) for the lumbar spine. Its development halted owing to problems with the material. These problems were dislocation of the material between the titanium plates and, thus, dislocation of the motion segment. Representative of subcategory C prostheses is the Bryan prosthesis (Medtronic Sofamor Danek, Memphis, TN) for the cervical spine, the construction of which allows function by virtue of the inclusion of polyurethane between two titanium plates. The M6 (SpinalKinetics, Sunnyvale, CA) is also being applied for the cervical spine, with an elastomeric polymer made of polycarbonate polyurethane in the center and a soft limitation of motion including rotation through an additional peripheral mesh of UHMWPE woven fibers. A peripheral polymer gasket prevents tissue ingrowth and contains wear debris, which is also described for the Bryan prosthesis (Fig. 3–11). The Physio-L (Nexgen Spine, Whippany, NJ) was developed for the lumbar spine. It consists of a multidurometer elastomeric core constructed using different grades of polycarbonate polyurethane between the two prosthetic plates of titanium (Fig. 3–12). The Freedom Lumbar Disc (AxioMed Spine Corporation, Cleveland, OH) was also developed for the lumbar spine. A viscoelastic polymer core is positioned between the two titanium plates (Fig. 3–13).
n
F I G U R E 3–13. Freedom Lumbar Disc.
CHAPTER 3
n
F I G U R E 3–14.
Classification of Spine Arthroplasty Devices
27
A and B, Theken eDisc.
The Theken eDisc (Theken Disc, Akron, OH) for the lumbar and cervical total disc replacement are also single-component discs, made of an elastomeric design between two titanium plates. The lumbar disc is additionally planned with electronics for the measurement of forces (Fig. 3–14A and B). Further prostheses of subcategory C are the CAdisc-L (Ranier Technology Limited, Cambridge, England) and the CAdisc-C (Ranier Technology Limited, Cambridge, England) made of polycarbonate polyurethane as compact versions without metal plates. Short-term fixation should be achieved by the compliant nature of the device applying a force to the vertebrae under low preload conditions. For long-term stabilization, a combination of a macrotexture, a microtexture, and a CaP coating that works like hydroxyapatite is alleged (Fig. 3–15). On the whole, the extent of elasticity in compact prostheses counteracts the stability and ultimately the intervertebral height, decisively depending on the biomechanical load on the spine. The subcategory C prostheses still have to prove their reliability in the clinical long term internally as implant material as well as in terms of the preservation of the height of the intervertebral space and of the segment stabilization. The mechanism of fixation of all three main categories of prostheses for the total replacement of an intervertebral disc shows, for primary safe fixation, the two main versions—spikes and keel—in different expressions. Some prostheses have additional screws or special shaped prosthetic surfaces corresponding with the vertebral bodies, for example, ridges or convex rough-coated surfaces. The
long-term fixation takes place through a bioactive coating with different chemical compositions. All clinically approved TDRs displace fusion surgeries and have to be implanted via a ventral approach. The contraindications are always to be considered. After implantation, TDRs lead to pain reduction immediately postoperatively, thus bringing a great benefit to patients. The first prospective randomized comparative studies between different TDRs are being carried out (Kineflex in comparison to the CHARITÉ Artificial Disc). To reduce potential complications, TDRs intended for a ventrolateral or lateral approach to the lumbar spine are being developed. The question arises for prostheses of a ball-and-socket type as to whether in the postoperative course a parallel positioning of adjacent vertebral body end plates in the anteroposterior view can be achieved if a small intervertebral space existed preoperatively, the extensive preparation only took place from one lateral side, and in addition, there was an intervertebral distraction. Prostheses for a total replacement of the intervertebral disc of the lumbar spine intended for implantation from a dorsal approach are also being developed. In this case, it needs to be considered whether a potential complication caused through a ventral approach should be exchanged by a definitely arising approach morbidity resulting from a dorsal approach. A dorsal prosthetic implantation is not possible without extensive resection of muscles, bone, ligamentous structures, and possibly even parts of the facet joints, resulting in instability. Furthermore, there is a greater risk of postoperative subsidence of the implant as a result of a lesser surface area for load transfer of the dorsally introduced prostheses. In summary, further developments are necessary for a TDR to completely replace the human intervertebral disc of the lumbar and cervical spine, as a physiologic partner within the three-joint complex. Nucleus Replacement
n
F I G U R E 3–15.
CAdisc-L.
Spherical prostheses made of steel or vitalium were implanted approximately 40 to 50 years ago into the cervical and lumbar spine from the dorsal approach. Probably because of the approach, these prostheses were retrospectively described as nucleus replacements.
28
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
However, despite considerable design differences, there are no clear biomechanical differences among present total disc replacements. It was soon realized that the surface area for the load transfer of the spheres was too small, leading to a large degree of subsidence into the adjacent vertebrae.14 Instilling soft materials such as polyurethane15 and silicon16 into intervertebral discs is regarded as the predecessor of the present nucleus replacements, although both of these treatment methods were abandoned by the end of the 1970s. The nucleus pulposus takes about 30% to 50% of the whole area in the cross-section of an intervertebral disc. This surface area is completely replaced by a so-called TDR of the present generation and only partly replaced by various types of nucleus replacements, to different extents. The nucleus replacement of the smallest area can be achieved by injecting a fluid in the early degenerative stages of the intervertebral disc.17 The fluid, acting as spare material, enters into cleavages within the nucleus pulposus tissue. After changing the consistency of the material, it is intended to reduce or eliminate pain and discomfort resulting from micro-instability. There is not much time in the course of the painful degeneration of the intervertebral disc to perform optimal nucleus replacement, that is, with the annulus fibrosus still intact. Along with the loss of stability in the region of the nucleus pulposus, there are increased shear forces and higher biomechanical load in the annulus fibrosus, leading to wear and tear of the annulus fibrosus. However, the complete functions of the nucleus pulposus may only be ensured in cooperation with the annulus fibrosus. This leads to the conclusion that the annulus fibrosus should be intact for a nucleus replacement. However, because this is increasingly not the case in the course of intervertebral disc degeneration and TDR is regarded as the final treatment alternative (also because of the risk of the surgical approach), lumbar devices for nucleus replacement are being applied with increasing stiffness, depending on the degree of degeneration of the intervertebral disc, in an attempt to compensate for the lack of stability of the annulus fibrosus. As a result of this nucleus replacement, implants were developed, which show a transition to present TDRs, an example of this being the PEEK-on-PEEK articulating implant NUBAC (Pioneer Surgical Technology, Marquette, MI). However, the consequence of a stiffer nuclear implant is to increase the share of load acting on the central area of the vertebral end plate. A nucleus replacement may result in adaptations as well as risks that apply to a total replacement of the intervertebral disc, for example, reactions of the vertebral end plates as well as intravertebral subsidence of the device. But there are also other influences that nucleus replacement may have on motion in other planes such as extension, flexion, lateral bending, and rotation. Because of the high biomechanical stress on the intervertebral disc, nucleus implants and replacement materials without specific fixation mechanisms are particularly at risk of postoperative horizontal or vertical dislocation. To reduce the risk of dislocation, developments of mechanism and chemical methods for an intradiscal fixation of the nucleus replacements have already been carried out; further developments with respect to material and implants are still necessary. A stable integration of a nucleus implant or material into the intervertebral space not only increases the indication span, but also increases the indication safety of surgeons clinical application.
The degree of nucleus replacement required depends on the degree of degeneration. However, it can be assumed that with less original nucleus tissue, an increasingly stable or quantitatively more extensive nucleus replacement will be required, also for the protection of the annulus fibrosus. During the surgical introduction of a device for nucleus replacement, the function of the annulus is impaired depending on the size of the incision. In order to preserve the motion-modulating and motion-stabilizing function of the annulus fibrosus, the approach through the annulus and the incision should be small as possible. The risk of dislocation of a material or implant for nucleus replacement is usually also reduced with a small incision. The following questions must be answered for a successful application of an implant or material for nucleus replacement: l
l
l
l
l l
What is the indication? Is the procedure intended to restabilize the area, as in the case of an early intervertebral disc degeneration, or is it intended as a prophylactic measure against a postdiscectomy syndrome during microscopic discectomy? What is the condition of the annulus fibrosus? Is it suited for the planned device? Can the nucleus replacement be safely placed into the intervertebral space, so that it cannot dislocate dorsally toward the adjacent nerves? Is the risk of subsidence into the adjacent vertebral bodies so low that the nucleus replacement can be implanted? Is it a biocompatible material with strongly delayed resorption? Are wear debris from the implant or material itself that may be in contact with the surrounding tissue a concern?
A classification of nucleus replacement devices can be made following different criteria. The loading capacity and physiologic balance of the intervertebral disc within the motion segment are to be improved or normalized, with the expectation of reducing or eliminating the patient's pain. Because there has been no success to produce a complete and permanent conjunction between the implant or material and the surrounding nucleus and/or annulus tissue and the cartilaginous plates, the direct loading capacity of the implant or material is important in relation to the known intradiscal pressures. For the evaluation of the suitability of a material or an implant for nucleus replacement, the consistency of the device may not be considered in comparison to normal nucleus tissue outside any load. Because this is a nucleus replacement aimed at replacing a physiologic nucleus pulposus including the protective function for the present annulus fibrosus, the biomechanically effective consistency of the implant material that withstands the mean intradiscal pressure during everyday activities, or the mean external load that is needed to compress the disc of a functional spinal unit with a physiogically adapted nucleus replacement, is of importance. Based on this definition, a classification of the nucleus replacement is feasible. At present, however, no data are available. The following subcategories of nucleus replacement are classified according to the kind of application, the protection of materials, substance groups, material characteristics, stiffness of the material, and mechanical criteria (Table 3–2). The classification focuses on presently known devices for nucleus replacement and brings them into coherent order. Because of the large variety of nucleus replacements, other classifications are also feasible,
CHAPTER 3
TABLE 3–2.
Nucleus Replacement
A. Injectable, in situ curing materials 1. Uncontained a. Hydrogel adhesive (examples: NuCore, BioDisc) b. Nonhydrogel nonadhesive (example, Sinux) 2. Contained a. Hydrogel (no example at present) b. Nonhydrogel (examples: DASCOR, PNR, PDR) B. Preformed implants 1. Nonarticulating a. Hydrogel (examples: PDN-SOLO), Hydraflex, NeuDisc, Aquarelle) b. Nonhydrogel (examples: Newcleus, NeoDisc, Regain) 2. Articulating a. Same material of components (example: NUBAC) b. Different materials of components (no example at present)
for example, according to the degree of degeneration of the motion segment with derived indication for a specific nucleus device. For this approach, spinal diagnostics are necessary with emphasis on the annulus fibrosus so that with the knowledge of its constitution including its stability as well as other factors, the appropriate device can be selected. Tests still need to be conducted on this measure. For an adapted therapy depending on the condition of the motion segment, the composition, load resistance, and other factors of the device must be examined to evaluate their suitability to the patient. If the intervertebral loads could be balanced with the implanted nucleus replacement to reduce or eliminate the patient's pain, a treatment algorithm within the group of different nucleus replacements could be developed. Following the soft nucleus replacement materials, which were described in 1977 (polyurethane) and 1978 (silicone), the Prosthetic Disc Nucleus (PDN) (Raymedica, Inc., Minneapolis, MN) prosthesis was first implanted in 1996. It comprises a polymeric
Classification of Spine Arthroplasty Devices
hydrogel encased in a high-tenacity polyethylene jacket that allows the device to absorb fluid and expand in height.18 In the meantime, there has been an evolution from the stiff PDN to the PDNSOLO (Raymedica, Inc., Minneapolis, MN) and lastly HydraFlex (Raymedica, Inc., Minneapolis, MN). The present group of nucleus replacements varies widely. In the limited context of this chapter, only a few devices can be described. If only the field of spine arthroplasty materials and implants is considered, the treatment of early degeneration of the intervertebral disc begins with injections. The percutaneous computed tomography–guided injection of NuCore (Spine Wave, Inc., Shelton, CT) into the nucleus pulposus has been used.19 The NuCore Injectable Nucleus is an adhesive, protein polymer that is curable in situ. The components are silk-elastin copolymer solution and diisocyanate cross-linker. NuCore is one of the least stiff materials used in nucleus replacement devices. Biomechanical testing in cadavers has shown that implantation of NuCore restores function and stability to the spine after a destabilizing discectomy procedure (Fig. 3–16).20 Another material in the group of injectable, in situ curing materials is the DASCOR Disc Arthroplasty Device (Disc Dynamics, Inc., Eden Prairie, MN). It involves the insertion of an in situ– polymerizing polyurethane, cured within a polyurethane balloon in the disc space. This device was first implanted in 2002 through an open procedure that required a 5-mm hole in the annulus (Fig. 3–17). The NeoDisc (NuVasive, Inc., San Diego, CA) is an implant with a smooth transition to TDR, because it is meant to partially replace the annulus fibrosus and even the ventral longitudinal ligament in the cervical spine. NeoDisc consists of an elastomere core in a polyester annular jacket with anterior flanges for fixation to the anterior surface of adjacent vertebrae.
B n
F I G U R E 3–16.
29
A and B, NuCore injection.
30
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
TABLE 3–3. Posterior Dynamic Stabilization—Screws and Connectors A. Mobile screw parts B. Mobile connectors (beginning from retainer on the screw) and rods made of 1. Plastic material 2. Metal 3. Combination of metal and plastic material C. Combination of A und B
of the different elasticity modules between bone and implant and also because of the biomechanical dorsal strains, especially regarding shear forces. The design of the device influences the load of the implant–bone interface and the load of the implant itself. A balance between the extent of mobility of the implant to avoid an overload of the bone–implant interface and an effective stabilization of the spinal motion segment must be found. There are many ways in which dorsal mobility between adjacent vertebrae could be realized, partly under consideration of the COR. The Posterior Dynamic Stabilization classification is based on present knowledge (Table 3–3). n
F I G U R E 3–17.
DASCOR.
Mobile Screw Parts
The Percutaneous Nucleus Replacement (PNR) and Percutaneous Disc Reconstruction (PDR) devices (TranS1, Wilmington, NC) are introduced by the transsacral approach. When this device is implanted, there are vertical screw pistons in vertebral bodies L5 and S1 with an intervertebral disc replacement, including an enclosing balloon lying within the intervertebral space L5-S1.
The heads of screws are mobile with respect to the screw piston through a special mechanism. In the case of the Cosmic Posterior Dynamic System (Ulrich Gm bH & Co. KG, Ulm, Germany), the thread of the screw is connected by a hinge for a permanent movable connection between screw and rod. The complete system allows axial load distribution and prevents any rotation and translation. The device was first implanted in a patient in 2002 (Fig. 3–18). Mobile Connectors
Posterior Dynamic Stabilization—Screws and Connectors
From Retainer on the Screw—and Rods of Plastic Material
The basic principle for dorsal dynamic stabilization is mainly based on screw fixation in adjacent pedicles and vertebral bodies. Because the ventral sections of the spine carry out the segmental transfer of axial load up to about 80%, the replacement seems to be of less importance. In contrast to this, primarily dorsal procedures with the option of function-preserving stabilization may be indicated in cases of dorsal pathologies of the spine. Furthermore, a dorsal dynamic stabilization is particularly beneficial in older patients with a higher rate of morbidity with the anterior approach or decreased bone-loading capacity. In cases of intervertebral disc degeneration, the intervertebral disc remains a source of pain after implantation of dorsal dynamic implants. The implant–bone interface and the material stability are of particular interest because
Implants that exclusively have plastic material between the pedicle screws are always less resistant to shear forces, which have to be absorbed up to 80% by the dorsal lumbar spine. One of the earliest pedicle screw-based devices was the Graf Ligament System (Sem Co., Montrouge, France). This system used braided polyester bands looped around the screws instead of rods with the aim of providing stability while allowing motion. After the spine was exposed and pedicle screws inserted, the bands were connected under applied compressive force between the pedicle screws as a ligamentoplasty. In 1994, Dynesys (Dynamic Neutralization System for the Spine; Zimmer Spine, Inc., Warsaw, IN) was implanted for the first time as a dorsal dynamic instrumentation, which has cords
Cosmic screws n
F I G U R E 3–18. Cosmic screws.
CHAPTER 3
n
F I G U R E 3–19.
Classification of Spine Arthroplasty Devices
31
Dynesys.
of polyethylene terephthalate with a tube made from polycarbonate urethane slid over them and fixed to two adjacent pedicle screws with nuts. The screws are made of a titanium alloy and are coated with hydroxyapatite if required (Fig. 3–19). The pedicle-based CD HORIZON LEGACY PEEK Rod (Medtronic Sofamor Danek, Memphis, TN) has been developed as a new generation of rods made of the semi-crystalline thermoplastic polymer PEEK as a semi-rigid alternative to titanium rods. PEEK has a modulus of elasticity between that of cortical and cancellous bone, and it is radiolucent. From Retainer on the Screw—and Rods of Metal
On the one hand, these are rods made of “soft” metal, so that axial deformations of the rods are possible. Furthermore, this group includes rods that have outside notches, such as in a screw thread, and which are therefore shapeable. An example is the semi-rigid AccuFlex system with a double helical cut made within a standard 6.5-mm rod. AccuFlex (Globus Medical, Inc., Audubon, PA) is 50% as stiff as solid rods; it is stiff in anterior shear and rotation but allows motion primarily in the flexion extension mode. Systems with integrated loop are also counted to this group, for example, the BioFlex System (Bio-Spine Corp., Seoul, Korea).21 It is based on Nitinol, which is an alloy of nickel and titanium and has various characteristics such as high elasticity and high tensile force, and flexibility or rigidity according to temperature change.
n
F I G U R E 3–20. NFlex.
The Isobar TTL Dynamic Rod (Scient'x, Maitland, FL) is made of TiAlV alloy, and a dampener is fixed between monoaxial or polyaxial pedicle screws. Stacked washers within the dampening element provide dynamism. The rod allows extension, flexion, and dampening (Fig. 3–21). The rod of the CD HORIZON AGILE Dynamic Stabilization Device (Medtronic Sofamor Danek, Memphis, TN) is made of a thin rod of TiAlV alloy within a pure titanium rod and a bumper of polycarbonate polyurethane surrounding the thin rod. It is used as single level and adjacent to fusion motion–retaining implant with two different sizes of the kinetic center (Fig. 3–22). The AXIENT Dynamic Fixation System (Innovative Spinal Technologies, Mansfield, MA) has pedicle screws with polyaxial screw heads and articulating CoCr sliding rods with a component for damping during extension, made of polycarbonate urethane. The implant allows segmental movement with integral stops to avoid excessive motion in extension, flexion, axial rotation, and sagittal translation. This system is described as semiconstrained, and the AXIENT TOTAL Dynamic Fixation System with a greater range of motion is described as unconstrained,
From Retainers on the Screw—and Rods as Combination of Metal and Plastic Material
NFlex (N Spine, Inc., San Diego, CA) is a dynamic stabilization system, used with polyaxial titanium pedicle screws and titanium and polycarbonate urethane rods. The system allows an elongation of the posterior elements during spinal flexion and compression during extension. There is a layer of polycarbonate urethane between the titanium core and the pedicle screw attachment that allows a small translation perpendicular to the long axis of the rod. A rotational stop is provided by the titanium core within the piston. The first devices were implanted in 2006 (Fig. 3–20).
n
F I G U R E 3–21.
Isobar TTL dynamic rod.
32
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
The Stabilimax NZ is already being classified as a total facet replacement by Khoueir et al.22 Posterior Dynamic Stabilization—Facet Arthroplasty The function-retaining replacement of the facet joints needs to be viewed critically because the arthritis of these joints usually results from a degeneration of the intervertebral disc. In this case, a sole replacement of the facet joints leaves the intervertebral disc as a pain generator and in a combined total disc and facet joint replacement, the replaced facet joints could be exposed to threshold loads or overloads as the result of nonphysiologic motion amplitudes of currently available TDRs. Depending on the design of the facet joint replacements, a hypermobility of the whole FSU could also result, including an irritation of nerval structures or a subluxation within the FSU. In elderly patients with a severe narrowed spinal canal, the removal of facet joints can be necessary. It must be determined whether these patients would benefit from stabilization with motion preservation or rigid stabilization. The following classification in subcategories is feasible (Table 3–4). n
F I G U R E 3–22. CD HORIZON AGILE Dynamic Stabilization
Device.
where there is no integral stop of the segmental motion. With the incorporation of rotary bearings in the implants, stress and strain to the bone-screw interface should be reduced (Fig. 3–23). The Stabilimax NZ (Applied Spine Technologies, New Haven, CT) has a mobility beginning from the retainer of the pedicle screws, described for this region as a ball and socket. The implant uses two independent concentric springs on each side that allow motion in all directions as well as a damping function. The springs are incorporated into the system through connecting rods.
Partial Facet Replacement
The smallest device is a replacement with only one component between both facets of a facet joint, which is fixed by a cord through both facets. These devices can be implanted with little resection of body tissue and with little effort. The challenge lies in the constancy of position of this implant, made of metal, between the two facets during extensive segmental motion. Total Facet Replacement Metal
The Total Facet Arthroplasty System (TFAS) implant (Archus Orthopedics, Inc., Redmond, WA) totally replaces the facet joints of a motion segment. The system consists of two stems that are fixed in the pedicles and concave discs connected to them, which allow two spherical parts (one on each side) to move on their surfaces. The two spherical parts are connected to the stems in the pedicles of the adjacent vertebra. Because the implant is completely made of metal, there are metal-metal articulating surfaces. The stems have no threading; they are cemented. For the implantation, a full laminectomy including facet resection is necessary. Metal and Plastic Material
The Total Posterior Arthroplasty (TOPS) System (Impliant Spine, Princeton, NJ) is a pedicle screw–based system with a
TABLE 3–4. Posterior Dynamic Stabilization— Facet Arthroplasty
n
F I G U R E 3–23. AXIENT Dynamic Fixation System.
A. Partial facet replacement 1. Plastic material 2. Metal 3. Combination of metal and plastic material B. Total facet replacement 1. Plastic material 2. Metal 3. Combination of metal and plastic material
CHAPTER 3
Classification of Spine Arthroplasty Devices
33
TABLE 3–5. Posterior Dynamic Stabilization— Interspinous Implants A. Implant with function by itself 1. Primarily design dependent 2. Primarily material dependent B. Spacer without own function 1. Plastic material 2. Metal 3. Combination of metal and plastic material
metal-on-polymer articulation to cushion against hard stops. TOPS is designed for the treatment of spinal stenosis combined with severe facet arthritis. It has been determined that TOPS allows a nearly physiologic range of motion, including sagittal translation. Posterior Dynamic Stabilization—Interspinous Implants The interspinal implants are to serve for foraminal distraction, the off-loading of facets, and the dorsal disc unloading in extension. There are two basic types of interspinous implants: (1) implants whose function results from the design or the material characteristics, and (2) implants that solely act as interspinal spacers without a specific function. Those in the second group are not defined as spine arthroplasty devices. However, there is still a certain segmental motion after implantation even in stiff materials because of the mainly axial load transfer through the intervertebral disc, away from the implant positioning (Table 3–5). Implant with Function by Itself Primarily Design Dependent
A major representative of this type of device is the coflex implant (Paradigm Spine, LLC, New York, NY). coflex is a functionally dynamic interspinous implant for levels L1-L5, which is compressible in extension, allows flexion, and has slight rotational stabilization. It is made of a TiAlV alloy. The implant wings can be crimped to achieve sufficient fixation to the spinous processes (Fig. 3–24). Primarily Material Dependent
The Device for Intervertebral Assisted Motion (DIAM) (Medtronic Sofamor Danek, Memphis, TN) is an H-shaped polyester-covered silicone bumper that is placed between the spinous processes with a mesh band and is sutured to hold it in place.
n
F I G U R E 3–24. coflex.
between the spinous processes of the lumbar spine to limit extension. The first component is implanted next to and under the spinous process, and the second component is placed on the opposite side of the spinous process and is then attached to the first. The supraspinous ligament is retained. The Superion Spacer (VertiFlex Inc., San Clemente, CA) consists of a titanium alloy and is implanted percutaneously directly from dorsal between two adjacent spinous processes. The stabilization takes place by enfolding of the two wings surrounding the upper and lower spinous process (Fig. 3–26). Combination of Metal and Plastic Material
The In-Space Interspinous Distraction Device (Synthes, West Chester, PA) can be implanted from L1-S1 percutaneously laterally or through the unilateral posterior approach. It is a cylindrical distance holder made from PEEK with wings made from TiAlV alloy for fixation to the spinous processes. The device prevents extension and allows flexion, lateral bending, and rotation. The device was first implanted in 2006 (Fig. 3–27).
Spacer Without Own Function Plastic Material
The Wallis system (Abbott Spine, Inc., Austin, TX) is a rigid interspinous spacer which consists of PEEK and two woven polyester bands, through which the fixation to the spinous processes is achieved. The spacer is suited for a limitation of the extension and flexion, without influence on rotation and lateral bending. The first implantation of the device was performed in 1987; since 2002, a second generation of the implant has been produced (Fig. 3–25). Metal
The X-STOP Interspinous Process Decompression Device (Kyphon, Inc., Sunnyvale, CA) is a two-component titanium implant that fits
n
F I G U R E 3–25. Wallis.
34
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
An algorithm for the degree of degeneration of the FSU and for the device derived therefrom cannot be compiled at this time owing to a lack of basic science data as well as all-embracing clinical knowledge. As an orientation, the implant or material should always be selected, with which the individual physiologic condition of the respective patient can be preserved or repaired most precisely, with the least risk and effort for the patient, and with the best expected long-term effect. Therefore, it is necessary to exactly analyze in which topographic region the pathology is located to reconstruct a biomechanical regular three-joint complex as a precondition for a reduction of the patient's pain. ACKNOWLEDGMENTS
n
F I G U R E 3–26.
Superion spacer.
I would like to express my sincere gratitude to the companies who responded to my questionnaire. Implants and materials of these companies were preferentially explained and illustrated. My special thanks is given to Mr. Andrew J. Carter, PhD (Spine Wave, Inc. Shelton, CT), with whom I had the opportunity to discuss theoretical basics. At the beginning of the research, Mr. Bruno Paton, HIPPOCRAFT, LLC, and Ms. April C. Bright, Marketing Manager of KNOWLEDGE Enterprises, INC., supplied an overview of the companies with motion-preservation devices. REFERENCES
n
F I G U R E 3–27. In-Space.
FINAL REMARKS This is the first classification of all available spine arthroplasty devices. Concepts for replacement of the complete functional spinal unit have already been presented. These concepts have not been clinically tested and are therefore not considered in this chapter. For anatomic and biomechanical reasons, a total replacement of the annulus fibrosus also is not included. A partial replacement of the annulus fibrosus could close a gap in the annulus caused by surgery, but it remains to be proven whether the biomechanics of the complete annulus would be restored.
1. Junghanns H: Die Wirbelsäule in der Arbeitsmedizin. Teil I. Biomechanische und biochemische Probleme der Wirbelsäulenbelastung. Hippokrates, Stuttgart, Die Wirbelsäule in Forschung und Praxis 78, 1979. 2. White AA, Panjabi MM: Clinical Biomechanics of the Spine. Philadelphia, JB Lippincott Co., 1978. 3. White AA, Panjabi MM: Clinical Biomechanics of the Spine, 2nd ed. Philadelphia, JB Lippincott Co., 1990. 4. Johnson EF, Caldwell RW, Berryman HE, et al: Elastic fibers in the annulus fibrosus of the dog intervertebral disc. Acta Anat (Basel) 118:238–242, 1984. 5. Pope MH, Magnusson M, Wilder DG: Low back pain and whole body vibration. Clin Orthop Rel Res 354:241–248, 1998. 6. Putz R: Function-related morphology of the intervertebral disks. Radiologe 33:563–566, 1993. 7. Panjabi M, Malcolmson G, Teng E, et al: Hybrid Testing of Lumbar CHARITÉ versus Fusions. Spine 32:959–966, 2007. 8. Büttner-Janz K: The Development of the Artificial Disc SB CHARITÉ. Dallas, Texas, Hundley & Associates, Inc., 1992. 9. Harms J: Motion preservation in spine surgery—dream or reality. 7th Annual Meeting of Spine Arthroplasty Society, May 1–4, Berlin, Germany, 2007. 10. Moumene M, Harms J, Albert TJ: The implication of the implant core radius in cervical total disc replacement, 7th Annual Meeting of Spine Arthroplasty Society, May 1–4, Berlin, Germany, 2007. 11. Lemaire J-P, Carrier H, Sari Ali E-H, et al: Clinical and radiological outcomes with the Charité™ artificial disc. A 10-year minimum follow-up. J Spinal Disord Tech 18:353–359, 2005. 12. David T: Long-term results of one-level lumbar arthroplasty: Minimum 10-year follow-up of the CHARITÉ artificial disc in 106 patients. Spine 32:661–666, 2007. 13. Boaten K, Schmidt G, Büttner-Janz K: Metal-on-Metal: Results of a Prospective Study after 10 years, 25th Anniversary of Alloclassic™ Zweymüller™ Stem, June 17–19, Vienna, Austria, 2004. 14. Fernström U: Der Bandscheibenersatz mit Erhaltung der Beweglichkeit. In: Zukunftsaufgaben für die Erforschung und Behandlung der Wirbelsäulenleiden (Erdmann H). In Junghanns H (ed): Die Wirbelsäule in Forschung und Praxis. Stuttgart, Germany, Hippokrates, 55, 1972, pp. 125–130.
CHAPTER 3
15. Schulman Ch M: Metod kombinirowannogo chirurgitscheskogo letschenija kompressionnych form pojasnitschnogo osteochondrosa s alloprotesirowaniem porashennych meshposwonkowych diskow. Z Vopr Neirokhir 2:17–23, 1977. 16. Fassio B, Ginestié J-F: Prothèse discale en silicone. Etude expérimentale et premières observations cliniques. La Nouvelle Presse Médicale 7:207, 1978. 17. Pfirrmann CWA, Metzdorf A, Zanetti M, et al: Magnetic resonance classification of lumbar intervertebral disc degeneration. Spine 26:1873–1878, 2001. 18. Ray CD: Spinal fusion alternative: Device expands, fills disc space. Orthop Today Int 1:32–38, 1998.
Classification of Spine Arthroplasty Devices
35
19. Büttner-Janz K: Percutaneous application of nucleus replacements. BIOSPINE 2, Leipzig, Germany, September 20–22, 2007. 20. Boyd LM, Carter AJ: Injectable biomaterials and vertebral endplate treatment for repair and regeneration of the intervertebral disc. Eur Spine J 15(Suppl 3):S414–S421, 2006. 21. Kim Y-S, Zhang H-Y, Moon B-J, et al: Nitinol spring rod dynamic stabilization system and Nitinol memory loops in surgical treatment for lumbar disc disorders: short-term follow-up. Neurosurg Focus 22:1–9, 2007. 22. Khoueir P, Kim KA, Wang MY: Classification of posterior dynamic stabilization devices. Neurosurg Focus 22:1–8, 2007.
CHAPTER
4
Advanced Spinal Anatomy for Cervical and Lumbar Nonfusion Surgery Daniel R. Fassett, Shiveindra B. Jeyamohan, Alexander R. Vaccaro, and Peter G. Whang
K E Y l
l
l
l
l
l
P O I N T S
An in-depth understanding of the surrounding anatomy and vertebral anatomy is needed to accurately place motion-preserving spinal devices. Poor alignment of motion-preserving spinal implants can adversely affect the biomechanics of the spine. The functional spinal unit, which is composed of two adjacent vertebrae articulating at the intervertebral disc and facet joints, has unique features in cervical and lumbar regions of the spine that contribute to the biomechanical properties of these areas. In the cervical spine, the uncovertebral joints, in addition to often being the source of neural compressive pathology, contribute significantly to segmental stability. In the lumbar spine, placement and development of motion-sparing technologies should consider the larger vertebral size and end plate structure, the relatively narrow angulation between the axial center of rotation and the facets, the importance of ligamentous structures in stability, and the complexity of surrounding anatomy in surgical approaches to this area of the spine. Motion-sparing strategies are being developed and employed as an alternative to fusion procedures for the treatment of degenerative spinal conditions such as axial pain syndromes, radiculopathy, and occasionally myelopathy. Cervical and lumbar total disc arthroplasty procedures have been performed in Europe for more than a decade. A series of clinical trials on lumbar disc replacements have been completed in the United States, and a number of cervical clinical trials are under way. In addition, new technologies such as nucleus pulposus replacement and posterior motion-sparing devices are now being studied. The main argument for motion-sparing surgery is to reduce adjacentsegment degeneration, which has been thought to occur after fusion procedures due to increased stress on adjacent segments. Although adjacent-segment degeneration is likely multifactorial, there is some credible evidence that a fusion accelerates spondylosis at adjacent segments. In addition, motion-sparing technologies potentially avoid the morbidity associated with fusion such as bone graft harvesting and pseudoarthrosis.
To recognize the potential benefit of restoring normal biomechanics and preventing adjacent-segment stresses, it is imperative that motion-sparing devices be placed precisely and that the technologies work harmoniously with surrounding anatomy in a biomechanically sound manner. A keen understanding of the surrounding anatomy, 36
disc space anatomy, and radiographic anatomy are needed for accurate placement of motion-sparing devices. ANATOMY OF THE FUNCTIONAL SPINAL UNIT The functional spinal unit is created by the articulation of two adjacent vertebrae, the intervening intervertebral disc (IVD) anteriorly and facet complex posteriorly (Fig. 4–1). The IVD is a complex structure composed of an inner nucleus pulposus that is surrounded by a dense annulus fibrosis and is bounded superiorly and inferiorly by the cartilaginous end plates of the adjacent vertebral bodies. The central nucleus pulposus has a very sparse cell population and is composed largely of an extracellular matrix of hydrophilic proteoglycans and type II collagen. The hydrophilic nature of a healthy nucleus pulposus draws water into the nucleus to maintain hydrostatic pressure within the disc space and tense the surrounding annulus fibrosis and ligamentous structures. The annulus fibrosis is composed predominantly of layers of type I collagen fibers that are arranged in an alternating lamellar pattern (Fig. 4–2). Within a single lamellar layer of the annulus, all of the collagen fibers are aligned in the same direction at an approximate 30-degree orientation with the horizontal vertebral end plates. Consecutive annular layers are oriented in an alternating pattern with one layer oriented obliquely to the right and the next layer oriented obliquely to the left. This alternating lamellar pattern of the annulus has been compared with a radial tire with excellent tensile strength to resist shear and prevent disruption. The combination of nucleus pulposus and annulus fibrosus of the intervertebral disc work together to provide biomechanical stability at each spinal segment, with the nucleus supporting compressive loads and the annulus, together with the facet joints and ligaments, resisting shear forces. The cartilaginous end plates that form the cephalad and caudal borders of the IVD are initially composed of hyaline cartilage, which calcifies with aging. Beneath the cartilaginous end plates is subchondral bone that can vary in thickness depending on its level in the spine and location within the disc space. The vertebral end
CHAPTER 4 n F I G U R E 4–1. A functional spinal unit is composed of an anterior intervertebral disc and posterior facets that act together to allow motion between adjacent vertebrae while providing stability to maintain spinal alignment and protect the neural elements. (Adapted from Vaccaro AR, Papadopoulos S, Traynelis VC, et al: Spinal Arthroplasty: The Preservation of Motion. Philadelphia, WB Saunders, 2007.)
Advanced Spinal Anatomy for Cervical and Lumbar Nonfusion Surgery
Posterior lateral superior processes joint of Luschka
37
Intervertebral disc
Groove for spinal nerve
Functional spinal unit
plates are very important anatomic structures for anterior motionsparing technologies because end plate subsidence has been noted with some anterior motion-sparing devices. In the lumbar spine, the vertebral end plates are the strongest at the outer margins of the disc space and weakest in the central portion. In the cervical spine, the superior end plates are strongest at the dorsal margins and the inferior end plates are strongest at the ventral margins. The end plates are also the source of nutrition to the adjacent avascular disc space and may be the limiting factor in disc regeneration strategies to preserve motion. Anterior
NP
AF
The facets complete the functional spinal unit, and the size and orientation of the facets are dependent on their location within the spine. In the cervical spine, the facets have a more vertical orientation in the coronal plane and the facet-to-disc-surface area ratio is larger in the cervical spine in comparison to the lumbar spine. In addition, the cervical facets are based at a wider angle from the axial center of rotation (larger arc of influence) than in the lumbar spine, and, therefore, have a greater contribution to axial stability (Fig. 4–3). As a result of all these factors, the cervical facets are thought to contribute more to rotational stability, and, for this reason, it has been reported that a cervical disc replacement may be more biomechanically sound in comparison to lumbar disc replacement.1 The ligaments of the spine are also important stabilizing structures and should be considered in the application of motionsparing technologies. The anterior longitudinal ligament (ALL), posterior longitudinal ligament (PLL), ligamentum flavum (LF), facet capsular ligaments, and interspinous ligaments contribute to spinal stability (Fig. 4–4A). The ALL runs along the anterior aspect of the vertebral column beginning at the anterior margin of the foramen magnum and extending to the anterior surface of the sacrum. The ALL gets wider and stronger as it descends into the lumbar spine, and it is the thickest over the intervertebral disc spaces. At the disc spaces, the deep fibers of the ALL travel only
Posterior
Radial x3
153.6 degrees Circumferential x1
Axial x2 n
F I G U R E 4–2. The intervertebral disc is composed of a central gelatinous core, the nucleus pulposus (NP), surrounded laterally by dense annulus fibrosis (AF). The lamellar architecture of the annulus with an alternating pattern of collagen fibers provides excellent tensile strength to resist shear force.
n F I G U R E 4–3. The arc of influence (AOI) is the angle formed between the axial center of rotation and the posterior facets. In the cervical spine, the facets are wider in relation to the axial center of rotation (larger AOI) and provide more stability against axial shear.
38
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
Anterior atlanooccipital membrane Posterior atlanoocipital membrane
C2 spinous process
Ligamentum nuchae Ligamentum flavum
Intervertebral foramen Anterior longitudinal ligament
C7 spinous process
Vertebral body Intervertebral disk Posterior longitudinal ligament
A
Spinous process Ligamentum flavum
Interspinous ligament Supraspinous ligament
B
Thin lateral expansions Begins on undersurface of superior lamina
Thicker central region
Goes to superior aspect of interior lamina Posterior longitudinal ligament
C n
D
F I G U R E 4–4. The spinal ligaments contribute substantially to spinal stability. A and B, The anterior longitudinal ligament (ALL), posterior
longitudinal ligament (PLL), ligamentum flavum (LF), facet capsular ligaments (FCLs), interspinous ligaments (ISLs), and supraspinous ligaments are shown. C, The posterior longitudinal ligament has a very thick middle portion that extends laterally over the disc spaces for additional stability. D, The LF runs along the ventral aspect of the lamina, extending from the undersurface of the lamina above to top of the lamina below.
one interspace and anchor the adjacent vertebral bodies together. The PLL also runs along the entire spinal column along the posterior surface of the vertebral bodies. The PLL is narrow and very thick in the midline over the vertebral bodies and extends laterally over the disc spaces for additional support (Fig. 4–4B). The LF
(yellow ligament) runs between adjacent lamina extending from the undersurface of the inferior aspect lamina above to the top of the lamina below (Fig. 4–4C). In the lumbar spine, the ALL and PLL have the highest tensile strength, whereas in the cervical spine, the capsular ligaments and
CHAPTER 4
LF have higher tensile strength than the ALL and PLL. Destruction of one of these ligaments during a motion-sparing procedure can have adverse impact on the biomechanics of the functional spinal unit. It has been speculated that cutting the ALL during lumbar arthroplasty may contribute to rotational deformities that have occurred with some lumbar disc replacements.1 ANTERIOR CERVICAL SPINE ANATOMY Surrounding Anatomy The anterior approach to the cervical spine is one of the most commonly performed approaches in spinal surgery and can expose the ventral spine from C2 to T2. The dissection plane is carried down along the medial border of the sternocleidomastoid muscle (Fig. 4–5). The carotid artery can easily be palpated and retracted laterally with the contents of the carotid sheath (carotid artery, internal jugular vein, and vagus nerve). The trachea and esophagus are gently retracted medially, and the ventral surface of the spine can be palpated. The prevertebral fascia can be opened with a combination of sharp and blunt dissection techniques. Structures at risk for injury on this dissection include the contents of the carotid sheath, esophagus, sympathetic chain, recurrent laryngeal nerve, and superior laryngeal nerve, with dysphagia and laryngeal nerve palsies being the most common complications in anterior cervical spine surgery. With the use of careful blunt dissection and palpation of the carotid with lateral retraction, injury to the carotid sheath structures should be avoided. With lateral retraction of the carotid sheath, the inferior and superior thyroid vessels may bridge from
n F I G U R E 4–5. The anterior cervical approach to the spine uses soft tissue planes medial to the sternocleidomastoid muscle. The carotid sheath structures are retracted laterally, and the esophagus and trachea are retracted medially to expose the ventral surface of the spine. (Adapted from Albert T, Balderston R, Northrup B: Surgical Approaches to the Spine, Philadelphia, WB Saunders, 1997, p. 10.)
39
Advanced Spinal Anatomy for Cervical and Lumbar Nonfusion Surgery
lateral to medial and obstruct the surgical exposure. These vessels can be ligated, but care should be taken to avoid neural structures in this area because the superior laryngeal nerve can course with these vessels. The esophagus may be traumatized during the initial dissection, but it is more likely to be injured at the time of decompression when it can protrude around the medial-lateral retractor blades and be wound in the shaft of a high-speed drill. The sympathetic trunks pass along the ventral surface of the longus coli muscles and can be traumatized with excessive elevation or cutting of the longus coli muscles. The superior laryngeal nerve (SLN) arises from the vagus nerve just as it exits the skull and descends in the neck just medial to the carotid arteries. At the level of the hyoid bone, the SLN turns medially and divides into the internal (sensory) and external (motor) branches. Injury to the SLN can cause changes in the pitch of the voice (external branch injury) and have significant aspiration risk owing to loss of sensation above the vocal cords (internal branch injury). The SLN is at greatest risk on upper cervical approaches (C3-C4 and higher), and, thus, is likely an uncommon complication of arthroplasty procedures, which are most commonly performed in the middle and lower cervical spine. The recurrent laryngeal nerves (RLN) have a more complex course and are more vulnerable to injury in lower cervical spine surgeries in comparison to the SLN. On the left side of the body, the RLN, after arising from the vagus nerve, passes beneath the aorta at the level of the ligamentum arteriosum before ascending to the laryngeal structures in the neck. The left RLN has a more vertical and predictable course within the tracheoesophageal groove
Anterior
Sternocleidomastoid
Alar fascia Prevertebral fascia
Middle layer of deep cervical fascia
Platysma
Carotid sheath
C6
External jugular
Vertebral artery
Left
Right
Posterior
P A R T
40
I
Introduction to Motion Preservation Surgery of the Spine
in comparison to the right RLN. The right RLN takes a more oblique course in its ascent after passing beneath the right subclavian artery and is in a more vulnerable location outside of the tracheoesophageal groove throughout a large portion of its course. In addition, a variant called a nonrecurrent laryngeal nerve, which arises from the mid-cervical vagus nerve and passes directly to the laryngeal structures, is occasionally (40), ossification of posterior longitudinal ligament, and diffuse idiopathic skeletal hyperostosis8 (Table 9–3). Although many exclusion criteria do overlap, they are somewhat more distinct than inclusion criteria among studies. For example, in some studies, isolated axial neck pain without radicular symptoms is a criterion for exclusion. In others, prior fusion adjacent to the vertebral level being treated excludes patients from study participation. Although there is a significant amount of overlap among studies, a careful review of each cervical disc replacement inclusion and exclusion criteria is warranted before participation (Figs. 9–2 to 9–5). CONCLUSION Appropriate patient selection is the cornerstone of any successful surgery. A careful history that reveals evidence of degenerative disc disease, radiculopathy, myelopathy, or myeloradiculopathy is the
n
F I G U R E 9–3. Patients with fixed cervical kyphosis, particularly with 3.5 mm or more cervical subluxation, should not be considered for current versions of cervical disc replacements.
first step in patient selection for cervical disc replacement. Plain radiographs and MRI are the two most useful tools in confirming the diagnosis. Although there are no long-term studies on cervical disc replacements that compare with those of anterior cervical decompression and fusion, provided the surgical indications remain the
A A
B
n F I G U R E 9–2. Contraindications for cervical arthroplasty are cases with extensive cervical spine immobility including congenital fusions, redundant posterior longitudinal ligament, multiple prior anterior cervical decompression and fusion, dural ectasia, and Klippel-Feil syndrome. The above lateral plain film (A) and sagittal MRI (B) reveal such extensive cervical spine immobility.
n
B
F I G U R E 9–4. Post-traumatic kyphosis (A) and patients with
posterior ligamentous disruption cannot be considered for cervical arthroplasty. The main stabilizing ligaments after cervical total disc replacement are the two zygoapophyseal capsular ligaments, so they need to be functioning well for a stable construct. This lateral radiograph (B) reveals significant kyphosis, a contraindication for cervical arthroplasty.
P A R T
84
I
Introduction to Motion Preservation Surgery of the Spine
REFERENCES
A
B
n
F I G U R E 9–5. An absolute contraindication for cervical total disc replacement is ossification of the posterior longitudinal ligament (OPLL), the dura is ossified and cannot be mobilized. The ossification will recur and prevent motion. Furthermore, anterior decompression has a high incidence of cerebrospinal fluid leakage. The site of neurologic compression is behind the vertebral body (A) instead of at the level of the disc, so adequate decompression often requires a full corpectomy, resecting the bone required (B) for anchorage of the prosthesis.
same, it is hoped that results will also predictably improve patient pain and overall function. Careful evaluation of the inclusion and exclusion criteria of each cervical disc replacement investigational study is the final step in patient selection before study participation.
1. Bohlman HH, Emery SE, Goodfellow DB, Jones PK: Robinson anterior cervical discectomy and arthrodesis for cervical radiculopathy. Long-term follow-up of one hundred and twenty-two patients. J Bone Joint Surg Am 75:1298–1307, 1993. 2. Emery SE, Bohlman HH, Bolesta MJ, Jones PK: Anterior cervical decompression and arthrodesis for the treatment of cervical spondylotic myelopathy. Two to seventeen-year follow-up. J Bone Joint Surg Am 80:941–951, 1998. 3. Zdeblick TA, Hughes SS, Riew KD, Bohlman HH: Failed anterior cervical discectomy and arthrodesis. Analysis and treatment of thirtyfive patients. J Bone Joint Surg Am 79:523–532, 1997. 4. Hilibrand AS, Carlson GD, Palumbo MA, et al: Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg Am 81:519–528, 1999. 5. Hoppenfeld S, Hutton R: Physical Examination of the Spine and Extremities. New York, Appleton-Century-Crofts, 1976. 6. Evans RC: Illustrated Orthopedic Physical Assessment, 2nd ed. St. Louis, Mosby, 2001. 7. Boden SD, McCowin PR, Davis DO, et al: Abnormal magneticresonance scans of the cervical spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am 72:1178–1184, 1990. 8. McAfee PC: The indications for lumbar and cervical disc replacement. Spine J 4(6):177S–181S, 2004. 9. Goffin J, Van Calenbergh F, van Loon J, et al: Intermediate follow-up after treatment of degenerative disc disease with the Bryan Cervical Disc Prosthesis: single-level and bi-level. Spine 28:2673–2678, 2003. 10. Pimenta L, McAfee PC, Cappuccino A, et al: Clinical experience with the new artificial cervical PCM (Cervitech) disc. Spine J 4(6): 315S–321S, 2004.
CHAPTER
10
Quantitative Motion Analysis (QMA) of MotionPreserving and Fusion Technologies for the Spine* John A. Hipp and Nicholas D. Wharton
K E Y l
l
l
l
l
P O I N T S
Widely used computer assisted methods are available to measure intervertebral motion in research studies, and these methods are validated to provide more accurate and reproducible measurements than obtainable with manual methods. The true fusion or pseudoarthrosis rates, the true type and frequency of adjacent segment changes, and the actual range of motion provided by motion-preserving technologies will only be known if maximal patient effort and high-quality imaging studies are used. Intervertebral motion can be measured by the quantity of rotation or translation, and by the quality of the motion as assessed by parameters such as the center of rotation. Many research studies have provided data to help define success criteria for interpreting intervertebral motion measurements. These studies are summarized in this chapter. More advanced methods for assessing the quality of intervertebral motion, including multiplanar assessment techniques, are being developed and will lead to a more comprehensive understanding of motion-preserving technologies.
Dynamic imaging studies are frequently used in research and clinical practice. Accurate and reproducible measurements of intervertebral motion obtained from these studies are critical to proving that motion-preserving treatments provide the intended motion, as well as for assessing fusions, the preoperative condition of the spine, and any postoperative changes adjacent to the treated level(s). Many of the methods that have been used in clinical practice and in research studies to quantify intervertebral motion are now known to have limitations in accuracy and reproducibility. These limitations can compromise the identification of pre- or postoperative abnormalities and the detection of differences between treatments. New computer-assisted methods have been proven to provide accurate and reliable measurements. These methods have been used in many studies, and are achieving standardization in laboratory and clinical research. However, even with accurate
*There are more than one hundred peer-reviewed publications containing valuable information supporting the topics covered in this chapter, but not all could be included due to space limitations.
image assessment methods, the actual amount and type of motion that an implant provides can only be determined if the dynamic imaging is done with repeatable acquisition methods and sufficient gross motion through the spine. In addition, the true pseudoarthrosis rate with fusion surgery and the true incidence of adjacent segment disease will only be known with good quality control in the imaging studies. Recommended quality control guidelines to achieve this are provided in this chapter. These guidelines are important for reliably measuring both the quantity and quality of motion in the spine. Recommended guidelines for interpreting measurements related to both the quantity and quality of motion in the spine are also provided in this chapter.
REQUIREMENTS FOR ACCURACY AND REPRODUCIBILITY OF INTERVERTEBRAL MOTION MEASUREMENTS With many treatments for spinal disorders, it is important to know if pretreatment intervertebral motion is abnormal; in particular if excess (or insufficient) motion exists that must be addressed with treatment. This requirement necessitates a diagnostic test for intervertebral motion that can reliably detect the presence of motion abnormalities. Once the diagnosis is established, different treatment options can entail somewhat different requirements for intervertebral motion measurements. In particular, motion-preserving devices require a somewhat different emphasis on measurement technique and interpretation than with fusion devices. The critical question with respect to the functional definition of fusion has always been: “Is motion stopped?” There has been less concern about the magnitude of the motion than about the presence or absence of motion. Traditionally, motion has been measured only in the sagittal plane, since it is generally assumed that if motion is stopped in this plane, it is also stopped in other planes. A pseudoarthrosis is functionally defined as motion between vertebrae in excess of a specific threshold used to define a solid fusion. This threshold has been typically assigned based on the accuracy limits of the measurement technology and the estimated 85
86
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
amount of deformation that would be consistent with a solid fusion. The reported fusion rate in a research study will depend on the threshold chosen to define fusion.1,2 This threshold can have a dramatic effect on published fusion rates. The measurement techniques and thresholds used in many previous studies for assessing fusion were generally too inaccurate to detect most failed fusions, with the exception of gross failures.3 General concerns over measurement error and reproducibility have caused most investigators to select a high threshold of motion to avoid misclassifying a successful fusion as a failure. A threshold of 4 to 5 degrees of motion has been commonly selected to define a failed fusion even though many experts recommend a lower threshold.4 For a test to have a high true positive rate for failed fusion, the threshold of motion must be greater than the error in the measurements. This ensures that the test is sensitive for detecting failed fusions, but this approach makes the test nonspecific in that many failed fusions will be incorrectly classified as successful. In clinical practice, given the limited reliability of conventional measurement techniques, intervertebral motion measurements have been generally combined with subjective assessment of bridging bone to classify each level as fused or not fused. There is some evidence that the reproducibility, sensitivity, and specificity of diagnostic tests for spine fusion success can be substantially improved with computer-assisted analysis of flexion/extension x-rays.3,5,6 However, until recently, most assessments were made with less reliable manual techniques, such as the Cobb method.4 In contrast to fusion technology, the success of a motionpreserving technology is defined in part by the quantity of motion the technology provides. The goal of some devices may be to provide for normal motion, while the goal for other devices may be to provide for “stable” or “safe” or “sufficient” motion. To determine if a motion-preserving device provides for normal motion, it is important to answer the question: “What is the quantity of motion?” The quantity of motion can also be important when establishing equivalence of a new device to a previously approved device. Several motion parameters exist to describe quantity of motion. These parameters most commonly include: angular motion, translational motion, change in disc height, and change in functional spine unit (FSU) height.7 The measurements are usually produced from flexion-extension or lateral bending x-rays or both. With special techniques, these parameters can also be measured from dynamic magnetic resonance imaging (MRI) studies. In many cases, these measurements are produced at multiple levels of the spine, both at the treated and at the adjacent levels. Results are frequently reported in degrees and millimeters. Normalization of the displacement and translation data is often used to eliminate the effect of variable magnification in the x-rays, as well as anthropometric variations, thereby facilitating comparisons across patients and across studies. Many studies report the total amount of motion from flexion to extension, or from left to right lateral bending. However, it may also be important to evaluate quantity of motion using neutral as a reference point. Component motion (e.g., neutral to flexion and neutral to extension) can be important for expressing the balance of motion between flexion and extension. If, for example, an implant places the level in hyperlordosis, then this component analysis would likely show limited motion in
neutral to extension. Failure to reproduce the normal ratio of neutral to extension versus neutral to flexion may also be a risk factor for abnormal loading of the posterior or anterior structures of the spine, both at the implanted and adjacent levels. However, this analysis of the balance between flexion and extension is confounded by the variability in neutral posture between individuals. A second, and potentially more important question is: “What is the quality of motion with a motion-preserving device?” The design objective of many motion-preserving devices is to replicate the natural kinematics of the intact disc or motion segment. Such kinematics may not be adequately described by simple measurements of intervertebral rotation and translation alone. More sophisticated measurement techniques can describe and characterize complex patterns of motion and can identify more subtle abnormalities that may result in longer-term complications. Several parameters exist to describe quality of motion. These parameters include: center of rotation (COR), the ratio between translation and rotation, helical axis of motion (HAM), load-displacement curves, and the neutral zone (NZ). Load-displacement curves and the NZ are important factors for understanding intervertebral motion and for evaluating quality of motion in laboratory studies, but are difficult to measure in patients because of the need to know the applied loads across a motion segment. Quality of motion parameters can be monitored on a continuous or instantaneous basis, or at multiple points in the motion cycle. Evaluating motion on a continuous basis may yield more significant information than can be obtained from analyzing the extremes of motion alone. Video fluoroscopy is excellent for this purpose, although image quality may be sacrificed when imaging the lumbar spine. Continuous motion assessment can detect discontinuities in motion or abnormal motion paths that might not be detected by analysis of x-rays taken at the end range of motion.8 For example, it has been shown that the instantaneous axis of rotation moves slightly during flexion-extension.9 However, evidence-based guidelines supporting a positive risk/benefit ratio for continuous motion assessment have yet to be established, and the clinical significance of discontinuities in motion has yet to be established. With respect to patient selection criteria and for assessing changes adjacent to implanted levels, relationships have been documented between quantity and quality of motion parameters and forces in the spine, damage to soft tissues, and the state of degeneration in the spine. The quantity and quality of motion, when evaluated in combination, are important indicators of the normal balance of forces in the spine. Abnormal intervertebral motion measured from a flexion-extension study may indicate an inability of the spine to support normal forces and moments. Abnormal motion may also be associated with pain, neurologic symptoms, degeneration, and a worsening of the instability (see Schneider et al10 for a representative study and a list of supporting references in the lumbar spine, and Amevo et al11 and Ng et al12 for cervical spine references). A spinal motion segment is composed of a complex combination of joints, ligaments, muscles, and other structures that provide complementary and redundant motion constraints and together serve to modulate motion. Disruption of any one component of this complex system may alter kinematics in subtle ways that may be difficult to detect acutely, but may
CHAPTER 10
Quantitative Motion Analysis (QMA) of Motion-Preserving and Fusion Technologies for the Spine
affect the longevity of the segment. Measuring complex interactions in this system and the effect on intervertebral motion, especially in devices designed to mimic natural motion, requires sensitive and precise measurements, as well as data to facilitate interpretation of these measurements. ACCURACY AND REPRODUCIBILITY OF MANUAL METHODS FOR MEASURING INTERVERTEBRAL MOTION This chapter focuses on measurements made from x-rays and other noninvasive imaging modalities that can be applied to large scale clinical studies. It is possible to measure intervertebral motion by other methods, such as by using pins percutaneously implanted into vertebrae, by analysis of multiplanar x-rays taken after implanting metal markers into vertebrae, or by inferring motion from measurements of the orientation of the skin overlying the spine. However, these techniques are limited in accuracy or their applicability to large-scale clinical studies. The accuracy and reproducibility of any measurement of intervertebral motion made from x-rays, MRI, or computed tomography (CT) will depend on image quality. With x-rays in particular, image quality is dependent on the specific imaging technique used by the technologist (which can be strongly influenced by the physician that requests the test). Important variables include the exposure settings (kVp, mas), the amount of out-of-plane motion of the patient, distortion in the images caused by the parallax effect, and post-processing of the images. Typical out-of-plane motion includes lateral bending or left-right twisting during a sagittal plane flexion-extension study. Parallax occurs in images formed by x-rays originating from a point source and projecting through the patient onto a large x-ray film. This creates a difference in the apparent relative position between vertebrae that can occur only because the position of the vertebrae with respect to the center of the x-ray film has changed. Consistent attention to imaging technique will improve the quality of clinical radiographs and thereby improve the accuracy and reproducibility of measurements produced from them. Even with excellent image quality, the accuracy and reproducibility of measurements made from the images depend on the specific measurement technique. Conventional radiographic measurements have been most commonly based on manual line-drawing techniques and/or landmark selection methods. In general, an observer draws lines through the perceived plane of each end plate or along the perceived posterior wall of each vertebra.13 The angle measured between lines on each radiograph describes the relative position between adjacent vertebrae, and the change in these angles between flexion and extension is a measure of angular motion between two vertebrae. There are many variations on this theme. For example, specific anatomic landmarks can be identified on a series of radiographic images, and the change in position of these landmarks can be used to describe disc heights and translations. The Cobb method and the posterior tangent method, sometimes supplemented by superimposition of images, are common methods used in clinical practice to measure intervertebral rotation based on drawing lines either through the vertebral end plates or through the posterior aspect of the vertebral bodies. These methods have been documented to have limited accuracy and reproducibility.13–15 These techniques are dependent on the ability of the
87
observer to select a series of landmarks in one image and reproducibly select the identical series of landmarks in the next image. Irregularity of the vertebral end plates and variations in quality between images confounds this process. Poor visualization of the bony anatomy due to out-of-plane effects, parallax effects, overlapping tissues, exposure settings, surgery, or the presence of implants makes it difficult to achieve optimal reproducibility. The reliability of these methods may be improved by superimposing the x-rays, which helps to visualize changes in the relative position of adjacent vertebrae and provides additional visual cues to aid in landmark selection.16 Several investigators have reported on the errors in assessing sagittal plane intervertebral rotation in motion-preserving devices using methods that require the observer to draw lines along features of the implants. A recent reproducibility study measuring rotation between the superior and inferior components of a motion-preserving device found maximum errors of 5 to 13 degrees and average errors of 1.8 to 3.3 degrees (standard deviation: 2.2 to 4.2 degrees) depending on the specific implant feature used.17 Another study reported 95% confidence intervals for errors to be between 2 degrees and 4.3 degrees for intra- and inter-observer errors.18 The geometrically regular features and high-contrast of artificial discs should theoretically improve the reproducibility of standard measurement techniques.17,18 As long as there is no relative motion between the implant and bone, the rotation between vertebrae at an implant site will be the same as the rotation between superior and inferior components of the implant. To be certain that motion has occurred across an implant, the motion must be greater than the error in the measurement technique. It is therefore essential to select geometric features of the device that maximize reproducibility and ensure that the method is as sensitive to small motions as possible.17 ACCURACY AND REPRODUCIBILITY OF COMPUTERASSISTED TECHNIQUES Most modern techniques in radiographic image analysis use region of interest (ROI) methods to track the motion of spinal vertebrae. These methods include geometric templating and pattern matching.13,15,16,19,20 These methods are, by necessity, computerized and have been proven to enhance the accuracy and reproducibility of motion measurements.15,16,21 Unlike manual techniques, which require selecting a number of reference points to assess motion, these methods typically match contours or patterns of vertebrae on radiographs using all of the information about a vertebra, including its size, shape, and density variation. The process is to define a template, or model, of a vertebral pattern in one image (e.g., flexion), and then search for its corresponding pattern in a target image (e.g., extension). The search is conducted by digitally superimposing the model on the target image and transposing it until a best-fit match is found. The model may contain the anterior body and posterior elements, or it may contain the anterior body alone. The process produces a mathematical transformation matrix that describes the spatial relationship between a vertebral pattern in one image and its corresponding pattern in another image. This process is repeated for multiple vertebrae, and it may be aided with automated algorithms or computer-assisted techniques.
88
n
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
F I G U R E 10–1. Digital registration of two images through
superimposition.
Using the motion information obtained from the image analysis, radiographic images may be digitally superimposed and registered such that a particular vertebra occupies the same relative position in multiple images (Fig. 10–1). This process is conceptually similar to overlaying plain films on a light box until a specific vertebra is aligned in both images. Registration makes it possible to detect subtle differences in the position and orientation of adjacent or overlapping structures between two images. One published technique involves alternately displaying the registered images onscreen so that the registered vertebra appears to hold a constant position while the adjacent structures move relevant to it.6,15,22 This technique, known as Feature Stabilization, makes it easier to visualize relative motion between vertebrae. Feature Stabilization also provides a visual feedback mechanism for evaluating the accuracy of the registration and, if necessary, correcting it. Imperfect stabilization may suggest that an adjustment is required to fine-tune the registration. If so, the direction and magnitude of the adjustment can be visualized. This ability to objectively identify error in the analysis, and correct it, is an important element of computer-assisted techniques. The transformation matrix that describes how the position of a particular vertebra changes from one image to the next may be used to calculate the motion of selected anatomic landmarks. Based on matrix math, the position of landmarks selected in the first image may be recalculated in subsequent images. Calculating the position of the landmarks avoids the reproducibility errors commonly associated with manual techniques. The assumption of rigid body motion also assures that the relative position of vertebral landmarks remains constant between images. Radiographic projections of vertebrae rarely remain constant from image to image, which can compromise the accuracy of the calculated landmark positions. To ensure accuracy, computer-assisted techniques often minimize out-of-plane effects by tracking regions of bone, such as the spinous processes, that are least sensitive to these effects. Poor contrast may be improved by applying contrast enhancement filters, and magnification effects may be addressed with digital resizing methods.
Several published studies have been conducted to assess the accuracy and reproducibility of these computer-assisted techniques. Few studies provide definitive accuracy assessments because they require “gold standard” reference measurements, and the accuracy of some of these techniques is close to, or may surpass, the accuracy of available “gold standards.” For this reason, some accuracy studies use frozen spines imaged in different positions. In these studies, any intervertebral motion reported by the method represents error in the measurements. Reported average accuracies range from 0.4 to 2.8 degrees for rotations and 0.5 to 1.5 mm for translations.15,21,23 The reported average reproducibility of these measurements ranges from 0.4 to 2.6 degrees as measured by inter- and intra-observer variations.16,20,24 The 95% confidence intervals for these methods range from 0.8 to approximately 3 degrees.16,20,24 The accuracy and reproducibility of these methods may be reduced by poor quality images or vertebrae that or rotated or tilted out of the plane of imaging. There are multiple different computer-assisted methods that have been described and used for measuring intervertebral motion.13,15,16,19–21,24 When selecting a computer-assisted technique, it is important to critically assess the validation of the technique, since some methods are more thoroughly validated than others. Because no current computer-assisted method is fully automated, some operator intervention is needed with all published methods. It is therefore also preferable to ensure that the method has been validated by observers who have been blinded to the experimental design and experimental hypothesis, which minimizes sources of potential bias. PATIENT POSITIONING TECHNIQUES AND PATIENT EFFORT Dynamic imaging studies must be performed with repeatable patient positioning and excellent patient effort. Whether a level is unstable or degenerating preoperatively, or whether motion is preserved or eliminated postoperatively, can only be determined and compared to other studies if the imaging is done with consistent patient positioning and a committed effort by the patient to maximally flex and extend. It has been shown that sequential recruitment of spinal levels occurs during flexion-extension, and certain levels eventually contribute more than others to the total motion.25–27 This has been referred to as a phase lag in spinal motion.26 With insufficient patient effort, motion may be non-existent at levels that are in fact capable of providing motion, and the relative proportion of motion measured at any particular level may inadequately describe the amount of motion that can be provided by that level. Several studies have documented that the true amount of intervertebral motion that occurs at a treatment level when a patient is highly motivated to maximally flex and extend will be missed if the patient only makes a modest effort to flex and extend during the imaging. Miyasaka et al showed that maximal effort resulted in over twice as much intervertebral motion at the lower levels of the lumbar spine when patients were coached to flex and extend maximally versus being uncoached.25 Measurements made with inadequate effort would have underreported the true motion of the spine, particularly at the lower levels. Miyasaka also showed that with maximal effort, the standard deviations were half what they were with modest effort. Dvorak et al also found greater
CHAPTER 10
Quantitative Motion Analysis (QMA) of Motion-Preserving and Fusion Technologies for the Spine
motion and smaller variance in flexion extension studies of the cervical spine with methods that encouraged gross motion through the head and neck.28 The smaller variations suggest that the physiologic restraints to motion in the spine may be fairly similar between individuals and that much of the variability reported in data describing motion in asymptomatic people is due to variability in subject effort. The obvious implication to clinical practice and research studies is that true instability in the spine, and the true extent of motion provided by an implant will remain unknown without assuring sufficient patient effort. In addition, the smaller standard deviations achievable with excellent patient effort will require fewer patients to show differences (or the lack thereof) between treatments. Quality control (QC) criteria for assessing whether there is enough gross motion through the spine in a flexion-extension study to reliably assess intervertebral motion at specific levels have not been established but are clearly needed and can be estimated from existing scientific literature. Damage to the soft-tissue structures that normally control intervertebral motion will intuitively not be evident unless intervertebral motion is at least to the level where these structures normally begin to restrict motion. In the laboratory, this can be demonstrated by studying the neutral zone. The neutral zone is the range of motion between vertebrae that can be created with application of minimal loads. Damage to the soft tissues is evident as an enlargement of the neutral zone. The neutral zone and the proportion of the total range of motion included in the neutral zone are powerful parameters for understanding the biomechanics of the spine and spinal treatments. However, direct and objective assessment of the neutral zone, and the load-displacement behavior of the motion segment outside the neutral zone, requires knowledge of the loads being applied to the segment, which is generally difficult to obtain in clinical practice. Nevertheless, the size of the neutral zone has been described in several publications, and this provides guidelines for assessing motion from clinical x-rays. Based on the data from Panjabi et al,29 the average neutral zone for the mid to lower cervical spine is approximately 3 degrees between flexion and extension. Intervertebral rotation would have to be at least one degree greater than the neutral zone at most levels to begin stressing the soft-tissue restraints, and also to be above measurement error in most computer-assisted measurement techniques. An alternative QC guideline is to require the total amount of motion through the spinal region to be within the 95% confidence interval for asymptomatic people (approximately 30 degrees from C2 to C7 or 30 degrees from L1 to S1). However, this criteria may be inappropriate in the presence of a multi-level fusion, motion limiting treatment, or preoperatively, since one or more levels may not be contributing normally to total motion. Additional support for QC criteria can be derived from the observation of White and Panjabi that in the cervical spine, it is rare to have more than 11 degrees difference between adjacent levels in the angle between end plates defining the intervertebral disc spaces.7 This measurement is made from a neutral or flexion x-ray and is part of the widely used American Medical Association (AMA) disability assessment guidelines for evidence of structural impairment to the spine.30 Consistent with this observation, further analysis of data from the study by Reitman et al22 revealed
89
that less than 5% of asymptomatic people had intervertebral rotation from flexion to extension at any one level that was 11 degrees or more greater than at an adjacent level. It would clearly not be possible to apply these observations clinically unless there were at least 11 degrees of motion in at least one level in the cervical spine. This amount of rotation is only a modest requirement since it is below the average for most levels in the cervical spine (see Table 10–1). Based on these rationalizations, a reasonable and minimal QC guideline for intervertebral rotation in cervical flexion-extension x-rays is to require intervertebral motion to be at least 11 degrees at two non-fused levels, and at least greater than 4 degrees at an adjacent, non-fused level. Ninety-five percent of flexion/extension studies of 140 asymptomatic volunteers22 satisfied these guidelines, based on a reanalysis of the original data. In that study of asymptomatic volunteers, subjects were coached to flex and extend as much as they could. In contrast, 33% of 457 cervical flexion-extension x-rays from a typical clinical practice that did not use any specific quality control protocol failed these guidelines (unpublished retrospective study). Criteria for determining if a lumbar flexion-extension study has enough gross motion to be of diagnostic quality can also be justified based on the amount of motion seen between vertebrae in the asymptomatic population and on published data describing the lumbar neutral zone. Based on data from Panjabi et al,31 the average neutral zone in the lumbar spine is approximately 5 degrees. Rotation would need to be at least 6 degrees to be certain soft tissue restraints are being tensioned and to account for measurement error. In an ongoing study of asymptomatic volunteers by the authors of this chapter, there were at least 9 degrees of intervertebral rotation in at least two lumbar levels in 95% of 75 volunteers, and less than 5% of the volunteers had greater than 9 degrees more rotation at one level compared to adjacent levels. Data from Harada et al27 suggests that during flexion, motion first starts at the upper levels, with lower levels contributing to overall motion as flexion proceeds. At the point where motion was 9 degrees at any one level, their data suggest that all levels were contributing to motion. Thus, rational and minimal QC criteria for lumbar flexion-extension studies would be to require at least 9 degrees of intervertebral rotation in at least two levels and at least 6 degrees of rotation at an adjacent level. A drawback of these cervical and lumbar QC criteria in routine clinical practice is the need to first accurately analyze flexion-extension x-rays to determine if they are of sufficient quality. Nevertheless, these QC criteria can be easily applied to flexion-extension studies from clinical trials or research studies to avoid inclusion of data that are unreliable due to insufficient patient effort. To minimize the number of unsatisfactory flexion-extension studies, it is very important that the physician stress to their patients that: 1) it is safe to maximally flex and extend their spine, and 2) it is important that they do this since the true condition of their spine can only be assessed with this effort. Many patients either are concerned that they may cause damage to their spine if they move too much, or are seeking empathy from the technologist performing the study by demonstrating how difficult it is for them to move. Both of these psychological restrictions to motion in the spine can be mitigated by reassurance and instructions from their physician. Rehearsing the motion with the patient can help
90
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
assure that the patient understands what is required of them, but also allows the physician to order dynamic imaging studies only for those patients where they will obtain reliable information. The patient positioning technique used during a flexionextension study can also change the amount (and possibly quality) of motion that is measured between vertebrae. For both the cervical and lumbar spine, a variety of methods for patient positioning have been described in published studies. In selecting a method, it is important to consider how consistently the expected patient population will tolerate the method, whether rigorous reference data are available for the specific method, how good the method is at minimizing out-of-plane motion, and, most importantly, whether the test results will describe the true performance of the treatment. In the cervical spine, flexion and extension can be performed with the patient standing or seated, with no known difference between these positions. To ensure maximum effort in flexion, patients should be coached to bring their chin to their chest without tilting or twisting their head to the left or right. Maximum extension can be obtained by coaching the patient to point their chin as far up toward the ceiling as they can, again without tilting or twisting their head to the left or right. Reitman et al22 provided normative data for flexion-extension cervical range of motion using this technique. These data are summarized in Table 10–1. Lumbar flexion-extension studies have been performed using a wide variety of patient positioning techniques including standing, seated, and lateral decubitus. There are pros and cons of each method. A version of the standing technique may currently be one of the most commonly used techniques in clinical practice, although this is difficult to document. Standing flexion-extension imaging occurs using a variety of techniques (including from a free-standing position, with the pelvis stabilized by a variety of mechanical systems, or with the patient leaning against a table). Standing flexion-extension using pelvis stabilization methods has been described in many research studies but is generally not performed in clinical practice. For sites that use an unsupported, free-standing protocol (Fig. 10–2A), Miyasaka et al25 proved the advantage of one technique to ensure maximum patient effort. Miyasaka showed that patients who were coached to bring their hands as close to the floor as possible in flexion, and to extend as far back as they possibly could in extension, had 30% more motion at L1-L2 and 136% more motion at L5-S1. This was compared to patients who were asked to bend forward and backwards as is normally done in clinical practice without coaching.25 One previously reported positioning technique32 that is now being utilized in a large study of intervertebral motion in the lumbar spine of asymptomatic people involves a simple seated technique (Fig. 10–2B). A stable, four-leg, low back chair is used with a specially designed cushion (model 583990, Infab Corp, Camarillo, CA) that facilitates maximum extension and prevents the pelvis from being tucked under the body. Patients are asked to bend forward and grab the front legs of the chair as far down as they can for flexion, and then to extend back over the cushion for extension (Fig. 10–2B). This seated technique is tolerated by patients, is easy to implement, and improves range of motion while minimizing out-of-plane effects. Normative data describing intervertebral motion and 95% confidence intervals in asymptomatic patients using this technique are provided in Table 10–2.
TABLE 10–1. Individuals
Intervertebral Motion Data for 129 Asymptomatic Rotation (degree) N
Mean
Std. Dev.
LL
UL
129 129 126 121 78
9.4 14.6 16.6 16.5 13.7
2.9 3.7 3.9 5.0 5.1
3.7 7.5 8.9 6.7 3.8
15.0 21.8 24.3 26.3 23.7
Level
C2-C3 C3-C4 C4-C5 C5-C6 C6-C7
Rotation (% sum of rotation for all visualized levels) Level
C2-C3 C3-C4 C4-C5 C5-C6 C6-C7
N
Mean
Std. Dev.
LL
UL
129 129 126 121 78
15.8 23.7 26.1 24.8 18.7
7.2 6.5 5.2 5.5 5.9
1.7 10.9 15.9 14.1 7.1
29.9 36.6 36.3 35.5 30.2
Translation (% AP End Plate Width)—Motion of the Posterior-Inferior Corner of the Superior Vertebra in a Direction Parallel to the Superior End Plate of the Inferior Vertebra. Level
C2-C3 C3-C4 C4-C5 C5-C6 C6-C7
N
Mean
Std. Dev.
LL
UL
129 129 126 121 78
12.4 15.9 17.2 13.7 7.5
5.2 5.2 5.4 5.4 3.5
2.2 5.6 6.5 3.2 0.5
22.6 26.1 27.8 24.3 14.4
Center of Rotation—AP (% AP End Plate Width) Reported Relative to the Midpoint of the Superior End Plate of the Inferior Vertebra. Negative Values are Posterior to the Midpoint. Level
C2-C3 C3-C4 C4-C5 C5-C6 C6-C7
N
127 129 125 121 76
Mean
9.7 9.6 7.4 6.7 7.9
Std. Dev.
9.6 6.5 6.1 5.6 6.2
LL
UL
28.4 22.4 19.3 17.6 20.0
9.1 3.2 4.6 4.2 4.2
Center of Rotation—Cranial-Caudal (% AP End Plate Width) Reported Relative to the Superior End Plate of the Inferior Vertebra. Negative Values are Inferior to the End Plate. Level
C2-C3 C3-C4 C4-C5 C5-C6 C6-C7
N
127 129 125 121 76
Mean
49.0 37.4 35.0 24.9 8.9
Std. Dev.
20.3 14.9 12.7 12.1 11.6
LL
UL
88.9 66.6 59.9 48.6 31.7
9.2 8.1 10.0 1.2 13.9
These volunteers were selected from an original population of 140 subjects. Levels with moderate to severe degeneration or volunteers with multiple degenerated levels were excluded. The 129 volunteers included 55 males and 74 females (average age 39 for both sexes). C6-C7 and sometimes C5-C6 could not be visualized in some subjects. LL and UL, lower and upper limits of the 95% confidence interval.
REFERENCE DATA FOR INTERVERTEBRAL MOTION Exactly how much motion should be provided by motionpreserving treatments has not been definitively determined and will be different depending on the goals of the specific treatment. For example, a total disc replacement may have the goal of restoring normal motion, a dynamic posterior stabilization device may have the goals of stabilizing the spine while providing enough motion to prevent adjacent level degeneration, and fusion treatments may have the goal of stopping measurable intervertebral motion. If
CHAPTER 10
n
F I G U R E 10–2.
Quantitative Motion Analysis (QMA) of Motion-Preserving and Fusion Technologies for the Spine
91
A and B, Protocol diagrams illustrating good sitting and standing techniques with adequate patient effort.
the goal is to provide normal motion at the implanted level, several studies provide data that can be used as guidelines for minimum and maximum allowable motion. Assessing whether motion is normal involves evaluating not only whether range of motion is within normal limits at each level (quantity of motion) but also whether the device preserves the natural kinematics and hence the natural balance of forces in the spine (quality of motion). Assessing whether the quantity of motion is normal includes both a lower and an upper limit of normal. A device could fail the intended goal by not providing enough motion, or it could fail by providing too much motion. An alternative goal for motion-preserving devices could be to retain the same motion as was measured preoperatively. However, preoperative range of motion is at least theoretically not a good reference since patients were symptomatic preoperatively. Their symptoms and underlying spinal disorder(s) may have compromised motion (due to instability, pain guarding, muscle spasms, etc.). Another goal may be to ensure that treated and adjacent non-treated levels contribute proportionally to motion through the spine and that a difference in motion across adjacent levels is within a normal range. Dynamic stabilization devices require different criteria. The goals for these devices are primarily to provide stability to the treated level, but also to provide for some motion. Success criteria for the motion these devices provide may be based on the goals of preventing additional degenerative changes at the treated level and reducing adjacent level degeneration, with a possible additional goal of preventing fusion at the treated level. Table 10–1 provides the mean, standard deviations, and 95% confidence intervals for intervertebral rotation and translations in the cervical spine of asymptomatic volunteers. The data in this table are from one specific study, but the data are not statistically different from data published by Frobin et al,21 and there is good consistency with other published studies that provide intervertebral motion data for the asymptomatic cervical spine.21,33,34 The data in Table 10–1 were originally published22 after an adjustment was made to control for the effect of patient effort on the intervertebral rotations. The data in table 10–1 do not include this
adjustment for patient effort, since this adjustment is generally not practiced in clinical studies. There is a highly (P < 0.0001) significant difference between levels, with motion being greatest at the C4-C5 level. Based on analysis of data from both Reitman et al22 and Frobin et al,21 there is also significantly greater intervertebral rotations at the C3-C4, C4-C5, and C5-C6 levels in females compared to males (P < 0.05). This observation could be used to obtain an even more precise discrimination between normal and abnormal motion. Table 10–1 also provides the rotation data expressed as a percent of the total rotation for all visualized levels between C2 and C7, since this may reduce the influence of patient effort and generally reduces scatter in the data (the standard deviations are a smaller percent of the means). It should be noted that intervertebral rotations, translations, and changes in disc height are generally interrelated by the biomechanics of the spine, and also tend to be statistically correlated with each other.21,22,28,33 Descriptive statistics for center of rotation (COR) were not originally reported in the study by Reitman et al,22 but are presented in this chapter from the same subjects in the original study. Table 10–2 provides the means, standard deviations, and 95% confidence intervals for COR. The reference frame for this data is located at the midpoint of the superior end plate of the inferior vertebra. The x-axis is coincident with the end plate, and the y-axis is perpendicular to the end plate. The data are plotted in Figure 10–3A. The reference data reported in Table 10–2 are normalized to the anterior-posterior dimension of the superior end plate of the inferior vertebra. These data are consistent with data published by Amevo et al35 after adjusting for differences in the normalization scheme and reference frame used in the two studies. When interpreting data for COR, it is important to pay careful attention to the coordinate system used to report COR. Different published studies may report equivalent data in different coordinate systems, which may confound comparisons of the data. Similar attention should be paid to the method used to normalize the data. COR can be sensitive to technical errors in the technique used to compute it.36 It is important that COR be calculated using
92
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
TABLE 10–2. Intervertebral Motion Data for 63 Asymptomatic Individuals From an Original Study Population of 75 Rotation (degrees) Level
N
Mean
Std. Dev.
LL
UL
L1-L2 L2-L3 L3-L4 L4-L5 L5-S1
63 63 63 63 57
10.5 12.0 13.0 14.1 11.1
3.0 2.5 2.8 3.9 5.8
4.6 7.0 7.5 6.5 0.0
16.5 17.0 18.5 21.7 22.5
Rotation (% L1-S1 rotation) Level
N
Mean
Std. Dev.
LL
UL
L1-L2 L2-L3 L3-L4 L4-L5 L5-S1
56 56 56 56 56
17.4 20.0 21.6 23.5 17.5
4.2 3.8 3.1 4.2 7.9
9.1 12.5 15.6 15.3 2.0
25.6 27.4 27.6 31.7 33.1
Translation (% AP End Plate Width)—Motion of the Posterior-Inferior Corner of the Superior Vertebra in a Direction Parallel to the Superior End Plate of the Inferior Vertebra. Level
N
Mean
Std. Dev.
LL
UL
L1-L2 L2-L3 L3-L4 L4-L5 L5-S1
63 63 63 63 57
4.8 6.3 7.2 7.1 1.3
1.6 1.9 1.9 2.6 2.3
1.6 2.5 3.5 2.0 -3.3
8.0 10.1 10.9 12.2 5.8
Center of Rotation—AP (% AP End Plate Width) Reported Relative to the Midpoint of the Superior End Plate of the Inferior Vertebra. Negative Values are Posterior to the Midpoint. Level
N
Mean
Std. Dev.
L1-L2 L2-L3 L3-L4 L4-L5 L5-S1
63 63 63 63 45
5.4 6.5 6.7 6.8 12.5
6.2 5.0 5.6 4.7 5.7
LL
UL
17.5 16.4 17.7 15.9 23.7
6.6 3.3 4.2 2.3 1.4
Center of Rotation—Cranial-Caudal (% AP End Plate Width) Reported Relative to the Superior End Plate of the Inferior Vertebra. Negative Values are Inferior to the End Plate. Level
N
Mean
Std. Dev.
L1-L2 L2-L3 L3-L4 L4-L5 L5-S1
63 63 63 63 45
5.4 6.7 7.8 3.6 10.9
7.5 6.6 6.0 7.0 10.3
LL
20.0 19.6 19.6 17.3 9.4
UL
9.3 6.1 3.9 10.1 31.1
Moderately to severely degenerated levels, and individuals with multiple degenerated levels in this asymptomatic population were excluded. There were 37 females and 26 males. There were 37 subjects under the age of 45 years, 21 in the 45- to 63-year age group, and 5 that were over 64 years old. The percent L1-S1 rotation could not be calculated in some subjects because L1 or S1 could not be visualized. LL and UL, lower and upper limits of the 95% confidence interval.
validated and consistent methods, and that reproducibility be tested for various ranges of motion. The COR is unreliable when there is little actual rotation between vertebrae.37 In this situation, small measurement errors may lead to large differences in the computed location of the COR. Depending on the technique used to compute COR, a minimum of 2 to 5 degrees of intervertebral motion is required to obtain reliable results. However, once this minimum threshold has been achieved, the position of the COR becomes generally insensitive to the degree of overall motion in
the spine. Based on a re-analysis of previously published22 data, intervertebral rotation at a specific level was highly dependent on the overall motion in the spine (P < 0.001), whereas the position of the COR was not (P > 0.6). In most radiographic studies, the center of rotation is measured in the sagittal plane between the extremes of motion using maximum flexion and extension images. The center of rotation may also be measured continuously at intermediate stages of motion to assess the “instantaneous” center of rotation (ICR). Some investigators refer to an analogous measurement called the instantaneous axis of rotation (IAR). An axis of rotation is the axis perpendicular to the plane of motion along which every point of a moving body, or hypothetical extension of it, is stationary.7 The center of rotation is the intersection of this axis with the plane of motion. Radiographically, the center of rotation is a point on an x-ray, while the axis is perpendicular to the x-ray. To calculate the center of rotation reliably, the motion must be in the plane of the x-ray. Since flexion-extension motion takes place predominately in a single plane that is (ideally) coincident with the imaging plane, COR can adequately describe this intervertebral motion. Although COR is typically reported in the sagittal plane for flexion-extension motion, it is important to recognize that the position of the COR may be different for maneuvers in other planes. For example, in the mid to lower cervical spine, in which coupled motion is not as severe as in the upper cervical spine, attempts at estimating the gross position of the COR in left-right bending have suggested that the COR may be located within or superior to the disc space in lateral bending while inferior to the disc space for flexion-extension.38 This observation must be interpreted as an approximation since left-right bending motions are coupled with out-of-plane motions, such as axial rotation. Nevertheless, three-dimensional marker studies in cadavers have shown similar results.39 For coupled, multiplanar motions (representing the most common activities of daily living) a more general method to describe three-dimensional motion is called the helical axis of motion (HAM).7 The HAM is a unique axis in space that completely defines the three-dimensional motion of a rigid body. It may be described conceptually as a three-dimensional IAR. The HAM cannot be assessed from radiographic images, which are limited to two-dimensional representations of three-dimensional motion. Instead, three-dimensional kinematic analyses using marker-based methods or dynamic MRI may be the best way to evaluate complex motion patterns of the spine. Data related to the use of the HAM to evaluate spinal kinematics is beginning to emerge, especially with respect to motion-preserving devices.40 Table 10–2 provides the means, standard deviations, and 95% confidence intervals for intervertebral motion in the lumbar spine of 75 asymptomatic subjects from one ongoing study by the authors of this chapter. Intervertebral motion was measured from flexion-extension radiographs obtained using the sitting method previously described. All volunteers had never sought medical treatment for back-related symptoms. Levels showing moderate to severe degeneration, and patients with multiple degenerated levels were excluded from the analysis. The intervertebral rotation data in Table 10–2 are statistically equivalent to data from Dvorak et al where intervertebral motion was measured from
CHAPTER 10
93
Quantitative Motion Analysis (QMA) of Motion-Preserving and Fusion Technologies for the Spine
F I G U R E 10–3. Illustrations of idealized cervical (A) and lumbar (B) spine segments with means and 95% confidence intervals for center of rotation. The individual data points for each asymptomatic subject are shown as red dots. The dark red ellipses depict the 95% confidence interval for the data while the dark red circle in the center of the ellipse depicts the average coordinates of the COR. The COR data were expressed normalized to the AP width of the superior end plate of the inferior vertebra for each level. Since not all individuals have the same ratio of vertebral width to height, the cranial-caudal position of the COR may be somewhat different than depicted for patients with shorter or taller vertebrae than illustrated in these figures. The COR data are referenced to a specific anatomic coordinate system, and the landmarks used to define the coordinate system at each level are also marked by dark red circles on the anterior- and posterior-most aspects of the superior end plate of the inferior vertebra at each level in these figures.
n
flexion-extension x-rays of asymptomatic volunteers in a standing position.41 In the Dvorak study, the x-rays were taken as the investigators pulled (for extension) or pushed (for flexion) on the volunteer's shoulders to obtain maximum flexion and extension. The position of the COR provided in Table 10–2 and plotted in Figure 10–3B is consistent data with data from several previously published studies.9,10 The same coordinate system was used as for the cervical data in Table 10–1. COR results for levels exhibiting less than 3 degrees of intervertebral motion were excluded. For lateral bending studies of the cervical and lumbar spine, Table 10–3 provides the means, standard deviations, and 95% confidence intervals after combining data from two published studies of intervertebral motion in the cervical spine38,42 and data from two studies of the lumbar spine.43,44 ENDPOINTS AND SUCCESS CRITERIA FOR THE QUANTITY AND QUALITY OF INTERVERTEBRAL MOTION The radiographic endpoints and success criteria selected for a clinical study will depend on the intended goal of the treatment being studied. Preoperatively, excessive intervertebral motion could be identified as motion greater than the 95% confidence intervals
TABLE 10–3. Summary of Published Data for Intervertebral Rotation Measured in the Coronal Plane During Lateral Bending in the Cervical38,42 and Lumbar Spine43,44 Cervical Level
N
Mean
C2-C3 C3-C4 C4-C5 C5-C6 C6-C7
12 12 22 24 24
7.4 7.0 8.4 9.5 12.1
Level
N
Mean
Std. Dev.
LL
UL
L1-L2 L2-L3 L3-L4 L4-L5 L5-S1
14 14 30 14 10
9.9 11.1 11.2 7.1 3.0
2.8 4.2 4.8 3.4 3.8
7 7 5 1 1
15 18 12 9 6
Std. Dev.
LL
UL
4.0 2.8 3.1 2.6 3.3
0.0 1.5 2.4 4.5 5.5
15.2 12.5 14.4 14.6 18.6
Lumbar
The data from the published studies are pooled together in this table to the extent possible. Data from Plamondon et al44 and Ishii et al42 were published for one side bending only and were doubled for pooling with data from Wharton and Hipp.38 Wharton and Hipp only provided data for C4-C5 to C6-C7. The lower limits (LL) and upper limits (UL) for the lumbar spine are the minimum and maximum values reported by Pearcy et al.43
94
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
in Tables 10–1 and 10–2. This motion could be abnormal with respect to intervertebral rotation, intervertebral translation, or the position of the center of rotation. Clinically, the presence of intervertebral motion greater than that measured in asymptomatic subjects is only of immediate clinical relevance if the motion correlates to patient symptoms. It is again important to note that the lack of excessive motion in the spine measured from a flexionextension study can only be used to rule out excessive motion if there was sufficient patient effort. In particular, a patient who has motion limitations from muscle spasms and/or pain guarding can have gross instability that may not be evident from x-rays if the patient does not, or cannot, make a sufficient flexion-extension effort. In addition to excessive motion, hypomobility may also be important, as long as it is not due to insufficient patient effort. Hypomobility is characteristic of the advanced stages of degeneration, and may also occur with perched facets, severe facet disease, severe osteophytes, ossified ligaments, or muscle spasms, and may thereby provide clinically relevant information. Post operatively, the success and failure criteria that can be applied to quantitative measurements of intervertebral motion depend on the goal of the treatment. In many clinical investigations of motion-preserving devices, fusion is the treatment control. For fusion, where the goal is to minimize motion and achieve solid bridging bone, the success criteria will be motion that is no more than allowed by elastic deformation of bone that has formed between vertebrae (assuming accurate measurements). The postoperative time to apply this success criteria remains unclear and depends on the type of fusion surgery, the healing potential of the individual, and other factors that are beyond the scope of this chapter. However, once a follow-up time has been established for the final fusion assessment, specific intervertebral motion criteria must be applied to assess the mechanical success of the fusion procedure. These criteria must account for potential deformation that may occur through the bone bridging between vertebrae or through the posterior elements in cases of anterior fusion. The other important factor in selecting fusion success criteria is the error in the measurement technique. With manual methods used in many previous studies, the inter- or intra-observer error in the measurement technique has been high, so the motion criteria for failure have been high. With respect to establishing pseudoarthrosis rates, these high success criteria minimize the possibility that a successful fusion will be missed. However, failed fusions will be missed where motion at the fusion site is below the threshold. Some fusion failures may also be missed because the patient did not move enough between flexion and extension to generate detectable motion at the fusion site. Intervertebral rotation is generally more reliable than intervertebral translations, since the scaling factor in the image does not have to be determined, and because with methods that use pattern-matching technology, the rotations do not require identification of specific landmarks. With accurate computer-assisted motion analysis techniques, the accuracy is generally under 1 degree6,15 and the observer error is typically around 0.5 degree. Assuming the worst case of accuracy and observer error combined with a small amount of motion that could occur by elastic deformation of bridging bone, a threshold level of 2.0 degrees is a reasonable threshold to use in clinical studies. It may be necessary to combine this quantitative assessment of
motion with subjective, visual assessments of bony bridging to confirm the final status of fusion. The success criteria for motion-preserving or dynamic stabilization devices depend on the specific goal of the treatment. If the goal of the treatment is to provide for normal motion, then success criteria based on the lower and upper limits established for an asymptomatic population are appropriate. Assuming that comparable protocols were followed and patient effort was sufficient, the data in Tables 10–1 and 10–2 provide appropriate success criteria for sagittal-plane motion during flexion and extension. Intervertebral rotation, translations, and center of rotation are all important. In addition, a comprehensive assessment would also include quantitative analysis of lateral bending. In general, quantitative assessments of intervertebral motion should show that motion is within the 95% confidence intervals for the normal patient population after accounting for measurement error. This includes determining if each individual level has motion within the 95% confidence interval for that specific level, but also identifying levels that are moving disproportionately more than the adjacent levels. In the cervical spine, there should not be more than an 11 degree difference in the range of motion between adjacent levels, based on a reanalysis of previously published data for 140 asymptomatic individuals.22 Eleven degrees is also the difference in intervertebral angles between adjacent levels in the cervical spine that the AMA uses as evidence of a “loss of structural integrity” in the cervical spine.30 In the lumbar spine, there should not be more than a 9-degree difference in the range of motion between adjacent levels, based on the analysis of intervertebral motion in the lumbar spine of 75 asymptomatic individuals (from an ongoing study by the authors of this chapter). Tables 10–1 and 10–2 provide the 95% confidence intervals for the proportion of total motion that should be provided by each level in the cervical and lumbar spine. This last criterion is only strictly valid in the absence of any fused levels. It must also be noted that the data in Tables 10–1 and 10–2 are for a population of people who have never had any significant spine-related symptoms that required medical treatment. Consistent with many other scientific studies, there were cases in these asymptomatic populations where there were radiographic abnormalities. Preliminary analysis of data from an ongoing study of the lumbar spine suggests that if only radiographically pristine levels are included, the standard deviations are substantially less than reported in Tables 10–1 and 10–2. Not all motion-preserving treatments have the goal of providing for normal intervertebral motion. Some devices are intended to provide stability to an unstable segment while providing for enough motion to help minimize consequences to adjacent segments. For these devices, the success criteria are less clear. With respect to the goal of providing stability, this implies motion within normal limits, and these limits can be obtained from Tables 10–1 and 10–2. However, there are no data to conclusively define the lower limit of “sufficient” motion. Some amount of motion may be needed to prevent a spontaneous fusion, but the amount of motion required for this goal has not been scientifically proven. Since the asymptomatic subjects who have been studied in the cervical22 and ongoing lumbar studies clearly have not fused any level, the lower limit of the 95% confidence intervals
CHAPTER 10
Quantitative Motion Analysis (QMA) of Motion-Preserving and Fusion Technologies for the Spine
described in Tables 10–1 and 10–2 provide a reasonable estimate of sufficient motion. Some amount of motion is also hypothesized to be needed to minimize the effects on adjacent levels, but, again, the range of motion is not known. Once again, the lower limits for the 95% confidence intervals in Tables 10–1 and 10–2 provide a reasonable estimate of the motion required to avoid adjacent level effects. In addition to the quantity of motion criteria, it is also assumed that the quality of motion must be preserved. The most commonly used measure of motion quality is the center of rotation, and the 95% confidence intervals for the location of the center of rotation between flexion and extension are provided in Tables 10–1 and 10–2. Finally, a comprehensive assessment of intervertebral motion should also include assessment of measurements typically made from x-rays of the spine in a neutral position. Currently, a range of criteria are used in research studies. Typical examples that would be applied at each intervertebral level would be requiring that the disc angle (lordosis or kyphosis) changes less than 10 degrees from pre- to postoperative, and requiring that postoperative changes in disc height remain less than 33% of the height of an adjacent normal disc. However, it is important to note that these criteria have not been validated to be predictive of clinical outcomes. Migration over time of a motion-preserving device relative to the vertebrae, subsidence of a device into a vertebra, and screw loosening with posterior stabilization systems are additional measurements made from neutral x-rays that can complement intervertebral motion measurements. Validated success criteria for these measurements have not been established. CONCLUSIONS The technology to accurately and reliably measure both the quantity and quality of intervertebral motion from imaging studies of the spine is readily available and gaining widespread acceptance in research studies. Comparison of results between studies will be greatly facilitated by consistent use of validated methods. Even with accurate and reliable intervertebral motion measurements, the true fusion or pseudoarthrosis rates, the true type and frequency of adjacent segment changes, and the actual range of motion provided by motion-preserving technologies will only be known if highquality imaging studies are used. Sufficient patient effort is essential. Substantial data are available in the literature to assess the quantity and quality of intervertebral motion, both pre- and postoperatively. With adoption of quality control standards and validated measurement technologies, Quantitative Motion Analysis (QMA) of the spine can be a valuable tool in research studies and clinical investigations. In particular, high-quality imaging studies will be essential to proving whether motion-preserving technology can reduce adjacent segment changes. The true value of QMA in routine clinical practice will likely be established once treatment algorithms are validated that require these measurements. REFERENCES 1. Hipp JA, Reitman CA, Wharton N: Defining pseudoarthrosis in the cervical spine with differing motion thresholds. Spine 30(2):209–210, 2005.
95
2. Santos ER, Goss DG, Morcom RK, Fraser RD: Radiologic assessment of interbody fusion using carbon fiber cages. Spine 28 (10):997–1001, 2003. 3. Ghiselli G, Jatana S, Wong D, Wharton N: Pseudarthrosis in the cervical spine: What is the gold standard? CT scan vs. flexion extension quantitative motion analysis with intraoperative correlation. Spine J 6:60S–61S, 2006. 4. McAfee PC, Boden SD, Brantigan JW, et al: Symposium: A critical discrepancy—a criteria of successful arthrodesis following interbody spinal fusions. Spine 26(3):20–34, 2001. 5. Fassett DR, Apfelbaum RI, Hipp JA: Comparison of fusion assessment techniques: Computer-assisted versus manual measurement. Cervical Spine Research Society Proceedings of the 33rd Annual Meeting, 162, 2005. 6. Reitman CA, Hipp JA, Nguyen L, Esses SI: Changes in segmental intervertebral motion adjacent to cervical arthrodesis: A prospective study. Spine 29(11):E211–E226, 2004. 7. White AA III, Panjabi MM: Clinical Biomechanics of the spine, 2nd ed. Philadelphia, J.B. Lippincott, 1990. 8. Takayanagi K, Takahashi K, Yamagata M: Using cineradiography for continuous dynamic-motion analysis of the lumbar spine. Spine 26 (17):858–865, 2001. 9. Pearcy MJ, Bogduk N: Instantaneous axes of rotation of the lumbar intervertebral joints. Spine 13(9):33–41, 1988. 10. Schneider G, Pearcy MJ, Bogduk N: Abnormal motion in spondylolytic spondylolisthesis. Spine 30(10):159–164, 2005. 11. Amevo B, Aprill C, Bogduk N: Abnormal instantaneous axes of rotation in patients with neck pain. Spine 17(7):48–56, 1992. 12. Ng HW, Teo EC, Lee KK, Qiu TX: Finite element analysis of cervical spinal instability under physiologic loading. J Spinal Disord Tech 16:15–65, 2003. 13. Herrmann AM, Geisler FH: A new computer-aided technique for analysis of lateral cervical radiographs in post-operative patients with degenerative disease. Spine 29(16):795–803, 2004. 14. Harvey SB, Hukins DW: Measurement of lumbar spinal flexionextension kinematics from lateral radiographs: Simulation of the effects of out-of-plane movement and errors in reference point placement. Med Eng Phys 20:603–609, 1998. 15. Zhao KD, Yang C, Zhao C, et al: Assessment of noninvasive intervertebral motion measurements in the lumbar spine. J Biomechanics 38(9):943–946, 2005. 16. Penning L, Irwan R, Oudkerk M: Measurement of angular and linear segmental lumbar spine flexion-extension motion by means of image registration. Eur Spine J 14(2):63–70, 2005. 17. Lim MR, Girardi FP, Zhang K, et al: Measurement of total disc replacement radiographic range of motion: a comparison of two techniques. J Spinal Disord Tech 18(3):52–56, 2005. 18. Cakir B, Richter M, Puhl W, Schmidt R: Reliability of motion measurements after total disc replacement: The spike and the fin method. Eur Spine J 15(2):165–173, 2006. 19. Breen AC, Muggleton JM, Mellor FE: An objective spinal motion imaging assessment (OSMIA): Reliability, accuracy and exposure data. BMC Musculoskeletal Disorders 7(7):1–10, 2006. 20. Teyhen DS, Flynn TW, Bovik AC, Abraham LD: A new technique for digital fluoroscopic video assessment of sagittal plane lumbar spine motion. Spine 30(14):E406–E413, 2005. 21. Frobin W, Leivseth G, Biggemann M, Brinckmann P: Sagittal plane segmental motion of the cervical spine: A new precision measurement protocol and normal motion data of healthy adults. Clin Biomech 17(1):1–31, 2002. 22. Reitman CA, Mauro KM, Nguyen L: Intervertebral motion between flexion and extension in asymptomatic individuals. Spine 24:2832–2843, 2004. 23. Dvorak J, Panjabi MM, Grob D, et al: Clinical validation of functional flexion/extension radiographs of the cervical spine. Spine 18(1):20–27, 1993. 24. Champain S, Benchikh K, Nogier A, et al: Validation of new clinical quantitative analysis software applicable in spine orthopaedic studies. Eur Spine J 15(6):82–91, 2006.
96
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
25. Miyasaka K, Ohmori K, Suzuki K, Inoue H: Radiographic analysis of lumbar motion in relation to lumbosacral stability: Investigation of moderate and maximum motion. Spine 25(6):32–37, 2000. 26. Kanayama M, Abumi K, Kaneda K, et al: Phase lag of the intersegmental motion in flexion-extension of the lumbar and lumbosacral spine: An in vivo study. Spine 21(12):416–422, 1996. 27. Harada M, Abumi K, Ito M, Kaneda K: Cineradiographic motion analysis of normal lumbar spine during forward and backward flexion. Spine 25(15):932–937, 2000. 28. Dvorak J, Froehlich D, Penning L, et al: Functional radiographic diagnosis of the cervical spine: Flexion/extension. Spine 13(7):48–55, 1988. 29. Panjabi MM, Crisco JJ, Vasavada A, et al: Mechanical properties of the human cervical spine as shown by three-dimensional load-displacement curves. Spine 26(24):692–700, 2001. 30. Andersson BGJ, Cocchiarella L: Guides to the Evaluation of Permanent Impairment, 5th ed. Chicago: American Medical Association, 2002. 31. Panjabi M, Oxland T, Yamamoto I, Crisco J: Mechanical behavior of the human lumbar and lumbosacral spine as shown by three-dimensional load-displacement curves. J Bone Joint Surg 76-A:413–423, 1994. 32. McGregor AH, Anderton L, Gedroyc WM, et al: The use of interventional open MRI to assess the kinematics of the lumbar spine in patients with spondylolisthesis. Spine 27(14):582–586, 2002. 33. Lin RM, Tsai KH, Chu LP, Chang PQ: Characteristics of sagittal vertebral alignment in flexion determined by dynamic radiographs of the cervical spine. Spine 1:26(3):56–61, 2001. 34. Holmes A, Wang C, Han ZH, Dang GT: The range and nature of flexion-extension motion in the cervical spine. Spine 19(22):505–510, 1994. 35. Amevo B, Worth D, Bogduk N: Instantaneous axes of rotation of the typical cervical motion segments:—A study in normal volunteers. Clin Biomech 6(2):11–17, 1991.
36. Amevo B, Worth D, Bogduk N: Instantaneous axes of rotation of the typical cervical motion segments, 2: Optimization of technical errors. Clin Biomech 6(1):8–46, 1991. 37. Panjabi MM, Goel VK, Walter SD, Schick S: Errors in the center and angle of rotation of a joint: An experimental study. Journal of Biomechanical Engineering 104(3):32–37, 1982. 38. Wharton ND, Hipp JA: Intervertebral motion in the asymptomatic cervical spine during lateral bending. Spine Arthroplasty Society, Annual Meeting, 2006. 39. Crawford NR, Brantley AGU, Baek S: Three-Dimensional Cervical Axis of Rotation During Lateral Bending. Spine Arthroplasty Society Annual Meeting, 2007. 40. Kettler A, Marin F, Sattelmayer G, et al: Finite helical axes of motion are a useful tool to describe the three-dimensional in vitro kinematics of the intact, injured and stabilised spine. Eur Spine J 13(6):53–59, 2004. 41. Dvorak J, Panjabi MM, Chang DG, et al: Functional radiographic diagnosis of the lumbar spine: Flexion-extension and lateral bending. Spine 16(5):62–71, 1991. 42. Ishii T, Mukai Y, Hosono N, et al: Kinematics of the cervical spine in lateral bending: In vivo three-dimensional analysis. Spine 31 (2):55–60, 2006. 43. Pearcy MJ, Tibrewal SB: Axial rotation and lateral bending in the normal lumbar spine measured by three-dimensional radiography. Spine 9(6):82–87, 1984. 44. Plamondon A, Gagnon M, Maurais G: Application of a stereoradiographic method for the study of intervertebral motion. Spine 13 (9):27–32, 1988.
CHAPTER
11
Invasive Diagnostic Tools Ashish Sahai and Todd Alamin
K E Y l
l
l
l
l
P O I N T S
Invasive diagnostic tests are designed to gain physiologic information about pain that is lacking at the conclusion of an anatomic work-up. The key goal of an invasive diagnostic test is to link a patient's complaint of pain with a specific anatomic lesion. A diagnostic tool is useful to the extent that its use improves patient outcomes. Invasive diagnostic tests in the spine are difficult to analyze because they are typically directed at disease processes for which a gold standard for diagnosis does not exist. It is hoped that improvements in our ability as spinal clinicians to confirm diagnoses with certainty will result in both improved outcomes of treatment and a better ability to differentiate among newer treatment modalities for these conditions.
INTRODUCTION As the options for treatment of spinal disease continue to expand, the importance of knowing with precision the origin of pain or the location of the “pain generator” becomes increasingly important. In a healthy patient without other complicating medical or psychosocial issues who develops an acute complaint of leg pain in a specific dermatome with associated motor loss, a sensory deficit, and minimal back pain, the diagnostic connection with the disc herniation seen on the magnetic resonance imaging (MRI) scan is typically clear. The diagnosis is much less clear when confounding factors are present: complex medical issues, such as chronic pain, secondary pain issues, and psychological comorbidities. The chief complaint itself can also be much less specific as well: axial back pain, pelvic pain, or isolated buttock pain. In such common clinical scenarios, it may be extremely difficult to know with certainty the diagnostic significance of the anatomic lesions noted on imaging studies, and the clinician is often left wanting reassurance that a causal relationship between the imaging finding and the clinical complaint truly exists. The ability to interpret the clinical significance of an anatomic lesion noted on an imaging study in the face of a relatively nonspecific clinical complaint hinges on an understanding of the prevalence of the finding in an asymptomatic population with similar demographic features to the patient of concern. Such an
understanding rests upon studies of imaging including the spine performed for clinical concerns other than spinal pathology (e.g., KUB films in patients with abdominal pain, or chest x-ray studies in patients with pulmonary issues), or alternatively on imaging studies in asymptomatic subjects. In a study examining CT scans in asymptomatic patients, Wiesel et al found disc herniations identified in 19.5% of the patients under 40 years old. In the group older than 40 years of age, there was an average of 50% abnormal findings, with diagnoses of disc herniation, facet degeneration, and stenosis occurring most frequently.1 Boden's classic articles evaluating cervical and lumbar spine MRI scans demonstrated that degenerative findings are common in asymptomatic subjects, and become increasingly more common as the subject's age increases (Table 11–1 and Fig. 11–1). To further understand and confirm the causal relationship between the anatomic lesions seen on imaging studies and the patient's clinical complaint, many different invasive diagnostic tools have been developed. These tools have been studied, but in many cases, they have been used to address clinical problems for which a gold standard for the diagnosis does not exist. The most notable and prevalent of these conditions is axial spinal pain. This fact has made it impossible to establish with certainty the specificity and sensitivity of these diagnostic tests in these conditions, and has led to much controversy surrounding them. Furthermore, these tools involve a subjective response on the part of the patient, but also involve in many cases a physiologic response as well; it is this interconnection with the subjective experience of pain and with physiology that makes these studies typically compelling to the involved clinician and patient, and hard to examine objectively. This inherent subjectivity may also make these tests problematic or less compelling from the point of view of the external reviewer. With regard to the cause of axial back pain, it is easiest to group these causes anatomically. Based on several large studies, many authors have concluded that most chronic axial pain complaints originate at least in part in either the intervertebral disc (IVD) or the facet joints. There are some clues to the origin of pain on history and physical examination, but the sensitivity and specificity of these findings are not known. Classically, in helping to differentiate the source of pain by history, flexion increases 97
98
P A R T
Introduction to Motion Preservation Surgery of the Spine
I
TABLE 11–1. Abnormal Cervical Disc MRI Findings in Asymptomatic Subjects Age
Any Major Abnormality (%)
Herniated Disc (%)
Bulging Disc (%)
Degenerated Disc (%)
< 40 > 40
14 28
10 5
0 3
25 60
MRI, magnetic resonance imaging. Modified from Boden SD, McCowin PR, Davis DO, et al: Abnormal MRI scans of the cervical spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am 72:1178–1184, 1990.
loads on the disc, and therefore pain with flexion typically correlates with pain that arises in the disc. Extension loads the facet joints, and so pain that predominantly occurs with extension implicates the facet joints as a causative factor. Several studies have attempted to estimate the sensitivity and specificity of physical findings in predicting the origin of the complaint. These studies, however, have found only weak correlation between lateral flexion and rotation and back pain.2–4 Intervertebral Disc The IVD has been the major focus of back pain research for decades. Specifically in the degenerative spine, disc degeneration is believed to precede all other degenerative changes. The disc is the major anterior load-bearing element in axial compression and flexion. The primary load transmission in the normal disc is through the hydrostatic couple of the nucleus pulposus and the annulus. Axial load is transmitted from the end plate of the vertebra above to the nucleus, which is contained by the annulus. As the nucleus is loaded, it generates a radially directed force onto the annulus. Hoop stresses are generated and resisted in tension by the fibers of the annulus; this resistance allows the annulus/ nucleus complex to support the axial loads borne by it. The annulus is well suited to resisting torsion as well by the alternating 30degree lamellar arrangement of its fibers. The annulus (specifically, the outer third) is the most innervated aspect of the disc. The innervation is primarily the sinovertebral nerve from the dorsal
Percentage of patients with findings in at least one disc level
100
80
93
HNP Stenosis Bulging disc Degenerated disc
60
79
59
56 50
40
34 21
36 22
21
20 1 0 20–39
0 40–59
60–80
Age (years) n
F I G U R E 11–1. Abnormal magnetic resonance imaging scans
of the lumbar spine in asymptomatic subjects. (Adapted from Boden SD, Davis DO, Dina TS, et al: Abnormal MRI scans of the lumbar spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am 72:403–408, 1990.)
primary ramus.5,6 The second portion that is innervated and likely involved in pain perception in the painful degenerated disc is the end plate. Recent work has demonstrated that the end plate is innervated by the sinovertebral nerve; this pathway is thought to be a major source of nociceptive afferent flow from the disc.7 Facet Facet joint-mediated pain is a significant problem in patients with chronic low back pain. Its prevalence has been estimated at 15% in younger patients with chronic back pain, and at 53% of older patients.8–10 It has been well known that facet joints can be a significant source of pain as evident by isolated fractures, infections, and localized disease process affecting the facets. Each facet joint receives dual innervation from the medial branches of the primary dorsal ramus from the same level and segment above. The medial branch runs caudally and dorsally, lying against the bone where the transverse and superior articular processes join. The load on the facets varies between 3% and 25% of the total axial load.11 This load increases significantly with degenerative condition of the spine, including disc space narrowing and facet arthritis, to 80%. The clinical complaints associated with facet joint pain are nonspecific. Patients typically present with a deep, aching pain in the low back and buttock with occasional radiation into the posterior thighs. The pain is often nondermatomal in distribution as well.12 MECHANISM OF DIAGNOSTIC INJECTIONS Invasive diagnostic techniques work in two potential ways. The first is provocation, in which the testing method aims to reproduce the patient's primary complaint through a localized stimulus. The classic example of this is provocative discography. The logic of a provocative test is as follows: If a precisely localized stimulus can reliably reproduce a patient's symptoms, then it is much more likely that the locus of the stimulus is the site of the origin of the patient's pain. The provocation technique relies upon the ability of the patient to reliably recognize and differentiate amongst different potential sources of pain, and also upon the ability of the stimulus to re-create the patient's pain in a non-physiologic manner. The second is anesthetic, in which the effect of the localized infusion of anesthetic on the patient's primary complaint is assessed. This anesthetic technique depends on the ability of the clinician to both accurately place and contain the anesthetic, and on the patient's ability to reliably elicit his typical pain (both before and after the infusion of anesthetic so that the response can be assessed). The typical concern involving provocative tests regards the specificity of this sort of test; anesthetic tests are more commonly criticized with regard to their sensitivity (their specificity may also be called into doubt if the anesthetic is not contained).
CHAPTER 11
Selective Nerve Root Block The selective nerve root block (SNRB) is a useful invasive diagnostic test that can be used in situations in which the source of a patient's radiating pain is not clear but is believed to be radicular in origin. There are many different potential causes of pain radiating into the extremities, including radicular pain, large joint arthritis, sacroiliac pain, and facet-mediated pain. In the face of an equivocal clinical examination and history with moderate compressive findings on noninvasive imaging studies such as MRI, the clinician may be uncertain about the relationship of the spinal imaging findings and the patient's clinical complaint of pain. In such cases, a test to further confirm the significance of the imaging findings can be extremely useful. The terms SNRB and transforaminal block are often used interchangeably, but in the strictest sense, the two differ in two major ways that make the SNRB more of a diagnostic procedure.13,14 The SNRB is distal to the root sleeve, limiting the amount of injectate that may flow into the epidural space. The second is that only local anesthetic is used (no corticosteroid) and in limited quantity, usually less than 1 mL. More than this amount may result in spread to the epidural space, limiting the diagnostic accuracy of the injection. A transforaminal block is placed in the foramen proper, proximal to the end of the root sleeve, and thus often involves epidural spread of injectate (Fig. 11–2).13 Steroid is often also used in a transforaminal block. However, SNRBs remain an invaluable tool both diagnostically and therapeutically. Imaging studies can accurately demonstrate neural compression whether they are in the neural foramen or the thecal sac in the central canal. However, these studies cannot differentiate symptomatic from asymptomatic compression; this is particularly an issue when there are multiple sites and levels of compression that may be symptomatic. SNRBs help to localize
Invasive Diagnostic Tools
99
pain to a specific nerve root, and can also aid in the nonoperative management of radicular pain. Indications
As mentioned earlier, SNRBs have many uses from the diagnostic to the therapeutic. They can aid in the evaluation of atypical extremity pain as well as in evaluating the significance of multiple potential sites of symptomatic compression. Technique
Either a single- or double-needle technique is used to access the intertransverse foramen. Previously, the “safe triangle” of Pauza and Bogduk (Fig. 11–3) was thought to be an ideal location for needle placement15; however, recently this has been brought into question.16 A combination of steroid and anesthetic is subsequently injected. Newer techniques have been developed including direct nerve root stimulation. In this technique, needles are placed into the affected foramen, and into the one above and below. The nerve roots are sequentially electrically stimulated, and provocative responses are noted for concordancy. Local anesthetic is then delivered to the concordant level to assess the anesthetic response, and then steroid is delivered to the root.17 Injection Location
The location of injection has led to significant variability within the practicing community and has been reflected in the many Vertical line at lateral pedicle border
Triangle Transverse line parallel to inferior border of pedicle Hypotenuse tangent to medial curvature of pedicle
n
F I G U R E 11–2. Example of a transforaminal block; note epidural spread of contrast.
n
F I G U R E 11–3. “Safe triangle” described by Pauza and Bogduk.
100
P A R T
I
Introduction to Motion Preservation Surgery of the Spine
approaches to injections and the literature supporting it. Most recently, a retrospective study of more than 1,700 injections hypothesized that needle placement within or adjacent to the intervertebral foramen would result in similar immediate pain reduction to the theoretical “ideal” position or “safe triangle” of Pauza and Bogduk.15 The presumed ideal position is the safest position next to the nerve root. The foramen was divided into quadrants in both the anteroposterior (AP) and lateral views, and multiple needle positions in the quadrants were determined. The results demonstrated that the needle tip position did not significantly affect immediate pain reduction.16 Diagnostic Value
SNRBs can also be beneficial in predicting the outcome of surgery especially in combination with an imaging test. A comparison of surgical outcomes between MRI and SNRB 1 year postoperatively was conducted for cervical or lumbar radiculopathy. Ninety-one percent of patients with a positive SNRB had good surgical outcomes, whereas 60% with a negative SNRB had good outcomes. This was differentiated from those in which a positive MRI revealed 87% good surgical outcomes, and a negative MRI had surprisingly similar outcomes of 85%. A positive MRI was considered one in which there was clear evidence of neural impingement or severe central/lateral recess/foraminal stenosis, whereas a negative MRI had only mild-to-moderate stenosis, no clear evidence of impingement or equivocal findings. When the findings differed between selective nerve root injection (SNI) and MRI, surgery at a level consistent with the SNI was more strongly associated with a good surgical outcome. Of the patients with a poor surgical outcome, surgery was most often performed at a level inconsistent with the SNI finding. The authors went on to state the SNI was found to be most helpful when the MRI results were unclear or unequivocal.18 A recent systematic review graded all the recent studies on selective nerve root blocks based on strength of evidence as well as methodologic quality evaluation. Datta et al19 reviewed the current literature to assess the accuracy of SNRBs in diagnosing spinal pain. Of the initial 336 studies, they narrowed down the studies using the Agency for Healthcare Research and Quality (AHRQ) and Quality Assessment Studies of Diagnostic Accuracy (QUADAS) criteria. Studies were graded and evidence classified into five levels: conclusive, strong, moderate, limited, and indeterminate, and could be either prospective or retrospective. The conclusion of the study was that there is still limited evidence on the effectiveness of selective nerve root injections as a diagnostic tool for spinal pain. The major concern regarded the potential for significant epidural spread despite careful imaging; this likely limits its diagnostic utility.19 Castro et al showed epidural spread in 48% and spread to an adjacent nerve root in 27% of cases with their lowest injected volume (i.e., 0.5 mL).20 Wolff et al used a combination of fluoroscopy and electrostimulation to perform the selective nerve root block, but still found epidural spread in 47% of L4 and 28% of L5 blocks and spread to adjacent nerve roots in 5%.21 The authors went on to state that with the available literature there is moderate evidence for the use of SNRBs in the preoperative evaluation of patients with negative or inconclusive imaging studies.21
Schutz et al22 reported finding a corroborative lesion at the time of surgery in 87% of patients with a positive diagnostic block. Dooley et al23 reported a specificity of 94%. Van Akkerveeken24 attempted to establish the diagnostic value of selective nerve root injections by comparing 37 patients with confirmed lumbar radiculopathy to nine patients with pain due to metastases. With these small numbers, the calculated sensitivity for determining pain of spinal neural origin was 100%.24 The specificity was studied by comparison to a normal level on imaging and examination with an SNRB, and it was 90%.24 Of the 37 patients with lumbar radiculopathy, some declined surgery. The predictive value for a good outcome was determined with, and without, patients who did not want surgery. If all patients who declined surgery were included in the analysis as surgical failures, the positive predictive value of a good surgical outcome with a positive SNRB was 70%. The positive predictive value was 95% when patients who had surgery were the only ones included in the analysis. In this study, the authors concluded that SNRBs were a highly sensitive, specific test with high predictive value for surgical outcome.24 In a study in 2006, Anderberg et al25 concluded that for a block to be truly selective enough for diagnostic investigations, only 0.6 mL of total injectate is acceptable. However, these high levels of specificity and sensitivity have to be interpreted with caution because they have not been repeated in controlled trials. These results were compared with evidence shown by North et al in a prospective, randomized single-blinded study that nerve root blocks have sensitivities between 9% and 42% and a specificity of 24%.26 Therapeutic Value
Not only do SNRBs have a role in determining whether a patient is a surgical candidate, but for certain types of pathology, they can also be therapeutic. Riew et al27 in a prospective, randomized, controlled, double-blinded study looked at the effect of nerve root blocks on the need for operative treatment of lumbar radicular pain. They examined operative candidates who avoided surgery for a minimum of 5 years after receiving an SNRB with either bupivacaine or combination of steroid and bupivacaine. Of the 55 patients, 29 avoided an operation in the original study at a follow-up of 13 to 28 months. Twenty-one of 29 patients were reevaluated, and 17 of the 21 still did not have surgery at 5 years. There was no difference between the groups treated with steroid or those treated with bupivacaine and steroid despite the initial belief that steroid provided a better effect at the initial study with a minimum 1-year follow-up. In a disease process that has a favorable natural history, an intervention that affords relief for some period of time may allow the patient to tolerate their pathology until the natural history of spontaneous improvement occurs. Complications
Complication rates were reviewed in a large study of more than 2,000 injections. The similar needle tip positions were determined using the similar quadrant method described earlier in the needle location section. An overall 5.5% complication rate of minor transient complications, such as transient increased pain, transient leg weakness, light-headedness, vasovagal response, dural puncture and injection into the subarachnoid space that usually abated in 24 hours, was
CHAPTER 11
reported. However, the authors did not find that multiple injections rendered a higher complication rate than did single injections.28 Facet Blocks No independent gold standard exists to confirm a diagnosis of facet pain. The symptoms associated with facet pain are relatively nonspecific. In an attempt to better map pain-referral patterns, a map of facet-referred pain was developed by injection of hypertonic saline into facet joints of normal research subjects by Mooney and others. Wide overlap was demonstrated between the referral patterns of lumbar facet joints.12 A larger study demonstrated a wide range of referred pain with no consistent pattern of referral. These results have been reproduced in several subsequent studies (Fig. 11–4).12,29,30 The facet joints capsules are densely innervated with nociceptors that fire when the capsule is stretched or subjected to compressive force. In both patients with pain and volunteers, chemical or mechanical stimulation of the facets has been shown to elicit back and/or leg pain.29–31
Anterior n
Posterior
F I G U R E 11–4. Pain referral patterns from the lumbar facet
joints. In descending order, the most common referral patterns extend from the darkest (low back) to the lightest regions (flank and foot). The key at the bottom of the figure legend is listed in order of affected frequency (i.e., low back to foot). The facet levels next to each location represent the zygapophyseal joints associated with pain in each region. Low back: L5-S1, L4-L5, L3-L4; Buttock: L5-S1, L4-L5, L3-L4; Lateral thigh: L5-S1, L4-L5, L3-L4, L2-L3; Posterior thigh: L5-S1, L4-L5, L3-L4; Greater trochanter:L5-S1, L4-L5, L3-L4, L2-L3; Groin: L5-S1, L4-L5, L3-L4, L2-L3, L1-L2; Anterior thigh: L5-S1, L4-L5, L3-L4; Lateral lower leg: L5-S1, L4-L5, L3-L4; Upper back: L3-L4, L2-L3, L1-L2; Flank: L1-L2, L2-L3; Foot: L5-S1, L4-L5. (Data adapted from McCall et al., Marks, and Fukui et al. Adapted from Cohen SP, Raja SN: Pathogenesis, diagnosis, and treatment of lumbar zygapophysial (facet) joint pain. Anesthesiology 106:591–614, 2007.)
Invasive Diagnostic Tools
101
The ability to correlate a diagnostic facet block to specific findings on clinical history and physical examination is limited. In 1988, Helbig and Lee32 coined the term lumbar facet syndrome based on a retrospective study. Patients who responded to intra-articular facet injections (the full parameters of the injection were not included) were more likely to have back pain associated with groin/thigh pain, paraspinal tenderness. and reproduction of pain during extension-rotation maneuvers. Pain radiating distal to the knee was not associated with a positive response to facet blocks. However, since the early description, multiple larger studies have been unable to reproduce the consistent associations found in this study. Technique
The two commonly accepted techniques for performing a facet joint block are the facet intra-articular block and the medial branch block (MBB). Several studies have demonstrated that both blocks have similar efficacy rates.33,34 However, each has its particular problems. An intra-articular block would appear to be the more logical choice in diagnosing facet joint pain. From a technical standpoint, an intra-articular facet block involves advancing a needle into the facet joint, confirming the position via fluoroscopy and an injection of contrast, followed by injection of either anesthetic or a combination of anesthetic and steroid. Unfortunately, owing to its small size and complex anatomy, the joint is often difficult to enter; once accessed, it is difficult to ensure that the injected anesthetic stays within the joint. An injection of between 1 and 2 mL into the facet often causes the capsule to rupture, allowing the injectate to extravasate and potentially anesthetize adjacent structures. The other approach to anesthetizing a facet joint involves anesthetizing the nociceptive afferents of the joint. The MBB has the potential to achieve this; however, it, too, has significant limitations. To perform this block, it is essential to understand the anatomy of the innervation to the lumbar zygapophyseal (l-z) joint. For example, the L4-L5 l-z joint is innervated by the medial branches of both L4 and L3. Each medial branch traverses the transverse process of the vertebra below. Therefore, the L4 medial branch will cross the L5 transverse process. An MBB is further complicated by the fact that a small amount of local anesthetic has the potential to anesthetize not only the immediate facet by blocking the medial branch but also the intermediate and lateral branches of the nerve arising from the primary dorsal ramus. This could result in a false-positive response through a block of the nociceptive afferents of the surrounding paraspinal musculature, fascia, periosteum of the neural arch, and the overlying skin. The actual block is performed by advancing the needle to the upper third of the groove formed by the superior articular process and the transverse process (Fig. 11–5). In order to anesthetize one joint, both medial branches that innervate it must be blocked. Radiographically, the target point is the upper part of the “eye of the scotty dog,” which is seen with the patient in an oblique/prone position. Each nerve is infiltrated with the preferred anesthetic/steroid combination. The injection is performed slowly over approximately 30 seconds. There are only two studies comparing MBB to intra-articular blocks with regard to their therapeutic effectiveness. Marks et al30 randomly assigned 86 axial low back pain (LBP) patients to receive either
102
P A R T
I
Introduction to Motion Preservation Surgery of the Spine n
Ascending branch to facet joint Descending branch to facet joint
F I G U R E 11–5. Right lateral oblique of vertebral body and dorsal rami. (Adapted from Cohen SP, Raja SN: Pathogenesis, diagnosis, and treatment of lumbar zygapophyseal [facet] joint pain. Anesthesiology 106:591–614, 2007.)
Primary dorsal ramus
Lateral branch Intermediate branch Medial branch
intra-articular injections or MBBs. The authors found no difference in the immediate response between the two groups, although the intraarticular group experienced better pain relief at their 1-month follow-up but not the 3-month follow-up. Nash conducted a prospective study in 67 patients with axial LBP who were randomly assigned to receive either MBB or intra-articular injections. In the 26 pairs who completed the study, the MBB was shown to be more beneficial in 12 at the 1-month follow-up: 11 reported the intra-articular injection to be better, and three reported no difference between the two.35 Unfortunately, radiologic imaging has also not been conclusively shown to predict a response to diagnostic l-z joint blocks, but it has been conflicting at best. Some studies have found a positive correlation between CT, MRI, or other imaging studies and response to l-z joint blocks; however, a similar number have not. The largest study, which was conducted by Jackson et al, found no relation between radiographic evidence of l-z joint degeneration and the response to single intra-articular facet injections in 390 patients.36 In contrast, Carrera and Williams found that 73% of chronic LBP patients demonstrating CT evidence of lumbar facet disease experienced pain relief after large-volume (2–4 mL) facet blocks compared with only 13% in whom CT scans showed no pathology.37 False-Negative Blocks
When performing these injections, care has to be taken to recognize that the needle is in the correct position and that the possibility of aberrant anatomy can lead to a misleading conclusion. In 18 asymptomatic volunteers, Kaplan et al found that properly performed MBBs result in failure to anesthetize the corresponding
facet joint 11% of the time (as assessed by pain provoked by secondary intra-articular injections of contrast), even with the avoidance of venous uptake.38 They also found that inadvertent venous uptake occurred during 33% of nerve blocks. When the needle was repositioned to avoid venous uptake, analgesia was achieved only 50% of the time.38 The authors concluded that when venous uptake occurs, it may be advantageous to repeat the procedure on a separate occasion rather than redirect the needle to avoid false-negative results. False-Positive Blocks
Numerous studies have documented a high false-positive rate for lumbar facet blocks, ranging from 20% to 40% using comparative blocks or saline controls despite the type of block used (intra-articular or MBB).39,40 The reasons for false-positive facet blocks are multiple including the concern of spread of injectate to pain-generating structures other than those targeted, a placebo response to diagnostic facet interventions, and the concomitant use of other pain-altering interventions—either opioids or local/topical anesthetic. DISCOGRAPHY Pain originating from degenerated or injured discs may be the most commonly implicated cause of disabling chronic low back pain. Discography was described nearly 60 years ago as a method to radiographically demonstrate intervertebral disc herniations. Lindblom41 and Hirsch31 reported this method of identifying herniated discs in the lumbar spine through an intradiscal injection of contrast, with subsequent radiographs demonstrating the contrast extending
CHAPTER 11
around disc material into the epidural space. The authors noted in the report that a reproduction of the patient's typical low back pain was common during the injection. After this initial note of back pain as a side effect of the disc injection, reproduction of the patient's complaint of back pain rapidly became the predominant goal of the test. Discography has since become a standard method of attempting to correlate degenerative imaging findings with pain. It is used most commonly in the lumbar spine, but it is also in the cervical and occasionally in the thoracic spine. It is hoped that when discography is used, it can better determine that the disc is the predominant source of pain, and as such is sufficient to account for most of the patient's clinical complaints of pain. The extent to which provocative discography achieves these diagnostic goals determines the extent to which it is a clinically useful tool. This is, in the end, a heavy burden that is carried by a presurgical diagnostic test, as interventional treatment strategies are often based on its results. If the results of the test are not reliable, then even a 100% effective therapy that is based on the test is likely to fail at a rate that is at best the false-positive rate of the method of diagnosis (in this case, provocative discography), minus any nonspecific or placebo effects. It may be that at this stage of our ability to treat discogenic pain that precision in the presurgical diagnosis may affect overall patient outcomes more than variations in the surgical treatment strategy (e.g., fusion versus disc replacement). Indications The most common indication for discography is to further confirm an intradiscal source of pain in a patient in whom an invasive procedure is being considered. In 1995, the North American Spine Society (NASS) issued a position statement on discography recommending a minimum of 4 months' duration of symptoms before consideration of the procedure.42 Technique (Cervical Spine) The cervical disc is accessed from a right anterior approach to avoid the esophagus while the trachea and esophagus are pressed left of midline to keep them from being injured. A double-needle technique is used with an outer 18- or 20-gauge needle and inner 22- or 25gauge needle to access the disc itself. Contrast is injected with pressure recorded on initial dye flow, and then during the subsequent injection of contrast. The patient's pain intensity, location, and character are recorded throughout the process. When significant pain is elicited by the injection, the patient is asked to assess the degree of concordancy of the pain experienced during the injection with the patient's usual pain. A rating of “similar” or “exact” is typically believed to be significant. One or two discs around the suspected painful disc are usually injected to help serve as the “control.” Technique (Lumbar Spine) 41
The first reported approach for lumbar discography by Lindblom and Hirsch31 was an interlaminar, transdural approach to the disc. Over time. an oblique, extrapedicular approach was developed that minimized the concurrent damage to neural elements. With this approach, the exiting nerve root is the structure most often irritated— if significant leg pain is encountered during insertion of the needle,
Invasive Diagnostic Tools
103
it is repositioned. The double-needle technique similar to that of the cervical spine is used. In the lumbar spine, the approach is posterolateral instead of the anterolateral approach used to access discs of the cervical spine. Again, a control disc is typically analyzed as well. Criteria for a Positive Result After performing the test, there are several different elements that are used to interpret the result. The first is an assessment of the technique of the study: dye was injected into the nucleus and not the annulus, the patient was not overly sedated, and no other technical difficulties were noted. The second is whether by imaging studies done after contrast injection, there is some anatomic evidence of an annular disruption. The third element used to interpret the results of the study is the patient's response to the injection—the pressure at which the pain is noted, its rating on a VAS scale for pain, and its character. This response is an inherently subjective one; it is assessed by whether or not the criteria for a positive test are satisfied. Some evidence of pain behavior (grimacing, verbalization) is also important. The criteria for a positive test are not universally accepted; the most widely accepted criteria are 1. Low-pressure injection ( 0.05). Average flexion-extension motion went from 9 degrees preoperatively to about 1 degree (essentially no motion) at more than 24 months postoperatively in the fusion group but was well preserved at about 12 degrees in the disc replacement group. Side bending went from 6 degrees to less than 2 degrees (essentially no motion) in the fusion group versus about 5 degrees in disc replacement patients. There were no major complications, technique or device-related, in any of the cases. Figure 25–6 represents the sagittal angular motion measured radiographically at the operated segment. As expected, motion is effectively eliminated at fusion levels, whereas angular motion is well preserved with the prosthetic discs. So the conclusion of the U.S. IDE trial was essentially that ACDR preserves range of motion without compromising the results as compared with the current surgical standard of ACDF. It is hoped that in the long run, the preserved motion will decrease adjacent segment degeneration. OPERATIVE TECHNIQUE A standard anterior approach to the cervical spine is performed. Any operating table that allows supine positioning and fluoroscopy of the neck can be used. No external traction of the spine is necessary. A transverse skin incision is made over the level being operated, after localizing either by anatomic landmarks or with a lateral radiographic image (a fluoroscopy machine is used for the duration of instrumentation with the ProDisc-C). After incising through the skin, subcutaneous fascia, platysma muscle, and superficial layer of the deep cervical fascia, blunt dissection is performed first between the strap muscles medially and the sternocleidomastoid laterally then the tracheoesophageal bundle medially and the carotid sheath laterally. The prevertebral fascia is then split to expose the disc space. Before the discectomy, another localizing radiograph is performed.
TABLE 25–2. Patient Demographics Patient Characteristic Average age (years) Gender (% male: % female) Body mass index (BMI) Workman's compensation status Preoperative duration of neck pain (months) Active smokers
ACDF (allograft þ plate) (n ¼ 15)
ACDR (ProDisc-C) (n ¼ 33)
P-Value (NS is > 0.05)
42.5 33:67 24.9 27% 10.5
40.1 20:80 23.7 20% 10.3
NS NS NS NS NS
1 (6.6%)
4 (16.7%)
NS
ACDF, anterior cervical discectomy and fusion; ACDR, anterior cervical discectomy replacement.
CHAPTER 25 n
F I G U R E 25–3. Visual Analog Scale (VAS)
ProDisc-C Total Cervical Disc Replacement
217
10
neck pain scores.
ProDisc-C Fusion
VAS score (10 cm scale)
9 8 7 6 5 4 3 2 1 0 Pre op
6 wk
3 mo
6 mo
12 mo
24–36 mo
12 mo
24–36 mo
12 mo
24–36 mo
Assessment interval
n
F I G U R E 25–4. Visual Analog Scale (VAS)
10
arm pain scores.
ProDisc-C Fusion
VAS score (10 cm scale)
9 8 7 6 5 4 3 2 1 0 Pre op
6 wk
3 mo
6 mo
Assessment interval
n
F I G U R E 25–5. Oswestry Disability Index
100
(ODI) scores.
Oswestry score (100 pt scale)
90
ProDisc-C Fusion
80 70 60 50 40 30 20 10 0 Pre op
6 wk
3 mo
6 mo
Assessment interval
218
P A R T
I I I
Cervical Total Disc Arthroplasty
16 ProDisc-C ACDF
14 ROM (degrees)
12 10 8 6 4 2 0 Pre op n
6 wk
F I G U R E 25–6.
3 mo
F I G U R E 25–7.
12 mo
24–36 mo
Visit Sagittal angular motion in degrees.
Once the operative level is confirmed, self-retaining retractors are placed mediolaterally and superoinferiorly (Fig. 25–7). The fluoroscopy machine is positioned to allow anteroposterior and lateral imaging during the procedure. Midline in the anteroposterior plane is marked on the vertebral bodies spanning the operative disc, using the fluoroscopy machine. Manual discectomy is then performed. Specialized pin distractors are then placed in the spanning vertebral bodies. These pins are actually fastened onto the distractor using nuts, making this device a rigid fixator (similar to an external fixator) rather than simply a distractor (Fig 25–7 and 25–8). Not only does this provide distraction for easier removal of disc tissue but it also provides rigid stabilization during instrumentation so that the relative alignment of the vertebral bodies is maintained. This prevents excessive jolting movements of the vertebral bodies and neural elements during impacting of the implant, and it also ensures precise and symmetric placement of the device keels within both vertebrae (avoiding any listhesis, see Fig. 25–8). Disc resection is performed entirely manually, with minimal need for end plate preparation (e.g., burring, milling, and so on). The two keels and the porous-coated surfaces of the ProDisc-C
n
6 mo
provide enough initial fixation to obviate any end plate milling or preparation, which removes end plate bone and can risk loss of segmental lordosis and implant subsidence. Occasionally, osteophyte resection or end plate flattening with a bur or Kerrison rongeur may be necessary. The posterior longitudinal ligament may or may not be removed depending on the location of herniated disc or osteophyte. Once the disc space is adequately cleared, trial sizing with the help of fluoroscopy is performed (see Fig. 25–8). The implant size that maximizes end plate coverage is chosen. Appropriate disc height is selected based upon the tightness of fit and the relative heights of the unaffected adjacent segments. Overstuffing or undersizing of the implant can both compromise stability and range of motion. Once the size of the implant is selected, an osteotome is slid over the trial (which acts as a stop) and malleted through the previously marked midline on the vertebrae to create a channel for the keels (see Fig. 25–8). A sharp chisel is followed by a box osteotome to widen the slot for the keels. This prevents excessive stress on the vertebrae during implant placement and also prevents posterior displacement of bony fragments. After the chiseling, the actual implant is then carefully malleted into place
Anterior exposure and discectomy.
ProDisc-C Total Cervical Disc Replacement
CHAPTER 25
n
F I G U R E 25–8. Trialing and chiseling under fluoroscopy.
n
F I G U R E 25–9. Insertion of prosthesis under fluoroscopy.
under fluoroscopic guidance (Fig. 25–9). The fixator, pins, and retractors are then removed, and closure is performed. POSTOPERATIVE CARE A soft neck collar can be used for the first week or two to allow for wound protection. Otherwise there is no extensive postoperative protocol. Patients can return to work as soon as comfortable but should allow 6 weeks before returning to recreational sports or full duty (if the job is physically demanding). TABLE 25–3.
219
COMPLICATIONS AND AVOIDANCE No major technique- or device-related complications were observed. Table 25–3 lists the complications for patients undergoing ACDR and ACDF. There was one revision surgery in each group. No device migration or subsidence requiring further treatment has been seen to date. This may be related to the screening out of osteoporotic patients, who would be high risk for such issues. The incidence of prolonged dysphagia, defined loosely as patientsubjective swallowing difficulty lasting more than 6 weeks, was
Complications From the U.S. IDE Study of ProDisc-C Versus ACDF
Complication
ProDisc-C (%)
ACDF (allograft þ plate) (%)
Implant migration/subsidence Nonunion Revision surgery Prolonged dysphagia Postoperative swallow study, laryngoscopy, or other throat workup Superficial infection Deep infection New transient symptoms*
0/33 NA 1/33 2/33 0/33 1/33 0/33 2/33
0/15 1/15 1/15 1/15 0/15 1/15 0/15 2/15
ACDF, anterior cervical discectomy and fusion; IDE, Investigational Device Exemption. *Subsided by 6 weeks to 3 months after surgery.
(0) (3) (6) (0) (3) (0) (6)
(0) (6.7) (6.7) (6.7) (0) (6.7) (0) (13.4)
220
P A R T
I I I
Cervical Total Disc Arthroplasty n F I G U R E 25–10. Example of multilevel cervical artificial disc replacement, 12 months after surgery.
minimal in either group. Patients claiming dysphagia were still eating a regular diet with solid foods. No swallow studies or any other post operative throat evaluations were needed for further persistent swallowing, breathing, or vocal difficulties in either study group. Transient new symptoms were observed in two patients each in the ACDF and ACDR groups. These were defined as spontaneous new-onset radiculopathy, numbness, or subjective weakness. All new symptoms subsided by 6 weeks to 3 months after surgery. ADVANTAGES/DISADVANTAGES: ProDisc-C TOTAL CERVICAL DISC REPLACEMENT Advantages Semiconstrained motion is facet protective Good fixation features (keels, coating) Familiar cobalt-chrome and polyethylene materials No device protrusions beyond disc space Class I U.S. FDA clinical data available Multilevel disc replacements possible and clinical data also available (2 and 3 levels) Good experience: more than 6,000 implanted worldwide at time of writing (Aug 2006) Easily revisable (both approach and device extraction with designated instruments)
Disadvantages Greater device-bone interface loading Questionable risk of vertebral fractures (especially during multilevel use) Long-term data (5 to 10 years) from Class I data are still pending Questionable risk of polyethylene debris
FDA, Food and Drug Administration.
CONCLUSIONS/DISCUSSION Our experience with the ProDisc-C artificial cervical disc suggests that cervical disc replacement is a viable surgical alternative to fusion for cervical disc degeneration and herniation, with preservation of motion and alignment at the treated vertebral levels and without
compromising clinical outcomes. Although it is yet too early for the U.S. clinical trials to offer any definite proof of benefit against accelerated adjacent segment degeneration, the fact that normal intervertebral motion is preserved at the treated segment is encouraging. Longer term safety and efficacy studies are in progress. This particular device has been used extensively for multilevel use, with as good or better clinical outcomes as compared with single-level surgeries and with good preservation of motion at each replaced level, good preservation of spinal alignment at each replaced level, and great patient satisfaction scores (Fig. 25–10). The technique and instrumentation are facile and streamlined, but the instruments continue to be refined still. One of the salient improvements has been in the way the keel cuts are made in the vertebrae: A drill with a protective guide has been built to seamlessly drill the keel grooves on the bone, thus minimizing the risk of vertebral splitting fractures, especially when performing multilevel disc replacement. The drill also minimizes malleting on the spine.
REFERENCES 1. Goffin J, Van Calenbergh F, van Loon J, et al: Intermediate follow-up after treatment of degenerative disc disease with the Bryan Cervical Disc Prosthesis: single-level and bi-level. Spine 28:2673–2678, 2003. 2. Wigfield CC, Gill SS, Nelson RJ, et al: The new Frenchay artificial cervical joint: results from a two-year pilot study. Spine 27:2446–2452, 2002. 3. Delamarter RB, Pradhan BB: Indications for cervical spine prostheses, early experiences with ProDisc-C in the USA. Spine Art 1:7–9, 2004. 4. Panjabi M, Duranceau J, Goel V, et al: Cervical human vertebrae: quantitative three-dimensional anatomy of the middle and lower regions. Spine 16:861–869, 1991. 5. Yoganandan N, Lumaresan S, Printat F: Biomechanics of the cervical spine, Part 2: Cervical spine soft tissue responses and biomechanical modeling. Clin Biomech 16:1–27, 2001. 6. White A, Panjabi M: Clinical Biomechanics of the Spine. Philadelphia, J. B. Lippincott Co., 1990, pp 110–111. 7. Pradhan BB, Delamarter RB, Bae HW, et al: 2- to 3-Year Results with the ProDisc-C Cervical Disc Replacement from the US Clinical Trials. North American Spine Society annual meeting, Seattle, WA, September, 2006.
CHAPTER
26
The NeoDisc Elastomeric Cervical Total Disc Replacement Andre Jackowski, Alan McLeod, Christopher Reah, G. Bryan Cornwall, Lukas Eisermann, and Alexander W.L. Turner
K E Y l
l
l
l l
P O I N T S
The surgical technique for the NeoDisc device is less demanding than other cervical total disc replacement designs, in particular requiring less exactness in placing the device relative to the natural center of rotation. The materials used to construct the NeoDisc device are amenable to MRI investigation and do not obstruct x-rays. Biomechanically, the design of the NeoDisc device is closer to the natural cervical disc than a traditional articulating mechanical artificial disc. It will not impose a fixed center or arc of rotation on the disc level and offers an element of shock absorption. The NeoDisc device has very low wear rates in testing. Soft tissue ingrowth has been demonstrated in the NeoDisc device and has the potential to provide a secure long-term fixation.
INTRODUCTION The operation of anterior cervical decompression and interbody fusion has been a widely accepted surgical procedure in the treatment of radiculopathy or myelopathy since the original reports of Cloward in 19581 and Robinson and Smith in 1955.2 Since that time, surgeons have employed these two operative techniques and their variations in the treatment of patients who present with persisting radicular pain and evidence of cervical compressive myelopathy. Two considerations have led several researchers to investigate alternative forms of surgical treatment utilizing nonfusion technology. The first is that there is a small but significant number of patients who suffer complications in the form of either graft site donor pain or graft complications including collapse, expulsion, or nonintegration. The second consideration is the desire to maintain motion at the operated segment together with concern from a number of investigators that motion segment fusion can lead to accelerated adjacent-level degeneration and subsequent, adverse clinical sequelae.3–6 The rationale for developing and implanting a cervical disc prosthesis is that such a device may allow the following: 1. the maintenance of segmental mobility; 2. avoidance of altered adjacent segment mobility;
3. avoidance of or reduction in the incidence of adjacent motion segment disease; 4. the potential for more rapid functional recovery. The design of an elastomeric artificial cervical disc prosthesis is influenced by the fact that the cervical disc is not a lubricated articulating synovial joint but is a deforming fibrocartilaginous joint. Elastomeric materials offer the potential to closely mimic the behavior of a natural intervertebral disc in a way that more traditional ball-and-socket devices cannot. Extensive initial testing was performed on a variety of elastomeric materials in conjunction with enveloping textile jackets and an artificial annulus or ligament in order to find the combination of materials that most closely mimics the natural joint. The design concepts for the NeoDisc are for the device to not only replace the disc but to also replicate the functions of the anterior and posterior longitudinal ligaments. To reduce the learning curve associated with adopting a new technology, the surgeon should not have to radically alter his or her standard practice when doing an anterior cervical discectomy. That is, the ideally designed prosthesis fits a standard surgical defect following a microdiscectomy or Smith-Robinson decompression. There is no requirement for special jigs or templates and no need to mill the end plate or to alter the end plate to fit the device. By realizing these objectives, the operation time would be shortened rather than increased over current practice. Additionally, the ideal device is free of magnetic resonance imaging (MRI) and computed tomography (CT) scan artifact so as to allow for immediate and future patient investigations. To achieve satisfactory integration with the implanted motion segment, the implanted device should achieve the following objectives: 1. allow for normal spinal biomechanics and not impose its own variable instantaneous access and rotation but adapt to the native axis of rotation; 2. not block motion within the patients normal range of motion; 3. allow coupled motions; 4. allow for a shock-absorption function. 221
222
P A R T
I I I
Cervical Total Disc Arthroplasty
To achieve immediate stability, the implant design incorporates fixation flanges that are anchored by traditional screws to the adjacent vertebral bodies. For long-term stability, the enveloping embroidered jacket has a naturally porous three-dimensional surface to allow adjacent end plate fibrous ingrowth. An elastomeric device can incorporate an artificial ligament to replicate the function of the anterior longitudinal ligament to avoid expulsion of the device either anteriorly or posteriorly during motion. The biocompatibility of the materials used in the device is of great importance because any material that is to be implanted into the human body must have excellent biocompatibility to avoid failure of the device. The device, therefore, was manufactured using an elastomeric core and an enveloping jacket that are constructed of biomaterials that have an extensive clinical history within orthopaedic surgery, neurosurgery, and reconstructive surgery.
l
l
l l
l l
l
l
Patients with evidence of instability, defined as greater than 3.5 mm sagittal-plane displacement (anterior translation of the anteroinferior corner of the moving vertebra) either on resting or on flexion-extension radiographic examination. Patients who require surgical intervention at more than one level of the spine. Patients with any active infection at time of surgery. Patients who have been clinically diagnosed with whiplash syndrome. Female patients who are pregnant or could be pregnant. Patients who are known drug or alcohol abusers or with psychological disorders that could affect follow-up care or treatment outcomes. Patients who have participated in a clinical investigation with an investigational product within 6 months before treatment. Patients who are currently involved in any injury litigation claims.
INDICATIONS/CONTRAINDICATIONS At the time of this writing, the NeoDisc (Nu Vasive, Inc., San Diego, CA) is under clinical use in studies that require regulatory approval to be conducted. Thus, at this stage, the inclusion-exclusion criteria have been tightly controlled in order to collect good quality data to support regulatory approval of the device. It is anticipated that in broader clinical usage, the indications will become broader in their scope. The indications and contraindications selected for the U.K. clinical trial are as follows: Indications l l
l
l
l
l
Patients 18 years of age or older. Patients who, in the opinion of the clinical investigator, are suitable for the intended surgical procedure. Patients who have failed to adequately improve following at least an 8-week period of conservative management. Patients with evidence of spinal cord and/or nerve root compression causing myelopathy or radiculopathy that can be attributed to degenerative changes in a disc at a single cervical level corresponding with symptoms and confirmed using MRI, CT, or myelography within the last 6 months. Patients who are able to give voluntary, written informed consent to participate in this clinical investigation and from whom consent has been obtained. Patients who are, in the opinion of the clinical investigator, able to understand this clinical investigation, able to cooperate with the investigational procedures, and are willing to return to the hospital for all the required postoperative follow-up examinations.
Contraindications l
l
l
Patients who, in the opinion of the clinical investigator, have an existing condition that would compromise their participation and follow-up in this clinical investigation. Patients who have compromised bone quality by osteoporosis, infection, inflammatory arthritis, or tumor. Patients who have had previous cervical spine surgery at the level being considered, including posterior surgery.
DESCRIPTION OF THE DEVICE The NeoDisc device is a multipart artificial disc intended to replace the function of the native intervertebral disc. The implanted device is composed of an elastomeric nucleus contained within an embroidered polyester fabric, secured by four bone screws. This device has two integral fixation flanges, each of which has a pair of fixation holes (Fig. 26–1). Fixation of the device occurs through three steps: 1. implantation with a secure interference fit; 2. supplementary mechanical fixation through anchoring of the flanges using bone screws; 3. additional biologic fixation through the process of soft tissue ingrowth in to the encapsulating polyester jacket of the device. Embroidered Jacket The outer jacket of the NeoDisc device is a highly engineered textile, manufactured from polyester suture using a computercontrolled embroidery process. The use of embroidery allows for the individual placement of threads within the textile, where each thread size can also be chosen to optimize its strength/bulk characteristics. The complex design of the jacket includes an interdigitation of the fixation flanges. The jacket has two separate fixation flanges—the flange emanating from the superior surface passes through the flange that emanates from the inferior surface. The advantages of the NeoDisc textile design areas follow: 1. Conventional elastomeric artificial disc design concepts bond the elastomer to metallic end plates. In these designs, extension of the neck would subject the elastomeric core to a potentially damaging tensile load. However, for the NeoDisc with its textile encapsulation of the elastomeric core, as the flanges are pulled in opposite directions, it is a compression load that is passed to the core that the elastomer is ideally suited to withstand. 2. Secure fixation of the device combined with the ability to resist anterior device migration. It is during extension motions that grafts or implants are most likely to migrate anteriorly, but
CHAPTER 26
A n
The NeoDisc Elastomeric Cervical Total Disc Replacement
223
B
F I G U R E 26–1. NeoDisc device. A, Photograph of the device illustrating the fixation flanges. B, Cross-sectional view of the device showing
the construction.
for the NeoDisc, the more the neck is extended, the tighter the flanges are at the point where they interdigitate through each other. This has the effect of securely holding the main body of the device within the disc space. 3. Allow for a full range of motion. In extension, although a flat textile flange passing across or attached to the anterior surface of the device would help to prevent anterior migration, it would also act to limit extension. However, the interdigitated flanges not only prevent migration but they also still allow a natural extension motion. This is possible because in extension it is not only the flange that stretches but also the encapsulating fabric that is tensioned around the core. Polyester was chosen as the material to construct the textile element of the NeoDisc because of its extensive history of successful clinical use in various devices where tissue ingrowth in to the device is desired; it has an extensive clinical history, particularly in vascular applications as a strong, stable, biocompatible material. Elastomeric Core A medical implant-grade silicone is molded to form the compliant nucleus of the NeoDisc device. It is important to note that the silicone elastomer is solid and does not contain silicone gel. Solid silicone elastomers have been commonly used in a variety of biomedical devices for more than 40 years with a proven safety record.
BACKGROUND OF SCIENTIFIC TESTING AND CLINICAL OUTCOMES Bench Top Mechanical Evaluation of the NeoDisc Device Wear Testing
In order to test the size of implant most at risk of failure, the smallest available NeoDisc device was selected. Eight specimens of this size were tested. Fluids used in medical device testing are typically either saline, bovine serum, or fetal calf serum. Using either bovine serum or fetal calf serum while testing a polyester textile implant will result in protein deposition on the textile, which would render all weight loss measurements meaningless because, even with soak controls, it will be impossible to allow for this variable. Thus, the test medium used for the NeoDisc device testing was saline. Each sample was tested in its own individual environment from which the fluid could be collected for particle analysis purposes. The compressive load on the devices was maintained at an average of 100 N ( 5 N), as specified in ASTM standard F2423–05 as being a representative compressive load under which to conduct testing for cervical disc implants. The three motions that occur within the cervical spine are flexionextension, lateral bending, and axial rotation. For the ASTM test, the following motions are used: flexion-extension, 7.5 degrees; lateral bending, 6 degrees; and axial rotation, 6 degrees. The requirement
224
P A R T
I I I
Cervical Total Disc Arthroplasty
is for 10 million cycles of each load to be applied to the test sample. This can either be 10 million cycles of the motions combined together, or each motion can be applied individually. In order to allow for the potential difference in results if the motion patterns were reversed, the following test procedure was used: l Group A (devices 1, 2, 3, and 4) was tested for 10 million cycles of 7.5 degrees of flexion-extension motion, followed by a further 10 million cycles of 6 degrees of lateral bending, coupled with 6 degrees of axial rotation. l Group B (devices 5, 6, 7, and 8) was tested for 10 million cycles of 6 degrees of lateral bending, coupled with 6 degrees of axial rotation, followed by a further 10 million cycles of 7.5 degrees of flexion-extension. l Additional discs were set up as soak controls, with only a pure compressive load being applied throughout the testing. The result was that all of the test specimens received a total of 20 million cycles of wear testing. The test medium was collected from each station every 1 million cycles. The sample showing the greatest weight loss from each group in the test had its test medium analyzed at various timepoints. The analysis was performed in accordance with ASTM standard F1877–98. Particle analysis of the wear-tested NeoDisc devices has demonstrated that throughout the test, the particulate stayed approximately consistent in terms of size and distribution. The Feret diameters based on both particle numbers and volume did not change significantly as the test progresseed, and the two samples were reasonably consistent with each other. Therefore, there does not appear to be a difference due to the order in which the motion pattern is applied to the specimens. The roundness, aspect ratio, and form factor also stayed similar throughout the tests. There is no evidence that the change in motion after the 10 million cycle changeover caused another running-in period in which the particles changed in their morphology. The polyester particles tend to be irregular and fibrous or flaky, whereas the silicone particles are mainly spherical or spheriodal. At 20 million cycles, the average volumetric loss of material from the NeoDisc devices was 13.8 mg (approximately 11.2 mm3). This corresponds to an average of 0.69 mg per million cycles of testing (0.56 mm3 per million cycles). The 20 million cycle test can be considered to be a worst case test. It is unlikely that any device in clinical use will experience this quantity of large, exaggerated motions. Axial Compression Testing
To assess the safety of the implant under static and dynamic compression loading, a series of tests were conducted. Measurements of the loading required to create permanent height loss in the implant were performed. Also, the fatigue resistance of the implant design was evaluated in cyclic axial compression with dynamic loading. The discs were attached to the test frame by means of Delrin fixtures contoured to the geometry of the NeoDisc device as recommended by ASTM F2346. The smallest specimen was selected as the worst case in order to test the implant with the highest applied stress. The smallest
implant would be subjected to the highest stress because the same load applied to a smaller surface area leads to higher applied stresses. Two dynamic test specimens were tested in saline at 37 C. A sinusoidal waveform with a maximum compressive load of 500 N was applied for 10 million cycles at a 5 Hz test speed. In this test, failure was defined as a loss of device height equivalent to 50% of the height of the elastomeric core. Five static test specimens were tested in air at room temperature. Each sample had a 50 N preload applied. The device was compressed by half of the height of the elastomeric core and the corresponding force recorded. The load was then removed and the displacement at 50 N was determined. The dynamic test specimens reached 10 million cycles without functional failure. The samples lost 0.13 and 0.18 mm of core height. For the static fatigue test samples, the load required to compress the discs by half the core height was between 4728 and 6975 N. The residual core heights were almost unaffected, with losses from 0 to 0.02 mm. In conclusion, the endurance limit of the NeoDisc implant under dynamic axial compression testing was determined to be at least 500 N at 10 million cycles. Static axial compression to half of the core height did not cause damage or failure of the NeoDisc implant. The safety of the NeoDisc implant subjected to this loading mode was demonstrated by minimal height loss of the silicone core. Shear Compression In order to assess the safety of the implant under shear loading, devices were tested in shear compression according to ASTM F2346. The smallest specimen was selected as worst case in order to test the implant with the highest applied stress. The smallest implant would be subjected to the highest stress because the same load applied to a smaller surface area leads to higher applied stresses. The discs were attached to the test frame by means of fixtures contoured to the geometry of the NeoDisc, as recommended by ASTM F2346. Delrin was used in the case of the dynamic testing and stainless steel for the static testing. Two dynamic test specimens were tested in saline at 37 C. A sinusoidal waveform with a maximum compressive load of 250 N was applied at an angle of 45 degrees to the test specimens for 10 million cycles. As with the pure compression test, failure was defined as a loss of device height equivalent to 50% of the height of the elastomeric core. Five static test specimens were tested in air at room temperature. Each sample had a 50 N preload applied. The device was compressed by half of the height of the elastomeric core and the corresponding force recorded. The load was then removed and the displacement at 50 N was determined. The dynamic shear test specimens reached 10 million cycles without functional failure. The samples lost 0.10 and 0.18 mm of core height. For the static shear test samples, the load required to compress the discs by half the core height was between 574 and 829 N. The residual core heights were almost unaffected with losses from 0 to 0.03mm.
CHAPTER 26
In conclusion, the endurance limit of the NeoDisc implant under dynamic shear compression testing was determined to be at least 250 N at 10 million cycles. Static shear compression to half of the core height did not cause damage or failure of the NeoDisc implant. The safety of the NeoDisc implant subjected to this loading mode was demonstrated by minimal height loss of the silicone core. PUSHOUT (EXPULSION) TESTING Soft tissue integration of the NeoDisc acts as long-term fixation. However, it is important to be sure that the fixation method is secure to prevent migration or expulsion of the implant. Thus, testing was conducted to determine the force required to expel the disc from a simulated disc space. The discs were attached to the test jig by means of polyurethane foam blocks that formed a simulated disc space. As with clinical use, the devices were secured to the fixtures using four titanium alloy bone screws of the type used in the European clinical trial of the NeoDisc. The smallest specimen was selected as worst case in order to test the implant with the highest applied stress. The smallest implant would be subjected to the highest stress because the same load applied to a smaller surface area leads to higher applied stresses. Five pushout test specimens were tested in air at room temperature. A compressive load of 100 N was applied to the discs through the foam blocks to simulate the compressive axial load in the cervical spine. Loading was applied on the posterior surface of the disc in order to force it out of the simulated disc space. In this test, failure was defined as pushing the NeoDisc fixation flange off the bone screws. In the simulated pushout testing, an average of 238 N was required to dislodge the device and 293 N in order to expel it from the disc space. In each case, the failure mode was the polyester flange pulling over the head of the screw. No failures of this type have been seen in clinical application at the time of this writing; there has also been no evidence of anterior or posterior movement displacement of any implanted NeoDisc. The high load required to dislodge the NeoDisc and the stability of the device in clinical use have demonstrated that this potential failure mode is not a cause for concern. Biomechanical Testing
The NeoDisc device is intended to replicate as closely as possible the behavior of the natural disc. Biomechanical testing was initiated to examine the behavior of the implant under a variety of physiologic loading conditions. The results of biomechanical testing of the NeoDisc device have been presented at a number of international meetings and are the subject of preparation for publication in peer-reviewed journals (Yeh and Jackowski, 19967; Jackowski and Yeh, 19978; Jackowski and McLeod, 20029). To develop the cervical spine model, a total of 19 cadaveric cervical spines were harvested and fresh frozen (in an unembalmed state) at minus 18 C within 24 hours of the donor's death. Nine of the specimens were used to develop the in vitro cervical spine model that was the subject of a dotoral thesis entitled “The Preparation, Development and Analysis of an In Vitro Cadaveric
The NeoDisc Elastomeric Cervical Total Disc Replacement
225
Model of Cervical Spine Motion” (Yeh, 1997). The model monitored motion of vertebrae by the placement of small infrared reflectors mounted on the individual spinal vertebrae. MacReflex motion analysis equipment (Qualysis AB) monitored the output from an infrared camera that tracked the position of the markers. To examine the behavior of the NeoDisc, the testing was conducted to monitor the spine in the following motions; flexion-extension, lateral bending, and axial rotation. The test monitored the behavior both for a multilevel spine and also on the individual motion segment. Four biomechanical representations were studied; the baseline was the behavior of the intact disc; the readings were then repeated after a microdiscectomy, and then after the implantation of the device, and finally after the motion segment had been mock “fused” with the aid of an anterior spinal plate. The analysis demonstrated that the NeoDisc device retained natural motion to within 5% of the amount of movement that would be present with the natural disc. This compared with 58% more movement following a discectomy, in which the natural disc is removed and the disc space is left empty, and 68% less movement when a bone graft and plate were used. Analysis of the single motion segment demonstrated that the NeoDisc implant could preserve 93% of the mean proportional range of motion at the operated cervical levels and keep the increase in motion at adjacent cervical segments following surgery down to only 7%. Hysteresis loops for the single motion segment model were analyzed and demonstrated that the implanted NeoDisc device left a motion segment that was an extremely good match for the natural disc in terms of flexionextension, lateral bending, and axial rotation motions. The compliant nature of the NeoDisc device combined with mechanical function close to that of the natural disc offer the patient a very real possibility of avoiding long-term (specifically adjacent level) issues, if such issues can truly be attributed to mechanical alterations of the adjacent levels. ANIMAL DATA Implantation Data in an Ovine Model It is widely accepted that there is no ideal animal model to evaluate the mechanical and kinematic performance of artificial discs. However, one of the key features of the NeoDisc device is the ability of the polyester to incite the ingrowth of soft tissue into the device. An ovine model was chosen as being the best animal model to use to demonstrate this effect A total of 10 sheep received NeoDisc implants. The sheep were sacrificed at varying time points. In all cases, there were no signs of adverse reaction to the device and the sheep were both neurologically and physically normal during the length of the study. The following observations were noted: In all cases of animal implantation, the polyester jacket was variably integrated by either fibrocartilage or by fibrous tissue, a result consistent with the historic use of polyester. The result is that the implant becomes a silicone core surrounded by a composite structure of polyester reinforced soft tissue. The tissue integration will serve to prevent the fibers of the textile jacket from being able to wear against each other. Fibrocartilaginous integration was seen
226
P A R T
I I I
Cervical Total Disc Arthroplasty
at the anterior aspect of the intervertebral part of the implant. The fibrocartilage had undergone endochondral ossification on the external periphery of the implant to become continuous with the surrounding vertebral bone; in this sense, the tissue will serve to act as a biologic anchor to the device, rendering the fixation screws superfluous. Examination of the tissues by a histopathologist demonstrated that there was no organized lymphoid tissue, indicating a lack of immune/allergic response to the device or any wear debris that may have been generated following implantation. In all of the sheep, there was no evidence of an immune or allergic response to the implanted materials. The fibrous tissue contained a far greater macrophage/macrophage polykaryon infiltrate than the fibrocartilage, and these cells showed their typical foreign body reaction to shed silicone and polyester fibers, as shown in Figure 26–2. Bone resorption was noted in animals. At the 3-month time point, the definition of the end plates had been disrupted; by the 18-month time point, the end plates had clearly reconstituted and were in approximately the same position as preoperatively. The histopathologist hypothesized an initial fibroblastic activity with tissue penetration into the adjacent bone, followed by a process of maturation from fibroblastic to fibrocartilaginous tissue, which would be associated with a net gain in adjacent bone, thus explaining the difference in observations from the 3- to 18-month time points. The extent of the inflammation resulting from the presence of wear debris was consistent with the tissue reactions seen in other orthopaedic implants in the day-to-day practice of the histopathologist. The observed tissue responses (to the debris from the NeoDisc device) are thought to be normal for a medical device and are not seen to be a cause for concern. In summary, the animal trials provided macroscopic and histologic evidence of the tissue ingrowth. There were no unforeseen causes of wear debris, and the tissue response to the wear debris generated is seen to be in line with that in other medical devices.
A n
EPIDURAL APPLICATION OF PARTICULATE WEAR DEBRIS—AN IN VIVO RABBIT MODEL STUDY To investigate the possible histopathologic effects on neurologic tissue caused by the release of wear debris from the NeoDisc device, an in vivo rabbit model was used.10 It is unfeasible to collect significant volumes of particulate from the wear testing (and then subsequently clean and resterilize this particulate); thus, simulated wear debris was generated for use in the animal study. The NeoDisc device is manufactured from silicone and polyester. Both of these materials produced wear debris during laboratory testing. Therefore, to provide pure sample particulate for each of these materials, samples of each type were processed separately. Particulate of a similar size and morphology to that collected during wear testing was generated by cryopulverization of each of the materials. For both materials, the particulate is mainly in the 1- to 50-m size range. There were a significant number of submicron-sized polyester particles and some submicron-sized silicone particles. In line with particulate captured during wear testing, the polyester particles tended to be irregular flaky particles, and the silicone particles were mainly spherical or spheriodal. For the epidural application study, a total of 30 New Zealand white rabbits were used. The animals were approximately 1 year old and 5 kg in weight. The animals were split into groups of 10 animals (five to be sacrificed at the 3-month interval and the remaining five to be sacrificed at 6 months). The first group of 10 animals was used as surgical controls. These animals had a sham operation in which the epidural area was exposed but no implantation took place. The second group of 10 animals was used to assess the silicone debris; for each of these, 3 mg of silicone particulate were applied to the epidural area of the lumbar spine. The third group of 10 animals was used to assess the polyester debris; for each of these, 3 mg of polyester particulate were applied to the epidural area of the lumbar spine.
B
F I G U R E 26–2. Histology showing fibrous tissue encapsulation of the polyester jacket at a survival time of 6 months in an ovine model.
A, Overall view illustrating the elastomeric core and the fibrous tissue encapsulation around the polyester jacket. B, Higher magnification view showing the orientation of the fibroblasts adjacent to the elastomeric core at top and the relatively small number of macrophages.
CHAPTER 26
The surgical technique for the implantation of the particulate was conducted with the animal positioned prone, aseptically prepared, and draped in sterile fashion. A midline skin incision of 4 to 5 cm in length was centered over the L5-L6 operative level. The L6 spinous process and L5-L6 supraspinous/interspinous ligament were then exposed using a periosteal elevator and electrocautery, as necessary. Most importantly the operative level bilateral facet capsules and joints were not exposed. The L6 spinous process and ligamentum flavum at L5-L6 were excised, permitting interlaminar exposure of the dural sac (5 mm diameter surface area). The membranous coverings and neural structures of the spinal canal were then accessible. The treatment materials were implanted in a sterile, dry format. To ensure that the spinal cord, nerve roots, and extramedullary vessels were not damaged, the surgical approach and material application techniques were performed using operating loupes. The control group consisted of epidural exposure alone.10 Following surgery, all of the animals in every group demonstrated a normal recovery, with no complications, and their behavior was unaffected. The animals were sacrificed at 3 and 6 month intervals. At the time of necropsy, both experimental groups exhibited a greater amount of epidural fibrosis compared with controls. There were no signs of infection in any of the animals. Specimens taken from the epidural region were culture negative in every case. The spinal cord and overlying fibrous tissues were examined using immunocytochemistry techniques. At the 3-month time interval, the number of macrophage-expressing cytokines present indicated no statistical differences between the control and study groups. In addition, there were no significant differences between the experimental and control groups with respect to activated macrophages. In general, macrophage activity was reported to be very low and was reported to be almost not observed in both of the experimental cases. At the 6-month time interval, the cytokine response from tumor necrosis factor-a, tumor necrosis factor-b, interleukin-1b, and macrophages was essentially nonreactive. Interleukin-1a and interleukin-6 (IL-6) were reactive but not statistically different from the control groups. Histopathologic analysis at both 3- and 6-month time intervals indicated there were no significant pathologic changes in the systematic tissues. All the systematic organs and organ systems demonstrated no issues of concern. It was postulated that lymphoreticular dissemination of both the silicone and polyester probably did occur. However, no evidence could be found in the tissues analyzed. In the control and study groups both at 3 and 6 months, there were increased concentrations of IL-6 cytokines. However, there were no statistical differences in cytokine activity between the study groups and the control. There were mild changes in cytokine and macrophage activity for the experimental groups, although the indication was that there was a normal distribution of myelin and the intracellular neurofibrilla network—characterizing the treatments as “without significant pathological changes at the three-month time interval.” There was no evidence of cellular apoptosis, giant cell reaction, or other significant pathologic changes. The study concluded that overall, based on the postoperative time period evaluated, there was no evidence of an acute neural or systematic histopathologic response to either the silicone or the polyester.
The NeoDisc Elastomeric Cervical Total Disc Replacement
227
CLINICAL PRESENTATION AND EVALUATION The first two NeoDisc prostheses were implanted in two patients in July 2004 as part of a two-site European study regulated by the MHRA (the United Kingdom Regulatory Authority). The 1-year results on 14 patients treated were presented at the Spinal Arthroplasty Society meeting in Montreal in May, 2006. There were no adverse clinical or device-related events to report. Average blood loss during surgery was less than 15 mL, and the average operation time was less than 60 minutes. Neck pain assessed by Visual Analog Scale on a 100-point scale fell from an average of 38 preoperatively to 13 at 3 months and 8 at 12 months. The majority of patients presented with radiculopathy, and the visual analog arm pain score preoperatively was an average of 65, falling to 14 at 3 months, 9.5 at 6 months, and 11 at 12 months. Assessment of disability using the Neck Disability Index showed a reduction from an average of 24 on a 100-point scale preoperatively to 9 at 3 months, maintained at 10 by 12 months. Patients exhibited a range of flexion-extension segmental motion preopertively with an average of 8 degrees at the symptomatic level. Postoperatively, the average range of motion at the treated level was 6 degrees at the 3-month follow-up, 7 degrees at the 6-month follow-up, and 6 degrees of flexion-extension range of motion at the 12-month (1-year) follow-up. Postoperatively, all patients either maintained or recovered full neurologic function. Clinical studies are ongoing. At the time of this writing, three patients have returned for 2-year clinical and radiologic follow-up examinations. The demonstrated flexion-extension ranges of motion for the three patients at the 2-year follow-up are 10, 6, and 9 degrees. The fourth patient was not able to attend his 2-year follow-up visit because he is on active military duty (with the U.K. armed forces) in Iraq at the time of this writing. All three patients continue to maintain excellent outcomes in terms of their clinical and functional outcomes with respect to the cervical spine. Repeat MRI imaging at 2 years has been performed and confirms that repeat imaging of the cervical spine and neural structures not only at adjacent levels but also at the operated levels is entirely possible in the presence of the NeoDisc device. Figures 26–3 and 26–4 demonstrate two patients from the European study. The first patient has radiculopathy, and the second patient has rapidly progressive myelopathy. The radiologic results for these two patients are shown in Figures 26–5 and 26–6. At the time of this writing, a US IDE study is also under way. OPERATIVE TECHNIQUE As described earlier, the design philosophy for the NeoDisc device is to minimize the necessity for the operating surgeon to alter his or her surgical technique; the implantation of the device does not involve the use of any specialized reamers, centering tools, or jigs. The procedure is performed with the patient under a general anesthetic, and the patient is positioned in a supine position with moderate head up tilt to reduce bleeding but with the neck being placed in a generally neutral position, avoiding excess flexion or extension. The surgeon performs the standard anterolateral approach and anterior cervical discectomy/decompression procedure. Because fusion is not the aim of the surgery, there is no requirement to prepare a bleeding end plate, but instead, only disc removal and decompressive techniques are needed. It is important to remove all disc material and
228
P A R T
I I I
Cervical Total Disc Arthroplasty
n
F I G U R E 26–3. A C6-C7 NeoDisc arthroplasty performed for radiculopathy. This 37-year-old woman presented with severe right-sided radiculopathy affecting her C7 nerve root. A preoperative MRI scan demonstrated a C6-C7 predominantly right-sided compression due to a combination of osteophyte and soft disc prolapse. The one-year postoperative magnetic resonance imaging scan demonstrates removal of compressive pathology and restoration of disc height. The prosthesis is visible as a low-signal intensity within the C6-C7 disc space. There is minimal artifact present, attributable to the titanium anchor screws. The patient had complete relief of right arm radiculopathy.
n
F I G U R E 26–4. A C3-C4 NeoDisc arthroplasty performed for myelopathy. This 41-year-old man presented with myelopathy with bilateral arm and leg weakness and increased tone affecting all four limbs. The preoperative magnetic resonance imaging (MRI) scan demonstrates single-level cord compression at the C3-C4 level, predominantly due to a midline soft disc prolapse. Also apparent on the MRI scan is signal change within the spinal cord at this level. The two-year postoperative MRI scan demonstrates relief of spinal cord compression and return of normal signal within the cord. The NeoDisc prosthesis is visible as a low signal intensity within the disc space. There is minimal artifact due to the titanium anchor screws. The patient had complete relief of myelopathic symptoms and subsequently went on to become a personal fitness trainer and competed in half marathons.
CHAPTER 26
n
F I G U R E 26–5.
The NeoDisc Elastomeric Cervical Total Disc Replacement
229
Flexion-extension radiograph of a C6-C7 NeoDisc arthroplasty for the patient shown in
Figure 26–3.
n
F I G U R E 26–6.
Flexion-extension radiograph of a C3-C4 NeoDisc arthroplasty for the patient shown in
Figure 26–4.
compressive osteophytes or extruded disc to ensure relief of the patient's neurologic compression and symptoms. The posterior longitudinal ligament may be left or resected depending on the requirement to achieve satisfactory decompression of the compressed neural structures. When the surgeon is satisfied with the decompression, the NeoDisc trial sizers are inserted in a sequential fashion until a
footprint and height of the trial are found that good fits as judged by visual examination, tactile feedback, and radiologic observation. Disc height should be restored to an appropriate level as judged by viewing adjacent disc spaces on an image intensifier. The sterile prosthesis is then selected to match the sizing trial and inserted into the disc space. This is achieved by the use of the insertion forceps, with
230
P A R T
I I I
Cervical Total Disc Arthroplasty
recommended gentle traction under the chin while the device is inserted into the empty disc space. The device should fit fully within the disc space and be seated 1 to 2 mm deep to the anterior edge of the vertebral body. The securing textile flanges are positioned flat against the vertebral body, and a bone awl is passed through the flange hole to penetrate the anterior cortex of the vertebral body beneath. The bone screws are then placed through the flange holes, and the screws are secured in the normal manner until finger tight. This surgical technique is summarized in Video clip # 2. After securing the device, it may be useful to passively perform gentle flexion and extension of the patient's head and neck while observing the disc under direct vision and on the image intensifier (shown in Video clip # 3). This gives confirmation that the correct size of device has been selected and that it is seated securely within the disc space. The wound is closed in the usual manner, and a wound drain is recommended for the first 12 to 24 hours after surgery. Postoperative Care Cervical collars are not required, and the patient is encouraged to move the head and neck as comfort allows and as normally as possible. It is recommended that a postoperative x-ray study is performed 24 to 48 hours after surgery to confirm correct position of the implant device. There are no unique recommendations or restrictions with regard to the NeoDisc device. Complications and Avoidance In general terms, the risks of complications with the NeoDisc device are largely those of a standard anterior cervical decompression and interbody fusion. The patients in whom this device would be contraindicated are described under the earlier section on indications and contraindications and, broadly speaking, encompass patients who are at significant risk of infection or with poor-quality bone stock. Evidence of significant instability at the operative level is also a contraindication for use of this device.
ADVANTAGES/DISADVANTAGES: THE NeoDisc ELASTOMERIC CERVICAL TOTAL DISC REPLACEMENT Advantages The surgical technique is extremely simple, with no end plate milling or reaming. Detailed postoperative MRI and CT imaging of the spinal cord, nerves, and foraminae are easily obtainable. The device does not impose its own access of rotation but allows the axis of rotation determined by the patient's own anatomy. The device is available in a wide range of sizes. Surgical time is shorter than a standard ACDF with the possibility of less retraction on the soft tissue including the esophagus. The elastomeric core provides both mobility and axial compliance, acting as a more physiologic disc replacement. The polyester embroidered casing is designed to provide both immediate and long-term fixation. The interdigitated flanges provide immediate stability, preventing the anterior or posterior migration of the device. The embroidery also allows soft tissue ingrowth providing long-term fixation at the operative level. Disadvantages The currently employed titanium screws are associated with a small degree of MRI artifact. At present, there are only limited data with 2-year follow-up on a limited number of patients. ACDF, anterior cervical discectomy and fusion; CT, computed tomography; MRI, magnetic resonance imaging.
CONCLUSIONS/DISCUSSION Cervical spine arthroplasty represents a developing and expanding area for designers and treating clinicians. Patient preference and pressures will increasingly lead to requests for artificial disc replacement in place of more traditional fusion procedures. Health economic pressures will require manufacturers and clinicians to demonstrate value for money invested and equal efficacy with the demonstration of preservation of segmental mobility. Long-term studies will be necessary before the full benefit of this technology can be determined. The NeoDisc device represents the only currently available elastomeric textile prosthesis for cervical implantation. The operative procedure is simple and does not require the surgeon to learn new or demanding techniques or instrumentation. The device is highly compatible with all imaging modalities. Also, the potential to replace the titanium screws with nonmetallic anchoring devices will enable the device to become completely free of artifact on postimplantation MRI studies.
REFERENCES 1. Cloward RB: The anterior approach for removal of ruptured cervical discs. Neurosurgery 15:62–67, 1958. 2. Robinson RA, Smith GW: Anterolateral cervical disc removal and interbody fusion for cervical disc syndrome. Bull Johns Hopkins Hosp 96:223–224, 1955. 3. Goffin J, Geusens E, Vantomme N, et al: Long term follow up after interbody fusion of the cervical spine. Presented at the annual meeting of the Cervical Spine Research Society, Charleston, SC, Nov–Dec 2000. 4. Goffin J, Van Loon J, V Calenberg F, et al: Long term results after anterior cervical fusion and osteosynthetic stabilisation for fractures and/or dislocations of the cervical spine. J Spinal Disord 8:500–508, 1995. 5. Hilibrand A, Carlson G, Palumbo M, et al: Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg Am 81:519–528, 1999. 6. Pospiech J, Stolke D, Wilke H, et al: Intradiscal pressure recordings in the cervical spine. Neurosurgery 44:379–385, 1999. 7. Yeh J, Jackowski A: Proceedings of the British Cervical Spine Society, 1996. 8. Jackowski A, Yeh J: 13th Annual Meeting of the European Cervical Spine Research Society. 1997, pp 26–28. 9. Jackowski A, McLeod A: Spine Arthroplasty II Symposia Montpellier, VT, 2002. 10. Cunningham BW, Orbegoso CM, Dmitriev AE, et al: The effect of spinal instrumentation particulate wear debris: An in vivo rabbit model and applied clinical study of retrieved instrumentation cases. Spine J 3:19–32, 2003.
CHAPTER
27
Mobi-C Jacques Beaurain, Pierre Bernard, Thierry Dufour, Jean-Marc Fuentes, Istvan Hovorka, Jean Huppert, Jean-Paul Steib, and Jean-Marc Vital
K E Y l
l
l
l l
P O I N T S
Mobi-C is a second-generation, three-piece nonconstrained cervical disc prosthesis. The implantation technique is simple and accurate without invasive preparation of the end plates, allowing a multilevel arthroplasty. Intermediate clinical and radiologic results were satisfactory at both early and late control, without device-related complication. Physiologic cervical motion is often restored after Mobi-C implantation. Additional long-term follow-up results are mandatory to further evaluate the effects of the motion restoration at both the implanted and adjacent levels.
INTRODUCTION Duplicating the cervical intervertebral disc's form and function with an artificial disc is challenging. Cervical disc arthroplasty offers several theoretical advantages compared with anterior cervical discectomy and fusion (ACDF). The stresses on adjacent levels above and below the fusion site may lead to higher incidence of degeneration and segmental instability. On the other hand, preservation of motion at the surgically treated level may potentially decrease the occurrence of adjacent-level degeneration.1,2 Several prostheses with different components and kinematic designs are now available for clinical use.3 Among them, Mobi-C (LDR Médical, Troyes, France) was designed by a French team of orthopaedic and neurosurgeons with two main objectives: (1) attempt to replicate the normal cervical intervertebral disc motion as much as possible and (2) develop a device with well-known materials that is easily placed with a simple and reliable technique. Mobi-C has a three-piece nonconstrained articulation with a polyethylene mobile nucleus moving between two Cr-Co plates that are designed to be as anatomic as possible.4 DESCRIPTION OF THE DEVICE The Mobi-C represents a metal-on-polyethylene device. It is composed of two spinal plates consisting of cobalt, chromium, 29 molybdenum ISO 5832-12 alloy, and an ultra-high-molecularweight polyethylene mobile insert (Fig. 27–1). The inner contact
surfaces of the superior and inferior plates are spherical and flat, respectively. The mobile insert is self-centering on the inferior end plate. Each movement of the superior plate induces the mobile insert to reposition on the inferior spinal plate. The inner contact surface of the superior plate is spherical, allowing a fully congruent contact surface with the convex spherical dome of the mobile insert. The inner contact surface of the inferior plate is flat and contains two lateral stops that limit the mobility of the mobile insert by contacting the lateral surface of the insert. The lateral stops reduce the potential for migration of the mobile insert. Both the superior and inferior spinal plates contain two teeth rows that are located laterally on each plate to ensure the primary fixation. A titanium and hydroxyapatite plasma spray coating is applied to the bony interface surfaces of the superior and inferior plates. Different plate sizes are available (13 15, 13 17, 15 17, and 15 20, depth by length in mm). Different insert heights are available (5 mm, 6 mm, 7 mm), to restore the physiologic height of the disc. The device allows for various degrees of mobility that include five independent degrees of freedom, two translational and three rotational. The five independent degrees of freedom are illustrated below (Fig. 27–2). BACKGROUND OF SCIENTIFIC TESTING Testing was conducted with success to determine the mechanical properties of the Mobi-C Cervical Disc Prosthesis in accordance with applicable American Society for Testing and Materials (ASTM) standardized test methods for the following: Durability/wear testing: The results demonstrated a 0.08% mass loss or a total of 0.24 mm3 volumetric loss after 10 million cycles (1 million cycles ¼ 1 year of clinical use; rate of wear ¼ 0.024 mm3/year). Static and dynamic axial compression fatigue testing: Specimens were loaded until failure in compression and exhibited a mean initial peak load of 4,752 309 N (normal axial loads in the cervical spine range from 70 to 150 N). Axial fatigue failure of the device is thus unlikely. 231
232
P A R T
n
I I I
Cervical Total Disc Arthroplasty
F I G U R E 27–1. The Mobi-C device and its components. A, B, Views of the three components of the device.
Static and dynamic shear compression fatigue testing: Testing to 10 million cycles was completed at an applied load of 450 N with no observed failure (physiologic shear loads: less than 120 N). Static subsidence testing: The devices exhibited a mean peak load of 1,090 46 N without implant failure. Normal axial loads in the cervical spine range from 70 to 150 N. Static expulsion testing for mobile insert and full device: An axial preload of 100 N was placed on the device, then the device and fixtures were rotated 90 degrees to test the expulsion resistance. The mobile insert was loaded with a posterior-to-anterior load. The inserts were loaded to failure (defined as movement of 3 mm) and exhibited a mean peak load of 143 18 N without implant failure. The observed loads are comparable with the expulsion loads required to displace the entire device. Therefore, expulsion of the mobile inserts from the end plates is not anticipated under physiologic loading condition. Mechanical testing demonstrated that the Mobi-C disc is adequate to provide an additional therapeutic option in maintaining motion segment position and spacing while preserving flexibility in the affected cervical vertebral level.
CLINICAL AND RADIOLOGICAL EVALUATION A multicenter clinical and radiologic prospective study is under way to assess the safety and efficacy of the Mobi-C implant in treating cervical degenerative disc disease (DDD). Ninety-two patients were enrolled across eight sites, between November 1, 2004 and January 31, 2006. Indications were DDD at one or several levels between C3 and C7, leading to radiculopathy or myelopathy, or both. Surgery was performed only after failure of appropriate conservative medical treatment. DDD was confirmed through anterioposterior (AP) cervical x-ray studies, computed tomography, or magnetic resonance imaging. All patients signed the informed consent form. Exclusion criteria are usual and include aging (>65 years old), noncompliance with the study protocol, osteoporosis, metabolic bone disease, congenital or post-traumatic deformity, infection, neoplasia, instability of the intersomatic space, or a narrow canal ( 2 degrees) at the last follow-up. Representative clinical cases are shown later (Fig. 27–5). There was no subluxation, no device migration, and no subsidence. Only one case of anterior ossification has been reported, which has been attributed to preoperative cervical kyphosis and to malposition of the device (too posterior) which led to a poor mobility after 1 year. No periprosthetic calcification was observed.
All of the surgical procedures are performed by senior surgeons. The patient is in supine position, under general anesthesia, with the neck in neutral position, and the head maintained in the straight position throughout the procedure. A standard rightor left-sided approach is taken to the anterior cervical spine. The midline is defined under AP fluoroscopic view. After a thorough discectomy, the intersomatic space is distracted by a vertebral distracter. Once the height restoration is obtained, the distraction is maintained by the Caspar retainer, providing access to the posterior disc space. After total disc material and osteophytes are removed and neuroforaminal decompression is completed, the end plates are prepared without any shaping or chiseling. Rather than recommending systematic and complete posterior longitudinal ligament (PLL) removal, we insist on performing a good release of the posterior part of the disc to ensure a parallel intervertebral space opening. To our opinion, a thin residual PLL layer has no mechanical influence on motion. Depth and width measurements allow the determination of the appropriate trial implant. The trials help to confirm the precise size of the implant, which is verified under fluoroscopic control. During this step, it is important not to exceed the height of the healthy adjacent discs and not to induce overdistraction of the facet joints. The prosthesis is gently impacted into the disc space using a specific inserter. An adjustable stop allows precise adjustments of the implant AP position. The primary anchoring optimization is obtained through compression with the Caspar. An x-ray control (AP and lateral view) must confirm the well positioning of the implant. POSTOPERATIVE CARE The patient was out of the bed the day following the surgery without cervical collar and discharged from the hospital as soon as possible. There was no systematic postoperative medical treatment or physiotherapy. COMPLICATIONS
TABLE 27 – 1 .
Intervention Cervical pain Arm pain
Satisfaction Index After the Mobi-C Surgery*
Very Satisfied (%)
Satisfied (%)
Not Satisfied (%)
Dissatisfied (%)
83 38
14 48
3 14
0 0
37
48
11
4
*One year after the surgery, patients were asked to quote their satisfaction degree toward intervention benefit, cervical pain and arm pain.
Postoperative complications were related to the cervical anterior approach: Two local hematomas and one cerebrospinal fluid leak have been reported without clinical consequence or reintervention, transient dysphagia (n ¼ 2), bitonal voice (n ¼ 2), and dysphonia (n ¼ 1) have also been noted and were resolved within a few months. There was no device-related intraoperative complication. One prosthesis had to be removed 4 months after the initial placement owing to persistent cervical pain and excessive mobility
CHAPTER 27
Mobi-C
Case 1: Mobi-C C5-C6 Female, 41 years, active Intractable neck and right arm pain Weakness and numbness in right arm
100
95
93
84
pre-op 12 mo
80 60
A
40
28
20
3
C
5
0 VAS neck VAS arm
NDI (%)
D B
12 mo post-op Case 2: Mobi-C C5-C6/C6-C7 Male, 58 years, active Continuous and severe neck pain (1987) Bilateral discontinuous arm pain with weakness
80
73
pre-op 12 mo
60 40
26
24 20 0
8 0
0
VAS neck VAS arm
NDI (%)
F
E
J
H
C5-C6
L G
I n
C6-C7
K
F I G U R E 27–5. Clinical cases. VAS, Visual Analog Scale; NDI, Neck Disability Index.
12 mo post-op
235
236
P A R T
I I I
Cervical Total Disc Arthroplasty
n
F I G U R E 27–6. Physiologic restoration of the mean center of rotation (MCR) after Mobi-C surgery at the C5-C6 level. A, Calculation of MCR before surgery. B, MCR calculation 1 year after the surgery.
below a congenital cervical fusion block. The patient underwent fusion with an anterior cervical cage, resulting in a good clinical outcome. CONCLUSIONS/DISCUSSION The Mobi-C prosthesis received approval to affix the CE Mark in 2004 and was first implanted in France in November 2004. The Mobi-C prosthesis is currently distributed in 20 countries in Europe, Asia, and Africa. As of September 2006, nearly 1000 Mobi-C devices have been implanted worldwide without any unanticipated adverse device effects. In our study, more than 1-year follow-up demonstrates an excellent safety profile with no reported device-related complications, surgical re-interventions, or radiographic failures. Moreover, the data from the follow-up visits also demonstrate initial efficacy of Mobi-C with improvements in both pain and function as demonstrated by VAS and NDI scores, which confirms our preliminary results.4 The MCR study5 shows that the Mobi-C device can restore the physiologic MCR position as defined by previous anatomic study, for each index level.6 Calculation of the MCR at each follow-up timepoint indicates that sometimes this functional restoration may require a few weeks, even several months, depending probably on pain history duration.
This normal restoration of the index level segmental motion and the preservation of the adjacent level motion support the concept that arthroplasty may reduce adjacent degenerative disc disease compared to fusion.5 By comparison to constrained artificial discs, which require a definitive precise placement to reproduce a normal cervical kinematic, the nonconstrained Mobi-C device has a theoretical advantage: By this way, it should provide a better preservation of the MCR, and limit facets and ligament stresses. According to the MCR results of our study and the cumulated experience with Mobi-C implantation, we advise to center the device in the intersomatic space, rather than to push it to the posterior wall, as recommended for lumbar arthroplasty. Whereas constrained devices require strong fixation systems such as keels or screws, the stability obtained with the four teeth rows and the coating of Mobi-C is proven to be sufficient. This design allows easy multilevel placement, without any risk of vertebral fracture or device conflict, even in small vertebral bodies. Intermediate results on Mobi-C cervical disc prosthesis are very encouraging in this prospective ongoing study. Of course, long-term follow-up will be needed to fully assess functionality of cervical disc replacement and its efficacy in preventing adjacent level early degeneration. We hope to answer these challenging questions in pursuit of our study, and with
CHAPTER 27
a prospective, multicentric, and randomized controlled clinical trial, which began in the United States in May 2006, in the context of an Investigational Device Exemption submission. REFERENCES 1. DiAngelo DJ, Robertson JT, Metcalf NH, et al: Biomechanical testing of an artificial cervical joint and an anterior cervical plate. J Spinal Disord Tech 16:314–323, 2003. 2. Goffin J, Geusens E, Vantomme N, et al: Long term follow-up after interbody fusion of the cervical spine. J Spinal Disord Tech 17:79–85, 2004.
Mobi-C
237
3. Phillips FM, Garfin SR: Cervical disc replacement. Spine 30:S27–S33, 2005. 4. Bernard P, Vital JM, Dufour T, et al: A new mobile cervical prosthesis: Mobi-C. Preliminary results of a prospective study. Poster session, Global Symposium on Motion Preservation Technology, Spine Arthroplasty Society, 2005, New York. 5. Dufour T, Delecrin J, Beaurain J, et al: Calculation of centers of rotation in the cervical spine, before and after implantation of the Mobi-C cervical disc prosthesis. Global Symposium on Motion Preservation Technology, Spine Arthroplasty Society, 2006, Montréal, Canada. 6. Bogduk N, Mercer S: Biomechanics of the cervical spine, I: Normal kinematics. Clin Biomech 15:633–648, 2000.
CHAPTER
28
The CerviCore Cervical Intervertebral Disc Replacement Jonathan R. Stieber, Jeffrey S. Fischgrund, and Jean-Jacques Abitbol
K E Y l
l l l
l
P O I N T S
The CerviCore Intervertebral Disc is a cervical total disc replacement featuring metal-on-metal, saddle-shaped bearing surfaces. Indicated for the treatment of cervical radicular symptoms. Specialized insertion instrumentation facilitates accurate implantation. The CerviCore Intervertebral Disc is designed to replicate the motion of a native intact cervical disc. The procedure may be converted to a standard anterior cervical fusion if necessary at any stage of the procedure.
l
l l
CONTRAINDICATIONS l
INTRODUCTION The CerviCore Intervertebral Disc (Stryker Spine, Allendale, NJ) prosthesis is an investigational device currently in a U.S. Food and Drug Administration (FDA)–approved Investigational Device Exemption (IDE) clinical trial. It is a semiconstrained cervical total disc replacement incorporating unique metal-on-metal, saddle-shaped bearing surfaces (Fig. 28–1). Following a standard anterior cervical decompression, the device is designed to preserve intervertebral motion by replicating the kinematics of the native cervical functional spinal unit. The CerviCore Intervertebral Disc is indicated for the treatment of cervical radicular symptoms associated with loss of disc height, disc/osteophyte complex, or herniated disc at a single level of the subaxial cervical spine from C3 to C7 resulting in upper extremity pain or neurologic deficit, or both. Specialized insertion instrumentation is designed to facilitate device implantation and minimize misalignment of the implant and other technical errors that may be associated with modular assembly and implantation. INDICATIONS l
l
238
Significant loss of disc height (>50% decrease), disc/osteophyte complex, or herniated disc at a single level between C3 and C7 on magnetic resonance imaging (MRI) or computed tomography, or both Clinical symptoms including radicular paresthesias in a dermatomal distribution or a progressive or acute onset functional neurologic deficit in a single nerve root distribution
Failure of conservative treatment consisting of a minimum of 6 weeks of medical management, pain medication or injection therapy, physical therapy and/or physician-prescribed activity modification Clinically asymptomatic adjacent cervical levels Skeletally mature patients aged 18 to 65 who are nonsmokers
l
l
l l
l
l
l
l l
l l
Poor bone quality, osteopenia, osteoporosis, or a clinically compromised vertebral body structure at any cervical level due to acute or past trauma End plate incompetence at the level to be treated diagnosed on x-ray study or MRI such as a Schmorl's node, end plate fracture, or end plate herniation Metabolic bone disease or current medications that may interfere with normal bone metabolism Congenital or spontaneous fusion at the target or an adjacent level Cervical levels with less than 3 mm of preoperative disc height, evidence of ankylosis or localized kyphosis Pathology requiring concurrent posterior surgical treatment (Patients who have undergone prior laminectomy or laminotomy at a level other than the target level or adjacent levels may be indicated for treatment.) Significant radiographic instability at the target or an adjacent level, defined as >3.5 mm of translation, >11 degrees of angulation when compared with an adjacent level, and >3 mm spondylolisthesis, retrolisthesis, or spondylolysis Patients with severe facet joint arthritic changes at the target or an adjacent level Ossification of the posterior longitudinal ligament at any level Cervical spondolytic myelopathy (Myelopathy due to a recent [2 years not yet available Long-term wear characteristics unknown
CoCrMb, cobalt, chromium, and molybdenum alloy; IDE, Investigational Device Exemption; UHMWPE, ultra-high-molecular-weight polyethylene.
CONCLUSIONS/DISCUSSION Our pooled multisite clinical experience demonstrates early favorable outcomes associated with the SECURE-C Cervical Artificial Disc. This new treatment for degenerative disorders of the cervical spine is a viable alternative to traditional fusion, which uses known methods of approach and discectomy. Early results suggest that motion is preserved and disc height is improved at the operative level. Long-term clinical safety and effectiveness of the device are yet to be determined and will be reported as enrollment concludes and patients reach their 2-year follow-up. Evaluation of adjacent level degeneration is of high interest and will be studied.
SECURE-C Cervical Artificial Disc
253
REFERENCES 1. Goldberg EJ, Singh K, Van U, et al: Comparing outcomes of anterior cervical discectomy and fusion in workman’s versus non-workman’s compensation population. Spine J 2:408–414, 2002. 2. Bohlmann HH, Emery SE, Goodfellow DB, Jones PK: Robinson anterior cervical discectomy and arthrodesis for cervical radiculopathy: Long-term follow-up of one hundred and twenty-two patients. J Bone Joint Surg 75(A):1298–1307, 1993. 3. Bolesta MJ, Rechtine GR, Chrin AM: One- and two-level anterior cervical discectomy and fusion: The effect of plate fixation. Spine J 2:197–203, 2002. 4. Samartzis D, Shen FH, Lyon C, et al: Does rigid instrumentation increase the fusion rate in one-level anterior cerical discectomy and fusion? Spine J 4:636–643, 2004. 5. Hilibrand AS, Carlson GD, Palumbo MA, et al: Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg 81A:519–528, 1999. 6. Robbins MM, Hilibrand AS: Post-arthrodesis adjacent segment degeneration. In Vaccaro A, Anderson DG, Crawford A, et al. (eds): Complications of Pediatric and Adult Spinal Surgery. New York, Marcel Dekker, 2004, pp 63–86. 7. Hilibrand AS, Carlson GD, Palumbo MA, et al: Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg 81A:519–528, 1999. 8. Ishihara H, Kanamori M, Kawaguchi Y, et al: Adjacent segment disease after anterior cervical interbody fusion. Spine J 4:624–628, 2004. 9. Baba H, Furusawa N, Imura S, et al: Late radiographic findings after anterior cervical fusion for spondylotic myeloradiculopathy. Spine 18:2167–2173, 1993. 10. White AA, Panjabi MM: Clinical Biomechanics of the Spine, 2nd ed. Philadelphia, Lippincott Williams and Wilkins, 1990, pp 85–125. 11. Sengupta DK, Demetropoulos CK, Herkowitz HN, Serhan HA: Instantaneous axis rotation and its clinical importance in a healthy lumbar functional spinal unit. Roundtables in Spine Surgery 1:3–12, 2005.
CHAPTER
30
Cerpass Cervical Total Disc Replacement Scott H. Kitchel, Lukas Eisermann, Alexander W.L. Turner, David Cutter, and G. Bryan Cornwall
K E Y l l l
P O I N T S
Fixed-bearing total disc replacement Ceramic bearing All non-ferrous metal construction
INTRODUCTION To a typical patient, degenerative disc disease of the cervical spine manifests itself as a combination of neck and arm pain. It is wellaccepted that the arm pain can be treated by decompression surgery. When the decompression is achieved through anterior discectomy, the surgeon is left with what may be a structurally unsound condition; thus following the decompression procedure, a stabilization procedure must be performed secondarily. The wellknown solution is to place a structural graft or cage into the intervertebral space. Commonly, this construct is also associated with the placement of a titanium plate to enhance stability and thereby increase the chances for the level to successfully fuse. The solution to the neck pain component has proved to be more elusive. Although the removal of the intervertebral disc does remove a potential pain generator, it does not remove all potential pain generators. The possibility that other nearby tissues are damaged— either as effects secondary to the degenerative condition (for example, arthritic facets due to excess loading secondary to disc collapse) or as separate primary conditions (for example, a ligament sprain) is not eliminated. Furthermore, the possibility that new neck pain is caused by the rigid stabilization should also be considered (for example, muscle atrophy at the fused level or increased stresses at adjacent levels). Finally, the possibility that all neck pain generators cannot be accurately diagnosed is an additional variable in the process. It is thought that by replacing the rigid stabilization (the fusion construct) with a mobile stabilization device that allows bending motions while maintaining appropriate disc height, that some of the pain generators secondary to rigid fixation may not arise. For example, some authors have shown that the tissue pressure in adjacent intervertebral discs is lower after mobile stabilizations than after fusion stabilizations.1,2 An added potential benefit may be that because the stresses and tissue pressures 254
at the adjacent levels are decreased, there exists the potential that the rate of degeneration at the adjacent levels remains at a normal rate, as opposed to an accelerated rate adjacent to rigid stabilizations.3 Although all of these potential benefits make disc replacement procedures seem to be very appealing treatments in comparison to fusion procedures, they do not come without their own set of risks. The typical patient requiring a decompression treatment (and, thus, following stabilization treatment) is 35 to 45 years of age; thus, the chosen stabilization procedure must either last for several decades of life or must be designed in a manner that allows another treatment to be reasonably and safely performed. When evaluating the potential risks associated with mobile stabilization versus the relatively known risks associated with rigid stabilization, the factor that stands out is the overall rate of re-operation of the cervical spine at any level. This is somewhat different than survivorship curves that have been commonly studied in the peripheral joints, because all levels must be taken into consideration, not only the operated level. To that stated end, the key risk variable that the Cerpass design attempts to remove is complications related to wear of the bearing surfaces. Borrowing from the total hip experience, the Cerpass prosthesis (NuVasive, San Diego, CA) has been designed with toughened ceramic bearing surfaces to minimize the potential for wear debris generation. Alternative bearing surfaces for total hip replacement have been examined in many simulator studies and clinical retrieval studies. Although the specific details of individual designs may vary, the articulating surface is essentially a ball and socket joint. Thus, comparisons of the various articulating pairs or wear couples are more appropriate than in the disparate designs of spine arthoplasty. Greenwald and Garino4 reviewed several different wear couples and found orders of magnitudes of difference in the amount of wear from these combinations of bearing surfaces. For purposes of comparison, the data are presented as volume of wear in cubic millimeters per year of loading: 3 l Metal on polyethylene had 55.7mm /year 3 l Ceramic on polyethylene had 17.1mm /year 3 l Metal on metal had 0.9mm /year 3 l Ceramic on ceramic had 0.04mm /year
CHAPTER 30
Ceramic-on-ceramic wear surfaces thus represent significantly reduced amounts of wear than other comparable systems. Another distinct advantage of ceramic articulating surfaces is the improved biocompatibility of ceramic wear debris compared with that of metal or polymer wear debris. Warashina et al.5 performed a comparative study examining the biological effects of placing micron sized wear debris of two types of ceramics: alumina (Al2O3) and zirconia (ZrO2), titanium alloy (Ti6Al4V), and high-density polyethylene (HDP) compared with a control in a murine calvarian defect. The study involved measuring the release of cytokines in response to the wear debris to examine the potential effects of osteolysis. The cytokines or bone resorbing mediators examined included tumor necrosis factor (TNF) and various interleukins including (interleukin-6 [IL-6], interleukin-a [IL-1a], and interleukin-1b [IL1b]) from macrophages. Particles of HDP had the highest reactivity with a threefold increase in levels of IL-6, and the Ti6Al4V group had a twofold increase; both of these were significantly greater than the control. Although the Al2O3 and ZrO2 had slightly elevated levels of IL-6, these were not significantly greater than the control. The biologic reactivity of ceramic wear particles is lower than that of metals and polymers, and because the amount of wear is lower as well, ceramic-on-ceramic bearing surfaces is an attractive solution for cervical spine arthroplasty. INDICATIONS/CONTRAINDICATIONS Cervical total disc replacement is generally indicated for degenerative conditions in the absence of significant structural insufficiencies at the symptomatic level or levels. From a mechanical standpoint, the disc replacement will serve to maintain the intervertebral disc height, while allowing bending and rotational motions. Because the prosthesis restores only a portion of the function of the anterior column, structural deficiencies of the posterior elements at the affected levels should be carefully reviewed and generally be regarded as contraindications for disc replacement surgery. Because the disc replacement allows bending motions to be maintained, adjacent level stresses are generally lower when compared with the stresses adjacent to fusion stabilizations. For the patient who presents with one clearly symptomatic degenerative level and one level that is degenerative and asymptomatic, disc replacement of the symptomatic level only may be an excellent treatment in comparison to either fusing the symptomatic level (which may hasten the degeneration of the adjacent level), fusing both levels, replacing both discs, or a hybrid construct having one level fused and one level replaced. Patients who are thought to not benefit from disc replacement in the cervical spine include those having idiopathic deformities, iatrogenic deformities secondary to removal of portions of the bony elements of the posterior spine, and instabilities related to trauma. Care should be taken to review the types of mechanical conditions that can be treated with different types of devices. Not every disc replacement will have the same indications and contraindications as every other device. For example, one of the theoretical advantages of fixed bearings as opposed to mobile bearings is the relatively higher degree of joint stability that is achieved by the reconstruction. Hence, degenerative deformities may be corrected by a fixed bearing device but perhaps not by a mobile bearing device.
Cerpass Cervical Total Disc Replacement
255
DESCRIPTION OF THE DEVICE The Cerpass device consists of titanium alloy end plates and ceramic bearing inserts (Fig. 30–1). In order to create a stable interface between the titanium and the ceramic, the ceramic component is swaged into the titanium end plate. This places the components into intimate mechanical contact with one another so that the ceramic bearing is fully supported by the metal end plate. The device is provided in three footprints (denoted as small, medium, and large) and five heights per footprint (5, 6, 7, 8, and 9 mm) for a total of 15 devices in the basic kit. Owing to the fixed bearing articulation, the device has a superior and inferior orientation. Titanium alloy and yttria-stabilized zirconia toughened alumina comprise the primary materials from which the device is constructed. A layer of plasma-sprayed hydroxyapatite is applied to the bone-contacting surface of each end plate in order to encourage bony ongrowth and thus appropriate device fixation. A key advantage of the materials chosen in this construction is that there are no materials that can grossly interfere with MRI examinations postoperatively. While there is expected to be some image scattering related to the titanium in the device end plates, the effect compared to prostheses containing stainless steel or cobalt chrome in their construction is negligible. BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES Testing activities related to the Cerpass device are primarily centered on two areas of key interest. First, an evaluation of the ceramic construction and assembly process with the titanium end plates to ensure that the ceramic is of the highest quality and has not been damaged in-process; and second, bench-top mechanical evaluation of the prosthesis for fatigue and wear performance. During the assembly process, the ceramic bearing components are inserted into a pocket in the titanium end plate, and then the pocket is deformed around the ceramic component to firmly capture and secure it by means of a swaging process. The swaging process works by applying several tons of pressure to the titanium in a controlled manner to produce a predictable deformation. This
n
F I G U R E 30–1. Photograph of the Cerpass device.
P A R T
256
I I I
Cervical Total Disc Arthroplasty
were tested according to the American Society for Testing and Materials (ASTM) standard guide for evaluating spinal disc replacement wear (ASTM F2423) and showed on average 0.17 mm3 wear debris generation at 5 million cycles, which is an extraordinarily low amount. To put this into perspective, in a very similar test format, the Bryan prosthesis (Medtronic Sofamor Danek, Memphis, TN) produces wear debris volume at a rate 28 times greater than the Cerpass prosthesis.6,7 The testing reported by Anderson et al showed that in a 10 million cycle test, the Bryan prosthesis produced 9.6mm3 of wear debris. The ASTM standard test is run with a constant load of 100 N to a total range of motion of 15 degrees. CLINICAL PRESENTATION AND EVALUATION
n
F I G U R E 30–2. Cross-section of a test part intentionally cracked to determine manufacturing parameters. The swaging process essentially squeezes the ceramic bearing diametrically.
deformation in turn applies pressure to the ceramic. Some pressure is required in order to prevent the ceramic component from moving around in the pocket (e.g., micromotion), but too much pressure could crack the ceramic (Fig. 30–2). To ensure that the process applies the correct amount of pressure, acoustic emissions are recorded from the devices during assembly. Cracks in the ceramic that are induced by the assembly process are readily identifiable in the acoustic spectra. Properly assembled components have been tested in axial compression and for wear performance. Axial compression tests show that the prosthesis is expected to carry the in vivo load indefinitely without failure. The most compelling property of the ceramic bearing becomes evident upon an examination of the wear data (Fig. 30–3). Devices
Like many articular bearing-based total disc replacements, the Cerpass prosthesis is designed for use in patients having degenerative disc disease of the middle and lower cervical spine. Radicular symptoms are generally treated by decompressive techniques, then the prosthesis maintains the disc height and allows continued motion of the joint. OPERATIVE TECHNIQUE Anesthesia Standard anesthetic techniques for invasive cervical surgery should be used. There are no differences in anesthetic technique from traditional anterior cervical discectomy and fusion surgery. Position The patient is positioned supine on the operating table, with the neck in a neutral position. The C-arm is positioned to allow lateral and anteroposterior views to be obtained during the procedure. Procedure l
TOTAL VOLUMETRIC WEAR (CUMULATIVE) Flexion-Extension (n = 2)
l
Wear volume (mm3)
0.25
l
0.20
l
0.15
l
0.10 0.05 l
0.00 0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
5.5
Million cycles n
F I G U R E 30–3. Wear data. Devices were tested according to the ASTM standard guide for evaluating spinal disc replacement wear (ASTM F2423) and showed on average 0.17 mm3 wear debris generation at 5 million cycles.
l
A standard anterolateral approach is used to expose the operative level. Self-retaining retractor blades are used to retract the longus colli muscles and surrounding tissues. The centerline of the disc is determined and marked. It is helpful to use anteroposterior (AP) fluoroscopy to determine the midline. A pin-based spinal distractor is employed to apply distraction to the operative level. The discectomy is peformed. Kerrison rongeurs and curettes are used to remove bony spurs and excess osteophytes. There is no evidence to suggest that there exists a measurable or functional difference in clinical outcomes if the posterior longitudinal ligament is retained or resected. End plates are flattened for receiving the prosthesis by means of a rasp. Distraction is removed (it is not necessary to remove the pins, only to remove the tension from the distractor) and sizing trials are sequentially placed into the disc space from smallest to largest until a firm, snug fit is found. The largest footprint size possible should always be selected in order to minimize the chance for subsidence.
CHAPTER 30 l
l l
The corresponding size prosthesis is opened and loaded onto the insertion tool. Distraction is reapplied to the operative level. The prosthesis is inserted into the disc space, taking care to align the device with the marked midline. AP fluoroscopy is used to check the midline position. Lateral fluoroscopy is used to verify the position of the posterior border of the prosthesis.
POSTOPERATIVE CARE The postoperative care should be considered routine for the Cerpass prosthesis. The surgeon may choose to prescribe a soft collar to be worn for several weeks postoperatively although this is not generally considered a requirement. If magnetic resonance image (MRI) scanning is desired postoperatively, no special precautions need to be taken. The construction of the prosthesis is entirely titanium and ceramic materials, so there is no chance of dislodging the device with the magnetic field. Image scatter is limited to that typically seen adjacent to titanium implants.
Cerpass Cervical Total Disc Replacement
257
CONCLUSIONS/DISCUSSION Cervical disc replacement appears to be a promising treatment in the armamentarium of the spine surgeon. However, its key promise, the reduction of adjacent level degeneration, compared to fusion treatments remains yet to be proven. The key measure for determining whether cervical disc replacement is truly an improvement over fusion procedures will be the long-term survivorship rate of both the device itself and of the intact intervertebral discs adjacent to it. We expect that the adjacent-level survivorship rates will be reasonably similar for the articulating disc replacements that are available or in development today. However, the survivorship rates of the devices themselves will vary. Devices will be removed for many reasons, some directly related to the device, and some not. One of the key reasons that we anticipate device removal to occur is the generation of wear debris leading to osteolytic reactions. By reducing the wear debris generation to an extremely low level, we believe the Cerpass device has the potential to have a very high survivorship rate. REFERENCES
COMPLICATIONS AND AVOIDANCE Patients having high degrees of instability or nondegenerative (e.g idiopathic) deformities are likely poor candidates for treatment with the Cerpass prosthesis. Additionally, patients having low bone density are at risk of subsidence if treated with an articulating total disc replacement. Patients reporting high levels of neck pain in the absence of radicular pain rarely have good clinical outcomes from surgery. It is unlikely that the only (or indeed the primary) pain generator in these patients is the intervertebral disc, so treatment by removing and replacing the disc generally misses other potential pain generators. ADVANTAGES/DISADVANTAGES: CERPASS Advantages Extraordinarily low wear debris generation Lower bioreactivity of ceramic wear debris than metallic or polymeric systems Good MRI imaging properties Standard, familiar surgical technique Disadvantages As with any ceramic implant, the potential for fractured bearing components exists. Since the device-bone interface is limited to a roughened surface with spikes, patients with high degrees of instability should not be treated with the Cerpass prosthesis.
1. Dimitriev AE, Cunningham BW, Hu N, et al: Adjacent level intradiscal pressure and segmental kinematics following a cervical disc arthroplasty: an in vitro human cadaveric model. Spine 30:1165–1172, 2005. 2. Wigfield CC, Skryzpiec D, Jackowski A, Adams MA: Internal stress distribution in cervical intervertebral discs: The influence of an artificial cervical joint and simulated anterior interbody fusion. J Spinal Disord Tech 16:441–449, 2003. 3. Cummins BH, Robertson JT, Gill SS: Surgical experience with an implanted artificial cervical joint. J Neurosurg 88:943–948, 1998. 4. Greenwald AS, Garino JP: Alternative bearing surfaces: The good, the bad, and the ugly. J Bone Joint Surg Am 83A:68–72, 2001. 5. Warashina H, Sakano S, Kitamura S, et al: Biological reaction to alumina, zirconia, titanium and polyethylene particles implanted onto murine calvaria. Biomaterials 24:3655–3661, 2003. 6. Anderson PA, Sasso RC, Rouleau JP, et al: The Bryan Cervical Disc: Wear properties and early clinical results. Spine J 4:303S–309S, 2004. 7. Anderson PA, Rouleau JP, Bryan VE, Carlson CS: Wear analysis of the Bryan cervical disc prosthesis. Spine 28:S186–S194, 2003.
CHAPTER
31
KineflexjC Cervical Artificial Disc James Robert Rappaport
K E Y l
l
l
l
l
P O I N T S
KineflexjC is a total artificial disc designed to relieve symptoms and restore disc height while maintaining motion in the cervical spine. KineflexjC is a metal-on-metal disc that has a semiconstrained core that allows for translation during flexion-extension, lateral bending, and axial rotation. Early clinical experience with the KineflexjC artificial disc shows promising results for patients with degenerative disc disease (DDD). The implant technique for KineflexjC is straightforward and similar to anterior cervical discectomy and fusion (ACDF) and to the Kineflex lumbar disc arthroplasty. KineflexjC is a promising alternative to cervical disc fusion because it may minimize adjacent-level disc disease.
INTRODUCTION Cervical neck pain and neurologic symptoms associated with degenerative disc disease (DDD) are a significant cause of lost work time and account for very substantial costs to the healthcare system. Although the majority of simple disc herniation cases may resolve spontaneously or with conservative management, the progressive nature of cervical DDD is underscored by the research that finds 97% of patients with pre-existing DDD demonstrate progression 10 years later.1 Although a variety of therapeutic interventions are available for cases that do not respond to conservative therapy, including discectomy and various methods of interbody fusion, each presents limitations. Because of the potential for long-term complications associated with traditional surgical treatments, extensive research has been conducted to develop an intervertebral disc prosthesis. CURRENT TREATMENT MODALITIES Traditionally, options for management of cervical DDD were principally limited to either conservative treatment (e.g., rest, therapy, analgesics) or removal of the disc through discectomy, with or without fusion at the affected level. A well-established, commonly performed cervical fusion procedure is anterior cervical discectomy and fusion (ACDF), which may be supplemented with anterior cervical plating or rigid internal fixation to further promote fusion. 258
In recent years, particularly due to increasing concern regarding the incidence of adjacent segment disease after fusion,1 considerable research has been focused on the development of cervical arthroplasty. Each of these surgical treatment modalities is discussed in the following section. Discectomy When conservative management does not yield adequate symptom relief, the first surgical measure that often may be attempted is a discectomy. Because damaged disc material may impinge on the spinal nerves, removal of this material frequently alleviates symptoms by eliminating the compression of spinal nerves. Satisfactory clinical results following cervical discectomy have been reported in the literature.2 However, discectomy is not designed to resolve the patient's underlying pathology. As a result, discectomy is often combined with spinal fusion. Spinal Fusion Spinal fusion effectively eliminates the motion segment between two vertebrae by use of a bone graft, thereby providing improved stability and decreased pain. A variety of graft materials may be used, including autograft bone, allograft bone, and others. Fusion may also involve use of instrumentation to stabilize the affected level and/or contain the graft, such as interbody cages and plating. Clinical success rates, using measures such as the Oswestry Disability Index (ODI) or the Neck Disability Index (NDI), are generally lower than fusion rates. One study, for example, found a 95% fusion rate at 6 months in patients treated with ACDF and a titanium fusion cage, but the success rate dropped to 70%, using an ODI score of 40 or less as the measure of success.3 A complication that has been the subject of increasing concern in the spinal literature is the possibility of the acceleration of adjacentsegment disease after cervical fusion, due to increased stress on adjacent unfused levels. Hilibrand et al4 reported that 2.9% of patients per year required surgical intervention for symptomatic new onset adjacent-segment degeneration following ACDF; these researchers estimated a 25% cumulative rate of symptomatic adjacent-segment disease 10 years after ACDF.
CHAPTER 31
Although the use of allograft material eliminates donor site complications, it carries a small risk of transmissible disease, as well as a risk of rejection reactions. Bone morphogenetic protein is another potential grafting option that became available in 2002 for use in the lumbar spine when U.S. Food and Drug Administration (FDA) approved Medtronic's InFUSE (Medtronic Sofamor Danek, Memphis, TN). Although the results of clinical trials are promising, this product is not yet approved in the United States for use in the cervical spine. Rigid internal fixation devices have been used increasingly in addition to bone grafting in order to increase fusion rates. Anterior cervical plate fixation has been shown to improve successful arthrodesis rate after single-level ACDF. One study reported that fusion rates improved from 90% to 96% when cervical plate fixation was added to single-level fusions with allograft.5 The same study also showed improvement from 72% to 91% in fusion rates for two-level cases with plating. The implantation of metallic spinal cages into the disc space between two vertebrae is another method used to stabilize the spine and promote fusion. The cage also may be filled with graft material. Although fusion rates with the use of cages is generally greater than 90% in one-level procedures, complications do occur. Moreover, success rates with fusion cages decrease when the definition of success includes other factors in addition to fusion. The Investigation Device Exemption (IDE) clinical study of 202 patients who received the AFFINITY Anterior Cervical Cage System reported an overall success rate of only 68% at 24-month follow-up. Overall success was defined as fusion of the operative segment, successful pain and disability outcome, neurological success, and no revisions, removals, or supplemental fixation. Complications included 15 neurologic events, 29 spinal events, and 35 reports of neck or arm pain. Total Disc Replacement The limitations of discectomy and spinal fusion procedures have resulted in significant interest in developing a total disc replacement. The goal is to develop an artificial disc that alleviates the pain associated with disc degeneration while preserving segmental range of motion and restoring stability. KineflexjC (Spinal Motion, Inc., Mountainview, CA) is a metalon-metal, semiconstrained cervical artificial disc designed to relieve pain and maintain motion for the treatment of degenerative disc disease of the cervical spine. It has been used clinically since 2004, with more than 1,000 discs implanted worldwide. KineflexjC is being investigated under an IDE clinical study in the United States that began in the summer of 2005, with completion of enrollment expected in the second quarter of 2007.
l
l
KineflexjC Cervical Artificial Disc
259
Have at least 6 months of prior conservative treatment, the presence of progressive symptoms, or signs of nerve root compression Have a NDI demonstrating moderate disability (40 or higher)
Exclusion Criteria Overview l
l l
Marked cervical instability on lateral or flexion-extension X-ray study l Nondiscogenic neck pain or nondiscogenic source of symptoms l Radiographic confirmation of severe facet disease l Bridging osteophytes l Less than 2 degrees of motion at index level Severe myelopathy Metabolic bone disease
DESCRIPTION OF THE DEVICE The KineflexjC Spinal System is designed to be used as a replacement for a degenerated or diseased cervical disc at one level from C3-C7 that is unresponsive to conservative management in subjects with single-level degenerative disc disease (DDD) with related pain. The spinal system is a three-piece modular design consisting of two cobalt chrome molybdenum (CCM) end plates and a fully articulating CCM core (Fig. 31–1). The system is available in two footprint sizes (size 1, 14 mm 16 mm; and size 2, 16 mm 18 mm). The end plates are designated in three thicknesses relative to the disc (core) centerline and allow for assembled heights of 5.7 mm, 6.3 mm, and 7.1 mm (Fig. 31–2). Each end plate exterior has a keel with two holes, a serrated edge surface, and a titanium plasma spray coating for bone ingrowth. The interior of each end plate has a polished concave bearing surface for evenly distributed contact with the convexshaped core. The inferior end plate has a retaining ring to prevent extrusion of the core during movement (Fig. 31–3). The KineflexjC core is manufactured of CCM to which has been applied a highly polished finish. Only one size core is used with all the system combinations. The spinal system is implanted as one unit by means of the insertion instrument.
INDICATIONS/CONTRAINDICATIONS Appropriate patient selection is important for achieving optimal clinical outcomes. A summary of patient selection criteria from the IDE clinical study follows: Inclusion Criteria Overview l l l
Between 18 and 60 years of age Symptomatic disc at only one cervical level from C3-C7 Have symptoms of radiculopathy in neck, one or both shoulders, and one or both arms
n
F I G U R E 31–1. KineflexjC cervical disc. The preassembled disc is shown, with core sitting in the inferior end plate. The disc is inserted assembled as one piece.
260
P A R T
n
I I I
Cervical Total Disc Arthroplasty
F I G U R E 31–2. KineflexjC cervical disc. Various viewpoints of the disc are shown as an assembled unit.
Static Shear
As in the compression test above, the test specimen was loaded until functional failure. Load and displacement data were recorded. No mechanical failures of the cores or end plates were observed. Dynamic Testing Compression Fatigue Testing
n
F I G U R E 31–3. Cervical disc core in inferior end plate. The lower lip of the core sits in the inferior end plate, which has a matching interior surface designed to prevent the core from extruding out in situ.
BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES A substantial body of preclinical mechanical testing has been performed on the KineflexjC Spinal System, including static testing, monoaxial fatigue testing, and wear testing. These tests were performed in order to simulate the load and movement to which the discs would be exposed under in vivo conditions and to verify that the prosthesis could withstand static and fatigue load conditions, as well as to determine the wear characteristics of the prosthesis. All of the tests were conducted in accordance with the protocols reviewed by FDA, and were based on the draft American Society for Testing and Materials (ASTM) standard for artificial disc testing. Static Testing Static testing was performed in two loading conditions, axial compression and shear, in accordance with the ASTM Artificial Disc Testing Draft Standard. Static Compression
The assembled disc was placed in an Instron machine with a 0- to 100- kN load capacity. The device was then loaded until either the maximum load permitted by the test fixture was reached, or mechanical failure of one of the device components occurred. Load and displacement data were recorded. The results demonstrated that there was no height reduction in any of the samples. The KineflexjC substantially exceeded the strength of the vertebral bone and, therefore, was sufficient to withstand the worst case compressive forces anticipated in clinical use.
Samples were tested under cyclic axial compressive loading to assess the suitability of the fatigue resistance of the device for in vivo use. Loads varied cyclically. The results demonstrated no measurable dimensional or mass changes in either the end plates or the core. Shear Fatigue Testing
Samples were tested under cyclic shear compressive loading to assess the suitability of the fatigue resistance of the device for in vivo use. Loads varied cyclically. The results demonstrated no measurable dimensional changes or mass changes in either the end plates or the core. Wear Testing
To evaluate the amount and size of wear particles generated by the KineflexjC in vivo, test samples were cyclically loaded in a multiaxial motion simulator for 10 million cycles. A custom-loading fixture was constructed to test the prosthesis under a combination of cyclic flexion-extension, lateral bending, and rotation, corresponding to the types of in vivo movement that may be encountered. To simulate in vivo cyclic loading conditions, ranges of motion were selected to conform to the ASTM Artificial Disc Testing Draft Standard. Weighing and dimensional measurements of the prosthesis (end plates and core) performed after every million cycles showed an average volumetric loss for the prosthesis was approximately 3.84 mm3 over the entire test, or 0.384 mm3 per million cycles. Mass loss was approximately 32 mg, for an average of 3.2 mg per million cycles. This represents loss of only approximately 0.08% of the total prosthesis mass over 10 million cycles. When the the observed wear rate for the KineflexjC was compared with other results that have been reported in the literature, the wear rates are similar to other metal-metal discs, and significantly below volumetric wear rates reported for total hip arthroplasties (THAs). For example, Oskouian et al6 reported a volumetric wear rate of 0.96 mm3/million cycles in testing of an all-metal artificial disc. The authors note that this is approximately two orders of magnitude below the wear rates reported for metal THAs, which
CHAPTER 31
range from 50 to100 mm3/million cycles. The KineflexjC volumetric wear rate of 0.384 mm3/M cycles is approximately one-third of the rate reported by Oskouian et al.6 Thus, both the materials used in the KineflexjC and the amount of wear debris generated are consistent with previous prostheses that are currently undergoing clinical studies. In addition, there are in vivo animal data demonstrating that wear debris of this type and quantity are unlikely to generate an adverse biologic response. Cunningham et al evaluated the neural and systemic tissue response to cobalt alloy particulate debris in an in vivo rabbit model up to 6 months. They placed 4 mg of cobalt alloy particles directly on the dura and compared the results with rabbits who had a sham procedure of dural exposure alone. At 3 months, the number of macrophage-expressing cytokines localized within the spinal cord and overlying tissues indicated no significant differences compared with the control group. Despite regions heavily laden with metallic particulate, histiocyctic reaction, and cytokine activity, the spinal cords indicated normal distribution of myelin and the intracellular neurofibrilla network. There was no evidence of cellular apoptosis, and all specimens were characterized as without significant histopathologic changes. Finally, metal-on-metal THAs have been used clinically for more than 10 years with a strong safety record. Tipper indicated that histologic studies of periprosthetic tissues have not shown an inflammatory reaction to metal wear particles, and that MOM bearings show considerable potential as an osteolysis-free solution for younger patients. In summary, the wear testing of the KineflexjC demonstrates a wear rate that is similar to but generally lower than other allmetal disc prostheses that have been under clinical evaluation. Prior in vivo animal testing demonstrates that direct application of particles of the same material, and in extreme doses, to the dura does not trigger a significant adverse biologic response. Therefore, the observed wear characteristics of the KineflexjC were determined to be appropriate for its intended use. CLINICAL PRESENTATION AND EVALUATION Of the group of patients who are candidates for total disc replacement, there is a subset of those patients who would normally be considered candidates for ACDF. Just as when fusion is considered, there are multiple factors that can cause patients with similar pathology and physical findings to be categorized into radically different prognostic groups. Some of these factors are well known such as: workers’ compensation issues, litigation, and socioeconomic factors. Other risk factors relate to physiology, such as a history of smoking. Unlike fusion, the total disc replacement does not stabilize the entire motion segment, so patients with preoperative instability may not be as effectively treated with a total disc replacement as with fusion. ACDF also effectively immobilizes the entire motion segment, including the facet joints. Therefore fusion has the potential to address both disc pathology and facet pathology. Cervical disc replacement addresses disc pathology and may improve facet pain to some degree, but it does have the potential to actually worsen any facet-mediated pain through increasing mobility at the motion segment. The advent of the disc replacement has caused a shift in the consideration of the role of the facet joint in the pain mediation
KineflexjC Cervical Artificial Disc
261
process. When evaluating a patient's candidacy for disc replacement, it is imperative that the surgeon first rule out facet-mediated pain and pathology. A total disc replacement will not immobilize or replace these joints. Plain x-ray studies, including flexion-extension views, may help rule out gross instability. The advent of total disc replacement has caused a greater scrutiny of the role of facetmediated pain and has made clear the need for a grading system for degrees of facet pathology. It is hoped that such a grading system can distinguish between acceptable facet disease and facet disease that is too far advanced to be treated with total disc replacement. At present, a determination of clinically significant facet disease can be attempted by 1. Evaluation of the patients’ symptoms, the nature of their pain, and aggravating factors. For example, pain with flexion usually indicates pain mediated by loading of the disc. Pain with extension may indicate pain mediated by facet joint loading. 2. Facet anesthetic injection or median branch blocks may also help determine if the facets are a significant part of the pain generator for an individual patient. Preoperative evaluation of bone quality is just as important if not more so in the patient work-up for total disc replacement than it is for fusion. Bone quality must be adequate to support the disc end plates. All patients with risk factors for osteoporosis are checked preoperatively with a dual-energy x-ray absorptiometry scan. Patients who have had previous anterior cervical surgery, whether related to the spine or unrelated, such as thyroidectomy, should undergo an ear, nose, and throat consultation to rule out an asymptomatic unilateral vocal cord paralysis. Doing so may help avoid the catastrophic complication of bilateral vocal cord paralysis. If a unilateral vocal cord paralysis is noted, then the surgical approach is planned from that side in order to avoid the unaffected recurrent laryngeal nerve on the opposite side. In the patient with bilateral normal functioning vocal cords, the side of approach is at the surgeon's discretion. Traditional teaching and older literature has stated that the left-sided approach was safer because the recurrent laryngeal nerve on the left side was anatomically somewhat less at risk. More recent studies have shown no difference between right- and left-sided approaches.7 Specific patient selection criteria from the KineflexjC IDE clinical study were presented earlier in this chapter. OPERATIVE TECHNIQUE The basic soft tissue dissection and exposure of the spine is the same for one-level ACDF and total disc replacement. The exposure required for total disc replacement may be somewhat less than that required for fusion. Therefore, a slightly smaller incision is generally made. There is less need for the extensive elevation of the longissimus coli muscles in a total disc replacement as compared with an anterior cervical discectomy and fusion in which exposure is required for an anterior plate. Although the third generation of cervical plates may be low profile, the artificial disc is a no-profile device. There is also no anterior footprint. There is no need for soft tissue dissection anterior to the vertebral body such as is required for cervical plating. Many of the techniques that ensure a safe ACDF are also used when performing a total disc arthroplasty.
262
P A R T
I I I
Cervical Total Disc Arthroplasty
Anesthesia Anesthesia is general endotracheal. Fiberoptic intubation is considered when a patient's range of motion while awake results in unacceptable neurologic symptoms. When self-retaining retractors are put into place, the endotracheal cuff pressure should be readjusted. This helps decrease the incidence of dysphonia.7 These procedures have very little blood loss. There is no need for the additional risk of hypotensive anesthesia. Position The patient is positioned with the neck in a neutral position. The posterior aspect of the neck is supported with a firm contoured roll. The chin and neck are taped to secure their position. The shoulders are taped in a depressed position to allow intraoperative lateral x-ray study. The tape has been shown to be more secure than wrist band pulls and prevents over-distraction, which can result in nerve root or brachial plexus injury. Casper distracter pins are used to apply distraction and control lordosis or kyphotic movement at the involved disc space. Blunt self-retaining retractor blades are used against the esophagus. The small sharp-toothed blade is used towards the carotid artery but care is taken to place both blades underneath the elevated longissimus coli musculature.8 Accurate anteroposterior (AP) and lateral fluoroscopic views are crucial to technique even more so than with performing an anterior discectomy and fusion. Points specific to the KineflexjC surgical technique include an instrumentation set and implant design, which are very similar to the Kineflex lumbar disc design. This allows a surgeon to gain competency and confidence much quicker than would happen if he or she were required to learn two distinctly different techniques. The modified mid-height keel allows for a more accurate initial intraoperative placement than a non-keeled device, yet avoids the potential complication of vertebral body fracture and compromise that has been associated with full-height keeled implants. Because the disc goes in as one unit, there is no necessity for over-distraction to place the central core. This helps to reduce injury to soft tissues and ligaments, which otherwise could result in increased postoperative instability and pain. PROCEDURE On completion of the disc space preparation, the implantation of the KineflexjC artificial disc requires a straightforward six-step process: S T E P 1 : The first step is the determination of the vertebral body footprint using the end plate sizing instrument. Two footprint sizes are available (size 1, 14 mm 16 mm; and size 2, 16 mm 18 mm). This step is confirmed under fluoroscopy in a lateral view to confirm the posterior aspect of the end plate, and viewed visually from the anterior aspect to confirm boney coverage of the end plate (Fig. 31–4). S T E P 2 : The second step determines the disc height of the prosthesis using the distraction wedges. Three choices are available (5, 6, and 7). This step is performed under fluoroscopy in a lateral view, comparing the distraction wedge with the adjacent
n
F I G U R E 31–4. End plate sizing. Measurement of vertebral body for maximum and optimal coverage.
levels, with the intent of restoring the disc space back to a normal height (Fig. 31–5). S T E P 3 : The third step uses the midline verification instrument. This instrument is constructed from radiolucent material with metal markers on each side of the instrument shaft. Under fluoroscopic visualization in the AP view, the markers are used to locate the midline position by lining them up with the pedicles of the vertebrae above and below. The midline is then marked with a bovie (Fig. 31–6). S T E P 4 : The fourth step is the midline slot cut, performed under fluoroscopy in the lateral view, which is impacted until the required depth is achieved (Fig. 31–7). S T E P 5 : The fifth step is the initial insertion of the KineflexjC prosthesis. The prosthesis is assembled and inserted as one piece. The initial insertion places the disc halfway into the disc space, ideally with one of the keel holes embedded in the vertebral body, with the other still exposed (Fig. 31–8).
n
n
F I G U R E 31–5. Distraction. Disc height is determined.
F I G U R E 31–6. Midline verification. Specialized instrument determines midline location before cutting keel space.
CHAPTER 31
n
F I G U R E 31–7.
KineflexjC Cervical Artificial Disc
Midline slot cut. Keel space is cut using midline
location.
n
F I G U R E 31–10. Final placement. Optimal final placement of
disc over end plate.
n
263
F I G U R E 31–8. Initial insertion. Disc is partially inserted into
disc space. n F I G U R E 31–11. Final placement. Optimal orientation and location. End plates are parallel.
STEP 6: The final step places the prosthesis within 1 mm of the
posterior aspect of the vertebral body. The placement device is used to gently move the disc posteriorly with minute movements, either as a whole disc assembly, or one end plate at a time, for an accurate final placement (Figs. 31–9, 31–10, and 31–11). The following MRI and x-ray images represent the treated disc space both before and after implantation. The disc herniation is seen at L6-L7 (Figs. 31–12 and 31–13). After implantation of the KineflexjC disc, disc height and mobility are restored. X-ray images of the disc are shown in lateral bending and flexionextension positions (Figs. 31–14 and 31–15).
n
F I G U R E 31–9. Final placement. Final advancement for optimal orientation and location of the disc.
n
F I G U R E 31–12. The disc herniation is seen at C6-C7.
264
P A R T
I I I
Cervical Total Disc Arthroplasty
existed preoperatively. Follow up x-ray studies are done on a routine basis. COMPLICATIONS AND AVOIDANCE
n
F I G U R E 31–13. The disc herniation is seen at C6-C7.
POSTOPERATIVE CARE Patients use a soft collar for comfort, psychological support, and minimal physical support. Early active range of motion is encouraged. Passive range of motion through physical therapy is not required and is avoided. Patients are placed on a mechanical soft diet for 10 days or until swallowing becomes relatively normal.9,10 Postoperative physical therapy is prescribed on an individualized basis and is dependent upon the functional deficits that
A recent retrospective study using the Nationwide Inpatient Sample reported the incidence of complication with cervical surgery for spondylosis to be overall 3.93%. The incidence of complication with an anterior approach alone was lower than that with posterior fusion or combined anterior and posterior approaches.11 For anterior cervical discectomy in the State of California, complication rates include 1.8% infection, 0.3% cerebral spinal fluid leak, 0.09% recurrent nerve palsy. These complications can be avoided or reduced by surgically respecting the soft tissues. Infection can be decreased by use of prophylactic antibiotics and copious use of irrigation. Dysphonia can be avoided by adjusting the endotracheal cuff tube pressure during the procedure. Potentially devastating bilateral recurrent laryngeal nerve injury can be avoided by the vocal cords of any patient who has had previous anterior spine or neck surgery and choosing the side of approach appropriately. Dysphagia and dysphonia can be reduced by placement of self-retaining retractors under the longissimus coli, and recalibration of the endotracheal tube cuff pressure after placement of the retractors. Complications specific to disc replacement include subsidence of the implant through the bony end plate. Careful preservation of the end plate and placement of the implant can lower this complication. Carefully balancing the soft tissue release around the disc space both anteriorly and posteriorly can help to avoid extrusion of the
n F I G U R E 31–14. X-ray images of the disc are shown in lateral bending position. (Courtesy of David B. Musante, MD, Triangle Orthopedic Associates, P.A., Durham, NC.)
CHAPTER 31
KineflexjC Cervical Artificial Disc
265
n
F I G U R E 31–15. X-ray images of the disc are shown in flexion-extension positions. (Courtesy of David B. Musante, MD, Triangle Orthopedic Associates, P.A., Durham, NC.)
disc replacement implant anteriorly (i.e., the watermelon seed phenomenon). Microscopic ACDF at one level is very safe and effective. But the risk of adjacent-segment disease after cervical fusion is an unacceptably high 2.9% per year in the first 10 years after treatment. The real advantage of total disc replacement may be seen in the long run if the incidence of adjacent-segment disease is decreased.12–14 Ironically some patients who highly value a return to normal cervical motion, such as an athlete in contact sports, may not benefit from the increased motion that a total disc replacement provides. A single-level ACDF has been documented to provide the stability needed to safely return to these types of activities.15 The edge of the indication envelope for total disc replacement remains to be defined. Additional challenges include increasing
ADVANTAGES/DISADVANTAGES: KINEFLEXjC. Advantages Ease of implant Restoration of disc height Preserve motion CCM is a durable material Plasma titanium sprayed for bone in-growth fixation Reproducible procedure Eliminate need for bone graft/complications such as nonunions Disadvantages Imaging can be obstructed at the implant level Lack of long-term clinical data
MRI compatibility. Challenges already met by the KineflexjC include the following: l Development of an easily reproducible safe and reliable implantation technique. l Development of an implantation technique, which reliably restores disc space height and provides for a quick postoperative recovery. l Development of a total disc implant that restores near-anatomic motion to the diseased spinal segment.
REFERENCES 1. Hilibrand AS, Carlson GD, Palumbo, MA, et al: Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg 81A:519–528, 1999. 2. Laing RJ, Ng I, Seeley HM, et al: Prospective study of clinical and radiological outcome after anterior cervical discectomy. Br J Neurosurg 15:319–323, 2001. 3. Moreland DB, Asch HL, Clabeaux RT, et al: Anterior cervical discectomy and fusion with implantable titanium cage: Initial impressions, patient outcomes and comparison to fusion with allograft. Spine J 4:184–191, 2004. 4. Hilibrand AS, Yoo JU, Carlson GC, et al: The success of anterior cervical arthrodesis adjacent to a previous fusion. Spine 22:1574–1579, 1997. 5. Kaiser MG, Haid RW, Subach BR, et al: Anterior cervical plating enhances arthrodesis after discectomy and fusion with cortical allograft. Neurosurgery 50:229–238, 2002. 6. Oskouian RJ, Whitehall R, Sami A, et al: The future of spinal arthroplasty: a biomaterials perspective. Neurosurg Focus 17:E2, 2004.
266
P A R T
I I I
Cervical Total Disc Arthroplasty
7. Apfelbaum RI, Kriskovich MD, Haller JR: On the incidence, cause and prevention of recurrent laryngeal nerve palsies during anterior cervical spine surgery. Spine 25:2906–2912, 2000. 8. Watkins R: Surgical Approaches to the Spine. 9. Riley LH, Skolasky RL, Alvert TJ, et al: Dysphagia after anterior cervical decompression and fusion: prevalence and risk factors from a longitudinal covert study. Spine 3022:2564–2569, 2005. 10. Rhyne AL, Siddiqi F, Darden DV, et al: Incidence of Postoperative Dysphagia Following Total Disc Replacement Versus Anterior Discectomy and Fusion with Instrumentation. Presented at the Cervical Spine Research Society, December 2005. 11. Marjorie C, Wang MD, Leighton Chan MPH: Complications and mortality associated with cervical spine surgery for degenerative disease in the United States. Spine 32:342–347, 2007.
12. Hillibrand AS, Robbins M: Adjacent segment degeneration and adjacent segment disease: The consequences of spinal fusion? Spine J 6:190S–194S, 2004. 13. D'Mitriev AE, Cunningham BW, Hu N, et al: Adjacent level intradiscal pressure and segmental kinematics following a cervical total disc arthroplasty: and in vitro human cataveric model. Spine 30: 1165–1172, 2005. 14. Eric JC, Humphries SC, Lynn PH, et al: Biomechanical study on the effect of cervical spine fusion on adjacent level intradiscal pressure and segmental motion. Spine 27:2431–2434, 2002. 15. Voccaro AR, Klein GR, Ciccoti M, et al: Return to play criteria for the athlete with cervical spine injuries resulting in stinger and transient quadriplegia/paresis. Spine J 2:351–356, 2002.
CHAPTER
32
DISCOVER Artificial Cervical Disc* Douglas G. Orndorff, Kornelis A. Poelstra, and Todd J. Albert
K E Y l
l
l
l
P O I N T S
The DISCOVER Artificial Cervical Disc is a ball-and-socket design consisting of superior and inferior end plates manufactured from titanium alloy. The DISCOVER Artificial Cervical Disc features a spherical bearing surface between the titanium end plates and the ultra-highmolecular-weight polyethylene (UHMWPE) core. This bearing surface allows motion in all rotation directions—flexion, extension, lateral bending, and axial rotation. Each DISCOVER Artificial Cervical Disc assembly has been designed to provide 7 degrees of lordosis when implanted in the cervical spine and will be available in heights from 6 mm to 9 mm in 1-mm increments to accommodate varying patient anatomy. Previous neurotoxicity studies and reported clinical literature assessments for risk of osteolysis suggest that the wear rate and particle sizes in the DISCOVER Artificial Cervical Disc will be well tolerated by the body.
spondylotic disease, at this time, this technology is probably best suited for patients with degenerative changes limited to a single spinal segment.1 Contraindications to cervical disc replacement include any history of tumor or infection, significant osteoporosis, or any kyphotic deformity. In addition, arthroplasty is not recommended for patients with restricted preoperative range of motion (40) Pregnancy or expected pregnancy within 3 years Active infection Medications that retard healing (e.g., steroids) Autoimmune diseases (e.g., rheumatoid arthritis) Systemic diseases (e.g., AIDS, HIV, hepatitis) active malignancy
AP, anteroposterior; DEXA, dual-energy x-ray absorptiometry; FDA, U.S. Food and Drug Admmistration. ML, mediolateral, MRI, magnetic resonance imaging.
A
B
C n
F I G U R E 39–1. A to C, The ProDisc-L artificial lumbar disc. (Synthes, West Chester, PA.)
319
320
P A R T
I V
Lumbar Total Disc Arthroplasty
BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES The first ProDisc-L implantation was performed in 1999. Since then, more than 16,000 prostheses have been implanted worldwide at the time of writing of this chapter. Multilevel disc replacements have also been performed (Fig. 39–2). In the original European studies, there have been no device-related failures reported. In the United States, FDA-supervised multicenter clinical trials and 2-year follow-up have been completed, culminating in full FDA approval for human implantation in the United States in August 2006. The U.S. Investigational Device Exemption Trial Table 39–1 lists the eligibility criteria for the U.S. IDE study on spinal arthroplasty with the ProDisc-L device versus lumbar fusion. Table 39–2 lists some of the demographic characteristics of the patients enrolled in the U.S. multicenter IDE study. Results for the IDE multicenter study 2-year results were first reported at the American Academy of Orthopaedic Surgeons annual meeting.8 This included 162 patients who underwent disc replacement and 80 patients who underwent fusion. Randomization was performed at a 2:1 ratio to disc replacement versus circumferential fusion. Pain, disability, and ROM were evaluated at preoperative, 6 weeks, and 3, 6, 12, 18, and 24 months follow-up visits. Table 39–3 summarizes the results in terms defined by the FDA. Although pain on the Visual Analog Scale (VAS) decreased significantly in both disc replacement and fusion, there was no defined success criteria based on pain relief alone. Based on a 15% reduction in the Oswestry Disability Index (ODI), the success rate with disc replacement was 77% versus 65% with fusion. Although the study was designed to show at least equivalency in the two techniques, this showed that patients with ProDisc-L did significantly better. Based on a 15-point reduction in ODI, the success rate with ProDisc-L was 68% versus 55% with fusion. These showed an even greater margin of success of ProDisc-L over fusion. The failure rate, defined by reoperations, revisions, and removal or addition of devices, was low and no different between the ProDisc-L and fusion cases. Success as defined by an improvement in SF-36 showed a 79% success rate with ProDisc-L versus 70% with fusion, another benchmark that approached statistical significance. Finally, by radiographic definition (no migration, no subsidence, no loss of disc height, and ROM), the success rate in ProDisc-L was 92% versus 86% for fusion. With this class I data showing equivalency and, in some cases, superiority of ProDisc-L over fusion, it must be kept in mind that this technology was designed to preserve motion, with the theoretical long-term benefit of retardation of accelerated adjacent-segment degeneration. Table 39–4 lists the sagittal ROM (flexion-extension) at the different follow-up time points. At 24 months, 94% of patients had motion with the physiologic range. The conclusion of the FDA IDE trial was essentially that ProDisc-L preserves ROM without compromising the results as compared with the current surgical standard of fusion,
with the potential upside of decelerating adjacent-segment degeneration. OPERATIVE TECHNIQUE A standard anterior left-sided retroperitoneal approach to the lumbar spine is performed. Any operating table that allows supine positioning and fluoroscopy of the lumbar spine can be used. A small bump or inflatable support may be placed under the small of the patient's back for adjustment of lordosis during surgery to open up the disc space anteriorly. In our institute, we use a mini-incision less than 6 cm for one-level cases and about 8 cm for two levels. Intraoperative fluoroscopy is used throughout the operation to verify the placement of the prosthesis. Once exposure is obtained, an anteroposterior (AP) view confirms the level and identifies the midline, which is then marked with the cautery or osteotome (Fig. 39–3A). A complete discectomy is then performed (see Fig. 39–3B, C). Cartilage is removed from the vertebral end plates. If herniated disc material is identified on the preoperative magnetic resonance imaging scan, this may be removed through the anterior approach. In some cases, the posterior longitudinal ligament may have contracted, preventing re-expansion of the disc space, so this must be released from the posterior vertebral body with a forward-angled curette. Once the normal anatomic height has been restored with distraction under fluoroscopy, a trial is placed to help select the proper disc size, angle, and height (Fig. 39–3D). A sagittal groove is then cut in the vertebral end plates in the exact midline using a chisel placed over the trial (Fig. 39–3E). This groove will accept the central keel of the implant. The trial is removed, and the final implant is then gently impacted into place with an insertion tool (Fig. 39–3F). The insertion tool allows distraction of the disc space for placement of the UHMWPE inlay, which snap-fits into position in the inferior end plate (Fig. 39–3G). After the insertion instrument is removed, gross inspection is made to ensure the UHMWPE inlay is properly flush against the inferior end plate (Fig. 39–3H), and final fluoroscopic views are taken to confirm correct position of the prosthesis. POSTOPERATIVE CARE A soft back brace can be used for the first week or two to allow for wound protection. Otherwise, there is no extensive postoperative protocol. Patients can return to work as soon as they are comfortable, but they should allow 6 weeks before returning to recreational sports or full duty (if the job is physically demanding). COMPLICATIONS AND AVOIDANCE No major technique- or device-related complications were observed. Table 39–5 lists the complications for both the ProDisc-L and fusion patients. There were four cases of device migration, subsidence, or loose polyethylene requiring revision surgery
B
A
C n
F I G U R E 39–2.
A to C, One-, two-, and three-level lumbar disc replacement with ProDisc-L.
322
P A R T
I V
Lumbar Total Disc Arthroplasty
TABLE 39–2. Patient Demographics Patient Characteristic
Fusion (n ¼ 80)
ProDisc-L (n ¼ 162)
P-value (NS is P > 0.05)
Average age in years (std dev) Sex (% male:% female) Body mass index (std dev) Preoperative Oswestry Disability Index (std dev) Target level at screening L3-L4 L4-L5 L5-S1
40.2 (7.6) 46:54 27.4 (4.3) 62.9 (13.4) 10.5 3 (3.8%) 27 (33.8%) 50 (62.5%)
39.6 (8.0) 51:49 26.7 (4.2) 63.4 (12.6) 10.3 3 (1.9%) 54 (33.3%) 105 (64.8%)
NS NS NS NS NS NS NS NS
n, number; NS, not significant; std dev, standard deviation.
TABLE 39–3.
Components of Overall FDA-Defined Success at 24 Months
ODI success By 15% improvement criteria ODI success By 15-point improvement criteria Reoperations/revisions/removal/supplemental fixation Maintenance or improvement of neurologic status SF-36 success (improvement over baseline) Radiographic success (fusion or >5 deg ROM at L3-L4, L4-L5, and >4 deg at L5-S1)
Fusion
ProDisc-L
46/71 (64.8%) 39/71 (54.9%) 2/75 (2.7%) 57/70 (81.4%) 49/70 (70.0%) 59/69 (85.5%)
115/149 (77.2%) 101/149 (67.8%) 6/161 (3.7%) 135/148 (91.2%) 118/149 (79.2%) 131/143 (91.6%)
deg, degree; ODI, Oswestry Disability Index; ROM, range of motion FDA, U.S. Food and Drug Admmistration.
TABLE 39–4. Fusion ProDisc-L
A n
Time Course of Mean Flexion-Extension Range of Motion (Degrees) Month 3
Month 6
Month 12
Month 18
Month 24
1.0 6.3
0.9 6.1
0.9 7.0
0.8 7.1
0.7 7.7
B
F I G U R E 39–3. A, Marking of midline. B, Discectomy performed all the way back to the posterior longitudinal ligament.
CHAPTER 39
C
E
ProDisc-L Total Disc Replacement
323
D
F
C, Discectomy performed all the way back to the posterior longitudinal ligament. D, Trialing for size, height, and lordosis. E, Chisel cut for the keels. F, Placement of end plates in collapsed form. n
F I G U R E 39–3. Cont'd
324
P A R T
I V
Lumbar Total Disc Arthroplasty
G
H
G, Distraction of end plates and locking of the polyethylene inlay. H, The construct is inspected to ensure that there is no step or gap between the polyethylene inlay and the inferior end plate. n
F I G U R E 39–3. Cont'd.
TABLE 39–5.
Complications From the U.S. IDE Trials of ProDisc-L Versus Fusion
Complication
Fusion
ProDisc-L
Clinically significant blood loss (>1,500 mL) Dural tear Edema Gastrointestinal (e.g., ileus) Genitourinary Infection (all superficial) Migration, not requiring surgery Migration, requiring surgery Motor deficit at index level Numbness at index level Reflex change Retrograde ejaculation Subsidence, not requiring surgery Subsidence, requiring surgery Venous thrombosis, deep Vessel damage/bleeding
2 (2.5%) 2 (2.5%) 3 (3.8%) 22 (27.5%) 4 (5.0%) 2 (2.5%) 1 (1.3%) 0 (0.0%) 0 (0.0%) 1 (1.3%) 0 (0.0%) 1 (1.3%) 1 (1.3%) 0 (0.0%) 1 (1.3%) 6 (7.5%)
0 (0.0%) 0 (0.0%) 8 (4.9%) 32 (19.8%) 14 (8.6%) 0 (0.0%) 3 (1.9%) 4 (2.5%) 4 (2.5%) 0 (0.0%) 1 (0.6%) 2 (1.2%) 2 (1.2%) 0 (0.0%) 2 (1.2%) 5 (3.1%)
IDE, Investigational Device Exemption.
CHAPTER 39
in the ProDisc-L group. There were four cases of motor deficits at the index level with ProDisc-L, and this may be related to the slightly more meticulous access and retraction necessary for the device compared with femoral ring allograft insertion. There was a very low rate of retrograde ejaculation in both surgery groups. ADVANTAGES/DISADVANTAGES: PRODISC-L TOTAL DISC REPLACEMENT Advantages Semi-constrained motion is facet protective Good fixation features (keels, coating)—stays where you put it, so salvage would simply need posterior fusion. Familiar cobalt-chrome and polyethylene materials No device protrusions beyond disc space Class I FDA clinical data available Multilevel disc replacements possible and clinical data also available (2 and 3 levels) Good experience: over 16,000 implanted worldwide at time of writing (Aug 2006)
Disadvantages Greater device-bone interface loading Would be difficult to remove, both from a revision approach standpoint and because of fixation Long-term data (5 to 10 years) from class I data are still pending Questionable risk of polyethylene debris Anecdotal reports of vertebral fractures (in small patients)
FDA, US Food and Drug Admmistration.
CONCLUSIONS/DISCUSSION This particular device has been used extensively for multilevel use, with as good or better clinical outcomes as compared to singlelevel surgeries, and with good preservation of motion at each
ProDisc-L Total Disc Replacement
325
replaced level, with good preservation of spinal alignment at each replaced level, and with great patient satisfaction. The technique and instrumentation are facile and streamlined. The experience with the ProDisc-L Artificial Disc and class I data now released by the FDA suggest that lumbar disc replacement is a viable surgical alternative to fusion for disc degeneration, with preservation of motion and alignment at the treated levels, and without compromising clinical outcomes. Although it is yet too early for the U.S. clinical trials to offer any definite proof of benefit against accelerated adjacent-segment degeneration, the fact that normal intervertebral motion is preserved at the treated segment is encouraging. Longer term safety and efficacy studies are in progress. REFERENCES 1. Marnay T: Lumbar disc arthroplasty: 8–10 year results using titanium plates with a polyethylene inlay component. American Academy of Orthopaedic Surgeons Annual Meeting, San Francisco, CA, 2001. 2. Marnay T: Lumbar disc replacement. Spine J 2:94S, 2002. 3. Delamarter RB, Fribourg DM, Kanim LE, Bae H: ProDisc artificial total lumbar disc replacement: Introduction and early results from the United States clinical trial. Spine 28:S167–S175, 2003. 4. Delamarter RB, Bae HW, Pradhan BB: Clinical results after lumbar total disc replacement: An Interim Report from the United States Clinical Trial for the ProDisc-II Prosthesis. Orthop Clin North Am 36:301–313, 2005. 5. Zigler JE: Clinical results with ProDisc: European experience and U.S. investigation device exemption study. Spine 28:S163–S166, 2003. 6. Zigler JE: Lumbar spine arthroplasty using the ProDisc II. Spine J 4:260S–267S, 2004. 7. White AA, Panjabi MM: Clinical Biomechanics of the Spine, 2nd ed. Philadelphia, Lippincott Williams & Wilkins, 1990. 8. Delamarter RB, Zigler J, Spivak JM, et al: The US Multi-Center IDE Results of the ProDisc-L Artificial Lumbar Disc versus Fusion. American Academy of Orthopaedic Surgeons Annual Meeting, Chicago, IL, 2006.
CHAPTER
40
Mobidisc Disc Prosthesis Jean-Paul Steib, Lucie Aubourg, Jacques Beaurain, Joe¨l Dele´crin, Je´rome Allain, Herve´ Chataigner, Iohan Bogorin, Marc Ameil, Thierry Dufour, and Jean Stecken
K E Y l l l l l
P O I N T S
Intervertebral motion results in many instantaneous centers of rotation. There is translation during the intervertebral rotation. Mobidisc is a second-generation non-constrained total disc prosthesis. Mobidisc has a mobile inlay. The surgical insertion of Mobidisc is easy and precise.
INTRODUCTION Lumbar disc degeneration is a major cause of back pain with huge economical consequences.1 When all treatments have failed, fusion can be indicated, and a good result can be expected.2 Like every treatment, arthrodesis has its specific drawbacks and its own complications. The idea to cure the pain while respecting the mobility of the joint with the help of an artificial disc is not recent.3,4 The first clinical trials were conducted in the 1990s with the SB CHARITÉ (DePuy Spine, Raynham, MA) and ProDisc I (Synthes, West Chester, PA) prostheses.5 Since then, many different concepts and designs were developed to address the discogenic pain.6 Despite the available nucleus prosthesis7 and semi-constrained8 total disc prostheses, we believe that there is a place for non-constrained devices.9 DESIGN THEORY Total disc arthroplasty can be considered as a treatment when it is certain that the pathologic disc is the only cause of the pain and when all conservative treatments have failed. To be successful, the artificial disc must restore the disc height and provide physiologic movement. The centers of rotation of intervertebral movement are determined by geometric construction based on the shape of the posterior articular processes. Thus, in flexion-extension, the center of rotation is located in the posterior one third of the intervertebral space and underneath the surface of the superior plate of the inferior vertebra. This is an average center of rotation derived from many instantaneous centers of rotation. Intervertebral rotation is imposed by the articular processes forming a circular surface of contact (Fig. 40–1). The center of rotation is posterior at the level of the spinal process. Thus, there is automatically a translation of the vertebral 326
n
F I G U R E 40–1. Intervertebral rotation is imposed by the articular processes forming a circular surface of contact.
body. In its function, an artificial disc has to respect these conditions of movement, and this is the leading concept behind the development of the Mobidisc prosthesis (LDR, Troyes, France). DESCRIPTION OF THE DEVICE Mobidisc is a second-generation disc prosthesis made of two vertebral end plates and a polyethylene core (Fig. 40–2). The plates are manufactured from cobalt chrome and have a truncated elliptical form with a plasma-sprayed porous titanium coating covered by hydroxyapatite to facilitate bone integration. There are three sizes available with 0, 5, or 10 degrees of lordosis. A keel provides for primary fixation. The superior end plate has a concave inferior surface adapted to the convexity of the core. The superior side of the inferior end plate is flat to receive the polyethylene inlay captured by four stops. The stops
CHAPTER 40
A n
Mobidisc Disc Prosthesis
327
B
F I G U R E 40–2. The Mobidisc prosthesis. A, Assembled prosthesis. B, The three components of the device: superior plate, mobile inlay, and
inferior plate.
provide for controlled translation of the core on the inferior plate in all directions. The mobility of the prosthesis is physiologic and follows the instantaneous centers of rotation, allowing a translation during the intervertebral rotation. Different sizes of the polyethylene core allow for proper disc height selection. BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES During development, biomechanical tests were performed at the CRI laboratory in Charleville Mézières and at the Laboratory of Biomechanics of ENSAM (Paris, France). Repeated sliding movements of the polyethylene core on the inferior plate up to 5 million cycles did show either fracture or significant wear. The prosthesis was also tested in physiologic conditions on a cadaver to more than 15 million cycles with no or few significant modifications resulting. The first implantation was performed in November 2003. CLINICAL PRESENTATION AND EVALUATION A prospective clinical case series of 149 patients were enrolled and underwent surgery by one of eight surgeons participating in the
study. Demographic data and index levels are summarized in Table 40–1. The mean operative time was 161 minutes (range, 60 to 350 minutes), with a mean blood loss of 292 mL (range, 50 to 3,600 mL). Clinical endpoints are summarized in Table 40–2. After 2 years, back pain improvement on Visual Analog Scale (VAS) averaged 60%; VAS leg pain improvement was 36% on the right and 62% on the left. Improvement on the Oswestry Disability Index (ODI) was 55%. Ninety percent of patients were satisfied or very satisfied following the surgery, 77% for the back pain and 70% for the leg pain result. The mean duration of sick leave after the index surgery was 4 months (range, 1 to 11 months). The mobility of the affected level changed from 4 degrees (range, 0 to 15 degrees) preoperatively to 9.4 degrees (0 to 18 degrees) postoperatively, as observed from 76% of all prostheses had mobility at the last follow-up. OPERATIVE TECHNIQUE The patient is positioned supine (Da Vinci position) and is under general anesthesia. The disc is exposed in a traditional way by a retroperitoneal approach. A pin is placed in the adjacent vertebra, which marks the midline of the disc. After discectomy, removal
328
P A R T
I V
Lumbar Total Disc Arthroplasty
TABLE 40–1. Summary of Demographic Data in 149 Patients Undergoing Mobidisc Total Disc Replacement Demographic Data Number of patients
149
Sex Male Female
46 (31%) 103 (69%)
Follow-up Mean (months) SD 3 months 6 months 12 months 12 months
156.9 149 142 100 35
Mean age (yr) SD Age range (yr)
41.07.2 19–56
Modic Sign Modic 0 Modic I Modic II
26% 54% 19%
Single-Level Cases L3-L4 L4-L5 L5-S1
130 (87.2%) 7 34 89
Double-Level Cases L2-L3-L4 L3-L4-L5 L4-L5-S1
18 (12.1%) 1 3 14
way as the chisel, using charts. Following removal of the guide, but before closing the patient, an x-ray study is performed to confirm satisfactory position of the prosthesis at the posterior wall of the vertebral body. POSTOPERATIVE CARE
Triple-Level Cases L3-S1
1 (0.7%)
Previous Surgery None Discectomy Fusion
56% 31% 5%
The patient is ambulatory the following day without a corset and is discharged from the hospital as soon as possible. COMPLICATIONS There have been a total of 12 reported complications resulting in three reoperations. Ten cases of subsidence (5.9%) have been reported, five were clinically satisfied, and one was converted to fusion. Additionally one laterally malpositioned prosthesis was reoperated and successfully implanted with another Mobidisc device. Finally, one improperly sized prosthesis was converted to fusion. CONCLUSIONS/DISCUSSION Mobidisc is a unique second-generation disc prosthesis that includes a mobile inlay allowing maintenance of the natural movement created by the posterior articulation. When the indications and operative technique are followed, the clinical results can be very satisfying.
ADVANTAGES/DISADVANTAGES: MOBIDISC Disadvantages Not indicated for: instability posterior arthritis
Advantages Non-constrained arthroplasty Simple, reproducible placement
SD, standard deviation
TABLE 40–2.
Summary of Clinical Scores Before and After the Mobidisc Implantation* Preoperative
VAS lumbar VAS left leg VAS right leg ODI (%) SF-36 MCS SF-36 PCS
6.7 4.2 3.8 49.7 31.7 33.9
0.2 0.3 0.3 1.3 0.9 0.6
3 months 2.9 1.6 1.9 24.4 42.1 43.1
0.3 0.2 0.3 1.8 1.1 0.9
6 months 2.3 1.4 1.7 23.9 41.7 44.4
0.3 0.2 0.3 1.9 1.3 1.0
12 months
24 months
2.3 1.7 2.1 20.0 42.5 48.0
2.7 1.6 2.4 22.5 41.7 49.2
0.3 0.3 0.4 2.3 1.4 1.2
0.6 0.5 0.6 4.0 2.5 1.8
MCS, Mental Scale; ODI, Oswestry Disability Index; PCS, Physical Scale; VAS, Visual Analog Scale. *Evaluation was performed preoperatively, and 3, 6, 12, and 24 months after the index surgery. Clinical endpoints are VAS (0–10 scale) for lumbar and leg pain, Oswestry and Short-Form 36 (SF-36) quality-of-life score. Results are expressed as mean SEM.
of osteophytes, thorough cleaning of the end plates, and the depth and the height of the disc are measured: These values are used to determine the size of the prosthesis to insert. Although the height is maintained by a spacer, a guide centered on the pin is inserted. An adapted chisel is placed in the rails of the guide to prepare the end plates for the keels. The Mobidisc prosthesis is then assembled: plates, keels, and core. The prosthesis is released in the guide and punched home with the impactor that is adjusted the same
REFERENCES 1. Rothman RH, Simeone FA, Bernini PM: Lumbar disc disease. In Rothman S (ed): The Spine, 2nd ed. Philadelphia, WB Saunders, 2001, pp 508–645. 2. Fritzell P, Hagg O, Wessberg P, Norwall A: Lumbar fusion versus nonsurgical treatment for chronic lumbar pain: a multicenter randomized controlled trial from the Swedish lumbar spine study group. Spine 26:2521–2534, 2001.
CHAPTER 40
3. Cleveland DA: The use of methylacrylic for spinal stabilization after disc operations. Marquet Med Rev 20:62–XX, 1955. 4. Fernström U: Arthroplasty with intercorporeal endoprothesis in herniated disc and in painful disc. Acta Chir Scand 357(suppl):154–159, 1966. 5. Buttner-Janz K: Development of the Artificial Disc SB Charité. Ann Arbor, MI, Hundley & Associates, 1992. 6. Szpalski M, Gunzburg R, Mayer M: Spine arthroplasty: a historical review. Eur Spine J 11(suppl 2):S65–S84, 2002.
Mobidisc Disc Prosthesis
329
7. Sagi HC, Bao QB, Yuan HA: Nuclear replacement strategies. Orthop Clin North Am 34:263–267, 2003. 8. Tropiano P, Huang RC, Girardi FP, et al: Lumbar total disc replacement: seven to eleven years follow-up. J Bone Joint Surg Am 87:490–496, 2005. 9. Steib JP: A new approach to lumbar disc prosthesis. In Szpalski M, Gunzburg R, LeHuec J-C, Braydo-Bruno M(eds): Nonfusion Technologies in Spine Surgery. Philadelphia, Lippincott Williams & Wilkins, 2006, pp 187–190.
CHAPTER
41
The FlexiCoreW Intervertebral Disc Jonathan R. Stieber and Thomas J. Errico
K E Y l
l
l
l
l
P O I N T S
The FlexiCoreW Intervertebral Disc is a lumbar disc replacement featuring a metal-on-metal, single-unit design. Tension-bearing construction is designed to prevent separation and/or dislocation. Fixed center of rotation is designed to minimize relative translation of implant components. Metal-on-metal (cobalt-chromium alloy) bearing surfaces avoid polyethylene wear and creep. Implantation is through anterior or anterolateral angles and intraoperative repositioning.
INTRODUCTION The FlexiCoreW Intervertebral Disc (Stryker Spine, Allendale, NJ) is a metal-on-metal mechanical total disc replacement device designed to treat lumbar axial back pain resulting from degenerative disc disease. The implant is intended to replace the painful intervertebral disc while maintaining or restoring the motion of the diseased functional spinal unit, thus replicating the native physiology of the spine and preventing the cascade of degenerative disease. DESCRIPTION OF THE DEVICE The FlexiCoreW Intervertebral Disc is a single-unit, all-metal device composed of superior and inferior baseplates articulating about a centrally located, stationary center of rotation (Fig. 41–1). The baseplates are sprayed with a titanium plasma coating to promote bone ingrowth for long-term fixation and are each flanked by short spikes that promote initial stability. The baseplate morphology features a central dome that is engineered to approximate the concavity of the native vertebral body end plates in order to maximize surface area contact, minimize subsidence, and enhance stability. This design also serves to facilitate osseous fixation and to optimize the device’s center of rotation. The two baseplates articulate through a central ball-and-socket joint that establishes the device’s center of rotation at a midpoint between the baseplate domes. The constrained design permits a physiologic range of motion while preventing translation of the superior and inferior vertebral bodies, minimizing forces conferred on the 330
posterior spinal elements. The joint allows 15 degrees of flexion and extension and lateral bending exceeding the normal physiologic range of motion of the lumbar motion segment. An internal rotational stop is designed to avoid pathologic facet loading by preventing supraphysiologic axial rotation beyond 5 degrees. The FlexiCoreW is available in six disc heights (13 to 18 mm in 1-mm increments) and two baseplate footprint sizes (28 35 mm and 30 40 mm, depth width). The spherical metal-on-metal bearing surface of the joint, manufactured of highly polished cobalt-chromium, is intended to maximize durability, minimize wear debris, provide overall strength, and augment the life expectancy of the device. Cobaltchromium has been shown in the orthopaedic joint replacement literature to have a lower coefficient of friction and decreased wear when compared with other bearing surfaces including polyethylene.1,2 Moreover, metal-on-metal bearing surfaces are not subject to the “creep” phenomenon observed with metal-on-polyethylene interfaces. During manufacture, the ball of the joint is permanently captured in the socket, creating an articulating one-piece device. This prevents the device from dislocating or separating under tension loads and permits it to be held, manipulated, and inserted by the surgeon as a single unit as it is implanted in the disc space. Specialized insertion instrumentation facilitates device implantation, and permits the device to be inserted from multiple anterior angles. The instrumentation and insertion technique minimizes implantation misalignment and other technical errors associated with modular assembly and implantation. Repositioning is possible using the dedicated repositioners after the device has been seated in the disc space. Removal is also possible using the repositioner/extractor tools. INDICATIONS l l
l
l
Skeletal maturity Failure of a minimum of 6 months of conservative treatment for degenerative disc disease Single-level lumbar disc disease (L1-S1) resulting in axial back pain of discogenic origin Radiographic evidence of degenerative disc disease
CHAPTER 41
n
F I G U R E 41–1. FlexiCoreW Intervertebral Disc.
CONTRAINDICATIONS l
l
l
l
l
l
l l
l
l l
History of previous lumbar fusion or bilateral open decompressive procedures Poor bone quality or clinically compromised vertebral body structure Significant end plate incompetence at the level to be treated (e.g., Schmorl's node or vertebral end plate herniation) Involved vertebral end plate less than 37 mm in the coronal diameter or 30 mm in the sagittal diameter Instability at the level to be treated (defined as any grade 1 or greater degenerative spondylolisthesis; more than 4 mm of translation [total excursion] on flexion and extension; spondylolysis, or isthmic spondylolisthesis) Lumbar deformity presenting as scoliosis of more than 15 degrees Significant facet joint disease Moderate to severe spinal stenosis at the level to be treated or an adjacent level Known infection, hepatitits, rheumatoid arthritis, autoimmune diseases, or malignancy. Patients taking medications that may interfere with bone or tissue healing. Morbid obesity or pregnancy Allergy to one of the implant materials
The FlexiCoreW Intervertebral Disc
331
generally used. A Pfannenstiel mini-laparotomy transverse incision (usually 5 to 8 cm long) may be used for a single-level disc replacement from L3-S1. A midline longitudinal incision may be used if a previous midline scar is present. The necessary access to the disc space generally requires appropriate identification, ligation, and mobilization of vascular structures. For exposure of the L5-S1 interspace, the middle sacral artery and vein are identified and ligated. For exposure of the L4-L5 disc space, the iliolumbar vein should be exposed and ligated to allow the iliac artery and vein to be mobilized medially. The entire disc space must be exposed from each lateral margin, to permit centralized placement of the prosthesis. The midline of the disc space should be determined intraoperatively and marked before commencing the discectomy by placing a metal marker in the vertebral body and verifying its location with respect to the spinous processes on an anteroposterior (AP) fluoroscopic view. For the L5-S1 disc space, the Ferguson view may allow for improved visualization. Discectomy and End Plate Preparation After a localizing radiograph has been obtained to identify the proper disc space, a central annulotomy is performed approximating the width of the implant. The use of disc space distraction early in the procedure may facilitate the necessary disc removal, decompression (if indicated), and annular release to permit adequate distraction. The disc and the end plate cartilage are excised exposing subchondral bone with punctate bleeding. The lateral recesses and posterior margin of the disc space should be cleared of osteophytes or soft tissue that may prevent full distraction of the posterior disc space or proper positioning of the device. The posterior margin of the S1 end plate may exhibit an osseous ridge that may require flattening before the implant may be properly seated. The posterior longitudinal ligament and the remaining posterior annulus may be preserved so long as adequate distraction of the disc space is achieved. It may be necessary to resect the annulus and posterior longitudinal ligament in order to elevate and mobilize the posterior disc space. Loupe magnification or use of an operative microscope may facilitate disc space preparation according to surgeon discretion.
OPERATIVE TECHNIQUE Preoperative Planning and Surgical Approach
Distraction (Disc Height Restoration)
Cross-sectional imaging on computed tomography or magnetic resonance imaging scans should be reviewed preoperatively to determine the appropriate footprint of the FlexiCoreW Intervertebral Disc so that the end plate coverage does not exceed 90% of either the anterior-to-posterior depth or the lateral width of the end plate. The appropriate implant height is later determined intraoperatively with the use of distraction spacers. Implantation of FlexiCoreW Intervertebral Disc is performed with the patient under general anesthesia in the supine Trendelenburg position for an anterior or anterolateral approach to the lumbar spine. Slight elevation of the lumbar spine using an inflatable pad may be appropriate to enhance lumbar lordosis and facilitate disc space distraction, particularly in the presence of a collapsed disc space. An open or mini-open retroperitoneal approach is
Once the end plates have been exposed, the next step in disc space preparation is the restoration of disc space height. Adequate mobilization of the posterior intervertebral space is essential to success of the procedure. Restoration of disc space height should proceed slowly, employing suitable techniques generally used for parallel distraction of a collapsed disc space. Serial distraction may be initiated with large, thin periosteal elevators and a gentle twisting motion. When sufficient disc height has been achieved, the round distraction spacers may be used. These distraction spacers are cylindrical in shape and are available in heights ranging from 8 to 18 mm in 1-mm increments (Fig. 41–2). The round distraction spacers are used to gradually increase the height of the intervertebral space while preserving end plate integrity. The beveled geometry of the round distraction spacers coupled with the long
332
P A R T
I V
Lumbar Total Disc Arthroplasty
n
F I G U R E 41–2. Insertion of the round distraction spacer with the distraction spacer handle.
lever arm of the distractor handle permits powerful but gentle distraction of the intervertebral disc space by rocking the spacer in a cephalad-caudal direction. The FlexiCoreW instrumentation set also includes distraction spacers that may be used once the disc space has been distracted to 13 mm to further distract the disc space and to determine the proper implant size (Fig. 41–3). These spacers are broader in their geometry and are shaped more similarly to the vertebral body end plates than the round distraction spacers. The design of the static distraction spacers helps to distribute the load during distraction. The distraction spacers range in height from 12 to 18 mm in 1-mm increments and are available in baseplate footprints, 28 35 mm and 30 40 mm. Each distraction spacer has a 5-degree lordotic angle such that the posterior height of the spacer is 1.8 mm less than the labeled height, and has a beveled posterior edge to ease insertion between the vertebral end plates.
The 12-mm distraction spacer is gently inserted into the disc space, rocked to loosen the surrounding ligaments, and removed. When inserted, each distractor should be centered and positioned within 2 to 3 mm of the anterior vertebral end plate margin as well as 1 to 2 mm of the posterior margin. Progressive distraction continues with sequential insertion and removal of distraction spacers of increasing size in 1-mm increments until the appropriate height of the intervertebral disc space has been restored. Resection of the remaining lateral annulus may also gradually facilitate restoration of disc space height, and this step should be performed following the removal of each successive distraction spacer. Disc space height should be maximally restored without overtensioning of the remaining annulus and ligaments or damaging the vertebral end plates. The position of the spacer footprint in relation to the vertebral body end plates should be verified in both the coronal and sagittal planes using fluoroscopy. The static distraction spacer may be disengaged from the inserter handle in order to enhance imaging. Again, the appropriate spacer width and depth should not exceed 90% of the depth or width of the end plates. The domes of the baseplates should sit within the concavities of the vertebral end plates. In some patients, the superior end plate of the inferior vertebral body may have been flattened secondary to sclerotic changes. It is important to confirm that the inferior surface of the distraction spacer is flush with the end plate. If intraoperative imaging reveals a gap between the space and the end plate, a burr may be used to contour the end plate so that it may conform to the dome of the distraction spacer and subsequent insertion of the FlexiCoreW Intervertebral Disc. Following removal of the distraction spacer that has restored the appropriate disc space height and annular tension, the dynamic distractor (Fig. 41–4) is employed to confirm the final implant size and to ensure that the intervertebral space has sufficient height to accommodate the baseplate spikes. It is inserted between the vertebral bodies and expanded to the height of the final distractor by turning the distal knob as required. The dynamic distractor provides tactile feedback for the assessment of appropriate ligamentous laxity and confirms the optimal height of the implant. Once the optimal height has been determined, the dynamic distractor is used to further distract the disc space by an additional 1.5 to 2.0 mm in order to permit passage of the implant's baseplate spikes. This final distraction maneuver also prepares the native vertebral end plates to receive the domed baseplates of the FlexiCoreW device. Care should be exercised when distracting the intervertebral disc space as excessive or aggressive distraction may result in damage to the end plates, facet joints, or soft tissues. Insertion of the FlexiCoreW Intervertebral Disc
n
F I G U R E 41–3. Insertion of the static distraction spacer with the distraction spacer handle.
Once the appropriate implant size has been selected, the FlexiCoreW Intervertebral Disc is secured to an impactor (Fig. 41–5). Two insertion techniques are available, each with its own impactor. The FlexiCoreW may be inserted either with or without the aid of insertion ramps. For insertion of the implant without the use of ramps, the fixed-angle impactor is selected incorporating a head with a ledge that holds both baseplates of the implant firmly in a fixed angle of lordosis and prevents premature engagement of
CHAPTER 41
n
333
F I G U R E 41–4. A, Dynamic distractor. B, Dynamic distractor in situ.
the baseplate spikes. The fixed-angle impactor also features a positive depth stop on both the top and bottom of the head in order to prevent overinsertion of the implant. The fixed-angle impactor should not be used with the ramps because it may result in overdistraction. For insertion of the FlexiCoreW using the insertion ramps to guide insertion, the flat plate impactor is selected. The flat plate impactor features a flat face that is held firmly in contact with the implant's lower baseplate and permits the upper baseplate to articulate during the insertion maneuver. The upper baseplate is allowed to follow the angle of the ramps during insertion, minimizing the amount of disc space distraction necessary to implant the device. The flat plate impactor has one depth stop on the bottom of the head to prevent overinsertion. To load either of the specialized impactors, the instrument's J-hook is extended so as to engage one of the three holes in the implant's baseplate. The particular hole is selected to facilitate the desired angle of insertion. The central hole is used for a direct anterior surgical approach, whereas the offset holes are used for
F I G U R E 41–5. Insertion of the FlexiCoreW Intervertebral Disc with the fixed-angle impactor.
n
The FlexiCoreW Intervertebral Disc
anterolateral approaches. When released, the spring-biased J-hook will retract to grasp the angled perimeter of the baseplates against the fitted mouth of the impactor head, preventing axial rotation during insertion. Both impactors include a locking nut on the handle shaft that is tightened to secure the implant to the impactor during insertion. The fixed-angle impactor can be used alone to insert the implant into the intervertebral space, similar to the manner in which a femoral ring allograft is inserted during an anterior lumbar interbody fusion procedure. The posterior edges of the baseplates are positioned between the vertebral end plates, and the implant is advanced into the space. The locking nut is loosened, and the flange is advanced to disengage the J-hook. Moderate force may then be used against the impactor handle to insert the implant to the appropriate AP depth. Excessive posterior insertion of the device is prevented by the positive stops on the impactor heads. The domes of the baseplates should be seated within the native vertebral end plate concavities. The final position of the FlexiCoreW device should then be confirmed radiographically. Alternately, the ramps may be used to distract the intervertebral space as the implant is inserted (Fig. 41–6). The ramps are provided in two sizes that correspond to the 35- and 40-mm baseplate widths. The pair of ramps is coupled at their proximal ends with a C-Clip connector. The ramps converge toward one another, such that their proximal ends are separated to receive the implant between them (previously loaded on the flat plate impactor). The distal ends converge toward one another to meet at the intervertebral space and are inserted between the vertebral end plates. The implant is inserted between the ramps and advanced into the intervertebral space. The implant spikes ride in grooves on the inner surfaces of the ramps, guiding the implant linearly. As the implant is advanced, the height of the implant forces the distal ends of the ramps to separate, distracting the intervertebral space to the requisite height for implantation of the device. As the implant is inserted using the ramps, it is important to ensure that the impactor handle is parallel to the lower map so that the locking nut has sufficient clearance as it nears the end of the ramps. Following insertion, the ramps are removed sequentially.
334
P A R T
I V
Lumbar Total Disc Arthroplasty
C-clip
A n
F I G U R E 41–6.
Insertion of the FlexiCoreW Intervertebral Disc with the aid of the insertion ramps.
Verification of Implant Size and Placement The FlexiCoreW Intervertebral Disc is appropriately positioned when the baseplate domes are seated within the concavities of the adjacent vertebral body end plates (Fig. 41–7). The implant position should be confirmed with fluoroscopy so that it is centered in the sagittal plane, and positioned 2 to 3 mm within the anterior margin and 1 to 2 mm within the posterior margin of the end plate. The implant should not protrude outside the anterior margin of the vertebral end plates. Ideal positioning entails the device being seated against the posterior annulus, parallel alignment of the baseplates with the adjacent end plates, and a central position in relation to the vertebral bodies. The end plates should be visually inspected for fractures. If fractures are present, consideration should be made of revising the surgery to fusion. Repositioning (or Extracting) the FlexiCoreW Intervertebral Disc
Closure Standard closure procedures for anterior or anterolateral spine surgery are followed. POSTOPERATIVE CARE
If the lateral or AP images indicate suboptimal positioning of the implant, repositioning can be performed using the repositioner/
n
extractors (Figs. 41–8 and 41–9). Anterior, right offset, and left offset instruments are available, each with a pair of pins on its distal end that can be inserted into any pair of holes in the upper or lower baseplates, thus providing multiple surgical approach angles. The repositioner/extractors may be used individually or in tandem to grasp the upper or lower implant baseplates. Once the pins are engaged, the device can be repositioned in the lateral and AP planes, as well as rotated axially. The repositioner/extractors can also be used to extract the implant by similarly engaging implant baseplates and withdrawing the handle's flange.
W
F I G U R E 41–7. Lateral view of the FlexiCore in situ.
The goal of postoperative rehabilitation is to return the patient to normal activity as expeditiously as possible without jeopardizing healing. The patient should wear a lumbar corset for 2 to 3 weeks to support healing of the abdominal incision, depending on patient comfort. The patient’s rehabilitation program should be individually tailored to his or her needs, taking into account age, stage of healing, general health, and physical condition. Return to work is dependent on specific vocation. Patients should be restricted from heavy lifting until full rehabilitation has been completed, typically after 8 to 12 weeks.
n
F I G U R E 41–8. Extraction of the FlexiCoreW Intervertebral Disc
with the repositioner/extractor and slotted mallet.
CHAPTER 41
n
The FlexiCoreW Intervertebral Disc
F I G U R E 41–9. FlexiCoreW Intervertebral Disc anteroposterior and lateral radiographs.
CLINICAL DATA Clinical results have been presented from four of the study sites under the IDE protocol on 103 patients of the 400 patients randomized for the treatment of symptomatic lumbar DDD. Patients were randomized in a 2:1 fashion (FlexiCore: Fusion) yielding 66 patients treated with the FlexiCoreW (F) and 37 patients treated with anteroposterior fusion (C). Disability and pain were assessed using the Oswestry Disability Index (ODI) and the Visual Analog Scale (VAS), respectively (Figs. 41–10 and 41–11). Prospective data were collected preoperatively and postoperatively at 6 weeks and 3, 6, 12, and 24 months. The mean ODI scores favored the FlexiCoreW at all time points, and the mean VAS score favored
the FlexiCoreW at both 12 and 24 months. The mean ODI scores were 60(F) and 60(C) preoperatively, 37(F) and 48(C) at 6 weeks, 29(F) and 32(C) at 3 months, 27(F) and 30(C) at 6 months, 25(F) and 32(C) at 12 months, and 22(F) and 28(C) at 24 months. The mean VAS scores were 86(F) and 83(C) preoperatively, 30(F) and 36(C) at 6 weeks, 34(F) and 29(C) at 3 months, 33(F) and 30(C) at 6 months, 28(F) and 33(C) at 12 months, and 33(F) and 37(C) at 24 months. The average operative time was 82 minutes for the FlexiCoreW group and 170 minutes for the fusion group. The average estimated blood loss was 76 mL for the FlexiCoreW group and 99 mL for the fusion group. The average hospital stay was 2.5 days for the FlexiCoreW group and 3.3 days for the fusion group.3
VAS Comparison
ODI Comparison
100
100 90
FlexiCore® Fusion
80 Avg. VAS score (%)
Avg. ODI score (%)
FlexiCore® Fusion
90
80 70 60 50 40 30
70 60 50 40 30
20
20
10
10 0
0 Pre-op (66, 37)
6 wks (65, 35)
3 mo (64, 34)
6 mo (61, 34)
12 mo (64, 35)
Pre-op (66, 37)
24 mo (48, 25)
F I G U R E 41–10. Oswestry Disability Index comparison.
6 wks (65, 35)
3 mo (64, 34)
6 mo (61, 34)
12 mo (64, 35)
Follow-up
Follow-up n
335
n
F I G U R E 41–11. Visual Analog Score comparison.
24 mo (48, 25)
336
P A R T
I V
Lumbar Total Disc Arthroplasty
COMPLICATIONS AND AVOIDANCE Complications can stem from inappropriate selection of a surgical candidate. Patients who do not have true discogenic back pain, who have unrecognized instability, or who suffer from multilevel symptomatic disc degeneration are poorly indicated for disc replacement surgery. Despite optimal patient selection, as with all major surgery, disc replacement is not without its own risks and concerns. Potential complications of disc replacement surgery include those traditionally associated with anterior lumbar interbody fusion. Intraoperative complications may include inadvertent peritoneum violation as well as vascular injury. The common iliac vessels, aorta, iliolumbar vein, and inferior vena cava may be at risk for injury from exposure, retraction, or loosening of the applied ligature. In a recent, large series of patients undergoing anterior lumbar spinal surgery, the risk of major vascular injury was found to be relatively low at 1.9%.4 Deep vein thrombosis can occur from manipulation of the great vessels and requires treatment with anticoagulation. In men, injury to the superior hypogastric plexus from dissection or the use of electrocautery during the surgical approach can lead to retrograde ejaculation in a reported 0.4% to 17.5% of cases.5–7 This can be a serious complication for a man of reproductive age, and preoperative sperm donation is often recommended. The transabdominal or retroperitoneal approach can yield a prolonged postoperative ileus necessitating insertion of a nasogastric tube. As with many abdominal surgeries, the approach also carries the risk of denervation of the abdominal wall with resulting muscular atony as well as the risk of incisional hernia. As with all surgery, especially those involving implantable instrumentation, infection can occur but can be minimized with perioperative antibiotics and meticulous sterile techniques. In addition, device-specific complications can occur with disc replacement surgery. The importance of radiographic confirmation of final device positioning in both the AP and lateral planes cannot be overemphasized. Misplacement of the implant can disrupt the replication of normal spinal kinematics. Failure to adequately replicate the physiologic kinematics of the lumbar spine may predispose the patient to early facet joint degeneration. Anterior malpositioning can result from inadequate resection of posterior disc annulus or posterior osteophytes. Conversely, posterior malpositioning can result from excision or rupture of the posterior longitudinal ligament allowing posterior implantation. An oversized or undersized prosthesis can permit insufficient or excessive motion at the specific level. Without true motion preservation, whether due to heterotopic ossification or other causes, these devices will merely act as interbody spacers and lead to spontaneous fusion. The implant may also subside or migrate owing to overzealous end plate preparation or unrecognized instability. Prosthesis migration or extrusion, may lead to device failure and serious vascular complications. DISCUSSION The FlexiCoreW Intevertebral Disc incorporates a number of novel design characteristics. The single-unit design of the FlexiCoreW total disc replacement has a number of advantages. The tension-bearing structure is designed to prevent separation of
the bearing surfaces as well as dislocation or extrusion of the bearing surface elements. Contrary to other designs employing modular components that have exhibited the potential to dissociate, the entire device must become loosened before is likely to migrate or extrude. Implantation as a single unit facilitates proper alignment and is anticipated to reduce technical implantation error. The one-piece configuration should also serve to minimize the necessary device inventory. The baseplate domes and the central ball-and-socket joint are designed to establish a stationary center of rotation that is centrally located between the end plates and slightly posterior to the midline, closely matching the center of rotation of a healthy, natural, native intervertebral disc. An internal stop that limits axial rotation beyond the natural range of motion is included to prevent overrotation and pathologic facet loading. The metal-on-metal bearing surfaces of highly polished cobalt-chromium are anticipated to maintain sphericity over the life of the device and minimize wear debris. Because the device design is free of polyethylene, the theoretical risk of osteolysis is minimized. The titanium plasmasprayed bone ingrowth domes are engineered to fit flush within the concavities of the vertebral end plates and to assist in optimal early and late bony fixation. Initial device fixation is attained by low-profile spikes flanking the central dome. This feature is designed to maximize end plate and vertebral body preservation in order to prevent subsidence and facilitate revision if necessary. When performed in an appropriately indicated patient, total disc replacement has a number of potential advantages over spinal fusion. Both procedures have the ability to aid or eliminate axial back pain. Disc arthroplasty, in particular, is designed to restore the functional biomechanics of the spine and to preserve spinal motion. By preserving or reestablishing motion at the functional spinal unit, the goals are twofold. Lumbar flexion-extension is preserved or improved, and the cascade of adjacent spinal segment degeneration may be impeded or possibly prevented entirely. Following implantation, a disc replacement articulates immediately, with recovery consisting of abdominal wound and tissue healing alone. This period of recovery is substantially shorter than that required for solid spinal fusion. Moreover, disc replacement surgery is a stand-alone anterior procedure compared with many spinal fusions that require a posterior approach either instead of, or in addition to, an anterior exposure. Thus, there is no additional recovery required from a posterior procedure or need for the patient to wear a postoperative orthosis. Total disc replacement is a promising development in lumbar spine surgery for the treatment of degenerative disc disease. The FlexiCoreW Intervertebral Disc incorporates a number of important advances in motion preservation technology. REFERENCES 1. Chan FW, Bobyn JD, Medley JB, et al: The Otto Aufranc Award: Wear and lubrication of metal-on-metal hip implants. Clin Orthop Relat Res 36:10–24, 1999. 2. Dorr LD, Wan Z, Longjohn DB, et al: Total hip arthroplasty with use of the Metasul metal-on-metal articulation: Four to seven-year results. J Bone Joint Surg Am 82:789–798, 2000. 3. Tibbs R, Sasso R, Miz G, Theofilos C: Prospective, randomized trial of lumbar metal-on-metal total disc replacement: Initial treatment of
CHAPTER 41
degenerative disc disease. IMAST 14th Annual International Meeting on Advanced Spinal Techniques, Paradise Island, Bahamas 2007. 4. Brau SA, Delamarter RB, Schiffman ML, et al: Vascular injury during anterior lumbar surgery. Spine J 4:409–412, 2004. 5. Flynn JC, Price CT: Sexual complications of anterior fusion of the lumbar spine. Spine 9:489–492, 1984.
The FlexiCoreW Intervertebral Disc
337
6. Mayer HM: A new microsurgical technique for minimally invasive anterior lumbar interbody fusion. Spine 22:691–699; discussion 700, 1997. 7. Tiusanen H, Seitsalo S, Osterman K, Soini J: Retrograde ejaculation after anterior interbody lumbar fusion. Eur Spine J 4:339–342, 1995.
CHAPTER
42
Kineflex Ulrich Reinhard Ha¨hnle, Malan De Villiers, and Ian R. Weinberg
K E Y l
l
l
l
l
P O I N T S
The Kineflex disc is a recentering, unconstrained, metal-on-metal mechanical disc prosthesis. A cervical and a lumbar disc are currently Conformit Europeane (CE) certified and are also being evaluated in U.S. Food and Drug Administration (FDA) Pre-Market Approval (PMA) randomized, controlled trials. The Kineflex disc was designed primarily for patients with advanced motion segment degeneration using a simple insertion technique allowing powerful distraction and posterior placement within the disc space. The insertion of the assembled prosthesis enables free articulation of the end plates, allowing the superior and inferior end plates to be advanced independently. Good short-term clinical results have been achieved at a minimum follow-up of 2 years.
INTRODUCTION Adjacent-level degeneration is a major concern in lumbar fusion operations.1,2 Lumbar artificial disc replacement is an alternative to arthrodesis. The purpose of the intervention is to restore the intervertebral segment stability and mobility and protect the adjacent levels against nonphysiological loading conditions. Surgical insertions of lumbar disc prostheses using a steel ball were first published by Fernström.3 It failed clinically essentially because of subsidence of the implant into the bony end plate. Modern-type total lumbar disc replacement commenced in 1984 with the insertion of the first generation CHARITÉ (DePuy Spine Inc., Raynham, MA) disc prosthesis (CHARITÉ SB I).4 The mechanism of the prosthesis was carried through to the third-generation device that is still being used today (CHARITÉ SB III). Subsequently more constrained lumbar disc prostheses have been developed. Three of these prostheses are currently being evaluated in U.S. Food and Drug Administration (FDA) studies (ProDisc, Synthes, West Chester, PA; Maverick disc, Medtronic Sofamor Danek, Memphis, TN; FlexiCore disc, Stryker Spine, Allendale, NJ). Despite major advances in the disc insertion technique and design, difficulties persist with the correct midline and posterior placement of the prosthesis within the disc spaces, even in experienced hands.5 The Kineflex disc prosthesis was originally named Centurion disc. It was developed in Centurion, located between Pretoria and 338
Johannesburg in South Africa. The main objectives in the development of this prosthesis were an unconstrained/semiconstrained but recentering mechanism, to facilitate reliable midline and posterior placement of the implant within the disc space in severely degenerated disc spaces and to develop a simple and safe implantation technique, with the implantation being executed through a minimal invasive approach. INDICATIONS/CONTRAINDICATIONS Inclusion criteria at our center were age of 18 to 65 years, mechanical back and leg pain, symptomatic single or multilevel degenerative disc disease at the L2-L3, L3-L4, L4-L5, or L5-S1 levels confirmed on x-ray studies, magnetic resonance imaging, or provocative discography. Further inclusion criteria included recurrent disc herniation, broad-based central disc herniation without sequestration, and junctional failure after previous fusion. In all patients, supervised conservative treatment of at least 6 months had failed. Only the symptomatic levels on clinical examination and discography were replaced. Exclusion criteria were general contraindications, such as severe obesity, osteoporosis, tumor, or infection. Spinal exclusion factors were thoracic kyphosis of more than 60 degrees, idiopathic lumbar scoliosis of more than 30 degrees, previous wide laminectomy with destabilization of the facet complex, spondylolisis or spondylolisthesis, greater than Meyerding Grade 1 of the level to be replaced, bony spinal stenosis, and sequestrated disc prolapse tracking up or down behind the vertebral body. Other contraindications were previous retroperitoneal surgery, advanced vascular pathology, and single kidney. Advanced facet arthritis was not an exclusion criterion unless osteophyte formation from the facet resulted in bony canal or recess stenosis. Spinal or lateral recess stenosis caused by soft tissue (disc, ligamentum flavum, or joint capsule) was not considered a contraindication for disc replacement if proper decompression during surgery, by means of direct or indirect decompression, could be anticipated on preoperative imaging. DESCRIPTION OF THE DEVICE The Kineflex disc (Spinal Motion, Inc., Mountainview, CA) represents a disc prosthesis with a mechanism that is unconstrained but
CHAPTER 42
re-centering, resulting in a mobile center of rotation. The amount of constraint lies between the CHARITÉ disc prosthesis and Mobidisc (LDR, Troyes, France) on the unconstrained side and the ProDisc, Maverick, and FlexiCore disc prostheses on the constrained side. The mechanism comprises two metal end plates congruently articulating over a sliding core that is positioned posterior to the center of and in between the end plates (Fig. 42–1). The inferior end plate has a retaining ring that limits the excursion of the inferior articulation and prevents dislodgement of the core. The superior end plate has no retaining ring. The angle of motion allowed by the articulating mechanism from the neutral position is 12 degrees into flexion-extension, left or right side bending. This is true for the lumbar as well as for the cervical disc (Kineflex|C). The end plates and core are made of cobalt-chromiummolybdenum alloy (Biodur CCM Plus, Carpenter Technologies, Reading, PA). The integrating side of the end plate, facing the bony end plate, is flat and oval shaped, and has a small 1.5-mm wide midline fin with an oblique leading edge and two transverse holes which follows a pre-cut insertion groove but also allows self-cutting of a groove if required. The side of the end plate has multiple machined sharp serrations for primary fixation. Only the central portion of the surface adjacent to the center fin is smooth to allow riding of the adjacent prosthesis along the ‘slotted end plate distracter.' The entire leading edge of the end plates is beveled toward the bone side, to avoid cutting into the bony end plate during the insertion procedure. Both the inferior and superior end plates are manufactured in three different sizes (small, medium, and large). The inferior end plate is manufactured in three different angles coupled with three different heights (0 degree, 5.5 mm; 5 degrees, 6.5 mm; and 10 degrees, 7.25mm). The superior end plate has no angle but two different heights (5.5 to 6.5 mm). As a result, the overall height of the prosthesis ranges from 11 mm to 13.75 mm.
n
339
BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES Preclinical Testing Substantial preclinical mechanical testing was performed on the Kineflex arthroplasty, including static testing, monoaxial fatigue testing, and wear testing. All tests were performed using finished, sterilized devices, with the 5-40- mm end plate size and a core. As noted earlier, there is only one core size; therefore, the size of the load-bearing area on each end plate, which conforms to the core, is the same for all end plate diameters. Static Compression and Static Shear Testing
Static compression and shear testing were performed in accordance with American Society for Testing and Materials F-04.25.05.01 Draft I (February 2003) Item Z8924Z. The tests were conducted in an Instron machine, and compression of the six disc samples were increased until mechanical failure or a preset ultimate load was reached. The yield displacement, yield load, ultimate displacement, ultimate load, and stiffness were recorded. The results demonstrated that there was no height reduction in any of the samples as would be expected with a metal-on-metal construct. All loads applied to the assembled samples were in excess of 25 kN. Thus, because the maximum load-carrying capacity of vertebral bone is approximately 5 to 8.2 kN,6 the compressive strength of the Kineflex Prosthetic Disc (KPD) substantially exceeds that of the vertebral bone. Dynamic and Shear Fatigue Testing
Samples were tested under cyclic axial and shear compressive loading to assess the suitability of the fatigue resistance of the device for in vivo use. Samples were immersed in a physiologically buffered saline bath at 37 C throughout the test to simulate in vivo conditions. A test frequency of 5Hz was applied. Loads varied
B
A
Kineflex
F I G U R E 42–1. Kineflex metal-on-metal (A) mechanism. (B) Unit assembled.
340
P A R T
I V
Lumbar Total Disc Arthroplasty
cyclically between the maximum value of 2,000 N and 10% of the maximum, or 200 N. When viewed in terms of an average in vivo load, the results demonstrated no measurable dimensional changes in either the end plates or the core in what equates to 10 years of usage. Wear Testing
To evaluate the amount and size of wear particles that may be generated by the arthroplasty in vivo, five test samples were cyclically loaded in a multiaxial motion simulator for 10 million cycles. Testing was performed at a frequency of 5Hz. A constant 1,200-N load was applied throughout the wear test. A total cyclic range of motion of 7.1 degrees of lateral bending, a cyclic range of motion of 7.1 degrees of flexion-extension bending, and 4 degrees of axial rotation were applied. Test specimens were immersed in a 37 C saline bath throughout testing to simulate in vivo conditions. After every 1,000,000 cycles, the disc and meniscus were weighed, and a dimensional check of height and diameter was performed. Visual inspection was also performed, and a photographic record was compiled. Weight and dimensional measurements of the prosthesis (end plates and core) performed after every million cycles showed an average volumetric wear rate of 1.39 mm3 per million cycles. Mass loss was approximately 115 mg, for an average of 11.5 mg per million cycles. This represents a loss of only approximately 0.1% of the total prosthesis mass over 10 million cycles. Dimensional changes in the core were small, with an average height loss of less than 0.3 mm; no core exhibited any change in diameter. The degree of dimensional change observed with respect to core height does not interfere with the functionality of the device. The mean wear particle size was approximately 0.5 mm across all samples. Analysis of the form factor of the wear particles indicates that they were generally slightly elongated. Evaluation of the particles showed that they had a flake-like morphology when greater than 1 mm in size, but were granular when less than 1 mm in size. In vivo animal data demonstrating that wear debris of this type and quantity are unlikely to generate an adverse biologic response is available. Cunningham et al7 evaluated the neural and systemic tissue response to cobalt alloy particulate debris in an in vivo rabbit model up to 6 months. They placed 4 mg of cobalt alloy particles directly on the dura and compared the results with rabbits that had a sham procedure of dural exposure alone. All cobalt alloy particles used were less than 5 m in diameter, and 70% of all particles were between 2 and 3 m in diameter. It should be noted that although the majority of particles were in the 2- to 3-m range, more than 1.03 1010 submicron particles were injected. At 3 months, the number of macrophage-expressing cytokines localized within the spinal cord and overlying tissues indicated no significant differences compared with the control group. Despite regions heavily laden with metallic particulate, histiocystic reaction, and cytokine activity, the spinal cords indicated normal distribution of myelin and the intracellular neurofibrilla network. There was no evidence of cellular apoptosis, and all specimens were characterized as without significant histopathologic changes.
The quantity of particles applied in the Cunningham study, together with the one-time application of these particles directly to the dura, represents an extreme worst-case scenario compared with the anticipated wear of the Kineflex over time. The Cunningham study used a single application of particles, compared with the gradual release of particles over time that would be expected in vivo. Based on these results, there is low risk of any adverse biologic response to the wear particles generated by the Kineflex, even under the worst-case conditions of the in vitro wear test with respect to loads, motions, and the saline lubricant environment. Biomechanics of the Kineflex Lumbar Disc Prosthesis
The design of the Kineflex prosthesis allows for high congruence of articulating components over its full range of articulation. However, the three-part design allows for a self-centering but relatively unconstrained motion, which enables translation of end plates with respect to each other, as opposed to ball-and-socket (twopart) designs, which are highly constrained. The instantaneous axis of rotation of the functional spinal unit (FSU) does indeed also incorporate a translating component, which is essential to allow flexion-extension motion in the three-point support structure of the FSU. The in vitro biomechanics of the Kineflex prosthesis was evaluated in harvested spines and compared with untreated harvested spines by Denis Di Angelo et al,8 who concluded that there was no significant difference in the motion response between the Kineflex disc and the harvested spine conditions except in extension, where the motion was 73% of the untreated condition. The center of rotation of the Kineflex disc is located posterior of center, which also mimics the FSU condition. Clinical Testing The primary clinical outcome measures for this study were pain relief and functional improvement as assessed by the Oswestry Disability Index (ODI) and our own questionnaire. Questionnaires (ODI and our own questionnaire) were completed by the patient preoperatively, at 6 weeks, 3 months, 6 months, and yearly in conjunction with the regular follow-up examinations. In addition to the outcome data, general demographic information, operative data, and data pertaining to radiologic examination were collected. Our own questionnaire was designed by first and last authors and has not been validated. The patient is asked about his or her satisfaction with the outcome of the treatment operation (options: excellent, good, fair, poor) and to state if he or she would undergo the same operation again or recommend it to friends (options: yes, don't know, no). Furthermore, the patient is asked to gauge his or her pain in the last 2 weeks preoperatively and at the time of completing the questionnaire on a scale of 1(no pain) to 10 (pain as bad as it can be). No differentiation between leg and back pain is made. Clinical Presentation and Evaluation
Baseline characteristics of the study population are shown in Table 42–1. Our first 100 patients have reached 2-year follow-up. They constitute a heterogenous study population, with 41% having undergone previous open lumbar surgery (see Table 42–1).
CHAPTER 42
TABLE 42–1. Preoperative Data of Study Population (n 5 100 Patients) Factor Gender Male Female Age (years) Height (cm) Weight (kg) Pain duration (month) Nonoperative care Physical therapy Chiropractic Acupuncture Previous surgery Rizotomy Discectomy Laminectomy Fusion Smoking Nonsmoking
60
Number of Patients (n 5 100)
50
57 43 44.9 (23–63) 174.6 9.6 (154–196) 81.3 15.6 (47–138) 63.5 74.6 (5–400)
40
99 61 25
10
341
Kineflex
ODI 47.8 ± 16.0
30 16.7 ± 16.5
20
15.2 ± 16.8
14.6 ± 14.9
14.2 ± 14.0
0
15 32 20 13 39 61
Pre OP
3 mo
6 mo
1 yr
2 yrs
All disc patients n = 98
Disc post fusion n = 13
Single disc isolated n = 44
Disc + fusion n = 12
Double disc isolated n = 30 n
F I G U R E 42–2. Oswestry Disability Index (ODI) preoperatively and at different follow-up intervals.
Clinical outcome assessment was available for 99 of 100 patients at 1 year and at 97% at 2 years. Sixty-nine patients underwent single-level disc replacements, and 31 patients had two levels replaced. Three of the patients with singlelevel surgery later received a second-level lumbar disc replacement at 7, 9, and 10 months after the index procedure, respectively. Twelve patients underwent fusions at another level (hybrid cases) during the index procedure. Four of these patients had previous discectomies. Thirteen patients presented with adjacent level disc disease after previous instrumented posterolateral fusion surgery (one to seven previous operations). Another 28 patients in this series, who underwent singleor double-level disc replacements, had one to four previous discectomies or laminotomies, or both (see Table 42–1). Operative times, estimated blood losses, and postoperative hospital stays are shown in Table 42–2. Recovery and Patient Satisfaction
Hospital stay averaged 2.8 0.8 days (2 to 8 days). Eighty-six patients were in employment at the time of the operation (Fig. 42–2). All but one patient went back to their previous occupation at an average of 31.0 16.1 days.
The pain score (scale from 1 to 10) of all patients decreased from 9.16 1.00 preoperatively to 2.88 2.34 at 1-year follow-up (2.78 2.2 at 2 years). The ODI improved significantly from 47.8 16 to 14.6 14.9 at 1 year (14.2 14.0 at 2 years) (P 1) decreases or eliminates the contribution of the posterior elements to the stability of the functional spinal unit (FSU). Although the devices available for TDR differ in construction, many incorporate nonconstrained or semiconstrained designs. At present, no device has been clinically evaluated in the context of a spondylolisthesis of greater than grade 1. Failure of the posterior elements may have the consequence of increasing the shear forces on the implant, which may result in increased pain from the facet joints, pars interarticularis defect (Fig. 50–1), or failure of the device either at the device–end plate junction or intrinsically. If lumbar facet joint arthropathy is responsible for a substantial element of the patient's discomfort, treatment of the painful intervertebral disc alone will not address all pain generators. Moreover, the process of facet degeneration may continue or be exacerbated after treatment with disc replacement. For this reason, patients with advanced facet arthropathy are thought to be poor candidates for treatment with disc replacement surgery; however, concomitant posterior motion-preserving stabilization, such as facet replacement or enhancement, may ultimately expand indications to include these patients. Patients with unrecognized or untreated spinal deformity may also experience inferior outcomes when treated with disc replacement. TECHNIQUE-RELATED PROBLEMS If the procedure is incompletely or incorrectly executed, the patient may experience persistent symptoms despite the correct diagnosis and choice of treatment. As previously mentioned, patients with a central or paracentral herniated nucleus pulposus may be treated with disc replacement surgery with a concurrent adequate excision of the herniated fragment. Failure to recognize the offending pathology and decompress the neural elements, however, may result in continued radiculopathy or stenosis. Similarly, among the surgical goals in treating the intrinsic disc pathology of degenerative disc disease is to remove the pain generator, that is, the intervertebral disc. Insufficient resection of the disc annulus may leave a substantial remnant of the innervated pain generator. Care must be taken to prepare the end plates appropriately and to adequately resect posterior osteophytes. Failure to do so may result in impaction of bone fragments from the end plate or residual posterior osteophytes into the spinal canal or neural
foramen leading to iatrogenic compression of the neural elements (Fig. 50–2). Inaccurate insertion of the implant out of a parallel plane with the end plates may also create this type of bony fragment and resultant problem. Implantation of an oversized prosthesis may overdistract the disc space and place the nerve roots on stretch. This can cause radiculopathic and dysesthetic pain. Furthermore, oversizing the implant may convert the articulating disc replacement into a static intervertebral spacer, which loses its ability to allow for motion.9 The advent of lumbar TDR has also increased the complexity of the anterior spinal procedure when compared with anterior lumbar interbody fusion. Because the disc replacement is an articulating device designed to replicate and preserve spinal motion, precision of device placement is far more integral to the success of the procedure than in fusion.10 Device placement posterior to the posterior vertebral margin will result in spinal stenosis, but device placement too anterior will limit the range of motion (ROM) of the motion segment. Implantation of the device too lateral to midline may also result in inferior outcomes (Fig. 50–3).11 CHARITÉ discs (DePuy Spine Inc., Raynham, MA) placed greater than 5 mm from their ideal placement in the coronal or sagittal plane had less ROM than those less than 5 mm displaced.10 Because some of the devices currently available incorporate modular components, it is of the greatest importance that the device is assembled correctly with the components locked in place. Incorrect or incomplete assembly may result in migration or extrusion of the polyethylene component from the device and catastrophic failure. Preoperative and postoperative disc height has been demonstrated to influence sagittal ROM following a TDR with the ProDisc-L (Synthes, West Chester, PA). Although this did not affect shortterm clinical outcomes, one can only anticipate improved long-term benefit through the maintenance of intervertebral motion. FAILURE TO ACHIEVE SURGICAL GOALS Among the goals of lumbar TDR are to eliminate the diseased intervertebral disc as the pain generator, to confer stability to the FSU, to restore or maintain disc height, to maintain or improve sagittal alignment, and to regain or maintain motion of the FSU. This last goal of regaining or maintaining motion differentiates TDR from our experience with spinal fusion. ROM may be limited secondary to suboptimal device placement, oversizing of the device, or autofusion.10 Although a complete and solid spontaneous fusion is unlikely to result in pain, partial or incomplete autofusion of the segment secondary to heterotopic ossification may result in persistent pain and inferior patient outcomes.12 FAILURE OF THE PROSTHESIS Pain is likely to be the initial symptom of device failure or other device-associated complications. Various designs of TDR prostheses have been shown to migrate, disassociate, and fracture or extrude their polyethylene components. Such device migration or extrusion may place the vascular structures that lie anterior to the disc space at risk, causing pain of a vascular etiology in addition to mechanical pain.13 Additionally, TDR designs that incorporate fins for stability have been reported to result in end plate
CHAPTER 50
Persistent Pain After Lumbar Total Disc Replacement
391
B
A
C
D
n F I G U R E 50–1. These radiographs are from a patient who underwent lumbar disc replacement with the ProDisc-L. A, The pars defect was not obvious on preoperative radiograph. The patient had complete relief of his low back pain for 18 months following surgery. The relief of pain allowed the patient to return to work with frequent heavy overhead lifting. B, Follow-up radiographs demonstrated the spondylolysis. Two sets of pars injections each afforded the patient complete relief of his pain. Despite the posterior column instability, the ProDisc-L did not fail. C and D, Radiographs of pars repair that subsequently afforded the patient complete pain relief and a return to his heavy lifting job.
P A R T
392
I V
Lumbar Total Disc Arthroplasty
B
A
C n
F I G U R E 50–2. This patient awoke following surgery with new-onset leg pain and a neurologic deficit with a foot drop. A, Postoperative evaluation demonstrated that the L5 nerve root had become entrapped between a displaced fragment of the posterior inferior aspect of the L5 vertebral body and the L5 pedicle. B, Impaction of one of the teeth from the CHARITE´ end plate had caused the L5 end plate to fracture and a fragment to displace. Following return to the operating room, the fragment was removed and the nerve root decompressed. The postoperative leg pain as well as the preoperative low back pain resolved, but the motor deficit remained. C, The implant subsided and has no motion 2 years postoperatively.
fractures.14 A case report of bilateral pedicle fracture following TDR has also been reported.15 DISEASE PROGRESSION Transition syndrome occurs when adjacent-segment disease progresses adjacent to a previous fusion due to the increased load placed on the
neighboring motion segment and the consequent hypermobility of that spinal motion segment. In posterior spinal procedures, there may also exist an iatrogenic component because dissection or instrumentation may cause harm to the facet joints. Based on a Kaplan-Meier survivorship analysis, Ghiselli et al16 concluded that 37% of the patients in their cohort would require an additional surgical procedure for adjacentsegment disease of the lumbar spine within 10 years of their index
CHAPTER 50
A
Persistent Pain After Lumbar Total Disc Replacement
393
B
n
F I G U R E 50–3. Failure to align the implant in the midline may lead to diminished postoperative pain scores. A, A well-aligned ProDisc-L. B, An offset from the midline and a diminished outcome.
procedure. Adjacent-segment disease may include progression of degenerative disc disease, facet arthropathy, spinal stenosis, or a new or exacerbated herniated nucleus pulposus. Among the goals of TDR is to impede or prevent the cascade of adjacent spinal segment degeneration by normalizing the stresses placed upon adjacent motion segments when compared with spinal arthrodesis. Despite designing disc replacement prostheses to accomplish this specific goal, there is yet insufficient follow-up and data to fully support this hypothesis. Moreover, although lumbar total disc prostheses are designed to replicate the normal physiology of the intact native intervertebral disc, the complex physiology of the FSU makes achieving this task very difficult. Thus, adjacent-segment disease may still occur and serve as a source of persistent pain, especially when some element of degenerative disease is pre-existing at that level. EARLY-ONSET POSTOPERATIVE PAIN Almost immediately following surgery, the patient should report some relief of axial back pain despite the expected postoperative discomfort. If the patient's symptoms are unchanged, it is possible that the symptomatic pathology was inadequately addressed due to surgery at the wrong level or, more likely, missed pathology. Unchanged or worsened axial back pain may be due to incorrect device positioning or assembly, leading to limited ROM, device migration, or device dissociation. Leg pain in the context of temperature differences between the lower extremities warrants a vascular evaluation, because it is possible that the necessary retraction during surgery may have resulted in vascular injury or occlusion. Decreased or absent pulses should trigger an examination with
duplex ultrasound. More subjective patient complaints concerning temperature changes may arise from disruption of the sympathetics during the surgical exposure. In this circumstance, observation is indicated. Following exclusion of a vascular etiology, neurogenic leg pain should be investigated. Overseating of the TDR posterior to the posterior vertebral margin may cause neurogenic claudication due to incursion into the spinal canal. Such a finding necessitates revision of the prosthesis. New-onset central stenosis may also occur due to iatrogenic retropulsion of a bone fragment into the central canal. Bilateral radiculopathic symptoms may result from overdistraction of the intervertebral disc space stretching the nerve roots. In most cases, this phenomenon is self-limiting, with resolution over weeks to months without the need for surgical revision. Unilateral radiculopathy may result from migration of either disc material or a bone fragment into the lateral recess or neural foramen. Computed tomography–myelogram may be of diagnostic utility in evaluation of the neural elements and delineation of compressive pathology. SUBACUTE POSTOPERATIVE PAIN Following an initial period of postoperative improvement, a recurrence or new onset of pain may occur. It is common for patients to report new or recurrent discomfort with increased level of activity or participation in a physical therapy program. This discomfort is likely to be self-limiting and must be differentiated from more serious and persistent pathology. A history of trauma or acute onset of pain may be associated with device failure or fracture of the end plate. Such a history of acute new-onset pain requires prompt
394
P A R T
I V
Lumbar Total Disc Arthroplasty
radiographic evaluation. A more insidious onset may correspond to device migration or infection. Again, evaluation with plain radiographs is indicated. Insidious pain associated with constitutional symptoms should be investigated for infection. Laboratory testing for infection markers is obligatory, including a complete blood count, erythrocyte sedimentation rate, and C-reactive protein. Nuclear imaging may also be of diagnostic use because the accuracy of magnetic resonance imaging will be degraded secondary to artifact from the implant. Combined 67Gallium citrate and technetium-99m-MDP studies are the current gold standard. LATE POSTOPERATIVE PAIN Onset of pain 6 months after surgery is considered late postoperative pain. Again, infection should be excluded as a diagnosis. Late postoperative pain may be related to the implant itself. Wear and fracture of polyethylene components has been reported for certain implants, as has subsidence and migration.17 Progression of facet disease may also be responsible for gradually progressive mechanical back pain. Continued degenerative changes of the facet joints and ligamentum flavum may lead to spinal stenosis. Adjacentsegment disease may also occur, requiring treatment of additional levels of pathology. SUMMARY Preoperative psychological screening may help avoid unsuccessful surgery for patients with personality or emotional disorders that may hamper a successful surgical outcome. A vigilant and meticulous investigation and work-up is integral to the diagnosis of the cause of persistent pain after lumbar disc arthroplasty. A systematic evaluation of patients with persistent pain after disc replacement surgery should include identification of the location and character of the pain; the chronicity of the pain and any pain-free interval; and appropriate use of imaging, injections, and neurologic studies. REFERENCES 1. Trief PM, Grant W, Fredrickson B: A prospective study of psychological predictors of lumbar surgery outcome. Spine 25:2616–2621, 2000. 2. Waddell G, McCulloch JA, Kummel E, Venner RM: Nonorganic physical signs in low-back pain. Spine 5:117–125, 1980.
3. Carragee EJ, Lincoln T, Parmar VS, Alamin T: A gold standard evaluation of the “discogenic pain” diagnosis as determined by provocative discography. Spine 31:2115–2123, 2006. 4. Junge A, Dvorak J, Ahrens S: Predictors of bad and good outcomes of lumbar disc surgery: A prospective clinical study with recommendations for screening to avoid bad outcomes. Spine 20:460–468, 1995. 5. Spengler DM, Ouellette EA, Battie M, Zeh J: Elective discectomy for herniation of a lumbar disc: Additional experience with an objective method. J Bone Joint Surg Am 72:230–237, 1990. 6. Block AR, Ohnmeiss DD, Guyer RD, et al: The use of presurgical psychological screening to predict the outcome of spine surgery. Spine J 1:274–282, 2001. 7. Klekamp J, McCarty E, Spengler DM: Results of elective lumbar discectomy for patients involved in the workers' compensation system. J Spinal Disord 11:277–282, 1998. 8. Taylor VM, Deyo RA, Ciol M, et al: Patient-oriented outcomes from low back surgery: A community-based study. Spine 25:2445–2452, 2000. 9. DiAngelo DJ, Foley KT, Morrow B, et al The effects of over-sizing of a disc prosthesis on spine biomechanics. Spine Arthroplasty Society Sixth Annual Global Symposium on Motion Preservation Technology, Montreal, Canada, May 2006, p 97. 10. McAfee PC, Cunningham B, Holsapple G, et al: A prospective, randomized, multicenter Food and Drug Administration investigational device exemption study of lumbar total disc replacement with the CHARITÉ artificial disc versus lumbar fusion, Part II: Evaluation of radiographic outcomes and correlation of surgical technique accuracy with clinical outcomes. Spine 30:1576–1583; discussion E388–390, 2005. 11. Yaszay B, Quirno M, Bendo J, et al: Effect of intervertebral disk height on post-operative motion and outcomes following Prodisc-L lumbar disk replacement (in press) 2007. 12. McAfee PC, Cunningham BW, Devine J, et al: Classification of heterotopic ossification (HO) in artificial disk replacement. J Spinal Disord Tech 16:384–389, 2003. 13. Stieber JR, Donald GDR: Early failure of lumbar disc replacement: Case report and review of the literature. J Spinal Disord Tech 19:55–60, 2006. 14. Shim CS, Lee S, Maeng DH, Lee SH: Vertical split fracture of the vertebral body following total disc replacement using ProDisc: Report of two cases. J Spinal Disord Tech 18:465–469, 2005. 15. Mathew P, Blackman M, Redla S, Hussein AA: Bilateral pedicle fractures following anterior dislocation of the polyethylene inlay of a ProDisc artificial disc replacement: A case report of an unusual complication. Spine 30:E311–E314, 2005. 16. Ghiselli G, Wang JC, Bhatia NN, et al: Adjacent segment degeneration in the lumbar spine. J Bone Joint Surg Am 86A:1497–1503, 2004. 17. van Ooij A, Oner FC, Verbout AJ: Complications of artificial disc replacement: A report of 27 patients with the SB CHARITÉ disc. J Spinal Disord Tech 16:369–383, 2003.
CHAPTER
51
DASCOR Michael Ahrens, Anthony Tsantrizos, and Jean-Charles Le Huec
K E Y l
l
l
l
l
P O I N T S
The DASCOR device is a two-part in situ–cured polyurethane injected within an expandable polyurethane balloon to form the final nucleus replacement device that conforms to the nucleus cavity created. Pressurized device injection in the disc space is possible using a proprietary injection system. Ability to verify intraoperatively the volume and size of the resultant DASCOR device before implantation using fluoroscopic imaging procedures in conjunction with a proprietary DASCOR imaging balloon. This catheter-based nucleus replacement technology can be used not only for minimally open procedures but also with percutaneous and endoscopic procedures. The technology has the capability of implanting the DASCOR device using an anterior, anterolateral, lateral, or posterolateral approach.
INTRODUCTION Repairing a flat bicycle tire often requires replacement of the inner tube rather than complete wheel replacement. Similarly, this analogy can extend to the intervertebral disc, depending on its degenerative condition, can be subject to different treatment options. Total disc replacement (TDR) implants have been proposed as alternatives to arthrodesis for treatment of chronic low back pain (LBP) with degenerative disc disease as the etiology. In essence, TDR implants would be analogous to removal of the complete wheel. However by replacing only the nucleus, that is, the inner tube of the disc, the clinical goals of pain relief, anatomic and biomechanical function restoration, and limiting progression of adjacent-level disease can be reestablished with less traumatic surgical techniques compared with TDRs. Nucleus replacement techniques are rapidly reemerging as advanced treatment options for mild-to-moderate stages of degenerative disc disease since the days of the Fernstrom endoprosthesis.1 Renewed interest in nuclear replacement, a minimally invasive treatment alternative, has resulted from novel materials and technologic innovations. Current nucleus replacement technologies can be classified as preformed or in situ–formed implants, despite the differences in materials used (Fig. 51–1). The DASCOR device (Disc Dynamics Inc. Eden Prairie, MN) is intended to alleviate discogenic pain and
restore disc height and segmental mobility while preserving the patient's bony and axial connective tissue/muscular support anatomy. The DASCOR device has received CE-Mark approval for commercial sale in the European Union in July 2005. The system has been in clinical use in trials outside the United States since 2003. The implant is also continually being used in Europe in the scope of an ongoing postmarket approval trial. The company also received approval from the U.S. Food and Drug Administration (FDA) in July 2006 to begin a clinical trial of the DASCOR device in the United States. INDICATIONS/CONTRAINDICATIONS At present, the DASCOR device is advocated for implantation in patients with the following indications: l Age between 18 and 70 years l Single-level degenerative disc disease as confirmed by patient's history, physical examination, or computed tomography (CT) or magnetic resonance imaging (MRI) l Persistent pain and/or symptoms after at least 6 months of nonoperative care l Leg pain, back pain, or both in the absence of nerve root compression per MRI or CT scan, without prolapse or narrowing of the lateral recess l Minimum 5.5-mm disc height at the affected level l Scarring or thickening of ligamentum flavum, annulus fibrosis, or facet joint capsule l Condition involving only one lumbar disc level from L2 to the S1 l Oswestry Disability Index (ODI) score greater than 40 (based on 100-point scale) l Back pain score of at least 5 on a 10-point Visual Analog Scale (VAS) l Intact end plates at the affected disc level l Positive CT discography at the affected level with concordant pain l
Contraindications include the following: Disc herniation or reherniation with radicular pain requiring posterior discectomy 397
398
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
Pre-formed
l
l
l
l l
l l l
l l
In-situ formed (injectable)
Adjustable geometry
Limited expandable geometry
Fixed geometry
Non-contained (Free injection)
Contained (pressure injection)
NewCleus Nautilus
NeuDisc PDN/HydraFlex Aquarelle
Regain NUBAC Satellite
NuCore SINUX BioDisc
DASCOR
Prior invasive treatment of the disc at the implant level, such as discectomy Previous diagnosis of osteopenia, osteoporosis, spinal osteoarthritis, ankylosing spondylitis, advanced Scheurermann's disease, history of inflammatory bowel disease, psoriatic arthritis, Reiter's syndrome, rheumatoid arthritis or inflammatory arthritic disease, spinal tumor or other known malignancy, or arachnoiditis Isthmic spondylolisthesis or degenerative spondylolisthesis greater than 3 mm Prior lumbar fusion surgery at any level Moderate or severe central spinal, foraminal, or lateral recess stenosis, or cauda equina syndrome Facet joints that are absent, fractured, or severely degenerated Obesity, as defined by a body mass index of greater than 35 Active systemic or local infection in the area of the planned surgery Incompetent annulus History of communicable disease such as human immunodeficiency virus (HIV) or hepatitis
DESCRIPTION OF THE DEVICE The DASCOR device is fabricated by mixing a methylenediphenyldiisocyanate (MDI)–based polyurethane two-part reactive system and injecting this mixture under controlled pressure while it is still liquid, through a catheter to an expandable balloon that is placed in the prepared nuclectomy space (Fig. 51–2). The polymer is contained within the expandable balloon and cures to a firm, but pliable form within minutes while concurrently bonding to the expandable balloon, forming the final device. The temperature at the interface released during the curing process is within the range of acceptable physiologic temperatures (less than 50 C) and well below typical interfacial temperatures measured with bone cements used for vertebroplasty procedures and with intradiscal electrothermal therapy (IDET) procedures. The DASCOR device has the ability to contour and conform to any nucleus cavity created during nuclectomy because the liquid polymer is being injected under pressure using a customized injection system (Fig. 51–3). This feature creates a large implant footprint and a volume (average 4 to 5 mL) through a small annulotomy, thus enabling better axial load transfer capabilities through the implant as well as radially to the surrounding annulus. The ability to inject under pressure permits the DASCOR device to immediately distract the disc space and, on curing, obtain a final implant geometry and
n
F I G U R E 51–1. Classification of current nucleus replacement technologies.
n F I G U R E 51–2. The DASCOR device. (From Disc Dynamics, Inc. Reprinted with permission.)
disc space distraction without having to rely on hydration. On the other hand, hydrogel nucleus implants do rely on the material's intrinsic swelling pressure to fully hydrate with time to attain final implant geometry and disc space distraction. The final cured polymer inside the balloon is an incompressible isotropic aromatic polyether polyurethane with an optimal mixture of hard and soft segments to optimize resilience, flexibility, and mechanical strength. A typical compression modulus of the device per ASTM D575 ranges between 4 and 6 MPa, with an ultimate compressive stress and strain of greater than 25 MPa and 90%, respectively. The low balloon catheter profile offers the advantage for percutaneous implantation procedures. The balloon component is an aliphatic polycarbonate urethane that has excellent cavity expansion and conforming capabilities and is attached to a 5.5-mm delivery catheter. The balloon component has a large safety margin for preventing burst or leakage and is capable of holding more than 10 times the average injection volume and 2 times the maximal applied injection pressure. BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES Extensive preclinical studies have been conducted on the DASCOR device, including mechanical bench and biomechanical testing as well as biocompatibility and biologic response testing in animal models.
CHAPTER 51
DASCOR
399
INJECTOR SYSTEM DISPLAY PANEL Display panel
Full profile
150
125
Zoomed profile
Pressure (P–S:)
100
Implant volume 75
50
25
0
Injector head
Hand control
0:00
–10
9:25
10:00
10:45
11:25
Time
11:51
Implant curing Please wait
Cart
8.0 7.0 6.0 5.0 4.3 4.0 3.0 2.0 1.0 .0
5.3
DELIVERY COMPONENTS
Mixer
Foot control
Pressure transducer
Balloon
Catheter
Two-part Vacuum polymer line cartridge n F I G U R E 51–3. DASCOR-assembled injection system with injector head, polymer cartridge, mixer, and balloon catheter. The two-part liquid polymer is packaged in a dual chamber cartridge. The mixed pressurized polymer enters the containment balloon, which has been placed in the disc space, expanding the balloon to fill the entire space. (From Disc Dynamics, Inc. Reprinted with permission.)
Mechanical Bench Testing Axial Compressive Fatigue Strength
The device's fatigue strength was investigated per American Society for Testing and Materials (ASTM) F2346 in a custommade six-station mechanical testing apparatus using unconstrained device conditions (i.e., no simulated annulus) to create a worst-case test construct. Samples were tested using various cyclic axial compressive stress conditions ranging from 7.65 MPa to 2.0 MPa at an applied load frequency of 3 Hz. Testing was concluded once samples from a particular axial stress condition were able to sustain a 10 million cycle runout without structural deterioration leading to failure. The 10 million cycle axial compressive fatigue strength of the device after constructing an S/N plot was determined to be 2.94 MPa with a mean cyclic axial strain of 35.6% (Fig. 51–4). The axial compressive strength of the device is significantly beyond the daily axial compressive stress experienced by the lumbar motion segment. Durability and Particle Analysis
Similarly, the device's durability was investigated per ASTM F2346 in the same custom-made six-station mechanical testing apparatus using the same unconstrained device conditions (i.e., no
simulated annulus) to create a worst-case test construct. The durability of the device was assessed for mechanical loads simulating brisk walking for testing up to 25 million cycles with concurrent 41-month device aging. Cyclic loading conditions consisted of 5.5 degrees flexion-extension angular displacement applied at a load frequency of 3 Hz combined with 1.8 MPa axial compressive stress applied at a load frequency of 1.5 Hz. None of the DASCOR device samples tested for up to 25 million cycles showed signs of device degradation, whereas the compressive modulus did not significantly change from initial values. Aging did not influence the mechanical performance of the device. Furthermore, the average wear rates measured for devices tested during the first 10 million cycles was only 0.26 mg/million cycles, and no significant permanent set was observed (permanent set less than 3.2%). Progression of wear was slow, and the wear rates obtained were similar to those reported for other nucleus replacement devices2 but magnitudes less than wear rates reported for TDR devices composed of polyurethanes.3 Particle analysis of retrieved solutions conducted per ASTM F1877 using SEM demonstrated particle quantity and size, which was significantly less than what would be expected for ultra-high-molecular-weight polyethylene (UHMWPE) implants used in hips and knees.
P A R T
400
V
Lumbar Partial Disc Replacement: Nucleus Replacement n
F I G U R E 51–4. A plot of stress versus number of cycles (S/N plot) constructed in order to establish the endurance limit of the DASCOR device.
9.0 8.0
Axial stress (MPa)
7.0 6.0 5.0 4.0 3.0 2.0 1.0
Fitted S/N equation (R2 = 0.824): Stress (MPa) = –0.863 Ln (cycles) + 16.941
*
0.0 0
1
2
3
4
5
6
7
8
9
10
Number of million cycles
Biomechanical Testing Human cadaveric lumbar spines were used to study the biomechanics of the DASCOR device with two aims: (1) to determine the ability of the DASCOR device to restore the multidirectional segmental flexibility of a nuclectomy motion segment construct to that of an intact construct4 and; (2) to determine and compare the end plate contact stress (Fig. 51–5) and load transfer capabilities of an instrumented DASCOR motion segment construct during multidirectional flexibility loading to motion segment constructs instrumented with a water balloon implant simulating hydrostatic conditions.5 Axial compression (1,200 N) as well as axial rotation, flexion/extension, and lateral bending (all at 7.5 Nm combined with 500 N compression) were the loading conditions tested. The results demonstrated that the device was able to restore the
segmental flexibility lost after a nuclectomy while still preserving segmental level biomechanics to within approximately 5% of the intact motion segment behavior. Similarly, a relatively uniform contact stress distribution was observed during all segmental flexibility loads applied for both DASCOR and water balloon implanted constructs with no differences found between these constructs. The latter study results demonstrated that the modulus of elasticity of the DASCOR device was well suited to its intended purpose of nucleus replacement. Animal Model and Biomaterials Safety Testing Biocompatibility performed according to an extended ISO 10993 (12 tests) demonstrated that the DASCOR device was a biocompatible material. Extensive animal model testing was similarly
END PLATE CONTACT STRESS MEASUREMENT METHOD Transducer placement
Axial compression Each element within the transducer measures a range of stress identified with a color gradient. The averaging of color gradient is what defines stress uniformity.
Stress gradient (low → high) 0
12
24 35 47 58 70 81 93 104 116 127 139
121-element pressure transducer placed between the implant and end plate n
F I G U R E 51–5. A pressure-sensitive transducer was placed between the implants and the end plate in order to measure end plate–implant contact stress during the multidirectional flexibility testing. The resulting stress measurements consisted of an array of color-graded stress profiles measured within various locations of the pressure-sensitive transducer.
CHAPTER 51
conducted on the DASCOR device in (1) A RasH2 mouse model for carcinogenicity potential verification6; (2) a dorsal laminectomy rabbit model to determine the biologic response to wear debris7; and (3) a functional baboon model to determine the device's biodurability.8 Carcinogenicity Testing Hydrolysis of free MDI can form 4,40 -methylenedianiline (MDA), which has been reported to be a suspected carcinogen in humans. No scientific evidence has linked aromatic MDI-based polyurethanes to cancer in humans. Nevertheless, because the aromatic MDI-based polyurethane of the DASCOR device is cured in situ within the device's aliphatic balloon barrier, there is still a risk of releasing some free MDI/MDA into the surrounding tissues. Therefore, a study was undertaken to evaluate the carcinogenicity potential of the DASCOR device and its presumed byproducts using the rasH2 transgenic mouse model. A protocol for a 26-week subcutaneous implantation was developed using seven treatment groups. Methylcellulose Millipore filter discs with 0.05-m pores and a 20-mm2 diameter were used as vehicle carriers for each test article. The treatments included (1) an uncured DASCOR polyurethane; (2) DASCOR polyurethane wear debris; (3) UHMWPE wear debris as a material control; (4) a low- and high-dose of free MDA (50 times and 1 million times the maximum amount in a DASCOR device, respectively); (5) a filter control, and (6) an ethyl carbamate–positive chemical control group. Gross necropsy observations after sacrifice included number, size, and location of tumors and histopathologic examination of implant and major visceral organ sites. According to the gross necropsy and histopathologic findings, ethyl carbamate (positive control) demonstrated the expected carcinogenic response, supporting the validity of the mouse model. Otherwise, all other treatment groups confirmed no body weight changes, no significant clinical observations, and no organ weight differences. There was no increase in tumor incidence at the test site or in other tissues associated with the DASCOR uncured polyurethane and wear debris test treatments. Similarly, there was no evidence at the test site or in other tissues of carcinogenic changes associated with the free MDA dose treatments. Based on these experimental conditions and end points, it was demonstrated that the MDI-based nature of the DASCOR device does not cause any carcinogenic changes. Biologic Response to Wear Debris The biologic response to implant wear debris is an essential element in device safety testing. A dorsal laminectomy rabbit model was used to evaluate (1) the local and systemic biologic responses to DASCOR device wear debris placed in the epidural space, and (2) the potential for DASCOR device wear debris translocation and any biologic tissue response resulting from wear debris translocation. Seventy-two male New Zealand white rabbits were stratified into the following experimental groups: (1) two DASCOR wear debris groups (one for the balloon and one for the cured two-part polyurethane component); (2) an UHMWPE group as the
DASCOR
401
material control; and (3) a sham group that had only a dorsal laminectomy performed. Dried wear debris particles were generated in vitro with a size distribution ranging between 0.1 and 100 mm. At least 50 million particles had a size distribution ranging between 1 and 15 mm. A 20-mg dose of wear debris was applied over the exposed dura mater obtained from the cured two-part polyurethane component. Similarly, a 4-mg dose was applied for wear debris obtained from the balloon component and the UHMWPE samples. At 6 weeks, 3 months, and 6 months postoperatively, six animals from each experimental group were euthanatized. All animals were immediately necropsied and histopathologically examined, and the results were compared between groups. There were no macroscopic changes that could be attributed to wear debris. Microscopically, the local tissue reaction to wear debris was similar between the wear debris groups and consisted of a histocytic inflammatory response common to an inert, nonirritating material. Wear debris was present in all animals and wear debris groups and at all postoperative time points. There were no significant tissue alterations due to DASCOR or balloon wear debris implanted in direct contact with the spinal column and nerve roots. Most of the wear debris was located within the epidural space, with relatively less within the bone marrow and within the perivertebral region along the outer margins of the new bone formed in repair of the surgical defect. Wear debris particles were not detected internal to the dura or within the spinal cord in any of the groups. There was no evidence of neural tissue reaction to the test articles. No test article–related tissue alterations or particles were observed in distant organs that could be associated with wear debris from the implant site. Wear debris particles did not adversely affect the healing of the surgical defect. The results from the study confirmed that the DASCOR device does not pose a significant biomaterials safety concern and could be classified as an inert, non-irritant material. Functional Biodurability A baboon study was designed to evaluate the biodurability and elicited histopathologic response of the DASCOR device following long-term implantation. A total of 14 mature male baboons were randomized into a 6-month (n ¼ 7) and 12-month (n ¼ 7) postoperative time period. Each animal underwent a lateral transperitoneal surgical approach, followed by a complete nucleotomy at the L3-L4 and L5-L6 levels with a DASCOR device implantation at the latter level. The L3-L4 served as a surgical (nucleotomy) control in each case. Postmortem analysis for each of the 6- and 12-month postoperative time points included MRI and plain radiography assessment for implant position and end plate changes, subsequent multidirectional flexibility testing and analysis on retrieved motion segments, and local or systemic histology. All animals survived the operative procedure and postoperative interval without significant intraoperative or perioperative complications. Modic Type 1 changes were observed to a greater extent at the implanted versus control levels; however, these changes appeared to reduce from the 6- to 12-month intervals. Modic Type 1 changes observed at the implanted sites were not considered significant and have been documented in a number of other in vivo
402
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
non-human primate studies. Multidirectional flexibility testing indicated trends for restoration of segmental motion for the DASCOR implanted constructs under flexion-extension and lateral bending by the 12-month time interval. Finally, the vertebral and systemic tissues indicated no evidence of significant histopathologic changes. The study demonstrated that the DASCOR device was biodurable and able to withstand physiologic loading with an insignificant elicited local and systemic histopathologic response. However, it should be acknowledged that implant volumes obtained in the baboon study were significantly less than those used in humans, a study limitation expected with such an animal model. It also was noted in these studies that significant disruption or damage to the vertebral end plate cartilage during the discectomy procedure can lead to adverse changes not related to the device itself. The extensive preclinical testing of the DASCOR device served as the scientific basis for supporting initial and ongoing clinical investigations into the use of the device since investigations were initiated in 2003.
CLINICAL PRESENTATION AND EVALUATION The 2-year clinical experience of the DASCOR system evaluating safety and effectiveness is being investigated in two multicenter prospective nonrandomized European studies that use nearly identical investigational protocols.9 Whereas the first study is already closed, the second study will close its enrollment in 2007. A standardized retroperitoneal midline or lateral approach was used to implant the DASCOR device in both studies. As of October 2006, 60 eligible patients (30 women, 30 men) with a mean age of 399 years meeting the specific inclusion criteria described had the DASCOR device implanted at the L3-L4 (n ¼ 3), L4-L5 (n ¼ 26), and L5-S1 (n ¼ 31) levels. Clinical success was defined as a 2- and 15-point decrease in the respective VAS and ODI scores. As of October 2006, of the 60 patients who underwent surgery, 44 patients were followed for up to 6 months, whereas 22 and 17 patients were followed for 1 and 2 years, respectively. Mean operating time was 8929 minutes and average blood loss was 43.170.2 mL. Clinical outcome measures have been tracked from preoperative measures up through 24 months (Fig. 51–6A–C). Mean preoperative VAS and ODI scores improved significantly after 6 weeks postoperatively and throughout the 2 years. Clinical success criteria were all met at the time of write-up. Mean VAS backpain scores showed a 62% reduction from a preoperative mean of 7.5. Mean ODI results showed a 73% reduction from a preoperative mean of 58. Analgesic or narcotic drug use based on a three-point scale decreased 88% from a preoperative average of 1.7, with most patients experiencing significant improvement after 3 months. The patient analgesic medication or narcotic drug use at the 2-year follow-up was nearly none. MRI evaluations performed by an independent radiographic facility did not show any device subsidence or expulsion (Fig. 51–7). Anteroposterior and lateral radiographic films showed that disc height was improved. Overall, the 2-year clinical experience with the DASCOR device
demonstrated high clinical safety based on the significant postoperative pain reduction, functional improvement, and a low complication rate.
OPERATIVE TECHNIQUES Anesthesia The anterior or lateral implantation of the DASCOR device is performed under general anesthesia. Normally, a Foley catheter is introduced into the bladder for intraoperative drainage and collapse of the bladder, especially for the L5-S1 approaches. Nitrous oxide is not recommended for general anesthesia, especially when transperitoneal access is under consideration. For intraoperative monitoring, central venous and arterial lines for continuous arterial blood pressure management have been recommended for TDR. However, because the implantation of the DASCOR device does not require a retraction of the major vessels to the degree needed in TDR procedures, the necessity for such intraoperative monitoring is less. With the availability of the minimally invasive posterolateral procedures, the risk of bleeding might also be less; hence, normal monitoring as in any other dorsal minimally invasive surgical procedure should be sufficient. With the availability of an endoscopic implantation of the DASCOR device, even sedation without intubation seems possible, as in other endoscopic disc procedures.
Position Fluoroscopy is used during the surgery, and therefore, patient positioning should accommodate its use. For an anterolateral approach, the patient should be placed in a supine or lateral position. If a true lateral approach is planned, a lateral decubitus position may be used with appropriate supports. For a posterior/posterolateral approach, the patient should be placed in a prone position on a kneeling radiolucent lumbar spine frame.
Procedure The DASCOR prosthesis has the potential of being implanted using multiple surgical approaches (anterior, anterolateral, lateral, or posterolateral) with minimal variation in the surgical technique. Key to any approach, however, is achieving total nucleus removal (TNR) and minimizing annular and end plate cartilage disruption. The implantation of a nucleus replacement device should not be perceived to be a simple adjunct to a standard discectomy for herniated nucleus pulposus. The initial annulotomy is created by using a 5.5-mm trephine or preferably a slit incision with a No. 11 scalpel and subsequent dilation to 5.5 mm. Care is taken to avoid damage to the annulus and cartilaginous end plates. To create a DASCOR device with the most optimal size, end plate conformity, and load transfer capabilities, TNR10 must be attained while preserving the
CHAPTER 51
DASCOR
100 90 80 Maximum postoperative ODI to define clinical success
70
Score
60 50 40 30 20 10 0 PreOp (n = 60)
6 wk (n = 55)
3 mo (n = 53)
6 mo (n = 44)
12 mo (n = 22)
24 mo (n = 17)
A 10 9
Maximum postoperative VAS to define clinical success
8 7
Score
6 5 4 3 2 1 0 PreOp (n = 60)
6 wk (n = 55)
3 mo (n = 53)
6 mo (n = 44)
12 mo (n = 22)
24 mo (n = 17)
B High med use
3.0
2.5 Moderate med use
Score
2.0
1.5 Low med use
1.0
0.5 No med use
0.0 PreOp (n = 60)
6 wk (n = 55)
3 mo (n = 53)
6 mo (n = 44)
12 mo (n = 22)
24 mo (n = 17)
C n
F I G U R E 51–6. Clinical study results: Oswestry Disability Index (ODI) (A); pain measurement on Visual Analog Scale (B); analgesic medication use (C).
403
404
P A R T
V
Pre-op
Lumbar Partial Disc Replacement: Nucleus Replacement 6 weeks
12 months
24 months
n
F I G U R E 51–7. Magnetic resonance imaging scan of a typical implanted DASCOR device leading up to the 2-year follow-up. Images show the implant device centrally located without signs of subsidence or inflammatory reaction. (From Disc Dynamics, Inc. Reprinted with permission.)
annulus and end plates. Based on previous human cadaveric studies, a set of standard pituitary rongeurs were chosen and customized pituitary rongeurs were developed based on each instrument's ability to reach a particular predefined mapped region of the nucleus (Fig. 51–8). Using these instruments and a standardized method for removing tissue from predefined mapped anatomic regions, a preliminary TNR is performed until the surgeon believes that adequate nucleus has been removed. Following the preliminary TNR, the nuclectomy performed and the tentative implant size, shape, and volume resulting from the nuclectomy created are evaluated using proprietary imaging methods developed specifically as part of the DASCOR Disc Arthroplasty System. An imaging balloon catheter is inserted in the disc space, and contrast media is injected under pressure, filling the balloon within the nucleus cavity created. Fluoroscopic images in anteroposterior, lateral, and oblique views are then taken, and a two-dimensional reconstruction of the imaged cavity is then performed in order to assess the completeness and symmetry of the nuclectomy (Fig. 51–9). If the cavity is not symmetric and centrally located, additional steps for nucleus removal and fluoroscopic imaging are conducted. Posterior annular integrity can also be assessed at this time. The contrast media and imaging balloon are then removed, and the implant balloon is inserted. Once the balloon catheter is connected to the assembly components of the injector system, the balloon test and injection may be performed. The injection pressure profile applied by the injection
5 4
1 3
2
Anterior
4 Lateral
3
1
5
2
n F I G U R E 51–8. A step-by-step approach used to achieve a total nucleus removal (TNR) by nucleus removal of one zone at a time using specific instruments that target each anatomic zone. Specific TNR maps were created for an anterior or lateral or posterolateral approach. (From Disc Dynamics, Inc. Reprinted with permission.)
CHAPTER 51
DASCOR
405
0º
A
90º
B
45º
C
–45º
D n F I G U R E 51–9. A to D, Intraoperative imaging steps using a proprietary contrast-filled DASCOR imaging balloon with anteroposterior, lateral, and oblique fluoroscopic C-arm positions. The procedure is used to identify remaining nucleus and anticipate the volume, position, and geometry of the actual implanted DASCOR device.
system is computer controlled and developed by the company while taking into consideration the anatomic/biomechanical limitations of a lumbar motion segment. There are several safety features to start, control, and potentially abort the injection at any time during the procedure should the surgeon feel it necessary. Once the injection is complete and the optimal filling has been achieved, the system allows for a 15-minute curing time before the catheter shaft is cut and removed from the balloon and the wound can be sutured and closed.
avoided. Even during the second week after surgery, sitting is avoided for long periods of time, because this puts more stress on the spine compared with standing. Walking is recommended, but lifting should be restricted to less than 20 kg (approximately 44 lbs.) for the first 4 weeks. In some cases, wearing a soft brace might be useful, but is required only if the patient's occupation involves lifting or bending, or if the patient needs support. Braces are typically worn during daytime hours for 4 to 6 weeks after surgery. COMPLICATIONS AND AVOIDANCE
POSTOPERATIVE CARE The patients are gradually mobilized postoperatively for the first 3 weeks, but allowed to stand up the first day postoperatively for soft tissue healing. Increasing activities in the second week is advised, and sitting for long periods of time the first week should be
It is crucial that the integrity of both the annulus and end plates are in a biomechanical and somewhat viable state in order to contain the nucleus implant and avoid implant extrusions. In that respect, the preoperative assessment of annular integrity is obviously important and needs to be added to the preoperative
406
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
diagnostic algorithm when nucleus replacement is chosen as the surgical treatment option. The procedure for implantation of the DASCOR device incorporates an imaging balloon step as a very important complementary test that is comparable to a dress rehearsal. This test provides feedback for positioning, volume, shape, and geometry of the final implant, and also most importantly, the annulus. Only with a satisfactory outcome of this test can the final implantation injection be conducted. Equally important is the ability to reduce the initial implantation trauma as much as possible so that postoperative recovery time can be minimized. Therefore, minimally invasive surgery, as with a percutaneous/endoscopic implantation of a nucleus implant, should be the ultimate goal attained by nucleus replacement technologies.
CONCLUSIONS/DISCUSSION The concept of being able to achieve pain relief and disc function restoration using a nucleus replacement procedure is appealing. An ideal nucleus replacement technology should allow for restoration of healthy motion segment, disc height, kinematics, and load distribution. Although no perfect device has been identified for replacing the nucleus, the DASCOR device does address many, if not all, of the perceived requirements of an optimal nucleus replacement device. There are significant differences in design features between the DASCOR device and other nucleus devices that may influence clinical results. The injectable, contained DASCOR device results in a large cross-sectional area that contours the nucleus cavity owing to its pressurized injection procedure. The favorable modulus elasticity of the device creates a more uniform load distribution across the end plates and surrounding annulus that may avoid the severe end plate reaction and device subsidence reported with other nucleus replacement devices. The DASCOR device is introduced through a small annulotomy, creating a nucleus implant of a large volume and size, thus making device expulsion unlikely. This contrasts with the extrusion rates reported for preformed devices that require a large annular window for insertion. Appropriate patient selection is the key
factor for every treatment, as is also the case with the DASCOR device. Preoperative annulus assessment is required for proper patient selection to ensure clinical successful outcomes. REFERENCES 1. Fernstrom U: Arthroplasty with intercorporal endoprosthesis in herniated disc and in painful disc. Acta Chir Scand Suppl 357:154–159, 1966. 2. Brown T, Bao QB, Kilpela T: Wear and mechanical durability of the NUBAC disc arthroplasty device. Spine Arthroplasty Society, 2006. Montreal, Canada, May 9–13, 2006. 3. Anderson PA, Rouleau JP, Bryan VE, Carlson CS: Wear analysis of the Bryan Cervical Disc prosthesis. Spine 28:S186–S194, 2003. 4. Ordway NR, Tsantrizos A, Yuan HA, Bowman B: Restoration of segmental kinematics following nucleus replacement with an in situ curable balloon contained polymer. North American Spine Society, 2005. Philadelphia, September 27–October 1, 2005. 5. Tsantrizos A, Ordway NR, Bao QB, Yuan HA: Endplate contact stress of the DASCOR device during segmental flexibility. North American Spine Society, 2006. Seattle, WA, September 26–30, 2006. 6. Lacy S, Streicker MA, Wustenberg W, et al: A RasH2 mouse model to study the carcinogenic potential of the DASCOR device. In XXX (ed): Spine Arthroplasty Society, 2006. Montreal, Quebec, Canada, May 9–13, 2006, p 46. 7. Lacy S, Johnson C, Long PH, et al: A rabbit model to evaluate the biological response to DASCOR device wear debris. Spine Arthroplasty Society, 2007. Berlin, May 1–4, 2007. 8. Cunningham BW, Hu N, Beatson HJ, et al: An investigational study of nucleus pulposus replacement using an in-situ curable polyurethane: An in vivo non-human primate model. Spine Arthroplasty Society, 2007. Berlin, May 1–4, 2007. 9. Ahrens M, Donkersloot P, Maartens F, et al: Nucleus replacement with the DASCOR Disc Arthroplasty System: Two year follow up results obtained from two prospective European multi-center clinical studies. Spine Arthroplasty Society, 2007. Berlin, May 1–4, 2007. 10. Ahrens M, Sherman J, LeHuec JC, et al: Total nucleus removal from a posterior or posterolateral approach: Technical considerations and early clinical experience. Spine Arthroplasty Society, 2006. Montreal, Canada, May 9–13, 2006.
CHAPTER
52
PDN-SOLO and HydraFlex Nucleus Replacement System Reginald J. Davis
K E Y l
l
l
P O I N T S
One of the more unique features of the spinal disc is the ability of the nucleus tissue to hydrate and participate in the distribution of mechanical loads while providing a conduit for the transport of nutrients and waste materials. This hydrogel-based technology represents an optimization of the clinically proven PDN-SOLO device, with the HydraFlex device having shorter tab lengths, which allow for greater hydrogel volume, resulting in a more compliant implant. The HydraFlex device is indicated in skeletally mature adults between 25 and 70 years of age with symptomatic degenerative disc disease (DDD) at one level from L2-S1.
INTRODUCTION Replacement of the diseased nucleus pulposus portion of a spinal intervertebral disc represents a viable surgical solution to a known clinical problem. Although the concept seems straightforward, the ability to define an ideal solution has proven to be much more elusive.1 Many of the challenges associated with treating spinal disorders stem from the complexities of the disc itself. The disc's distinctive structure and physiology often make surgical intervention difficult. This is largely due to the fact that the disc form and function involves a combination of both biochemical and biomechanical processes. One of the more unique features of the spinal disc is the ability of the nucleus tissue to hydrate and participate in the distribution of mechanical loads, while providing a conduit for the transport of nutrients and waste materials. The natural hygroscopic glycosaminoglycan nucleus tissue cannot be reproduced and is extremely difficult to approximate by mechanical means.2,3 The HydraFlex (Raymedica, Inc., Minneapolis, MN) device offers an innovative solution to address the complexities of treating degenerative disc disease (DDD). This hydrogel-based technology represents an optimization of the clinically proven PDN-SOLO
device (Raymedica, Inc., Minneapolis, MN), with the HydraFlex device having shorter tab lengths, which allow for greater hydrogel volume, resulting in a more compliant implant (Fig. 52–1A and B). The implant design and material selection incorporate extensive knowledge of disc and nucleus biomechanics, physiology, and research evaluations.4 INDICATIONS/CONTRAINDICATIONS The HydraFlex device is indicated in skeletally mature adults between 25 and 70 years of age with symptomatic degenerative disc disease (DDD) at one level from L2-S1. (DDD is back pain of discogenic origin that has been confirmed by history and physical examination with degeneration confirmed radiographically.) Traditionally, such patients present with low back or leg pain, or both. Symptoms interfere with daily activities and have been present at least 6 months while not responding to conservative treatment. Severe symptomatic central spinal, foraminal, or lateral recess stenosis are contraindications for use of the device. Other contraindications include dynamic degenerative spondylolisthesis greater than grade 1 or lytic spondylolisthesis, fractured and/or symptomatic degenerated facet joints, pronounced Schmorl's nodules at the affected level, an incompetent annulus, disc height at the affected level less than 6 mm, significant osteoporosis or osteomalacia, tumors of the spinal cord, malignant tumors of the vertebral column, malignant or benign tumors at the adjacent or affected level, and surgery at the affected or adjacent level. Also contraindicated are symptomatic DDD requiring surgery at more than one lumbar level, active systemic or localized infection at the level where surgery will be performed, any trauma causing fracture of the bony element at the intended or adjacent level or spinal trauma leading to a neurologic deficit, congenital bony abnormalities at the affected level or congenital spinal cord abnormalities, gross obesity, disc height less than 6 mm, and allergy to polyacrylamide, polyethylene, or platinum iridium.
407
408
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement n
F I G U R E 52–1. A and B, HydraFlex vs. PDN-SOLO. The HydraFlex device’s shorter tab lengths allow for greater hydrogel volume, resulting in a more compliant implant.
HydraFlex
A
B PDN-SOLO
The HydraFlex device has three components: an inner copolymer hydrogel pellet that is designed to resume a biconvex shape once it is hydrated (Fig. 52–2A and B), an outer woven jacket of ultrahigh-molecular-weight polyethylene (UHMWPE) fibers, and platinum-iridium wire markers for radiologic identification.4 The HydraFlex device core is formed of a physiologically inert proprietary copolymer of polyacrilonitrile and polyacrylamide. The polymer components are pressure molded into a predetermined form and then dehydrated. This controlled dehydration reduces the device profile for ease of insertion, while preserving the HydraFlex implant's shape memory. The current hydrogel polymer formulation permits the HydraFlex device to absorb up to 80% of its dry weight in water more rapidly than the PDN-SOLO device (Fig 52–3). This provides a compliant device that maintains expansion and lifting forces. Several other formulations were evaluated extensively, but none were found to have these needed abilities. This rehydration-expansion begins immediately after insertion and slows exponentially over 7 to 10 days. The outer jacket of loosely woven UHMWPE fibers allows for rapid rehydration of the hydrogel core, while controlling device expansion to prevent possible damage to the end plates. The jacket also facilitates insertion and surgical manipulation. Small, short platinum-iridium wires are inserted into each end of the pellet, rendering visible the position of an implant by ordinary C-arm fluoroscopic or plain x-ray scans. When rehydrated inside the evacuated nucleus cavity, the implant properly fills the space. This process maintains or restores height and mechanical stability to the disc while maintaining reasonable segmental flexibility. BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES Extensive bench and animal testing were performed on the polymer core and jacket using scientific test methods and established
U.S. Food and Drug Administration (FDA) guidelines for longterm implant materials. The HydraFlex devices and individual components passed all FDA- and ISO-required animal, cytologic, and toxicity testing. These test outcomes demonstrate biologic safety and mechanical durability, and also served as a guide to subsequent prototype improvements. Animal studies were ethically performed in standard, approved experiments using large mongrel dogs, large goats, and transgenic mice, as well as in tissue culture preparations. In addition, a baboon study was performed to assess long-term histopathology. None of these tests or studies showed an adverse response to any of the components, the pellets and jackets, or intact miniaturized implants. Mechanically, the pellet and jacket were subjected to standard fatigue testing for up to 50 million normal-range compression cycles and to 10 million compression-translation cycles. All components passed without demonstrated deterioration. Following prolonged cyclic tests, the terminal burst strength of intact implants exceeded the 6-kN limit of the test machine. In cadaver segment biomechanical studies, denucleated segments showed loss of normal stiffness and decreased stability as compared with initially intact segments. Insertion and subsequent rehydration of the HydraFlex device restored normal stiffness and stability.4–6 CLINICAL PRESENTATION AND EVALUATION The clinical syndrome of the degenerating lumbar disc has gained increased recognition and significance with more precise magnetic resonance imaging (MRI) and better understanding of disc physiology. Controversy remains as to the precise nature of the disease
100 Percent hydration
DESCRIPTION OF THE DEVICE
80 60 40 20
PDN-SOLO XL HydraFlex
0
A
0
B
n F I G U R E 52–2. A and B, The inner copolymer hydrogel pellet is designed to resume a biconvex shape once it is hydrated.
n
F I G U R E 52–3.
50
100
150
200
Time (hours) Hydration versus time for HydraFlex device.
CHAPTER 52
with issues such as the origin of pain generators, mechanical versus chemical bases of pathology, interpretation of diagnostic data, and appropriate intervention. It is widely accepted that the degenerating disc undergoes a multifactor cascade that includes altered perfusion, loss of cellular and glycosaminoglycan components, desiccation, and compromised load transfer functions. The observed changes include the characteristic collapsed disc space with a dark or black disc on MRI and end plate changes on MRI and x-ray study. In those cases appropriate for nucleus arthroplasty, interventions occur during the mild-to-moderate phase of disease. Such patients typically present with axial back pain, with or without concomitant leg pain, which is unresponsive to conservative measures. The neurologic examination often fails to reveal discreet motor or sensory deficits. The diagnosis is confirmed on MRI demonstrating disc desiccation, loss of height, and Modic changes. Provocative discography is often used to confirm symptomatic levels. The hallmark of clinical success is pain relief. The FDA mandates a 25% reduction in pain scores for a device to be considered effective. In a prospective multicenter international trial conducted on the PDN-SOLO device, the predicate to the HydraFlex device, there was a 74% reduction in the Oswestry Disability Index as compared with published reports of a 41% and 42% reduction with the two leading fusion cages and a 50% reduction for a leading total disc replacement. Additionally, the Visual Analog Scale declined from a preoperative level of 7.5 to 2.7 at 12 months after surgery, a 64% reduction. Finally, a patient satisfaction survey of almost 300 PDN-SOLO recipients at multiple international sites demonstrated a much better or better response in 87% of patients, and 88% of patients said they definitely would or might undergo the same surgery again. OPERATIVE TECHNIQUE The surgical procedure for implantation of the HydraFlex nucleus replacement device is relatively straightforward and much less invasive than alternative treatments such as total disc replacement (TDR) or fusion. A typical procedure can be completed in as little as 1 hour. The procedure is performed with the patient under general endotracheal anesthesia. The patient is positioned supine, with a lumbar bolster positioned to augment normal segmental lordosis. Alternatively, an operating room table with flexion and extension capability can be used. C-arm fluoroscopy is required, and this must be kept in mind during patient positioning. Nerve monitoring is optional but highly recommended if the approach truly traverses the psoas muscle. The recommended surgical access to the disc for insertion of the HydraFlex device is accomplished using an anterior retroperitoneal approach (ARPA) and the provided HydraFlex Nucleus Arthroplasty System instrumentation. The PDN-SOLO device was also implanted using an anterolateral trans-psoatic approach (ALPA) and the traditional posterior approach. Use of the posterior approach produced a success rate that was less than optimal due to a multitude of factors addressed by use of the ARPA instrumentation and surgical technique. Instrumentation specific for use in the posterior approach is in development at this writing.
PDN-SOLO and HydraFlex Nucleus Replacement System
409
Using ARPA, the disc space is accessed by creation of a flap in the anterolateral annulus immediately adjacent to the anterior longitudinal ligament. The integrity of this flap is preserved for suture closure at the conclusion of implantation. The nucleus is totally evacuated by standard means. Care should be taken to avoid trauma to the end plates because this may lead to end plate edema, inflammation, and subsequent poor outcome. Damage or violation of the posterior annulus is to be avoided as well. Failure to preserve the integrity of the posterior annulus can lead to posterior extrusion of the device. Once the total evacuation of the nucleus is confirmed, the internal disc space is measured using the supplied instruments. The appropriate size implant and insertion system are selected. The disc space, already positioned in lordosis, is gently distracted. The distracting tool also serves as an introducer past the outer edges of the vertebral bodies. This facilitates introduction and minimizes potential damage to the end plates and implant. The implant is then guided into place with the introducing instruments. Proper location is in the posterior third of the space on lateral projection and in the mid-transverse portion of the space on anteroposterior projection. Rehydration of the device is initiated with saline irrigation after final positioning. The annular flap is then reapproximated with suture. POSTOPERATIVE CARE Postoperative management is critical to a successful clinical outcome and is designed to minimize potential implant migration. Device hydration in the early postoperative period results in increased device expansion in both the vertical and horizontal plane, thus accepting spinal loading. Patient activity restriction and back bracing are required for 6 weeks to allow further equilibration and healing. COMPLICATIONS AND AVOIDANCE Aside from the usual complications associated with surgery, there are those specific to use of the HydraFlex device. The most significant is device extrusion. This is best avoided by strict adherence to patient selection criteria, meticulous surgical technique, and proper postoperative management and rehabilitation.7,8 CONCLUSIONS/DISCUSSION In conclusion, the PDN technologies have enjoyed increasing clinical success for more than a decade. Recent worldwide experience supports claims of clinical safety and efficacy and widespread ADVANTAGES/DISADVANTAGES: PDN-SOLO AND HYDRAFLEX NUCLEUS REPLACEMENT SYSTEM Advantages Preservation of physiological motion Affords all secondary surgical options Disadvantages New/unfamiliar approach Dependent on intact annulus
410
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
acceptance. Current improvements promise even better outcomes and are being evaluated in prospective multicenter, randomized fashion. REFERENCES 1. Lee CK, Langrana NA, Parsons JR, et al: Development of a prosthetic intervertebral disc. Spine 16(suppl 6):S253–S255, 1991. 2. White AA, Panjabi MM: Clinical Biomechanics of the Spine. Philadelphia, Lippincott Williams & Wilkins, 1990. 3. Pope MH: Disc biomechanics and herniation. In Gunzburg R, Szapalski M (eds): Lumbar Disc Herniation. Philadelphia, Lippincott Williams & Wilkins, 2002, pp 3–21. 4. Ray CD, Schönmayr RS, Kavanaugh SA, Assell R: Prosthetic disc nucleus implants. Revista di Neurorad 12(suppl 1):157–162, 1999.
5. Ray CD: The artificial disc: Introduction, history and socioeconomics. In Weinstein, JN (ed): Clinical Efficacy and Outcome in the Diagnosis and Treatment of Low Back Pain, pp 205–225. 6. Ray CD: The PDN prosthetic disc-nucleus device. Eur Spine J Suppl 2:S137–S142, 2002. 7. Shim CS, Lee SH, Park CW, et al: Partial disc replacement with the PDN Prosthetic Disc Nucleus Device. J Spinal Disord Tech 16:324–330, 2003. 8. Jin D, Qu D, Zhao L, et al: Prosthetic disc nucleus (PDN) replacement for lumbar disc herniation. J Spinal Disord Tech 16:331–337, 2003. 9. Bertagnoli R: Review of modern treatment options for degenerative disc disease. In Kaech DL, Jinkins JR (eds): Spinal Restabilization Procedures. Amsterdam, Elsevier Science, 2002, pp 365–375.
CHAPTER
53
NeuDisc Artificial Lumbar Nucleus Replacement Anthony T. Yeung, Ann Prewett, and James J. Yue
K E Y l
l
l
l l
P O I N T S
Degeneration of the lumbar intervertbral disc, providing 80% of the support of the spinal segment, is responsible for chronic lumbar discogenic pain. Discectomy paradoxically induces natural and accelerated degeneration and subsequent back pain. The rate of degeneration may be directly related to the amount of nucleus removed. The nucleus pulposus is a natural hydrogel. NeuDisc synthetic hydrogel is a leading material for nucleus replacement.
INTRODUCTION Internal disruption of the intervertebral lumbar disc is a major source of back pain. Discectomy is the most frequently performed surgical procedure for back pain and sciatica. Although discectomy alone, with proper patient selection, may have a reasonably good short-term outcome in relieving radicular symptoms, less than optimal long-term relief usually results in further disc degeneration, instability, and pain. Therefore, discectomy may promote secondary sequelae, such as facet arthrosis and spinal stenosis, by altering the biomechanical properties of the intervertebral disc. Fusion has long been the standard method for treating refractory painful spinal disc disease. Fusion, however, is a nonphysiologic solution that may cause so-called fusion disease by damaging normal muscle and spinal anatomy in addition to creating adjacent-level problems. There is a great need for more effective and longer lasting treatments for low back and radicular pain from a scientific as well as an economic point of view. The surgical focus of this chapter, therefore, is on the feasibility of surgical replacement of the intervertebral disc nucleus as a means to slow the painful consequences of the degenerative process and discectomy. The Clinical Challenges of Nucleus Replacement Although it is assumed that replacement of the nucleus pulposus is desirable after nuclectomy, we must understand the natural history
of discectomy for disc herniation and when it is desirable to replace the nucleus as an appropriate procedure. At present, there is evidence that adverse results of disc degeneration are directly related to the amount of disc tissue removed,1 but there is no consensus among surgeons on how much disc to remove in a discectomy for disc herniation. The best we can assume is that the goals of disc replacement are to maintain or restore the normal height of the disc, to keep the annulus properly tensioned, and to provide shock and vibration absorption to the spinal segment. The clinical goal is to reduce or eliminate pain for as long as possible. The more realistic expectation may be for nucleus pulposus implants to maintain disc height, maintain motion, and preserve kinematics, insofar as satisfaction of these requirements is sufficient to ensure pain relief. The original concept of nucleus replacement centers on the effects that nuclectomy has on destabilizing segmental stability. Nucleus replacement, therefore, aims to reduce discogenic pain by restoring the function of a normal nucleus. Biomechanical Considerations of Implant Material and Design Many biomechanical and biochemical studies have focused on various biomaterials that provide support against compression and still maintain some of the physiologic properties of the nucleus pulposus. Hydrogel, with its biocompatibility properties that mimic the function of the nucleus pulposus, a natural hydrogel, has the ability to resist deformation when jacketed or supported by a Dacron or fabric mesh.2,3 Therefore, hydrogel is considered one of the leading materials being investigated. Implant size, the next consideration, affects the longevity of the implant. The size of the implant must be of sufficient surface area to support the spine segment and not become another loose body as residual nuclear material degenerates. The third consideration is prevention of implant migration and extrusion. If the implant is too soft, there is a greater chance of extrusion during axial loading inasmuch as the implant is able to move or migrate within the disc. Preservation of the annulus to act as a barrier to implant extrusion or reinforcement of the annulus is also an important factor, thus emphasizing the importance of a minimally invasive, tissue-sparing approach for 411
412
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
implantation. Alternatively, there is a need to accurately access the integrity of the annulus over the long term, because success for nucleus replacement may well depend as much on the stability and integrity of the annulus as the nuclear implant itself. The size of the annulotomy or herniation defect may also affect the ability of the annulus to contain the implant. The NeuDisc (Replication Medical, Inc., Cranbury, NJ) design and configuration make the posterolateral transforaminal endoscopic approach a theoretically very desirable insertional technique. This approach is excellent both for nuclectomy and prosthesis implantation. Other approaches, such as the lateral or anterolateral approach, are also appropriate as long as a thorough nuclectomy is accomplished through the surgical approach. The ability to compress the NeuDisc implant, coupled with its rapid hydration, gives it the flexibility to be implanted from any approach as long as a thorough nuclectomy is accomplished. The final consideration is to then insert the proper size implant to provide axial support to the disc. DESCRIPTION OF THE DEVICE NeuDisc by Replication Medical (Cranbury, NJ) mimics the physiologic function of the nucleus pulposus (Fig. 53–1). Its proprietary layered hydrogel structure is designed to distribute axial loads in the disc, and its hydrogel mimics osmotic properties of the nucleus pulposus. The vertically layered structure alternates soft hydrogel layers between Dacron knitted mesh to attain the proper stiffness required but should be sufficiently soft to avoid damage to the end plate. The acrylic copolymer hydrogel mimics the properties of the nucleus pulposus and allows osmotic intradiscal swelling to provide lift and shock-absorbing support.2,3 Anisotropic swelling resists bulging as it swells, providing mostly lift support without creating undesired pressure radially. The Dacron net stabilizes the soft implant to resist extrusion. This ensures that NeuDisc can be implanted through a small incision, avoid pressure on a weakened annulus, and quickly hydrate to resume a shape that is much larger than the annular incision. Unloaded, the NeuDisc implant expands in thickness from its desiccated, compressed state of 6.5 to 15 mm in its hydrated state (Figs. 53–2 and 53–3). Anisotropic expansion restricts lateral expansion to enlarge the footprint.
A
BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES The NeuDisc has also undergone extensive mechanical testing to verify that its mechanical properties closely match those of the native nucleus pulposus. Biomechanical bench testing of the NeuDisc hydrogel implant tested the capability of the implant to provide axial lift force under various pressures ranging from 150 to 3,000 Newton (N). Lift force builds up to maximum lift in about 2 days, then stays constant over time, responding only to variations in temperature. When lift force is plotted for various hydration levels, at 60% to 65% hydration, it provides lift parameters similar to young nucleus pulposus at 400 N. If exposed to compression in a fully hydrated state, NeuDisc resists compression beyond the physiologic limits in the disc. Moreover, when anisotropic expansion limits radial bulging, it decreases the danger of radial expansion pushing residual disc tissue out through an annular defect created either by a surgical annulotomy incision or herniation defect. Endurance Testing The endurance testing was performed in multiple motions, range-of-motion studies, expulsion testing, and fatigue testing in a cadaveric model. The testing subjected the NeuDisc implant to compression, flexion and extension combined with compression, lateral bending combined with compression, and axial torsion combined with compression. Each test was for 30 million cycles, representing a 30-year lifetime in each of the motions. No monomer elution was evident in the fluid gathered from any of the endurance-tested samples. The extensive endurance testing peformed demonstrated the device's ability to survive long-term strenuous use in multiple loading modes. The NeuDisc implant also remained physically intact and as functional as controls after 30 million compression cycles at 150% of maximum physiologic displacement. Initial fatigue testing also included cycling the hydrogel implants in unconfined compression between load levels of 200 and 800 N. Throughout this testing, it was demonstrated that lateral expansion and deformation of the implants resulted. The implant fatigue samples completed the 30 million loading cycles intact, with only minor
B n
F I G U R E 53–1.
The NeuDisc design.
CHAPTER 53
NeuDisc Artificial Lumbar Nucleus Replacement
413
n F I G U R E 53–2. Anisotropic expansion from 2 to 15 mm after hydration. Anisotropic swelling, 37 C the initial thickness is 2 mm. Note that the implant expands in thickness, but the footprint size does not enlarge.
surface cracking and no loss of function or change in water uptake. Initial in vitro results of testing suggest that the NeuDisc hydrogel implant may be a suitable nucleus pulposus substitute.2,3
Short-term and long-term implantation tests in rabbits demonstrated that the acrylic copolymer hydrogel Aquacryl (Replication Medical, Cranbury, NJ) was both nontoxic and biocompatible. The long-term and chronic toxicity of the material was analyzed by implantation in rat paravertebral muscles. The results were similar to control implantations and showed no chronic cytotoxicity.
failure in compression was end plate fracture. Ligament ruptures, annular tears, facet joint fractures, and vertebral compression fractures also occurred before implant failure. The average force at failure was 3,533 N, which agrees with literature values. The failures in the flexion specimens occurred at an average moment of failure of 52.2 Nm. The specimens that failed in lateral bending had five different failure modes. One specimen had an end plate fracture, three had fractures of the facet joint, six had ligament rupture, one had an annulus tear, and one implant expelled. The average moment at failure was 24.8 Nm, and the implant expulsion occurred at 47.9 Nm. The failure strengths in bending agree with published values.
Expulsion Testing
Clinical Outcomes
Expulsion testing demonstrated that the NeuDisc is unlikely to expel during normal activity. Following range-of-motion testing, 32 implanted functional spinal units were tested to failure. There were two implant expulsions. Both expulsions that were observed occurred at extremely high loads. The most common mode of
The 6.5-mm bullet configuration of the NeuDisc implant has been undergoing a two-arm pilot European prospective longitudindal study since June 2005. The first arm is an open approach using the anterolateral transpsoatic approach (ALPA). The second arm is a posterolateral endoscopic approach. In both arms, the primary
Biocompatibiliy Testing
A
B n
F I G U R E 53–3.
Bullet-shaped 6.5-mm implant.
414
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
n
F I G U R E 53–4. Preoperative images of L4-5 degeneration.
inclusion criteria is recalcitrant low back pain secondary to degenerative disc disease with or without leg pain. Disc height can be no less than 50% of normal height compared with adjacent levels. Levels of insertion have been L2-L5 for the ALPA approach and L2-S1 in the endoscopic approach. Successful insertion is dependent on complete nucleotomy and appropriate implant positioning and sizing. A total of 15 implantations have been performed. In cases of ideal disc space preparation and implant sizing, no revisions were necessary. Two revision surgeries for infection have been performed. Early clinical outcomes indicate
n
F I G U R E 53–5.
early resolution of symptoms with mainentance of improvement at 12 months (Figs. 53–4 and 53–5). OPERATIVE TECHNIQUE The ability and ease of insertion of the NeuDisc implant through a 6-mm fenestration in the annulus make this a viable implant for endoscopic as well as open use. The implant configuration makes it easy to insert through any discectomy portal, including a traditional dorsal or anterolateral microdiscectomy approach
Six-month postoperative images.
CHAPTER 53
NeuDisc Artificial Lumbar Nucleus Replacement
415
shown to have partially hydrated and is also demonstrated to provide vertical lift to the disc space in an unloaded cadaver trunk. ADVANTAGES AND DISADVANTAGES
n
F I G U R E 53–6. Before and after hydration.
(Figs. 53–6 and 53–7). The transcanal approach can be accomplished with a uniportal or biportal annular fenestration. The surgeon performs a nuclectomy with or without the aid of an endoscope, then measures the enucleated cavity and does a trial implantion with an implant template. A proprietary instrument to insert the NeuDisc implant is needed for the uniportal approach and facilitates the biportal approach. Current endoscopic systems use a 6-mm inner diameter/7-mm outer diameter cannula for discectomy and a 7- to 8-mm inner diameter/8- to 9-mm outer diameter cannula for foraminoplasty. A recently developed foraminoscope has a 4.2-mm working channel that also facilitates endoscopic visualized nuclectomy. The Yeung Endoscopic Spine System (YESS) system, developed in conjunction with Richard Wolf Surgical Instrument Company, has endoscopic cannulas, discectomy instrumentation, and an introducer that is ideally suited specifically for the NeuDisc implantation.4,5 Discectomy may be accomplished through the foraminal portal from T10-L5, with plenty of room for endoscopic instrumentation.6 A uniportal or biportal technique may be used for nucleus implantation and has the advantage of limited annular disruption to limit implant extrusion. The further development and application of this technique have broadened application to address many concurrent painful degenerative conditions of the lumbar spine.5,7–10 This technique is ideal for the implantation of the NeuDisc. The implant is inserted through a standard beveled 8-mm inner diameter cannula and visualized endoscopically as the implant is seated in place. After waiting 4 hours, the implant is
Nucleus prosthesis replacement has obvious advantages over total disc replacement. By replacing only the nucleus, it preserves the remaining disc tissues and, therefore, preserves function. Because the nucleus has a more uniform structure and function than the annulus and end plate, the design of the nucleus prosthesis is simpler and it can be designed to be implanted by minimally invasive methods with only a small incision in the annulus. Because the implant is not fixed to the vertebrae, no fixation component is required, and the surgical time is comparable or only slightly longer than a discectomy. Although implant extrusion remains a primary concern, it is less likely to cause permanent nerve injury because of its relatively small size and modulous of elasticity. In cases of prosthesis failure, it is relatively easy to remove the implant anteriorly and convert to a total disc replacement procedure or a fusion. The major limitation of the NeuDisc is that its application may be limited to patients with early or intermediate degeneration because it requires a relatively competent natural annulus. In a disc with severe annular delamination or loss of height, implantation of a NeuDisc may not be possible. Restoring disc height may have its own concerns if the spinal nerve and facet articulation have already adapted to its shortened state. Because of this concern, work is being considered for a less tall implant that is designed for narrowed, more degenerative discs. It is growing increasingly clear that artificial disc or nucleus implantation may become the treatment of choice for lumbar and cervical disc disease, especially if implanted in patients to mitigate the degenerative process. It has been estimated that this technology has the potential to be part of up to 50% of future spine surgeries. The potential of disc replacement may parallel that of joint replacement for the near future and offers an exciting development in minimally invasive spine surgery. CONCLUSIONS/DISCUSSION The native nucleus is composed of a hydrogel. Physiologic prerequistes for ideal functioning of the hydrogel include normal end plate metabolic diffusion and an intact annulus. The NeuDisc hydrogel implant promotes continued end plate metabolism and can be inserted with a minimal annulotomy. When it is implanted with a diameter of 6.5-mm, the implant quickly enlarges in height and width to fill the enucleated space and restore disc height. The implant can be inserted through traditional open or endoscopic techniques. Early clinical trials indicate implant stability and clinical improvement. REFERENCES
n
F I G U R E 53–7. Insertion tools.
1. Brinkman P, Grootenbore H: Change of disc height, radial disc bulge, and intradiscal pressure from discectomy. Spine 16:641–646, 1991. 2. Stoy V, Sabatino J, Gontarz J, et al: Mechanical testing of a hydrogel nucleus replacement implant. Spine Across the Sea 2003, Maui, Hawaii, July 27–31, 2003.
416
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
3. Yeung AT: “Lumbar artificial disc nucleus.” In Perez-Cruet, Khoo, Fessler (eds). An anatomic approach to minimally invasive spine surgery. Quality Medical Publishing, 2006. 5. Yeung AT, Yeung CA: Advances in endoscopic disc and spine surgery: Foraminal approach. Surg Technol Int 11:253–261, 2003. 6. Yeung AT: The evolution of percutaneous spinal endoscopy and discectomy: State of the art. Mt Sinai J Med 67:327–332, 2000. 7. Yeung AT: Evolving methodology in treating discogenic back pain by selective endoscopic discectomy and thermal annuloplasty. Journal of Minimally Invasive Spinal Technique 1:8–16, 2001.
8. Tsou PM, Yeung AT: Transforaminal endoscopic decompression for radiculopathy secondary to intracanal noncontained lumbar disc herniations: Outcome and technique. Spine J 2:41–48, 2002. 9. Yeung AT, Tsou, PM: Posterolateral endoscopic excision for lumbar disc herniation: Surgical technique, outcome, and complications in 307 consecutive cases. Spine 27:722–731, 2002. 10. Tsou PM, Yeung CA, Yeung AT: Selective Endoscopic Discectomy™ and thermal annuloplasty for chronic lumbar discogenic pain: A minimal access visualized intradiscal procedure. Spine J 2:563– 574, 2004.
CHAPTER
54
NuCore Injectable Nucleus: An In Situ Curing Nucleus Replacement Othmar Schwarzenbach, Ulrich Berlemann, and Thomas Wilson
K E Y l
l
l
l
P O I N T S
NuCore Injectable Nucleus is a suitable replacement for the natural nucleus pulposus. NuCore Injectable Nucleus treats partial nucleotomies, such as those following microdiscectomy treatment of disc herniation, and early-stage degenerative disc disease (DDD). Patients experience pain relief consistent with early results of standard microdiscectomy, with no extrusions or reactions to the material. Early data show an improvement in clinical stability in relation to predecessor implants.
INTRODUCTION The use of an injectable material for nucleus replacement allows the flexibility to treat partial nucleotomies, such as those following microdiscectomy treatment of disc herniation, as well as early-stage degenerative disc disease (DDD), in which complete nucleus removal and replacement may be required. An injectable biomaterial is ideal for restoration of disc volume removed during discectomy and for preventing loss of disc height. Flowable materials may be injected through a small incision, allowing minimally invasive, even percutaneous, access to the disc space when appropriate. Fluids can interdigitate with the irregular defects and may, depending on the material used, physically bond to the adjacent tissue. The use of an injectable material allows for complete filling of the disc, a task that is not possible with preformed implants. Complete filling allows pressurization of the annulus, ensuring load transfer and load sharing between the annulus and nucleus. Injectable biomaterials allow for incorporation and uniform dispersion of cells or therapeutic agents, or both. The addition of growth factors may be valuable in enhancing the repair process. Inclusion of inhibitors of inflammatory cytokines and proteases may act to retard matrix degradation and the potential effects of these cytokines on surrounding tissue and neural structures. Generally, these biomaterials are injected as viscous fluids and then cured through methods such as thermosensitive cross-linking, pH-
sensitive cross-linking, photopolymerization, or the addition of a solidifying agent to form a gel-like substance. The setting time should be long enough to allow for accurate placement during the procedure yet short enough to not prolong the length of the surgical procedure. Any heat generated during the cure process should not cause harm to the surrounding tissue. The viscosity or fluidity of the material should balance the need for the substance to remain within the disc with the ability of the surgeon to manipulate its placement and with the need to ensure complete filling of the intradiscal space or voids. Ease in accessing the disc space also needs to be considered. For example, polymers that cure through a photopolymerization procedure could pose a problem owing to a limited ability to access the small cavities of the disc space with the light needed to initiate cross-linking. Injectable biomaterials have been considered as an augment to discectomy for some time. As early as 1962, Nachemson suggested the injection of a vulcanizing silicone into a degenerated disc using an ordinary syringe.1 Later, Schneider and Oyen studied the use of silicone elastomer in the intervertebral disc.2,3 Since then, work has been put toward the development of various injectable materials, including cross-linkable silk elastin copolymer,4,5 polyurethane-filled balloons,6 collagen-polyethyleneglycol (PEG),7 chitosan,8 various injectable synthetic polymers,9 recombinant bioelastic materials,10,11 light-curable PEG polymers,12 and other multicomponent polymer systems.13 Several groups are actively pursuing the development of an injectable biomaterial for use in the intervertebral disc.14,15 One of those materials is a recombinant protein copolymer known as NuCore Injectable Nucleus (Spine Wave, Inc., Shelton, CT). NuCore consists of amino acid sequence blocks derived from silk and elastin structural proteins. This material appears to have ideal characteristics for the augmentation of the nucleus pulposus following discectomy procedures. Early development and characterization work on this protein polymer was performed by Cappello and co-workers.16,17 Further clinical development of this polymer is being pursued by the manufacturer. Successful development of these materials could drastically change the way back pain is treated in the future.
417
418
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
INDICATIONS/CONTRAINDICATIONS Degenerative Disc Disease Clinical symptoms resulting from herniated, protruding, and/or painful intervertebral discs are commonly treated by discectomy. Discectomy is the most commonly performed spinal surgical treatment, and is frequently used to treat radicular pain due to nerve impingement from a herniated or protruding intervertebral disc. During a discectomy, a substantial portion of the volume of the nucleus pulposus is removed and immediate loss of disc height and volume can result. This procedure is typically performed in a relatively young patient population, with a mean age between 25 and 40 years.18–20 The impact of altered biomechanics and long-term sequelae in this young patient population may be significant. Substantial disc height reduction following discectomy is evident soon after the discectomy procedure. Disc height loss has been found to be proportional to the amount of nucleus removed in an in vitro study.21 Clinically, the operated disc spaces of patients postoperatively are significantly narrower following discectomy than controls.22 Scoville and Corkill23 found a 50% incidence of narrowing following surgery at the 3-month follow-up examination. In another study, Tibrewal and Pearcy24 found disc space narrowing evident within 3 months following surgery as compared with control patients who had not undergone surgery. Proper disc height is necessary to ensure proper functioning of the intervertebral disc and spinal column. On the local (or cellular) level, decreased disc height can lead to a decrease in cell matrix synthesis and an increase in cell necrosis and apoptosis. It has been shown in other cartilaginous tissues that increased static loading decreases matrix protein biosynthesis.25–27 Animal models have shown that overloading of the intervertebral disc can initiate disc degeneration.28,29 In addition, the change in intradiscal pressure creates an unfavorable environment for fluid transfer into the disc, which can cause a further decrease in disc height. Decreased disc height also results in significant changes in the global mechanical stability of the spine, which may result in further degeneration of the spinal segment. With decreasing height of the disc, the facet joints bear increasing loads and may undergo hypertrophy and degeneration, which may act as a source of pain over time.30,31 Decreased stiffness of the spinal column and increased range of motion resulting from loss of disc height can lead to further instability of the spine.31 Excessive motion can manifest itself in abnormal muscle, ligament, and tendon loading, which can ultimately be a source of back pain. Radicular pain may also result from a decrease in foraminal volume caused by decreased disc height. Specifically, as disc height decreases, the volume of the foraminal canal decreases. This decrease may lead to spinal nerve impingement, with associated radiating pain and dysfunction. Finally, adjacent-segment loading increases as the disc height decreases at a given level.31,32 The discs that must bear additional loading are now susceptible to accelerated degeneration and compromise, which may eventually propagate along the destabilized spinal column. A further issue with microdiscectomy surgery is the occurrence of reherniation. Atlas et al33 reported a reoperation rate of 25% at 10-year follow-up of a study of patients with lumbar disc
herniation, with the median time to reoperation of 24 months. Carragee et al34 reported a 11.5% reherniation rate, with 6.5% reoperation at 5 years. The objectives of augmentation of the nucleus pulposus following disectomy are to prevent disc height loss and the associated biomechanical and biochemical changes resulting from reduced disc height and volume. Use of an injectable biomaterial to restore disc volume and prevent loss of disc height is currently being evaluated. The ability of an injectable material to seal the disc and prevent or reduce the incidence of reherniation is also being studied. DESCRIPTION OF THE DEVICE Technology has been developed for the production of synthetically designed protein polymers consisting of repeated blocks of amino acid sequence. Through a combination of biologic and chemical methods, block polymers are produced using gene template– directed synthesis. Using this method, the design and polymerization of a new polymer occurs once during the synthesis of the gene template. Through the construction of synthetic genes, it is possible to specify the sequence of protein blocks (the unit of repetition of a protein polymer) several hundred amino acids in length, many fold greater than the limit of sequence control of chemical synthesis. Molecular architecture is critical to biologic systems. The silk fibroin portion of the polymer is a closely packed structure that serves as a reinforcing scaffold for the surrounding protein, increasing the toughness of the material. The elastin portion of the polymer is a flexible structure that winds into a spiral, functioning as a flexible molecular spring.35 This gives the material its elastic properties. The protein polymer used in the NuCore Injectable Nucleus is a sequential block copolymer of silk and elastin, with two silk blocks and eight elastin blocks per polymer sequence repeat. One of the elastin blocks is modified to provide for chemical crosslinking. The protein polymer is synthesized using recombinant DNA techniques through Escherichia coli strain K12, a nonpathogenic strain of bacterium typically used in recombinant protein expression. Following batch fermentation, the cells are ruptured by homogenization and the protein polymer is purified from the lysate using precipitation and a series of filtration and adsorption purification steps. The identity and purity of the polymer are confirmed using amino acid composition, amino acid terminal sequencing, mass spectroscopy, and other biochemical tests. The NuCore material is composed of the protein polymer and a cross-linking agent and is formulated to closely match the properties of the human nucleus pulposus as shown in Table 54–1. The material closely mimics the protein content, water content, pH, and complex modulus of the natural nucleus pulposus. It is injected into the disc after being mixed with a very low concentration of a diisocyanate-based cross-linking agent and has approximately a 90-second working time before it becomes a viscous gel. The material reaches near-final mechanical strength approximately 30 minutes after addition of the cross-linker, but it is sufficiently gelled after 5 minutes to allow the surgery to be completed. As a consequence of the very low concentration of cross-linker used, there is no measurable temperature rise during the curing process.
CHAPTER 54
NuCore Injectable Nucleus: An In Situ Curing Nucleus Replacement
419
TABLE 54–1. Comparison of NuCore Injectable Nucleus and Natural Nucleus Property Protein content (%) Water content (%) pH Complex shear modulus, G* (kPa)
NuCore Injectable Nucleus
Natural Nucleus
19.4 79.1 7.1 26
13.6–21.9* 74–81 6.7–7.1 7–21{
*Kitano T, Zerwekh JE, Usui Y, et al: Biomechanical changes associated with the symptomatic human intervertebral disk. Clin Orthop Relat Res 293:372–377, 1993. { Iatridis JC, Weidenbaum M, Setton LA, Mow VC: Is the nucleus puposus a solid or a fluid? Mechanical behaviors of the nucleus pulposus of the human intervertebral disc. Spine 21:1174–1184, 1996.
BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES Extensive preclinical testing and characterization of the material were performed to establish its appropriateness for the intended application. Biocompatibility and toxicology testing following the ISO 10993 guidelines were performed on the material. Acute tests include cytotoxicity, sensitization (guinea pig), intracutaneous reactivity (rabbit), systemic toxicity (mouse), pyrogenicity, muscle implant evaluation and genotoxicity testing. The material was nontoxic and nonirritating in all of these test evaluations. Chronic toxicity testing was conducted in a rat subcutaneous model and evaluated at time points beyond 1 year with no toxicity seen. Neurofunctional testing in a rat model showed no neurotoxicity of the material when placed adjacent to spinal nerve roots. Mechanical characterization of the material was also carried out preclinically in benchtop testing. Testing of cadaveric anterior column units was done to determine how well the NuCore material resists extrusion under load. Segments were tested in axial compression and loaded to failure. In all cases, there was no extrusion unless preceded by bony failure or end plate failure of the model itself.36 The average maximum load imparted to the spinal segments was 3,555 N, well above the loads experienced in normal daily activities. Results demonstrated that the NuCore material integrated extensively with the surrounding disc tissue and did not extrude during any of the testing (Fig. 54–1). Human cadaveric spinal units were also tested to determine how well the NuCore material restores stability and function to a spinal unit.37 Anterior column units (no posterior structures) were tested in compression in the intact condition, then with a partial nucleotomy, and finally with NuCore material injected. Statistical analysis using a repeated measures analysis showed that the discectomy caused a significant loss of height during the test (P0.05) between the displacement of the intact condition and the NuCore material–treated condition.38 Functional spinal units (anterior and posterior structures intact) were also tested in flexion/extension, lateral bending, and axial rotation with similar results.38 This analysis indicates the NuCore Injectable Nucleus restores function and stability to the spine after being destabilized by a discectomy procedure.
n FIGURE 54–1. Sectioned test specimen showing interdigitation of NuCore Injectable Nucleus into natural intradiscal tissue.
Owing to limitations of cadaveric tissue, testing has been performed in a synthetic disc model to evaluate the effect of dynamic loading over long periods of time. The annulus fibrosus of the disc was simulated with a silicone elastomer. This silicone annulus was injected with NuCore material, and the model was subjected to cyclic physiologic loading up to 10 million cycles with no failure of the NuCore material or test model. This testing indicates the NuCore material to be durable, fatigue resistant, and capable of withstanding in vivo loads for an extended period of time. Early Clinical Results As reported at the Spine Arthroplasty Society meeting in Montreal in May 2006, a nonrandomized clinical study is underway in Switzerland to investigate the use of NuCore Injectable Nucleus as a tissue replacement in cases of microdiscectomy for herniated nucleus pulposus.39 Patients enrolled in the study suffered from a single-level symptomatic herniated nucleus pulposus in the lumbar spine that did not respond to conservative therapy. Data were collected preoperatively, postoperatively, at 6 and 12 weeks, as well as 6 and 12 months. The clinical data at each time point included a neurologic examination as well as an assessment of pain, disability, and function using Visual Analog Scale (VAS) for back and leg pain, Oswestry Disability Index (ODI), and SF-36 patient questionnaires. At each time point, standing lateral x-ray studies with a marker attached were taken. Disc height analysis was performed by an independent analysis group (Medical Metrics, Inc.) using a validated measurement method. Magnetic resonance imaging (MRI) obtained preoperatively, postoperatively, and at 6 and 12 months were qualitatively reviewed with regard to the signal in the disc and the postoperative behavior of the decompressed tissues. Eight of 15 enrolled patients had completed 1 year of follow-up. There were no dropouts over the follow-up period. Average age of the patients was 35.9 years. Eight were male, and seven were female. Of the 15 levels treated, 11 were at L5-S1 and four were at L4-S5.
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
10.0 9.0 8.0
VAS scores
n
F I G U R E 54–2. Average Visual Analog Scale (VAS) and Oswestry Disability Index (ODI) results from NuCore clinical study.
100.0 Avg. leg pain Avg. back pain Avg. ODI
90.0 80.0
7.0
70.0
6.0
60.0
5.0
50.0
4.0
40.0
3.0
30.0
2.0
20.0
1.0
10.0
0.0
ODI scores
420
0.0 Pre-op
6 wks
3 mo
6 mo
12 mo
Examination timepoint
The mean injection volume was 1.3 mL in these 15 patients. The goal was to replace the entire amount of tissue lost to the herniation and the partial nucleotomy, thus restoring the original volume of nuclear tissue within the disc. Improvements to the injection technique led to increased injection volumes in later cases. Before the development of the improved injection technique, the injection volumes matched only the volume of tissue removed from within the disc. The improved technique resulted in complete replacement of all tissue lost in the later cases. No device-related adverse events were reported. Results of all blood, serum, and urine tests were within normal ranges at all time points. There was no evidence of any immunologic reaction to the material. Pain and function scores are shown in Figures 54–2 and 54–3. Mean leg pain (VAS) reduced from 6.7 at preoperative assessment to 0.6 at 1 year after surgery. Mean back pain (VAS) reduced from 4.0 at preoperative assessment to 1.2 at 1 year after surgery. Likewise, mean ODI reduced from 44.0 at preoperative assessment to 7.9 at 1 year after surgery. All of these scores were consistent with those expected in microdiscectomy patients. Radiographic analysis of standing lateral x-ray studies showed good maintenance of disc height in the postoperative follow-up examinations. Discs treated with the early surgical technique averaged 93% of the original disc height at the 12-month follow-up. Discs treated with the improved technique showed disc height approximating 100% of the original height. Review of the MR images showed no migration of the material and no extrusions. As can be seen in Figure 54–4, the NuCore implant appears as a hyperintense signal within the disc. Consolidation of the posterior tissue remnants from the herniation can be seen over the follow-up period.
Following the discectomy, the amount of tissue removed was measured and used as a target volume for the NuCore material injection. The procedure was paused for 5 minutes following injection to allow the material to cure; then the wound was closed. No postoperative brace was used, but patients were reminded to restrict their activities for 6 weeks.
PF 100
MH
RP
50 25 0
RE
BP
SF
GH
VT Pre-op 6 weeks 3 months 6 months 12 months
OPERATIVE TECHNIQUE A typical interlaminar approach for microdiscectomy was used. The herniated tissue was removed, and the disc was further cleared of loose tissue by several washouts and gentle use of rongeurs.
75
n
F I G U R E 54–3.
study.
Average SF-36 results for NuCore clinical
CHAPTER 54
NuCore Injectable Nucleus: An In Situ Curing Nucleus Replacement
421
Pre-op
Discharge 6 months PO 12 months PO Representative magnetic resonance images from a NuCore clinical trial patient. Disc herniation at L5-S1, treated with microdiscectomy and nucleus replacement.
n
F I G U R E 54–4.
CONCLUSIONS/DISCUSSION Based on the preclinical and clinical data reviewed, NuCore Injectable Nucleus appears to be a suitable replacement for the natural nucleus pulposus. The preclinical characterization of the mechanical and biomaterial properties of the NuCore material is supported by the early clinical study data. Patients have seen pain relief consistent with the expected early results of standard microdiscectomy with no extrusions or reactions to the material. Disc height appears to be better maintained than that indicated by the literature. The long-term benefits will be borne out in longer term follow-up on these patients to determine whether improved biomechanics and decreased complications (reherniation, progression to fusion, recurrent pain, and so on) are indeed realized. These early data are promising at this point and show an improvement in clinical stability in relation to predecessor implants that have been tested in this application. The fact that an injectable flowable hydrogel has been successfully implanted and remained in place over considerable time is a significant step forward in the development of a viable nucleus replacement. Additional clinical evaluation of the use of NuCore Injectable Nucleus for treatment of early-stage DDD has just recently begun. Ongoing efforts are also characterizing the use of the material as a cell delivery vehicle for disc repair and reconstruction, as well as the use of the material for repair of articular cartilage defects. REFERENCES 1. Nachemson A: Some mechanical properties of the lumbar intervertebral disc. Bull Hosp Joint Diseases 23:130–132, 1962. 2. Schneider PG, Oyen R: Intervertebral disc replacement, experimental studies, clinical consequences. Z Orthop Ihre Grenzgeb 112:791– 792, 1974.
3. Schneider PG, Oyen R: Plastic surgery on intervertebral disc, part I: intervertebral disc replacement in the lumbar regions with silicone rubber: Theoretical and experimental studies. Z Orthop Ihre Grenzgeb112:1078–1086, 1974. 4. Cappello J, Stedronsky ER, inventors: Synthetic proteins for in vivo drug delivery and tissue augmentation. US patent 6,380,154, April 30, 2002. 5. Ferrari FA, Richardson C, Chambers J, et al, inventors: Peptides comprising repetitive units of amino acids and DNA sequences encoding the same. US patent 6,355,776, March 12, 2002. 6. Bao Q-B, Yuan HA, inventors: Implantable tissue repair device. US patent 6,224,630, May 1, 2001. 7. Olsen DR, Chang R, McMullin H, et al, inventors: Methods for the production of gelatin and full-length triple helical collagen in recombinant cells. US patent 6,428,978, August 6, 2002. 8. Chenite A, Chaput C, Wang D, et al: Novel injectable neutral solutions of chitosan form biodegradable gels in situ. Biomaterials 21:2155–2161, 2000. 9. Milner R, Arrowsmith P, Millan EJ, inventors: Intervertebral disc implant. US patent 6,187,048, February 13, 2001. 10. Urry DW, inventor: Polynanopeptide bioelastomers having an increased elastic modulus. US patent 5,064,430, November 12, 1991. 11. Urry DW, inventor: Polymers capable of baromechanical and barochemical transduction. US patent 5,226,292, July 13, 1993. 12. Hubbell JA, Pathak CP, Sawhney AS, et al, inventors: Photopolymerizable biodegradable hydrogels as tissue contacting materials and controlled-release microcarriers. US patent 5,626,863, May 6, 1997. 13. Hubbell JA, Wetering PVD, Cowling DSP, inventors: Novel polymer compounds. U.S. patent application US_2002/0177680_A1, 2002. 14. Bao Q-B, Yuan HA: New technologies in spine: nucleus replacement. Spine 27:1245–1247, 2002. 15. DiMartino A, Vaccaro A, Lee JY, et al: Nucleus pulposus replacement: Basic science and indications for clinical use. Spine 30:S16–S22, 2005. 16. Cappello J, Ferrman F: Genetically engineered protein polymers. In Domb AJ, Kost J, Wiseman D (eds): Handbook of Degradable Polymers. Amsterdam, Harwood Academic Publishers, 1996, pp 387–414.
422
P A R T
V
Lumbar Partial Disc Replacement: Nucleus Replacement
17. Cappello J, Ferrari F: Microbial product of structural protein polymers. In Mobley DP (ed): Plastics from Microbes. Munich, Carl Hanser Verlag, 1997, pp 35–92. 18. Hermantin FU, Peters T, Quartararo L, Kambin P: A prospective, randomized study comparing the results of open discectomy with those of video-assisted arthroscopic microdiscectomy. J Bone Joint Surg 81-A:958–965, 1999. 19. Mochida J, Toh E, Nomura T, Nishimura K: The risks and benefits of percutaneous nucleotomy for lumbar disc herniation: A 10-year longitudinal study. J Bone Joint Surg Br 83:501–505, 2001. 20. Yorimitsu E, Chiba K, Toyama Y, Hirabayashi K: Long-term outcomes of standard discectomy for lumbar disc herniation: A followup study of more than 10 years. Spine 26:652–657, 2001. 21. Brinckmann P, Frobin W, Hierholzer E, Horst M: Deformation of the vertebral endplate under axial loading of the spine. Spine 8:851–856, 1983. 22. Frymoyer JW, Hanley EN, Howe J, et al: A comparison of radiographic findings in fusion and nonfusion patients ten and more years following disc surgery. Spine 4:435–440, 1979. 23. Scoville WB, Corkill G: Lumbar disc surgery: Technique of radical removal and early mobilization. J Neurosurg 39:265–269, 1973. 24. Tibrewal SB, Pearcy MJ: Lumbar intervertebral disc heights in normal subjects and patients with disc herniation. Spine 10:452–454, 1985. 25. Buschmann MD, Gluzband YA, Grodzinsky AJ, Huziker EJ: Mechanical compression modulates matrix biosynthesis in chondrocyte/agarose culture. J Cell Sci 108:1497–1508, 1995. 26. Ohshima H, Urban JPG, Bergel DH: Effect of static load on matrix synthesis rates in the intervertebral disc measured in vitro by a new perfusion technique. J Orthop Res 13:22–29, 1995. 27. Rand N, Juliao S, Spengler D, Dawson J: Static Hydrostatic Loading Induces In Vitro Apoptosis in Human Intervertebral Disc Cells. Orlando, FL, Orthopaedic Research Society, 2000. 28. Iatridis JC, Mente PL, Stokes AF, et al: Compression-induced changes in intervertebral disc properties in a rat tail model. Spine 24:996–1002, 1999.
29. Lotz JC, Colliou OK, Chin JR, et al: Compression-induced degeneration of the intervertebral disc: An in vivo mouse model and finiteelement study. Spine 23:2493–2506, 1998. 30. Gotfried Y, Bradford DS, Oegema TR: Facet joint changes after chemonucleolysis-induced disc space narrowing. Spine 11:944–950, 1986. 31. Panjabi MM, Krag MH, Chung TQ: Effects of disc injury on mechanical behavior of the human spine. Spine 9:707–713, 1984. 32. Natarajan RN, Ke JH, Andersson GBJ: A model to study the disc degeneration process. Spine 19:259–265, 1994. 33. Atlas SJ, Keller RB, Wu YA, et al: Long-term outcomes of surgical and nonsurgical management of sciatica secondary to a lumbar disc herniation: 10 year results from the Maine lumbar spine study. Spine 30:927–935, 2005. 34. Carragee EJ, Han MY, Yang B, et al: Activity restrictions after posterior lumbar discectomy: A prospective study of outcomes in 152 cases with no postoperative restrictions. Spine 24:2346–2351, 1999. 35. Urry DW, Hugel T, Seitz M, et al: Elastin: A representative ideal protein elastomer. Philos Trans R Soc Lond B Biol Sci 357:169–184, 2002. 36. Walkenhorst J, Lee D, Spenciner D: Extrusion Resistance of an Injectable Nucleus Replacement in the Human Cadaver Spine. Proceedings of the International Meeting on Advanced Spine Techniques (IMAST). Bermuda, July 2004. 37. Mahar AT, Oka R, Whitledge J, et al: Biomechanical efficacy of a protein polymer hydrogel for inter-vertebral nucleus augmentation and replacement. Proceedings of the Fourth World Congress on Biomechanics. Calgary, August 2002. 38. Walkenhorst J, Kitchel S, Spenciner D: Effect of Injectable Disc Nucleus on Function of the Human Spine. Proceedings of the Annual Meeting of the International Society for the Study of the Lumbar Spine (ISSLS), Porto, June 2004. 39. Berlemann U, Schwarzenbach O, Etter C, Kitchel S: Clinical Evaluation of an Injectable, In-Situ Curing Nucleus Replacement. Spine Arthroplasty Society Proceedings, Montreal, May 2006.
CHAPTER
55
Aquarelle Hydrogel Disc Nucleus Qi-Bin Bao and Hansen Yuan
K E Y l l
l l
l
P O I N T S
Aquarelle was the first hydrogel nucleus device. Water content and swelling pressure are similar to that of the natural nucleus. It has low modulus yet strong mechanical strength. The replacement was subjected to comprehensive biocompatibility testing with favorable results. Implant extrusion is the main challenge due to a low elastic modulus and slippery surface.
INTRODUCTION The Aquarelle nucleus is the first hydrogel nucleus arthroplasty device, the development of which was started in 19901 by Howmedica (which was subsequently acquired by Stryker). This was almost 6 years before the development of the PDN-SOLO device by RayMedica,2 which has the most clinical experience for hydrogel nucleus arthroplasty devices. The definition of a hydrogel can vary slightly in different fields. According to Dorland’s Illustrated Medical Dictionary,3 a hydrogel is a gel that has water as its dispersion medium. From a biomaterial point of view, typically a hydrogel is a gel that has water dispersed in a polymer network. According to this definition, there is no question that the natural nucleus is a hydrogel. The disc has a composite structure consisting of the nucleus pulposus core circumferentially surrounded by the multilayered fibers of the annulus fibrosis and superiorly and inferiorly by the cartilage end plates. The structure of a disc is analogous to that of an automobile tire. The annulus, with successive layers of collagen fibers oriented in alternating directions, surrounds the nucleus. Similar to the multilayers of a tire, these fibers have a relatively low compressive stiffness and strength but high tensile stiffness and strength. This anisotropic structure allows the tire and the disc to support the load much better when they are at a fully inflated stage than at a semiinflated stage. The nucleus is a hydrogel with 80% water in youth and gradually desiccates with age. Water is drawn into the nucleus by the presence of hydrophilic proteins called proteoglycans (PG). Collagen fibers cross-link PG in an irregular fashion, creating a firm hydrogel matrix. It should
be mentioned that the water content of the nucleus is largely dependent on the concentration of the two chemical components in the proteoglycans, chondrotin sulfate (CS) and keratin sulfate (KS), both of which carry a negative charge. Other than the hydrophilic nature of collagen and other components of the PG that can hydrate and retain a certain amount of water, these negatively charged components are especially important in regards to the amount of water the nucleus can retain. The higher the negative charge density in the proteoglycans, the higher the water content for the nucleus. In the progression of disc degeneration, the concentration of CS and KS in the PG decreases and that leads to the decrease in water content. Another interesting phenomenon for the hydration of the nucleus is called swelling pressure. In simple terms, this means that the water content of the nucleus changes with the external pressure (swelling pressure). If the nucleus is placed in a closed container with at least part of the surface allowing the permeation of water diffusion, as the external pressure on the nucleus increases and sustains at this high level, part of the water will diffuse through the permeable surface until the new equilibrium of water content is reached. After the equilibrium of water content is achieved at a high swelling pressure, if the external pressure is reduced and maintained at a lower level, the nucleus can draw the water from the external tissue through the permeable surface. For the disc, the annulus has very small water permeability. The main permeable surfaces are the two thin layers of cartilage end plates through multiple capillaries but with very low water permeability. In other words, in the short term, as the external load on the disc changes, the annulus and end plates act like a totally sealed container for the nucleus. The adjustment of water content to the new pressure can take hours to occur. The function of the disc mirrors its compositional complexity. In a gross picture, the disc bears the load in a similar way to a tire. The annulus is inflated by the nucleus with the tire in a vertical position while the disc is in a horizontal position. Although neither the nucleus nor air has any significant mechanical strength on its own, by inflating the annulus or tire, the load on the disc or tire is supported by the radial stretch of the annulus or the tire. As mentioned earlier, because both the annulus and tire have a 423
P A R T
424
V
Lumbar Partial Disc Replacement: Nucleus Replacement
very high tensile (or stretching) stiffness, the inflated disc or tire becomes much stiffer than when it is less inflated. However, the major difference between the two is that instead of simply pumping a certain pressure of air into the tire, the degree of nucleus inflation is regulated by its negative charge density and the external load. A nucleus with a high negative charge density has a high swelling pressure. As the disc starts to degenerate or simply age, the negative charge density decreases and, consequently, the water content decreases. For a nucleus with a given negative charge density, there is a correlation between the swelling pressure and the equilibrium water content as shown in the solid line AB in Figure 55–1. As mentioned earlier, although the air in the tire is completely sealed, the water in the nucleus can be diffused slowly through the end plates in response to the external load change. However, the water diffusion rate of the end plates is so small that it makes the disc almost impermeable when the external load, or its corresponding swelling pressure, suddenly increases from B to D in Figure 55–1. This sudden load change creates a quasiequilibrium state. If the load is sustained, the water in the nucleus will slowly diffuse through the end plates and the water content will eventually reach the new equilibrium state of A. In the evening, when a person goes to bed and the external load decreases from point A to C, the swelling pressure and water contend again reaches a nonequilibrium state. The hydrogel nucleus will act like a water reservoir and imbibes water from the vertebrae until the new equilibrium B is reached. Biomechanically, this water diffusion regulates the load distribution between the annulus and nucleus. Owing to the incompressible nature of the nucleus, the small diffusion rate and the high tensile modulus of the annulus, when there is a sudden load increase, the nucleus takes a large portion of the load and subsequently a high compressive stress is applied to the end plates in the nucleus region. If this high load is sustained, water gradually diffuses through the end plates and the nucleus starts to lose volume. The loss of volume in the nucleus causes a deflation of the disc, which leads to more of the compressive load being shifted to the annulus. Like the tire, the annulus is strong in tension 0.45
Swelling pressure (MPa)
0.40 0.35
A
D
0.30 0.25 0.20 0.15 0.10 C
0.05
B
0.00 1.0
1.5
2.0
2.5
3.0
3.5
4.0
Hydration (g H2O/g dry weight) n
F I G U R E 55–1.
under the load change.
Expelling and imbibing of water by the hydrogel
and weak in compression. If the annulus is constantly under a high compressive load, it is more susceptible to damage. Other than regulating the load distribution, the water diffusion under cyclic loading provides the necessary nutrition for the disc. The intervertebral disc is the largest avascular tissue in the body, and the main cells in the disc are chondrocytes. Owing to its avascularity, nutrients for the chondrocytes and the metabolite byproducts generated in the disc must rely on the diffusion of body fluids across the disc through the end-capillaries in adjacent vertebra during cyclic loading. If this diffusion is interrupted or impaired, the disc tends to degenerate faster than the disc with proper fluid diffusion.4 INDICATIONS AND CONTRAINDICATIONS The indications for the Aquarelle nucleus should be the same as that for other nucleus devices, that is, discogenic back pain caused by degenerative disc disease (DDD), as confirmed by patient history and radiographic studies, with or without leg pain. The contraindications for the Aquarelle nucleus should also be similar to that for other disc arthroplasty devices: osteoporosis, osteopenia, scolosis, instability caused by isthmic spondylolisthesis, spondolysis, or retrolisthesis or anterolisthesis of greater than 3 mm, and so on. Because the nucleus device is more suitable for patients with early to moderated DDD, disc height less than 5 mm should also be a contraindication. DESCRIPTION OF THE DEVICE From the unique characteristics of the natural nucleus and its functions, it becomes very logical to select a hydrogel material for a nucleus replacement device. The Aquarelle nucleus was the first hydrogel nucleus device. Since the Aquarelle nucleus, there are many other nucleus devices currently under development using a hydrogel as the material of choice. These hydrogel nucleus devices include PDN, or its newer version HydraFlex (Raymedica, Inc., Minneapolis, MN); NeuDisc (Replication Medical, Cranbury, NJ); NuCore (SpineWave, Inc., Shelton, CT); BioDisc (Cryolife Inc., Kennesaw, GA); Geliflex (Synthes, West Chester, PA); and SaluDisc (SpineMedica, Marietta, GA). As mentioned earlier, hydrogel is a relatively broad term. Although the dispersion phase for these hydrogel materials is all water, the polymer network, which determines the hydrogel properties for these devices, can vary significantly from device to device. Owing to the polymer network difference, the mechanical and biologic properties for various hydrogels can be very different. What would be the optimal polymer network for a nucleus replacement device? It seems that the ideal hydrogel material for nucleus replacement should replicate all physical, mechanical, chemical, and physiologic characteristics of the natural nucleus. In reality, owing to the change in boundary conditions, it will not be the best option to choose a hydrogel with similar properties to the natural nucleus for nucleus replacement. When a patient becomes symptomatic with low back pain caused by the DDD, the annulus of the disc often has already had different degrees of damage, in addition to the dehydration of the nucleus. Even if the annulus damage is minimal, the procedure of removing the diseased nucleus and implanting the nucleus prosthesis often requires the creation of an annular window. Therefore, the annulus is no longer intact
CHAPTER 55
and cannot contain the loose gel under substantial pressure. If the hydrogel nucleus device is as deformable as the natural nucleus and inflates the annulus like the healthy disc, it would be easily extruded through the annular defect. If one tries to avoid the extrusion problem by substantially underinflating the disc space, it will not be able to share a substantial amount of load with the remaining annulus, restore its mechanical function, and protect the annulus from overcompression. Therefore, to minimize the risk of implant extrusion, the hydrogel nucleus implant should be much firmer, or stiffer, than the natural nucleus. Other than the material modulus, for the same reason, the mechanical strength and durability of the hydrogel should also be taken into consideration in selecting the material for nucleus replacement. Although the natural nucleus has almost no mechanical strength by itself and can be easily fragmented, the nucleus replacement has to maintain its integrity after repeated loading cycles. If the integrity of the replacement is not maintained, the fragmented implant can be easily extruded through the annular defect. In addition to fatigue strength, the biostability of the hydrogel should also be considered. Because it is a long-term implant for patients at a fairly young age, there should be reasonable assurance that the hydrogel material selected will not degrade over time. The degradation of the material not only will deteriorate the mechanical strength but might also release toxic degradation by-products. One of the main reasons for choosing a hydrogel elastomer over a nonhydrogel elastomer for nucleus replacement is to restore the water swelling pressure characteristics so that body fluid diffusion can be maintained. Although most hydrogel nucleus replacements do not contain living cells, it is at least a perceived benefit to restore the body fluid diffusion so that it will continue to bring nutrients for the remaining nucleus and annulus. Because a hydrogel can be made from different polymers that might have different swelling pressure characteristics, it would be desirable to use a hydrogel with a similar swelling pressure to the natural nucleus. Last but not least, biocompatibility is a necessity for any permanent implant. If the biocompatibility of a new material has not been established yet, a series of biocompatibility tests according to the International Organization for Standardization (ISO) 10993 standards have to be conducted. Even with favorable results from these tests, the long-term stability and potential carcinogenicity issues can still be raised. If there are reasonable doubts on potential carcinogenicity, a full-scale carcinogenicity study or a relatively shorter essay using a P53 transgenic mouse model might be required. BACKGROUND OF SCIENTIFIC TESTING/CLINICAL OUTCOMES The Aquarelle nucleus is made of polyvinyl alcohol (PVA) with the following chemical structure: [CH2CHOH]n. Chemically, PVA has a similar C-C backbone to that of polyethylene (PE), which has been widely used as a permanent implantable material due to its biocompatibility and stability. PVA itself also has a long history of being used or considered for various medical applications, spanning from contact lens to skin and cartilage replacements. Unlike PE, which is hydrophobic, the hydrophilic
Aquarelle Hydrogel Disc Nucleus
425
character of PVA is mainly from its hydroxyl group (OH), and under ambient conditions, the water content is about 83%. PVA hydrogels can be formed from either chemical cross-links or physical cross-links. The Aquarelle nucleus is physically crosslinked through repetitive freeze-thaw cycles. One of the advantages of having a physically cross-linked hydrogel is that it does not include any cross-link agents, which can often be toxic if they are not fully incorporated into the polymer structure. Biocompatibility Testing With the long history of PVA hydrogels for various medical applications, the full battery of biocompatibility testing per ISO 10993 standards has been performed for the Aquarelle Hydrogel Nucleus device. These tests include l Cytotoxicity Test in L929 Mouse Fibroblast Cells PVA Hydrogel Agar Averlay Assay l Mutagenicity Assay with a Saline Extract and a Polyethylene Glycol Extract of PVA Hydrogel l Cytotoxicity Test on PVA Hydrogel in vitro Hemolysis in Rabbit Whole Blood l Mutagenicity Test on Saline and PEG200 Extracts of PVA Hydrogel in the L5178Y KTþ/ Mouse Lymphoma Forward Mutation Assay l Mutagenicity Test on Saline and PEG200 Extracts of PVA Hydrogel Measuring Chromosomal Eberations in Chinese Hamster Ovary (CHO) Cells l Genotoxicity Test on Saline and PEG200 Extracts of PVA hydrogel in the Assay for Unscheduled DNA Synthesis in Rat Liver Primary Cell Cultures l Rabbit Pyrogen Study—Material Mediated l Systemic Toxicity Study in Mice (Extracts) l Intracutaneous Toxicity Study in the Rabbit (Extracts) l Delayed Contact Sensitization Study (A Maximization Method) in the Guinea Pig (Saline Extract) l USP Muscle Implantation Study in the Rabbit (90 days) The Aquarelle nucleus has passed all these tests. In addition, the National Toxicology Program has conducted both 30-day and 2-year toxicology and carcinogenesis studies on PVA in mice and concluded that there was no evidence of carcinogenic activity (NTP).5 Swelling Pressure Study As mentioned earlier, swelling pressure is an important characteristic of the nucleus. Owing to its unique swelling pressure, the equilibrium water content of the nucleus changes with the external pressure on the nucleus. This water content change in response to the load change in daily activities regulates the degree of disc inflation and functions as a driving force for the body fluid diffusion, which is essential to the biologic activities of the disc. To obtain the swelling pressure of the nucleus or a biomaterial, other than subjecting the nucleus or a biomaterial in a container with permeable membrane to various pressures, it can also be tested by subjecting the nucleus or a biomaterial in solutions to different osmotic pressures generated by different concentration
P A R T
426
V
Lumbar Partial Disc Replacement: Nucleus Replacement
of polyethylene glycol (PEG).6 The osmotic pressure (II) of the PEG solutions was found using the following equation: II=RT ¼ c=18; 500 þ 2:59 103 c2 þ 13:5 103 c3
90
Water content (% age)
80
70
60
50
40
30 0.0 n
0.5
F I G U R E 55–2.
1.0
1.5
2.0
2.5
3.0
.7
55-1
3.5
4.0
Swelling pressure (MPa) Swelling pressure of polyvinyl alcohol hydrogel.
.6
Pressure (MPa)
where c is the concentration in g PEG/mL of the solution and 18,500 is the molecular weight of the PEG. In this study, 12 different PEG solutions of different concentrations were prepared so that a curve of PVA equilibrium water content at different swelling pressures could be obtained. For each concentration of PEG solution, five replicates of PVA hydrogel were used. Specimens were weighed and then enclosed in a 1,000 molecular weight cut off (MWCO) dialysis membrane (Spectrum Medical Industrials Inc., Los Angles, CA) and immersed in the PEG solution for 1 week to reach equilibrium. The weight of each specimen was then measured again with an analytical balance. The samples were then dried in an oven to obtain their dry weight. The equilibrium water under different swelling pressures was then calculated using the dry weight and the wet weight at a given swelling pressure. The curve of the swelling pressure of the PVA hydrogel is shown in Figure 55–2.7 When comparing the swelling pressure of the PVA hydrogel with the swelling pressure of the natural nucleus, it was found that the PVA hydrogel has a very similar swelling pressure curve to that of the natural nucleus (Fig. 55–3). The results of this study showed that the Aquarelle Hydrogel Nucleus would have a swelling pressure similar to that of the human intervertebral disc nucleus. It releases water under load and imbibes water when the load is released or decreased. These properties of the PVA hydrogel might bring the benefit of preventing further degeneration of the disc by providing adequate nutrition to the disc when used as a nucleus replacement. To our knowledge, this is the only study that has demonstrated the proper swelling pressure for any hydrogel nucleus devices.
L3–L4 discs Age 39 37 91 14 44 42 68 25 PVA hydrogel
.5
.4
.3
.2
.1
1.0
2.0
3.0
4.0
5.0
6.0
7.0
Hydration (g H2O/g dry wt.) n F I G U R E 55–3. Swelling pressure comparison between polyvinyl alcohol hydrogel and nucleus.
Biomechanical Test Multiple biomechanical studies using a human cadaveric lumbar model have been conducted for the Aquarelle Hydrogel Nucleus to demonstrate the ability of the Aquarelle Hydrogel Nucleus in restoring the disc height and biomechanical function of the lumbar segment. The first biomechanical study for the Aquarelle nucleus was presented in the 1994 annual meeting of the North American Spine Society.8 In this study, nine fresh cadaveric lumbar functional spine units were used to assess changes in disc height, disc stiffness, facet load, and distribution of intradiscal pressure at three different stages; intact, postdiscectomy and post–Aquarelle implantation. The biomechanical tests consisted of three modes of loading: pure compression (PC), compression with 5 degrees of flexion (CF), and compression with 5 degrees of extension (CE). The maximum load applied was 1,000 N at a loading rate of 100 N/s. The intradiscal pressure was measured by inserting a strain guage pressure probe along five equidistant parasagittal paths under a constant load of 1,000 N. At the end of these three stages, the intervertebral disc was dissected and the dynamic loaddisplacement test was repeated between 50 N of tension to 300 N of compression on the posterior elements for each of the three loading modes. The results of this study showed that in comparison to the intact disc height at 1,000 N, the discectomy led to a disc height loss of 12.0 6.2%, 11.4 5.4%, and 10.1 6.1% for PC, CF, and CE, respectively. After the implantation of Aquarelle Hydrogel Nucleus, the average disc height was restored to the intact level under all the three loading modes. Stiffness, measured from 800 to 1,000 N, showed no significant difference between stages except for the implant stage in pure compression.
CHAPTER 55
The stiffness for this case was about 12% lower compared with both the intact and discectomy stages (P30 days) in contact with tissue and bone. Tests include cytotoxicity, acute systemic toxicity, subchronic toxicity, chronic toxicity, sensitization, irritation, genotoxicity, and implantation. In addition to the ISO 10993 testing requirements, the silicone materials are being tested for extractable and leachable components using saline, isopropyl alcohol, and bovine serum as extract media. Owing to the proximity of this device to the spinal canal, animal studies will be conducted to evaluate the response to particulate wear debris in contact with the dura. Because this device has no articulating surfaces, the quantity and size distribution of particles chosen for this study is consistent with studies conducted on other spinal devices.6–8
Between September 2005 and June 2006, 26 subjects who were radiographically and clinically confirmed with a diagnosis of DDD were implanted with the PNR. All patients had clinical, functional, and radiographic follow-up examinations. The study subjects (12 women and 14 men) averaged 43 years, ranging from 29 to 62 years of age. Mean operative time was 78 minutes, with average blood loss of less than 50 mL. Subjects were able to take their first unassisted walk in an average of 11 hours (7 to 13 hours) after implantation. Clinical outcomes measured pain (VAS) and disability related to pain (ODI). All subjects had a decrease in the severity of pain (89 mm to 52 mm) at the time of hospital discharge. At the 3week follow-up, their mean VAS had decreased to 31 mm and ODI had improved from a preoperative average of 56% to 27%. The results demonstrated that the PNR device was effective in most of the patients who underwent surgery. After implantation, most patients experienced relief of pain. Five patients had persistent pain after implantation of the PNR and underwent uneventful anterolateral or posterolateral interbody fusion. Improvements were noted in pain intensity, walking distance, lumbar mobility, and neurologic weakness, ODI scores, intervertebral disc height, and segmental ROM. No difference in work status after PNR implantation could be detected. Compared with the preoperative height, the patients who had the intervertebral disc had gained 18.3% (P 200 patients). This analysis indicated that the median load on the spinous process in extension is 45.8 N. Thus, the endurance limit of the coflex demonstrates that it can withstand approximately three times the expected loads exerted on the spinous process in vivo for 10 million cycles. Torsional Analysis (Wing Bending) To assess the mechanical performance of the coflex in torsion, it was determined that the key design feature of the device in torsion is in the wings where they provide resistance to rotation of the spinous processes and are affected by a perpendicular load directed on the wing. Static Torsion Testing
Static torsion testing on the coflex was not performed. Only the wings are expected to bear any load when the spine is in axial rotation. Owing to the low amount of axial rotation in the lumbar spine, it is expected that the amount of loading that will be seen on these wings is also very low. However, two tests were performed: a dynamic test placing a perpendicularly dynamic load on a wing and a biomechanical analysis evaluating the coflex in worst-case bending modes. Torsion Fatigue Testing
The purpose of this test was to determine the maximum torsional load the coflex can withstand for a run-out of 10 million cycles. The coflex device survived a maximum load of 75 N to run-out. Finite Element Analysis and Mechanical Evaluation of Physiologic Loading on coflex When evaluating the loading conditions of the coflex, one must consider the loading of the apex of the U because of its positioning near the physiologic loading of the facets. A finite element analysis (FEA) and a static compression test evaluating the stress on the coflex when a load is placed near the apex of the U were performed. This FEA model was validated in the load deformation analysis to the static compression testing. The stresses observed increase
CHAPTER 70
significantly. In the FEA model, the resultant load observed for the size 8 mm coflex was 717 N. This load is much higher than those seen physiologically (400 N) for the entire posterior column. The static compression test was performed for comparison by placing a load approximately 8 mm from the apex of the U portion of the coflex. This loading condition was set to mimic the loads placed on the spine by the posterior elements, such as the facets. When a load was placed toward the apex of the U, the yield load was approximately 1,212 N. This load greatly exceeds the loads cited in the literature and is the quoted axial load on the entire spine in the upright position. Biomechanical Wing Test Biomechanical testing was performed to determine the worst-case loads exerted on the wings of the implant. Strain gauges were placed at five locations in the device (one on each side of each wing, and another in the arch of the U), and then the coflex devices were implanted in cadaveric lumbar spines. Each specimen was then articulated in flexion-extension, lateral bending, and rotation to determine the maximum forces exerted on the wings. The tests were designed to produce a 7.5 Nm external moment in each mode, which was based on prior testing in the same laboratory, indicating that this represents a worst-case in vivo moment. The results demonstrated that under these test conditions, the largest forces exerted on the wings were in lateral bending. However, even in lateral bending with an external moment of 7.5 Nm, the maximum wing-bending moment was only 0.42 Nm, and the maximum force on the wing was only 42 N. In axial rotation, the maximum wing-bending moment was only 0.26 Nm, and the maximum force on the wing was 26 N. In flexion-extension, the maximum wing-bending moment was only 0.16 Nm, and the maximum force was 16 N. Thus, the endurance limit of 75 N provides adequate assurance that the coflex can withstand the worst-case moments that may be applied to the device in vivo. When comparing the resulting loads seen in the study, specifically those in axial rotation, an endurance limit of 75 N at 10 million cycles is more than a 2.7 factor of safety. Wear Characteristics of the coflex The coflex device consists of one piece of milled titanium alloy without any articulating surfaces. This design eliminates the possibility of fretting corrosion and the generation of wear debris. Therefore, in vitro wear testing is not warranted for the coflex. Biomechanical Evaluation of the Effect of the coflex on Range of Motion Biomechanical testing was performed in a series of tests on cadaveric specimens to assess the impact of the coflex on spinal segment motion in flexion-extension, lateral bending, axial rotation, and compression.6 The following testing conditions were evaluated sequentially in each of eight cadaveric spine specimens: (1) the intact spine, (2) the partially destabilized spine (resection of all ligaments, ligamentum flavum, facet capsules, and bilateral 50% resection of inferior facets), (3) the partially destabilized spine then implanted with coflex, then (4) the complete destabilization of the spine via
coflex Interspinous Implant for Stabilization of the Lumbar Spine
537
total laminectomy (coflex removal), and finally (5) the completely destabilized spine restabilized with pedicle screw fixation. In the first test, for purposes of characterizing the effect of the coflex on motion, eight cadaveric specimens (L4-L5) were tested to represent a range of potential anatomic presentations. The center of rotation was established as part of the test set up to minimize offaxis bending, and the specimens were preconditioned via dynamic compressive loading for 1,000 cycles at 500 N. After determining the center of rotation, compression (900 N at 0.25 cm/minute), flexion-extension (12 Nm), lateral bending (12 Nm), and axial rotation (9 Nm) were applied to each cadaveric spine under a 600-N preload. Each specimen and device was then tested in each mode three times, and a load-deformation curve was generated. The results of this test demonstrated that the coflex principally serves to stabilize motion in flexion-extension and initially in axial rotation. In each mode, the motion permitted by the coflex was normalized to the amount of motion permitted by the intact spine. Similar results were observed in axial rotation, with a significant increase in the range of motion observed after partial destabilization, but restoration of near-normal rotational motion following implantation of the coflex. Because the coflex device is located centrally between the spinous processes, it is not designed to provide any significant limitation on lateral bending, which was confirmed by the results of the cadaveric testing. In Vivo Loading Environment of the coflex The coflex is intended to be implanted with the superior and inferior wings facing the sides of the cranial and caudal spinous processes, respectively. The apex of the U is inserted in line with the facets and leads to a slight distraction of the facets. The loads exerted on the coflex in vivo are applied initially at the apex of the U at the level of the facets and to the arms of the U in extension (and the wings at the contacting interfaces with the cranial and caudal spinous processes). Therefore, relatively little load is expected to be exerted on the spinous process, as the majority of the load on the lumbar spine is carried by the intervertebral disc and the facets. In flexion, the loads on the spinous processes are generally expected to be reduced, as flexion causes the spinous processes to separate further. If bony ingrowth or crimping causes the coflex to become fixed to both the upper and lower spinous processes, the arms of the U could be separated slightly in flexion, placing the device in tension. As each spinous process moves between the wings in flexion, low bending forces are also applied to the wings. However, in extension, the spinous processes move closer together, increasing the compressive load that is exerted on the U. When the coflex is properly inserted, low loads are exerted on the spinous processes generally. This fact is coupled with the limitations on the extent of compression that are imposed anatomically by the facets. The typical compressive loads on the coflex in extension are not expected to exceed approximately 45.8 N as extrapolated from direct in vivo x-ray measurements in more than 200 patients. In lateral bending, the spinous processes transmit load to the coflex via contact with the wings. Direct measurement of the forces on the device when implanted in cadaveric specimens has been performed using strain gauges. These measurements have demonstrated that even under lateral-bending moments of 7.5 Nm, the
538
P A R T
V I I
Lumbar Posterior Dynamic Stabilization: Interspinous Based
degree of force applied to the wings of the coflex is relatively low, averaging approximately 23 N. A maximum bending force on the wing is estimated at 42 N for extreme bending moments of up to 10 Nm and overall very little bending (torsional) load is transmitted to the U. There is no significant “twisting” force at the apex of the U, and instead the wings are designed to bend at the point of attachment to the U arms. Based on the direct measurement of forces on the device in cadavers, the magnitude of this load under a 7.5-Nm externalbending moment is less than in lateral bending with an average of approximately 16 N (a maximum bending load of 27 N at extreme-bending moments of 10 Nm). The degree of torsional loading transmitted to the U is thus minimal. Maximum compressive loading of the device is expected to occur when the spine is in full extension, and maximum-bending loads will occur in lateral bending. The mechanical testing, together with the extensive prior clinical experience with the device, demonstrates that it can withstand the worst-case physiologic loads that are expected to occur in vivo without permanent deformation or breakage of the device and without fracture of the spinous processes. In Vivo Deflection and Loading Analysis To assess the maximum load that may be exerted on the device in vivo, a comprehensive review of the degree of implant deflection was performed by an independent Radiographic Analysis Core Laboratory (Medical Metrics, Inc., Houston, TX) using radiographs obtained in the retrospective study. Review of the neutral, extension, and flexion radiographs for the 209 subjects with stenosis, which was the population eligible for participation in the U.S. IDE study, was performed. The collected ranges were then inserted into a validated finite element analysis. Based on this analysis, a load/deformation curve was developed. Medical Metrics, Inc. took the validated FEA and applied the formula to the retrospective data patients. The median load in extension was 45.8 N. This demonstrated that the maximum estimated load exerted on the implant was less than the 150 N endurance limit found for 97.1% of the implants, with only one case (0.5%) in which the calculated load on the coflex device in extension was greater than the reported yield load for the device (239 N). This load deformation analysis demonstrates that the compressive static and fatigue strength of the coflex is adequate to withstand the worst-case dynamic loading conditions that are expected in clinical use. In addition, Medical Metrics, Inc. conducted an analysis of the range of motion of the coflex and adjacent levels in 180 patients. This number of patients was based on the original cohort of patients (209). The analyses determined that the range of motion for one-level coflex implantations at preoperative time was 4.3 degrees, 2.1 degrees at 1 year, 2.3 degrees at 2 years, and 2.1 degrees beyond the 2-year point. The range of motion (flexion-extension) for the two-level patients was 4.0 degrees at the preoperative time, 2.1 degrees at 1 year, 1.6 degrees at 2 years, and 3.4 degrees beyond the 2-year point. CLINICAL PRESENTATION AND EVALUATION The company has gathered retrospective data on the clinical outcomes of 429 coflex patients, as well as contemporaneous clinical
and x-ray data on the same cohort of patients. These patients represent a portion of the 589 patients treated with the coflex at four clinical sites from which data were available. Of these 429 patients, 209 patients were treated for spinal stenosis at a single level or two adjacent levels, a population that is substantially similar to the population for the protocol that is the subject of the IDE study. In this population, the coflex demonstrates a clear record of safety as well as preliminary evidence of efficacy. Patient data for the retrospective study were gathered via a questionnaire that captured the following information: gender, preoperative diagnosis, preoperative clinical evaluation, previous conservative therapy, previous spinal therapies, concomitant medical conditions, operative data, radiographic and diagnostic tests, postoperative tests, postoperative clinical examination, and qualitative postoperative x-ray analysis. All patients from the four sites who were more than 6 months past operation were given the option of participating in this data collection, which the company believes helped to minimize selection bias. Of the 589 patients identified by the surgeons, 429 (73%) responded and agreed to participate. All patients were asked to return for contemporaneous history, clinical examination, and dynamic x-rays. These results were compared to available patient records and pre-existing x-rays pertaining to their quality of life and implant survivorship. The patient case report forms and x-rays from 429 patients were reviewed by an independent orthopaedic spine surgeon who identified 209 patients with spinal stenosis. The remaining patients were treated for various indications such as “topping-off ” of spinal fusions, use of the device with other spinal implants, disc herniation, and other diagnoses. Table 70–1 evaluates patients’ clinical success outcomes, satisfactions, and adverse events relative to the three time intervals of follow-up. For the follow-up groups of 6 to 12, >12 to 24, and >24
TABLE 70–1. Time* Postoperative Outcomes (n ¼ 209)
Improvement in moderate or severe preoperative low back pain Improvement in preoperative claudication Improvement in preoperative walking distance Patient satisfaction “Would have surgery again” Adverse events (n ¼ 17) “Device-related” issues (n ¼ 7)
Composite Pertinent Patient Outcomes Over Overall
Follow-Up Time Intervals
n ¼ 72 6 to 12 mo
n ¼ 69 >12 to 24 mo
n ¼ 68 >24 mo
75%
73%
82%
72%
87%
90%
85%
87%
74%
83%
75%
66%
89% 92%
90% 96%
91% 90%
88% 91%
8.1%
8.1%
0%
0%
3.3%
1.9%
0.5%
0.5%
*Percentage calculations relative to total spinal stenosis population per side.
CHAPTER 70
months the mean follow-up times were 8.8 1.5, 18.4 1.7, and 53.7 9.4 months, respectively, with the median follow-up times being 9, 19, and 33 months, respectively. Again, no differences were noted between one- and two-level decompressions relative to clinical outcomes, and the incidence of adverse events was low and not different between these two groups. Therefore, Table 70–1 was constructed to see the composite pertinent patient outcomes overall and their relationship over time. Overall, 75% of the patients had improvement in their moderate or severe preoperative low back pain, and this improvement remained constant over time. Claudicatory symptoms improved overall in 87% of the patients and again remained reasonably constant over time. Postoperative walking distance improved overall in 74% of the patients with a slight tapering effect at the longer term follow-up period (a decrease to 66%). This observation could be related to the increasing age/activity of the patient. Patient satisfaction was positive in 89% of the patients overall, and 92% of the patients stated they “would have surgery again.” Both of these results remained constant during all three time intervals. Non-device–related adverse events (8.1%) occurred uniquely in only the short-term follow-up group (6–12 months). “Device-related issues” were low (3.3% overall), and although few, the majority were noted less than 13 months postoperatively (2.4%). Only one patient in the long-term follow-up (>24 months) was noted to need a second operation and removal of the U with subsequent conversion to fusion after a fall (no device fracture or deformation, but a new L4-L5 spondylolisthesis). The retrospective data demonstrated an extremely low incidence of breakage or deformation of the U component, with the primary mode of mechanical failure at the wing (two cases). Wing breakage was noted intraoperatively secondary to opening and closing of the wings with standard surgical pliers.
coflex Interspinous Implant for Stabilization of the Lumbar Spine
and any bony overgrowth of the spinous process that may interfere with insertion is resected. Microsurgical Decompression The ligamentum flavum is resected and microsurgical decompression is performed. This procedure includes bilateral laminotomy or hemilaminectomy with appropriate resection of overgrown facets and foraminotomies as needed in order to relieve all points of neural compression. Implant Site Preparation Trials of the device are utilized to define appropriate implant size. Selection of the appropriate implant size is essential in obtaining proper function of the device and good clinical results. The trial instrument is placed to evaluate proper contact with spinous process and the amount of interspinous distraction. Some bony resection of the spinous process may be needed to ensure proper contact of the implant. Distraction is considered to be appropriate to prevent any settling of the interspinous distance after successful decompression of the spinal stenosis. The appropriate sized implant will fit securely on the spinous processes and lead to a 1- to 2-mm slight distraction of the facet joints. To ensure proper depth of implant insertion, a small portion of the laminar surface may need partial resurfacing before implantation of the device (Fig. 70–2).
OPERATIVE TECHNIQUE Preparation The patient is placed in a prone position on the surgical frame, avoiding hyperlordosis of the spinal segment(s) to be operated on. A neutral position or a slight kyphosis may be advantageous for surgical decompression as well as for the necessary interspinous distraction. A routine (midline) skin incision is performed. The muscle is sharply dissected lateral to the supraspinous ligament, preserving the entire thickness of the supraspinous ligament. Resection of the supraspinous ligament should be avoided, if possible. If maintenance of the supraspinous ligament is impossible, however, it may be resected. If resection is needed, it should be reconstructed following insertion of the device. Paraspinal muscles are then stripped off the laminae while preserving the facet capsules. The supraspinous ligament is dissected subperiosteally and preserved as a thick cuff and retracted laterally. If resection is necessary, a small portion of the bony tip can be resected together with the supraspinous ligament. This method will aid in faster healing after reconstruction of the ligament. Depending on the pathology, a microsurgical unilateral decompression can then be performed, and the supraspinous ligament together with the fascia and muscle from the opposite side can be mobilized. Completion of the microsurgical decompression can then be performed. The interspinous ligament is sacrificed,
539
n
F I G U R E 70–2. Implant site preparation.
P A R T
540
V I I
Lumbar Posterior Dynamic Stabilization: Interspinous Based
Implant Insertion First, the surgeon must determine if the wings have to be bent open at all by holding the coflex over the spinous processes and checking the width. The wings do not need to be opened if it is felt that in the process of insertion, they will make good bony contact and be placed deep into the interspinous space, ensuring appropriate distraction. Bending pliers are available to bend the wings open prior to application while avoiding any damage to the spinous processes during implant insertion. If the spinous processes are felt to be too thick, or will impede safe insertion, special pliers are utilized for controlled bending of the implant wings. Use only the special wing-bending pliers provided. The wings are bent open (or crimped together) until both sets of wings are approximately 2 mm smaller in width than the receiving spinous processes. The bending result should correlate with the actual thickness of the spinous processes and allow easy implantation without any stress on the adjacent spinous processes. The implant is introduced via impaction utilizing a mallet. Proper depth is determined if a beaded tip probe can be passed freely leaving 3 to 4 mm of separation from the dura (Fig. 70–3). If the implant is not seated appropriately, further resurfacing or slightly more impaction force may be utilized. For appropriate bony contact after insertion, additional stability is achieved by crimping the wings, utilizing the coflex crimping pliers. The crimping pliers close the wings in a controlled manner after insertion. It is important to note that crimping has to correlate with the initial bending. After impaction, crimp the implant until the teeth on the contact surface of the wings penetrate into substantia corticalis. Appropriate stability is achieved if surrounding fat tissue starts to squeeze around the wings. In the event that there is wing
breakage during the insertion process, the coflex device should be explanted and replaced with a new one. In the event of a spinous process fracture, the coflex should remain explanted. In the case of ligament reconstruction a figure–of-eight suture can be passed through two bone holes in the superior spinous process and the supraspinous ligament. If, due to Baastrup's disease, the patient has minimal or no supraspinous ligament, tight closure of the muscle and fascia around the spinous processes should be performed. A surgical drain may be placed as per surgeon preference. Paraspinal muscles are reattached to the supraspinous ligament. Skin is closed in the usual manner. Alternatively, the fascia and the supraspinous ligament can be closed in one layer over the spinous processes. Double-Level Implantation If a two-level decompression is mandated, the implants must be sequentially placed to the appropriate depth, avoiding any overlap (contact) of one set of wings on the other. The inferior implant should be inserted first, followed by the superior implant. After appropriate anteroposterior and lateral radiographs are taken intraoperatively, the wings should be crimped. POSTOPERATIVE CARE Compliance with postoperative care is important to the success of the coflex device. A soft corset or brace may be used for patient comfort. The patient should avoid excessive lumbar flexion and extension as well as lifting heavy objects for 6 weeks after surgery. Gradual return to daily living activities and exercise, such as walking, can be initiated as soon as the surgical incision is healed. CONCLUSION The coflex interspinous stabilization device is ideal for spinal stabilization after surgically addressing neural decompression from soft or bony stenosis of the spinal canal. The device is functionally dynamic, is compressible in extension, allows flexion, and increases rotational stability. Insertion is a less invasive, tissue-sparing procedure. The coflex has up to 11 years of clinical history and more than 15,000 implantations worldwide to prove its safety. REFERENCES
n
F I G U R E 70–3. Implant insertion.
1. Arnoldi CC, Brodsky AE, Cauchoix J, et al: Lumbar spinal stenosis and nerve root entrapment syndrome: Definition and classification. Clin Orthop 115:4–5, 1976. 2. Yong-Hing K, Kirkaldy-Willis WH: The pathophysiology of degenerative disease of the lumbar spine. Orthop Clin North Am 14:501– 503, 1983. 3. Delamarter RB, Howard M: Lumbar spinal stenosis. In Hochschular S, Cotler H, Guyer R (eds): Rehabilitation of the Spine, Science and Practice. St. Louis, Mosby, 1993, pp 443–456. 4. Deyo RA, Rainville J, Kent DL: What can the history of physical examination tell us about low back pain? JAMA 268:760–765, 1992. 5. Fast A, Greenbaum M: Degenerative lumbar spinal stenosis. Phys Med Rehabil St Art Rev 9:673–682, 1995. 6. Tsai KJ, Murakami H, Lowery GL, Hutton WC: A biomechanical evaluation of an interspinous device (coflex) used to stabilize the lumbar spine. J Surg Orthop Adv 15(3):167–172, 2006.
CHAPTER
71
X-STOP Interspinous Process Decompression for Lumbar Spinal Stenosis Cary Idler, James F. Zucherman, Kenneth Y. Hsu, and Matthew Hannibal
K E Y l
l l
l
l
P O I N T S
The X-STOP, FDA approved November 2005, can be used for neurogenic intermittent claudication at one- or two-level lumbar spinal stenosis. The X-STOP preserves the interspinous and supraspinous ligaments. The X-STOP preferred approach is performed with local anesthesia as an outpatient procedure. The 1-, 2-, and 4-year outcomes are at least as good as those for laminectomy, with lower cost and lower surgical risk. The X-STOP can be used in low-grade spondylolisthesis.
The X-STOP (Kyphon, Inc., Sunnyvale, CA), an interspinous process spacer, is a promising surgical treatment alternative for neurogenic intermittent claudication (NIC) caused by lumbar spinal stenosis (LSS). The device provides an unloading distractive force to the stenotic middle column of the motion segment and can relieve claudicatory symptoms of central, lateral, and foraminal stenosis. Other devices currently being tested are suggested for degenerative disc disease, adjacent level syndromes, lumbar spinal stenosis, and herniated disc.1 Some spacers require either the supraspinous ligament or interspinous ligament to be significantly altered or removed before they can be inserted, and some spacers require the spinous processes themselves to be either modified or shaped. Some spacers are designed to function as stand-alone devices while others incorporate an artificial ligament as an integral part of the design. The artificial ligament helps to limit flexion and it may also decrease the laxity of the motion segment, which could be an important component in treating certain pathologies such as degenerative disc disease. NIC is the most common symptom seen in lumbar spinal stenosis. Patients typically obtain relief with sitting or positions of flexion and are exacerbated with standing or walking. Moreover, elderly patients tend to be osteopenic and at risk for osteoporosis so any shaping or remodeling of the spinous process would reduce bone strength and should be avoided. In fact, care should be taken to avoid decorticating or damaging any bone surrounding the spinous process.
The first interspinous process decompression device approved in the U.S. by the FDA for general use (Nov. 21, 2005) was the X-STOP. It was approved for use in Europe in July 2002. Since then, over 10,000 devices have been implanted. Placement of this device requires preservation of the spinous process and interspinous and supraspinous ligaments. This chapter will describe current treatment options, patient selection, biomechanical studies, the technique for performing interspinous process decompression (IPD) with the X-STOP, as well as outcomes from all clinical, biomechanical, and radiographic studies published to date. CURRENT TREATMENT Treatment for NIC involves conservative and operative measures. Conservative treatment usually begins with activity management as well as non-steroidal anti-inflammatory medications, physical therapy, and a short course of oral steroids. Trunk stabilization and core muscle strengthening is typically the goal in physical therapy. However, bracing and physical therapy alone have little proven efficacy.2 Hurri and associates showed 44% had some improvement with nonoperative treatment.3 Atlas and associates found that 45% of patients had improvement in leg pain,4 and Johnsson and associates reported 32% considered their symptoms improved with conservative treatment.5 Epidural steroid injections are often used as an adjunct in patients with severe or unremitting radiculopathy or NIC. In about one third of cases, this treatment can result in enough relief to avoid surgery for a short period of time. However, long-term relief is less likely.6 If conservative treatment fails to provide relief or the condition is worsening, operative treatment is indicated. Traditionally, the surgeries include decompressive laminectomy, laminotomy, or foraminotomy, depending on the anatomic region of the stenosis. Moreover, fusion may be indicated where the motion segment is unstable. The success of these surgeries varies with severity, surgical technique, patient selection, and outcomes measures.
541
542
P A R T
V I I
Lumbar Posterior Dynamic Stabilization: Interspinous Based
A meta-analysis of 74 studies related to surgery for spinal stenosis reveals a rate of good to excellent results of 64% at 1 year.7 Prospective studies such as the Maine Lumbar Study have shown superior outcomes for operative treatments of symptomatic lumbar stenosis compared to nonoperative treatment.8 Surgical decompression, while offering the potential to improve the quality of life for the patients, also has the potential for significant complications, especially when a fusion is performed and in revision surgery. Postoperative complications may include the cardiovascular and pulmonary complications of general anesthesia, infection, iatrogenic instability, pseudarthrosis, hardware failure, and the need for future surgery due to the development of disease at adjacent levels.9 A meta-analysis of the literature of spinal stenosis surgery by Turner and associates in 1992 showed the following complication rates for lumbar decompressive surgery: perioperative mortality (0.32%), dural tears (5.91%), deep infection (1.08%), superficial infection (2.3%), deep vein thrombosis (2.78%), and any complication (12.64%).10 In a study by Yuan and co-workers, 2% to 3% of patients undergoing lumbar decompression and arthrodesis, with or without internal fixation, suffered an infection, and the risk of nerve root injury from placement of pedicle screws was 0.4%.11
INDICATIONS AND CONTRAINDICATIONS Patient selection criteria include leg, buttock, or groin pain with or without back pain which is relieved with sitting or flexion. Once the diagnosis is confirmed with either magnetic resonance imaging (MRI) or computed tomography (CT), at one or two levels, and the patient has undergone a trial of conservative management (typically up to 6 months), placement of the device can be considered. Moreover, patients should be able to sit for at least 50 minutes without pain. Contraindications include cauda equina syndrome, scoliotic Cobb angle greater than 25 degrees, gross instability at the motion segment, fragility compression fracture or severe osteoporosis, Paget's disease, metastasis to the vertebrae, ankylosis at the affected segment, spinal anatomy that would cause the device to become unstable (such as aplastic spinous process or spina bifida occulta), isthmic spondylolisthesis, olisthesis, and degenerative spondylolisthesis greater than Meyerding Grade 1. Spondylolisthesis up to Grade 1 is indicated and described in more detail later. Patients with prior spinal surgery were excluded from the study trials, however, patients who have had prior laminotomy from a microdiscectomy may be considered, assuming the interspinous and supraspinous ligaments are intact. Although, extensive prior laminectomy would be a contraindication. X-STOP may also be indicated for the patients unable to undergo general anesthesia. Although in the clinical study conducted in the U.S., patients with symptomatic stenosis at L5-S1 were excluded, the X-STOP has been successfully implanted at the L5-S1 level in Europe in patients with sufficiently sized S1 spinous processes. Approximately one third of patients in the United States have received implants at two levels, but three-level procedures were not performed in the U.S. study. As with L5-S1 procedures, triple-level procedures are performed in Europe, but less frequently.
DESCRIPTION OF THE DEVICE The X-STOP was developed to provide a safer, less invasive treatment option for those who fail conservative management and those needing the riskier decompression surgery. The X-STOP was designed specifically to reduce extension only at the individual level(s) that provokes symptoms, while allowing unrestricted movement in flexion, axial rotation, and lateral bending of the treated as well as untreated level(s).12 Because the implant was designed to be placed without removing any bony or soft tissues, the technique is minimally invasive and is usually performed with the patient under local anesthesia. Thickened lamina, hypertrophied ligamentum flavum, spondylolisthesis, disc bulges, and facet hypertrophy can concomitantly lead to canal, foraminal, and lateral recess stenosis.13 NIC is often related to position so that symptoms such as pain, numbness, tingling, and weakness are elicited with extension of the lumbar spine and relieved in flexion or sitting.14,15 The level affected primarily is L4-L5, followed by L3-L4, L5-S1, L2-L3, L1-L2.16 Several key design features allow for the straightforward implantation of the X-STOP. The oval spacer separates the spinous processes and limits extension at the implanted level (Fig. 71–1). The oval spacer helps distribute the load along the generally concave shape of the spinous processes and, by eliminating any sharp edges, reduces the likelihood of damaging the cortical bone. The two lateral wings prevent migration laterally, the lamina prevents migration anteriorly, and the supraspinous ligament, as well as the concave space between the spinous processes, prevents the implant from migrating posteriorly. The tapered tissue expander facilitates lateral insertion, from right to left, allowing the supraspinous ligament and its insertions to be preserved (Fig. 71–2).
n
F I G U R E 71–1. An image of the X-STOP depicting the adjustable universal wing, tissue expander, fixed wing, and spacer. The tapered tissue expander allows for easier insertion between the spinous processes. The universal and fixed wings limit anterior and lateral migration. The spacer limits extension of the treated spinous processes.
CHAPTER 71
X-STOP Interspinous Process Decompression for Lumbar Spinal Stenosis
n
F I G U R E 71–2. Posterior and lateral views of a lumbar motion segment with an implanted X-STOP. The implant is placed posterior to the lamina and away from the nerve roots and spinal cord. The supraspinous ligament is retained to prevent posterior migration. The implant is not fixed to any bony structures.
The L5-S1 level may present a difficult challenge. Most people lack an S1 spinous process large enough to support the device. Those who do have a large S1 spinous process usually are those who have a lumbarized S1 segment. In these cases, the X-STOP can be placed in the same manner as the proximal segments. SCIENTIFIC TESTING Biomechanical studies have shown that the X-STOP significantly prevents narrowing of the spinal canal and neural foramina, limits extension, and reduces intradiscal pressure and facet loading. In an MRI cadaver study, Richards and associates reported that the X-STOP increased the neural foramina area by 26% and the spinal canal area by 18% during extension. In addition, foraminal width was increased by 41% and subarticular diameter by 50% in extension.17 In a kinematics cadaver study, Lindsey and associates18 showed terminal extension at the implant level was reduced by 62% following X-STOP placement, although lateral bending and axial rotation range of motion were unchanged. In a cadaveric disc pressure study, Swanson and associates19 reported that the pressures in the posterior annulus and nucleus pulposus were reduced by 63% and 41%, respectively, during extension and by 38% and 20%, respectively, in the neutral standing position. Rohlmann and associates performed a similar study and found a slight increase in intradiscal pressure, which reduced dramatically with extension at the implanted segment. This finding, however, occurred only when the interspinous space was distracted more than 6 mm.20 They recommend not placing an implant much larger than the interspinous space. Finally, Wiseman and associates21 performed a cadaveric facet
543
loading study and reported that the mean facet force during extension decreased by 68%. In each of those studies, the adjacent level measurements were not significantly changed from the intact specimen state. These preclinical studies indicate that the X-STOP increases spinal canal and neural foramina space and produces significant unloading of the disc and facets. In a 6-month clinical follow-up MRI study, Wardlaw and associates and Siddiqui and associates reported equal results in their clinical studies evaluating positional MRI changes after X-STOP implantation.22,23 Siddiqui and associates found in 17 levels an increase in the dural sac from 77.8 to 93.4 mm2 (P ¼ 0.006). In a later study, Siddiqui and associates found in 26 patients no differences in disc heights, end plate angles, and segmental and lumbar range of movement.24 The supraspinous ligament is a substantial structure, and its presence, as well as the preservation of its original osseous insertion, prevents overdistraction of the segment. The ultimate load and tensile strength of the interspinous-supraspinous ligament complex are 203 N and 1.2 Mpa, respectively.25 In another biomechanical study, the supraspinous-interspinous ligament complex was the largest contributor to resisting applied flexion moments in the porcine lumbar spine.26 A relative contraindication for its use is in a patient with severe osteoporosis. Talwar, and associates showed that the spinous process is significantly weaker in those with low bone mineral densities, and therefore, care must be taken when implanting the X-STOP in these patients.27 CLINICAL OUTCOMES A multicenter prospective, randomized controlled trial was performed in the United States, comparing the outcomes of mild-tomoderate NIC patients treated with the X-STOP interspinous process decompression system to those of patients treated nonsurgically.28 There were 191 patients treated at nine centers. Inclusion criteria included patients older than 50 years old and patients with leg, buttock, or groin pain with or without back pain while walking, relieved with forward flexion, and able to sit for at least 50 minutes pain free. Exclusion criteria includes a fixed motor deficit, cauda equina syndrome, significant instability, previous lumbar surgery, dense peripheral neuropathy, scoliosis with Cobb angle greater than 25 degrees, spondylolisthesis greater than Grade 1 (25% slip or less), history of pathologic compression fractures including fragility fractures, severe osteoporosis, obesity (BMI >40), and active infection. Eligible patients were randomized to either the X-STOP group or the conservative care group. Those randomized to the conservative care group received one or more epidural steroid injections and had the option to receive non-steroidal anti-inflammatory drugs (NSAIDs), analgesics, and physical therapy and additional injections as needed. Physical therapy consisted of “back school,” which included stabilization exercises, pool therapy, massage, and cold/hot packs. Assessments were based on baseline (prior to initial treatment) values and findings at 6 weeks, 6 months, 1 year, 2 years, and 4 years. Assessment data were based on outcomes measured specifically for neurogenic claudication, the Zurich Claudication Questionnaire (ZCQ), as well as the SF-36.
544
P A R T
V I I
Lumbar Posterior Dynamic Stabilization: Interspinous Based
One hundred patients received the X-STOP, and 91 patients were treated nonoperatively. A total of 136 levels were implanted in 100 patients: 64 single levels and 36 double levels. One-level procedures took an average of 51 minutes, and two-level procedures took 58 minutes. Blood loss was negligible: 40 mL for one-level procedures and 58 mL for two-level procedures. The most common level implanted was L4-L5 (89/136), and the second most common level was L3-L4 (43/136). The most common implant size was 12 mm. There were five X-STOP sizes available during the trial, ranging from 6 mm to 14 mm. The procedure was performed under local anesthesia in 97 patients and under general anesthesia in 3 patients. The length of stay was, on the average, less than 24 hours. At 2-year follow-up, data from 93 of the 100 X-STOP patients and 81 of the 91 control patients were available for analysis. The X-STOP group had a significantly greater percentage of patients with an improvement in Symptom Severity Domain of ZCQ than did the control group at each posttreatment visit. At 2-year follow-up, 56/93 (60.2%) of the patients reported a clinically significant reduction in the severity of symptoms compared to the 15/81 (18%) of the control group. The X-STOP group also had a significantly greater percentage of patients with an improvement in Physical Function Domain of ZCQ than did the control group at each posttreatment visit. At the 24-month evaluation, 57% of the X-STOP patients reported a clinically significant improvement in their physical function compared to 15% of the control patients. At 2-year follow-up, 73% of the X-STOP patients were at least “somewhat satisfied” compared to 36% of the control group. Interestingly, patients with two-level X-STOPs had greater symptom relief than those with the single level, although not significantly greater. This is opposite to the trend seen in laminectomy and fusion cases, in which multilevel procedures tend to have less favorable outcomes. Results of the SF-36 scores showed no significant differences in the pretreatment enrollment scores between the X-STOP and control groups for any SF-36 domain. At all follow-up time points, the X-STOP group scored significantly better than the control group in every physical domain including the mean scores, whereas in the control group, none of the mean scores were better. More recently, based on the original 18 patient FDA pilot study group who all received X-STOP, Kondrashov and associates showed 78% had successful outcomes at 4-year follow-ups. They had a 15 point improvement in Oswestry Disability Index (ODI) compared with the baseline (Table 71–1).
TABLE 71–1. Zurich Claudication Questionnaire Success Rates at 1 and 2 Years Period
Category
One year N100/N91
Symptom Severity Physical Function Patient Satisfaction Symptom Severity Physical Function Patient Satisfaction
Two years N93/N81
X-STOP
Control
P Value
64% 70% 75% 60% 57% 73%
21% 21% 45% 19% 15% 36%
0.001 0.001 0.001 0.001 0.001 0.001
GERMAN REGISTRY In Germany, a registry is being maintained to gather prospective data on NIC patients who are treated with the X-STOP implant in general practice. Patients are assessed pre- and postoperatively using the validated, condition-specific ZCQ. The ZCQ is the only validated LSS-specific outcome measure. The questionnaire consists of three domains: Symptom Severity (SS), Physical Function (PF), and Patient Satisfaction (PS). To date, 212 patients have been evaluated 1 year after surgery with very good results (Table 71–2). Two patients had a reoperation because of lack of efficacy and one because of dislocation of the implant. EUROPEAN CLINICAL EXPERIENCE A prospective clinical evaluation of 15 patients at 3- and 6-month follow-ups was carried out by Wardlaw and associates in conjunction with pre- and postoperative positional MRI scan measurements. All cases demonstrated clinical improvement, and the X-STOP implant increased the cross-sectional area of the dural sac and exit foramina without affecting overall movement of the lumbar spine.29 Heijnen and Kramer reported on the satisfaction of 14 patients with NIC who were treated with the X-STOP implant. One patient died of a non-back-related disorder. Eleven of the other 13 patients expressed great satisfaction. They are free of NIC symptoms, and all but one would undergo the surgery again, if the choice had to be made again.30 PATIENTS WITH DEGENERATIVE SPONDYLOLISTHESIS Interestingly, 39 patients in the U.S. FDA study with Grade I degenerative spondylolisthesis were treated with the X-STOP and 22 patients were treated nonoperatively. Using 15-point improvement over baseline scores in the ZCQ as the criterion of clinical success, 69% of the 39 X-STOP patients had a successful outcome at 2-year follow-up, compared to 9% of the control patients. The mean improvement score for the 39 X-STOP patients was 26 points. There were no significant differences in the mean percentage of slip between X-STOP and control patients at baseline or at 2-year follow-up. Spondylolisthesis patients are often treated with spinal fusion and decompression. Moreover, Anderson and associates in a cohort of 75 patients, 42 X-STOP and 33 nonoperative control patients at 2-year follow-up, showed statistically significant clinical success in 63.4% of X-STOP patients and 12.9% of control patients using ZCQ and SF-36 outcome measures. Sagittal balance (listhesis and kyphosis) remained unaltered.31 The X-STOP represents a
TABLE 71–2. After Surgery ZCQ Category
German Registry Success Rates 6 and 12 Months 6 Months
12 Months
82% 81% 82%
82% 77% 82%
Symptom Severity Physical Function Patient Satisfaction ZCQ, Zurich Claudication Questionnaire.
CHAPTER 71
545
X-STOP Interspinous Process Decompression for Lumbar Spinal Stenosis
significantly less invasive alternative therapy for these patients, resulting in very good clinical outcomes and, most importantly, no evidence that the implant results in any instability of the motion segment. SAGITTAL BALANCE The requirement to maintain proper sagittal alignment and balance in patients receiving spinal implants is well understood. Experience with lumbar fusion procedures that cause a flat back has overwhelmingly resulted in unacceptable clinical outcomes. Three different radiologic studies were therefore undertaken to measure any possible effect of the X-STOP on sagittal alignment. In the U.S. study, radiographs were taken at each follow-up visit for both X-STOP and control patients, and measurements were made of the lumbosacral angle (L1 to S1) and the treated intervertebral angle. At 2-year follow-up, there were no significant differences in the mean angles between the two groups of patients. Preoperative x-rays from a subset of X-STOP patients were also compared to standing films taken at 2-year follow-up. In 23 patients with single-level implants, the change in the intervertebral angle was only 0.5 degree, and the change in the lumbosacral angle was 0.1 degree. A newly published report by Siddiqui and associates using a positional MRI scanner, in addition to confirming in vivo the increases in the area of the foramen and canal that were measured in the preclinical in vitro cadaver study, confirms a minor change in angulation for both the lumbosacral angle and intervertebral angle of approximately 1 degree.32 These studies confirm that the X-STOP results in only minimal changes to sagittal alignment. This is due to preserving the supraspinous ligament and its original insertions. This ligament is a very robust structure receiving the confluence of the lumbodorsal fascia, and its preservation prevents overdistraction of the segment. X-STOP VERSUS DECOMPRESSIVE LAMINECTOMY It is not easy to interpret X-STOP clinical results in the context of published outcomes of surgical treatment for stenosis. To date, no randomized controlled multicenter study has been performed for X-STOP versus laminectomy. The X-STOP was clearly superior to conservative treatment in the U.S. study, but it does not permit a comprehensive comparison between the X-STOP and laminectomy. Hannibal compared patients from the U.S. FDA Pivotal X-STOP Trial (June 2000–July 2001) with those who received laminectomy during the same time period at the same institution, using the same criteria used in that trial in a nonrandom manner. At 4 years after surgery and with a 15 point improvement from baseline ODI score as a success criterion, 80% (12/15) of X-STOP patients and 38% (5/13) of laminectomy patients had successful outcomes.33 Compared to literature-reported outcomes of decompressive surgery, there are significant differences in operative time, estimated blood loss, length of hospital stay, complication rate, and reoperation rates, favoring the X-STOP.34,35 The results of the X-STOP patients showed 59.8% statistically significantly improved in Symptom Severity, 56.5% improved in
Physical Function, and 72.8% were satisfied. During the course of the U.S. study, 24 patients in the control group underwent decompressive laminectomy for the relief of their stenosis symptoms, and outcomes are available for 22 patients at a mean follow-up time of 12.8 months. Sixty-four percent had clinically significant improvement in Symptom Severity Domain of ZCQ, 68.2% had clinically significantly improvement in Physical Function Domain of ZCQ, and 59.1% were satisfied with the outcome of their treatment. Furthermore, Katz and associates36published a large series of surgically treated spinal stenosis patients also using the ZCQ outcomes tool at 2-year follow-up. In that study, 63% of the patients significantly improved in Symptom Severity, 59% improved in Physical Function, and 72% were satisfied. Fokter and Yerby looked at pre- and post-laminectomy ZCQ scores at 12 to 54 months in 58 patients, and 63.8% of the patients had significant clinical improvement in Symptom Severity, 55.2 had significant clinical improvement in Physical Function, and 58.6% of the patients were at least somewhat satisfied (Table 71–3).37 There is striking similarity in outcomes between the X-STOP and laminectomy groups. However, there are some important differences between these procedures. The mean operative time for the X-STOP was less than an hour for two levels, compared with 72 to 278 minutes reported for laminectomies. Mean blood loss of 40.1 to 58 mL during the X-STOP was negligible compared with 115 to 1040 mL reported for decompression.38 Moreover, the X-STOP procedure can be performed under local anesthesia, thus nearly eliminating the risk associated with general anesthesia. Finally, a comparison of the incidence and severity of complications cited in the laminectomy literature with the X-STOP indicates that the X-STOP is a much safer procedure. OPERATIVE TECHNIQUE The patient is placed in the right lateral decubitus position on a radiolucent table (Fig. 71–3). The level(s) to be treated is identified by fluoroscopy using an 18-guage needle taped to the skin. An indelible ink mark is made at the appropriate level. The site is prepped with usual sterile technique and draped using shower curtain type drape. Two 22-guage spinal needles can be placed at the caudal and cephalad ends of the proposed incision to accurately identify the level(s) and length of the incision. The spinal needles may be used to instill local anesthetic with epinephrine
TABLE 71–3. Laminectomy
X-STOP, 2-year follow-up Katz laminectomy, 2-year follow-up Laminectomy after failed conservative management Fokter laminectomy
ZCQ Scores Comparing X-STOP with Symptom Severity
Physical Function
Satisfaction
59.8%
56.5%
72.8%
63%
59%
72%
63.6%
68.2%
59.1%
63.8%
55.2%
58.6%
ZCQ, Zurich Claudication Questionnaire.
546
P A R T
V I I
Lumbar Posterior Dynamic Stabilization: Interspinous Based
A
F I G U R E 71–3. Surgical technique. A, Patients are placed in a right lateral decubitus position, and a midsagittal incision of approximately 4 cm is made over the spinous processes of the stenotic level(s). B, The small curved dilator sizing instrument is then inserted at the most anterior margin of the interspinous space. C, The universal wing is attached. n
to block the posterior rami bilaterally. A midsagittal incision about 4 cm is made over the spinous process to the dorsal fascia. A Cobb elevator is used to sweep the subcutaneous tissue from the dorsal fascia. After further local anesthesia to the dorsal fascia, two longitudinal incisions are made through both layers of the dorsal fascia about 1 cm from the lateral aspect of the spinous processes. The paraspinal musculature is then subperiosteally elevated from the spinous processes and medial lamina bilaterally using a large Cobb elevator. A large Cobb is appropriate to ensure that the canal is protected, especially in cases with prior laminotomy. The spinal canal should never be violated and neither laminectomy nor laminotomy is performed. Removal of any portion of the ligamentum flavum is unnecessary. If the facets are hypertrophied, they may block proper insertion of the device, causing them to be positioned too posterior; thus, they may be partially trimmed medially with a rongeur to ensure
adequate anterior placement. Fuchs and associates found in a cadaveric biomechanical study that the facets can be safely trimmed without destabilizing the motion segment while using the X-STOP. However, one should avoid aggressive bilateral facetectomies.39 Prior to starting the insertion process, the patient is asked to curl up and flex the back by trying to place the chin to the knees. A small curved dialator is inserted across the interspinous space at the most anterior margin of the interspinous space. After the correct level is verified by fluoroscopy, the small dilator is removed and replaced with a larger curved one. A finger is placed on the left side at the point where the small dilator is removed to ensure placement of the large dilator as well as the sizing distractor, which are placed at the same location. The interspinous ligament is dilated, not excised. After the larger dilator is removed, the sizing distractor is inserted. Because the patient is in the flexed, pain-free
CHAPTER 71
X-STOP Interspinous Process Decompression for Lumbar Spinal Stenosis
position, the sizing distractor should be opened until it contacts the spinous process and slightly distracts the interspinous space. If the interspinous space is sized between two available X-STOP sizes, choose the next smallest size. The X-STOP is then implanted from right to left, again with a finger on the left side to help guide the beveled tip of the device through the appropriate point. The right wing should be flush against the side of the spinous process. The screw hole for the universal wing on the left side is directly visualized, and the wing screw is engaged. The two wings are approximated medially, and the universal wing screw is secured using a torque-limiting screwdriver. Anteroposterior and lateral fluoroscopy views are obtained to ensure proper placement. The two fascia incisions are closed separately along with subcutaneous tissue and skin. A drain is rarely indicated. The procedure can typically be performed in less than an hour, and most patients can be released from the hospital within 24 hours. POSTOPERATIVE CARE Patients are encouraged to get up and ambulate as soon as they feel comfortable. They should avoid hyperextension activities for 2 to 6 months. COMPLICATIONS Reported complications related to the X-STOP have been minor and resolved easily without further sequelae. In the U.S. clinical study, there was one wound dehiscence, one seroma that was aspirated, one hematoma, and one report of incisional pain. No spinous process fractures occurred during X-STOP implantation. There have been no reports of either vascular or neurologic complications, an outcome which is anticipated because the laminae are left intact and the spinal canal and neuroforamina are not entered. Device-related complications included one patient who fell, causing the implant to dislodge; it was removed without any sequelae. A review of the patient's radiographs showed a very prominent facet that prevented the implant from being properly positioned anteriorly. One patient reported worsening pain about 1 year after the procedure, which was determined to be possibly related to the implant. One implant was placed too posterior and was considered to be malpositioned. An asymptomatic spinous process fracture was diagnosed in another patient on routine 6-month follow-up radiographs. Revising the implant is rather uncomplicated. Once the set screw on the wing is removed, the implant can be easily removed or replaced. Should an adjacent level need to be instrumented with the X-STOP, there would be little added difficulty. Placement of the X-STOP adjacent to a prior fusion remains a subject for further testing. COST ANALYSIS Kondrashov and associates recently evaluated the cost-effectiveness of X-STOP patients versus those treated with laminectomy. They found X-STOP to be significantly more cost-effective. There were 18 X-STOP and 11 laminectomy patients. Average hospital costs for one-level X-STOP and one-level laminectomy groups were
547
$17,059 and $45,302, respectively. Average hospital costs for two-level X-STOP and two-level laminectomy groups were $24,353 and $45,739, respectively. The main savings in the X-STOP group (cost drivers) were in operating room costs (shorter operative time), hospital charges (X-STOP is an outpatient procedure), and anesthesia charges (X-STOP is placed under local or MAC anesthesia). The cost of the X-STOP implant and higher radiology charges due to use of fluoroscopy during X-STOP placement were significantly offset by those savings.40
ADVANTAGES/DISADVANTAGES: X-STOP Advantages Clinically proved to be an effective treatment for symptoms of LSS with or without degenerative spondylolisthesis Safe Short surgery time, implanted under local anesthesia Minimally invasive An outpatient procedure Immediate relief of NIC Cost-effective Easily be revised to other procedures No violation of the spinal canal No tissue removal Disadvantages Cannot be used with lytic spondylolisthesis No published prospective controlled studies comparing X-STOP with laminectomy
CONCLUSION Decompression of the lumbar spine with the X-STOP offers a well-proved, safe, effective, and cost-effective treatment of patients suffering from NIC secondary to LSS. The X-STOP can be implanted using local anesthetic, and many patients can return home within hours after surgery. The X-STOP implant offers the benefits of decompression, yet with a low-risk profile, for NIC patients. It utilizes ligamentotaxis to indirectly increase the foraminal and canal dimensions by reconstituting tension in the posterior ligamentous structures. The comparative analyses suggest that the outcomes of the X-STOP decompression may at least be comparable to outcomes reported in the literature for decompressive laminectomy. The X-STOP interspinous process decompression is indicated for patients 50 years of age or older with one- or two-level mildto-moderate LSS symptoms and back and lower extremity complaints that are relieved in flexion. X-STOP outcomes have been demonstrated to be vastly superior to nonoperative therapy in the U.S. multicenter prospective randomized trial in LSS patients with mild-to-moderate symptoms. Complications with the X-STOP are relatively minor and uncommon. X-STOP also prevents the risks of pedicle screw placement and pseudarthrosis. Most importantly, being a motion-sparing device, X-STOP does not increase the adjacent segment stresses and probably does not contribute to adjacent segment degeneration and adjacent segment disease.
548
P A R T
V I I
Lumbar Posterior Dynamic Stabilization: Interspinous Based
REFERENCES 1. Senegas J: Mechanical supplementation by non-rigid fixation in degenerative intervertebral lumbar segments: The Wallis system. Eur Spine J 11(2):S164–S169, 2002. Epub 2002 Jun. 2. Postacchini F: Management of lumbar spinal stenosis. J Bone Joint Surg Br 78(1):154–164, 1996. 3. Hurri H, Slatis P, Soini J, et al: Lumbar spinal stenosis: Assessment of long-term outcome 12 years after operative and conservative treatment. J Spinal Disord 11(2):110–115, 1998. 4. Atlas SJ, Keller RB, Robson D, et al: Surgical and nonsurgical management of lumbar spinal stenosis: Four-year outcomes from the Maine lumbar spine study. Spine 25:556–562, 2000. 5. Johnsson KE, Uden A, Rosen I: The effect of decompression on the natural course of spinal stenosis: A comparison of surgically treated and untreated patients. Spine 16(6):615–619, 1991. 6. Riew KD, Yin Y, Gilula L, et al: The effect of nerve-root injections on the need for operative treatment of lumbar radicular pain: A prospective, randomized, controlled, double-blind study. Bone Joint Surg Am 82-A(11):1589–1593, 2000. 7. Turner JA, Ersek M, Herron L, Deyo R: Surgery for lumbar spinal stenosis: Attempted meta-analysis of the literature. Spine 17:1–8, 1992. 8. Atlas SJ, Keller RB, Robson D, et al: Surgical and nonsurgical management of lumbar spinal stenosis: Four-year outcomes from the Maine lumbar spine study. Spine 25:556–562, 2000. 9. Wang JC, Bohlman HH, Riew KD: Dural tears secondary to operations on the lumbar spine: Management and results after a two-yearminimum follow-up of eighty-eight patients. J Bone Joint Surg Am 80:1728–1732, 1998. 10. Turner JA, Ersek M, Herron L, Deyo R: Surgery for lumbar spinal stenosis: Attempted meta-analysis of the literature. Spine 17:1–8, 1992. 11. Yuan HA, Garfin SR, Dickman CA, Mardjetko SM: A historical cohort study of pedicle screw fixation in thoracic, lumbar, and sacral spinal fusions. Spine 19:2279S–2296S, 1994. 12. Lindsey DP, Swanson KE, Fuchs P, et al: The effects of an interspinous implant on the kinematics of the instrumented and adjacent levels in the lumbar spine. Spine 28(19):2192–2197, 2003. 13. Verbiest H: A radicular syndrome from developmental narrowing of the lumbar vertebral canal. J Bone Joint Surg 36B:230–237, 1954. 14. Porter RW: Spinal stenosis and neurogenic claudication. Spine 21:2046–2052, 1996. 15. Willen J, Danielson B, Gaulitz A, et al: Dynamic effects of the lumbar spinal canal: Axially loaded CT-myelopathy and MRI in patients with sciatica and/or neurogenic claudication. Spine 22:2968–2976, 1997. 16. Jonsson B, Annertz M, Sjoberg C, Strömqvist B: A prospective and consecutive study of surgically treated LSS, Part II: Five-year follow-up by an independent observer. Spine 22:2938–2944, 1997. 17. Richards JC, Majumdar S, Lindsey DP, et al: The treatment mechanism of an interspinous process implant for lumbar neurogenic intermittent claudication. Spine 30:744–749, 2005. 18. Lindsey DP, Swanson KE, Fuchs P, et al: The effects of an interspinous implant on the kinematics of the instrumented and adjacent levels in the lumbar spine. Spine 28(19):2192–2197, 2003. 19. Swanson KE, Lindsey DP, Hsu KY, et al: The effects of an interspinous implant on intervertebral disc pressures. Spine 28:26–32, 2003. 20. Rohlmann A, Zander T, Burra NK, Bergmann G: Effect of an interspinous implant on loads in the lumbar spine. Biomed Tech (Berl) 50(10):343–347, 2005. 21. Wiseman CM, Lindsey DP, Fredrick AD, Yerby SA: The effect of an interspinous process implant on facet loading during extension. Spine 30:903–907, 2005.
22. Wardlaw D, Smith F, Pope M, et al: Change in spinal canal and nerve root foraminal measurements before and six months following insertion of the X-STOP Interspinous Process Distraction Device in relation to early clinical outcome. Presented at Trans ISMISS Zurich, Switzerland, 2004. 23. Siddiqui M, Nicol M, Karadimas E, et al: The positional magnetic resonance imaging changes in the lumbar spine following insertion of a novel interspinous process distraction device. Spine 30(23):2677–2682, 2005. 24. Siddiqui M, Karadimas E, Nicol M, et al: Effects of X-STOP devices on sagittal spine kinematics in spinal stenosis. J Spinal Disord Tech 19(5):328–333, 2006. 25. Iida T, Abumi K, Kotani Y, et al: Effects of aging and spinal degeneration on mechanical properties of lumbar supraspinous and interspinous ligaments. Spine J 2(2):95–100, 2002. 26. Gillespie KA, Dickey JP: Biomechanical role of lumbar spine ligaments in flexion and extension: Determination using a parallel linkage robot and a porcine model. Spine 1;29(11):1208–1216, 2004. 27. Talwar V, Lindsey DP, Fredrick A, et al: Insertion loads of the X-STOP interspinous process distraction system designed to treat neurogenic intermittent claudication. Eur Spine J 15(6):908–912, 2006. Epub 2005 May 31. 28. Zucherman JF, Hsu KY, et al: A prospective randomized multicenter study for the treatment of lumbar spinal stenosis with the X-STOP interspinous implant: 1-year results. Eur Spine J 13 (1):22–31, 2004. 29. Wardlaw D, Smith F, Pope M, et al: Change in spinal canal and nerve root foraminal measurements before and six months following insertion of the X-STOP Interspinous Process Distraction Device in relation to early clinical outcome. Presented at Trans ISMISS Zurich, Switzerland, 2004. 30. Heijnen SAF, Kramer FJK: Spinale distractie als therapie bij lumbale wervelkanaalstenose—De eerste resultaten. Ned Tijdschr Orthop 11(4):199–203, 2004. 31. Anderson PA, Tribus CB, Kitchel SH: Treatment of neurogenic claudication by interspinous decompression: Application of the XSTOP device in patients with lumbar degenerative spondylolisthesis. J Neurosurg Spine 4(6):463–471, 2006. 32. Siddiqui M, Karadimas E, Nicol M, et al: Effects of X-STOP devices on sagittal spine kinematics in spinal stenosis. J Spinal Disord Tech 19(5):328–333, 2006. 33. Hannibal M: Interspinous process decompression with the XSTOP device for lumbar spinal stenosis: a 4-year follow-up study. Journal of Spinal Disorder Technology 19, 2006. 34. Benz RJ, Ibrahim ZG, Afshar P, Garfin SR: Predicting complications in elderly patients undergoing lumbar decompression. Clin Orthop 384:116–121, 2001. 35. Timothy J, Pal D, Ross S, Marks P: Early experience with the X-STOP a lumbar spinous process distractor for the treatment of lumbar canal stenosis. Abstract Br J Neurosurg 2005. (in press). 36. Katz JN, Stucki G, Lipson SJ, et al: Predictors of surgical outcome in degenerative lumbar spinal stenosis. Spine 24:2229–2233, 1999. 37. Fokter SK, Yerby SA: Patient-based outcomes for the operative treatment of degenerative lumbar spinal stenosis. Eur Spine J 21:1–9, 2005. 38. Reindl R, Steffen T, Cohen L, Aebi M: Elective lumbar spinal decompression in the elderly: Is it a high-risk operation? Can J Surg 46:43–46, 2003. 39. Fuchs PD, Lindsey DP, Hsu KY, et al: The use of an interspinous implant in conjunction with a graded facetectomy procedure. Spine 30(11):1266–1272; discussion 1273–1274, 2005. 40. Kondrashov DG, Hannibal M, Hsu KY, Zucherman JF: XSTOP versus decompression for neurogenic claudication: economic and clinical analysis. The Internet Journal of Minimally Invasive Spinal Technology 1(2), 2007.
CHAPTER
72
TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System Larry T. Khoo, Luiz Pimenta, and Roberto Dı´az
K E Y l
l
l
l
l
P O I N T S
The TOPS device is a total posterior spinal arthroplasty system for the treatment of moderate-to-severe lumbar spinal stenosis. The TOPS device replaces not only the facets but also the soft tissue as well as bony elements that are removed at the time of decompression. The TOPS device permits complete decompression of the central and lateral zones of the spine, thereby permitting excellent direct decompression of compressed neural elements. The TOPS device can be used in conjunction with an adjacent level fusion. Adequate bone density should be preoperatively evaluated as in any motion-sparing device.
The pathophysiologic mechanisms of low back pain continue to be poorly elucidated and to be difficult to effectively study. Whereas pain resulting from neurologic compression has been traditionally treated with great success via decompressive procedures, treatment of mechanical or discogenic lumbar pain has proved far more problematic. For many patients who remain refractory to conservative or less aggressive modalities, spinal fusion continues to be the mainstay of surgical treatment for the relief of axial back pain.1–7 Unfortunately, clinical outcomes have been variable and inconsistent with regard to the efficacy of spinal fusion in relieving lumbago as measured by standardized measures such as the Oswestry Disability Index (ODI), Visual Analog Scales (VAS) for pain, and SF-36.1 To compound the problem, accelerated degeneration of the adjacent segment has also been observed in biomechanical laboratory investigations, on long-term radiologic studies, and in numerous retrospective clinical surgical series.8–15 Although the exact incidence of this “adjacent segment disease” (ASD) remains poorly defined, it is clear that ASD is one of the most dreaded long-term clinical sequelae after successful fusion. From biomechanical investigations and clinical radiographic studies, it appears that there is an alteration of load sharing with an increase in mobility, shear, strain, and pressure at the invertebral disc, uncovertebral joints, and facet joints of the adjacent segment(s) after rigid spinal fusion.13,15 With this in mind, many
postulate that preservation of motion and load sharing at the index pathologic level would help to mitigate or reduce the overall incidence of ASD. RATIONALE FOR POSTERIOR FACET REPLACEMENT AND ARTHROPLASTY In the hope of decreasing adjacent segment forces, total anterior disc replacement (TDR) devices such as the CHARITÉ III (DePuy Spine, Raynham, MA) and the ProDisc II (Synthes, Paoli, PA) were developed in an attempt to preserve motion at the etiologic intervertebral disc. From the clinical federal Food and Drug Administration (FDA) Investigational Device Exemption (IDE) study, the CHARITÉ III was able to provide equivalent relief of low back pain as compared to the randomized control arm of anterior fusion.2,16 However, numerous authors have cited that severe facet arthropathy, spinal stenosis, neurogenic claudication, significant canal disease, spondylolisthesis, and translational instability are all relative or absolute contraindications to placement of an anterior total disc replacement.11,17 In a study by Huang and Camissa examining the typical makeup of patients seen in a tertiary spinal clinic, there was a majority preponderance of such patients with dorsal disease, spinal stenosis, spondylolisthesis, and spinal instability. These patients were ideally suited for classical spinal decompression and, in many cases, posterior spinal fusion and not candidates for TDR.9 As such, it is clear that motion-preserving devices that can be used for patients requiring dorsal surgical treatment are needed. When we examine the issue of posterior spinal disease and spinal stenosis, it is clear that we face not only the natural history of the disease process but also the iatrogenic instability that results from surgical decompression of these patients. As a large majority of these patients are symptomatic from radicular or central canal compression, they require decompression of the paramedian lamina and at least the medial third or medial half of the facet complex. As a result, progressive resection for neural decompression can lead to progressive spinal instability when the facet orientation 551
552
P A R T
V I I I
Lumbar Facet Replacement
is more sagittal than coronal.1 In many patients with spinal stenosis who require aggressive decompression for extensive neural foraminal narrowing, spinal fusion is often necessitated after facet resection.5,18 In Fischgrund's analysis, patients with spondylolisthesis and stenosis overall did better with regard to their low back pain scores when they had a primary fusion in addition to decompression as opposed to those who had decompression alone.6,7 However, it is also clear that many patients with stenosis or stenosis with spondylolisthesis do well without fusion and do not go on to have gross or glacial spinal instability after decompressive surgery. As such, a motion-preserving technology that can be placed via a standard posterior approach can help to avoid fusion in the many stenotic patients who are either preoperatively only mildly unstable or made unstable after surgical decompressive destabilization of the facet complex. With this in mind, questions remain as to the ideal nature of such a posterior motion-preserving stabilizer of the spine. Whereas numerous theories regarding the etiology of low back pain exist, perhaps the most developed of these is the concept of the biomechanical neutral zone as postulated by Panjabi.19,20 In this useful heuristic system, a motion segment functions to share load, move, and impinge within a given set of mechanical parameters. During biomechanical testing, any given spinal motion segment will thus move a given amount per quantum of applied load as determined by the viscoelastic properties of the surrounding structures that bind the two verterbrae such as the intervertebral nucleus and annulus, facet joint and capsule, interspinous ligaments, spinal longitudinal ligaments, attached paraspinal muscles, and truncal musculature. Plotted in any of three dimensions, this leads to a classical load-displacement plot of the “neutral zone” (Fig. 72–1). Degeneration, acute injury, or other pathology alters the biomechanical limiters of the system, leading to laxity, altered load sharing, and a widening of the load-displacement curves, thereby altering the neutral zone itself. As movement of the spinal segment begins to exceed its intial set points, joint, nociceptive, and stretch receptors begin to activate and signal pain and injury
in the area, which may lead to progressive pain and inflammation in that area. As such, induced injury of the ligaments, facets, or disc complex will lead to a widened neutral zone as seen on load-displacement curves during biomechanical cadaveric testing (Fig. 72–1). This model thus provides a useful point of reference for the cause of mechanical back pain in patients. Whereas decompression will help to relieve radicular pain, surgical restoration of proper load-sharing and normalization of the “neutral zone” may help to decrease pain and inflammatory stimuli in the treated spinal segment(s). Thus, the efficacy of rigid spinal fusion may ultimately result from its ability to radically correct the load-displacement characteristics of the motion segment to near-zero movement for any applied load after rigid fusion and instrumentation. With this in mind, a proper motion-preserving stabilizing device must also be able to correct the load-displacement curves back to an anatomically natural neutral zone and still preserves some degree of native spinal mobility above that of rigid fusion. Additionally, by preserving load-sharing of the treated segment, it must also decrease the “stress-riser” effect on adjacent untreated levels to potentially minimize the incidence of ASD. Finally, the motion-preserving dorsal device must also be secured to the spine in such a way that the device-bone interface remains stable over several million cycles. For example, devices that are to be implanted through the vertebral pedicles must thereby minimize the stress at the screw-bone interface to prevent screw pullout. The motion-preserving device must be able to not only provide motion but must also do so in a way that does not significantly load the screws.
DESCRIPTION OF THE DEVICE Design Parameters The TOPS system (Impliant Spine, Princeton, NJ) is a unitary device (Fig. 72–2) composed of a titanium “sandwich” with an interlocking polycarbonate urethane (PcU) articulating construct.
Intact Injured
Moment Spine segment
ROM (intact) ROM (injured)
Back view
Stiffness
Side view
NZ (intact)
Rotation
NZ (injured) n
F I G U R E 72–1. Mechanical properties of intact and injured spine. NZ, neutral zone; ROM, range
of motion.
CHAPTER 72
TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System
553
made of flexible materials and titanium, allows for constrained bending, straightening, and twisting movements in the affected segment after surgery. As such, the TOPS system serves to achieve the goals detailed earlier of restoring the physiologic neutral zone, maintaining a degree of spinal motion over rigid fusion, decreasing abnormal load-sharing at the adjacent levels, and minimizing screw-bone interface stresses, through the dampeners of the PcU elements contained therein. SCIENTIFIC TESTING RESULTS Finite Element Analysis As part of the overall development program on early designs of the TOPS device, a finite element analysis (FEA) was performed on the implant using ANSYS computational software. The original model developed for this theoretical stress analysis was a half section representation of the device. The model was chosen because the device itself and the loading conditions on it were found to be symmetric about the central plane. This hemi-model analysis allowed for faster computation without loss of precision. The results of this assessment show the principal stresses acting on the model as a result of the applied loading. The principal stress generated in the device during maximum anticipated loading is well below the yield stress for the titanium alloy from which it is fabricated21 (Fig. 72–4). Biomechanical in Vitro Motion Segment Analysis n
F I G U R E 72–2. Artificial facet joint capsule. Polyurethane
capsule enhances stability and internal bumpers.
The flexible PcU elements within the construct allow relative movement between the titanium plates so as to create axial rotation, lateral bending, extension, and flexion. The internal construct mechanically restricts motion to 1.5 degrees of axial rotation, 5 degrees of lateral bending, 2 degrees of extension, and 8 degrees of flexion. The implant also blocks excessive posterior and anterior sagittal translation. The TOPS system uses four standard hydroxyapatite-coated (HA-coated) polyaxial pedicle screws for fixation to the vertebrae (Fig. 72–3). As the internal configuration of the PcU bumpers ultimately acts as the limiter of motion, the TOPS device has an inherent dampening property that serves to dissipate energy that is passed through it during standard loadsharing of the moving spinal motion segment. Furthermore, as the PcU elements also have some “shock-absorption” properties in the vertical axis, vertical load transmitted through the cross-bars through the centroid of the device is somewhat dampened as well. These features serve not only to allow nearly full spinal motion but also to decrease stresses at the adjacent levels and at the screwbone interface. The TOPS system provides patients suffering from degeneration or hypertrophy of the facet joint, Grade I degenerative spondylolisthesis, and spinal stenosis with three major advantages. The surgeon can perform a wide decompression to eliminate the pain generators. The procedure stabilizes the posterior spine. The procedure allows a controlled range of motion. The implanted device,
The TOPS system was tested on six frozen cadaver specimens to (1) evaluate the capability of restoring motion to the intact spinal segment and (2) evaluate the effects on motion to the adjacent spinal segment after stabilization. The test showed that the TOPS system almost ideally restores the motion behavior of a segment in left and right lateral bending (Fig. 72–5A) and left and right axial rotation (Fig. 72–5B) after facet removal compared to the intact segment. In flexion and extension (Fig. 72–5C), the range of motion was 55% of that of an intact segment. By way of comparison, these results are significantly better than the Dynesys system (Zimmer Spine, Inc., Warsaw, IN).21 There was no effect shown on the adjacent segments. This finding, however, has to be interpreted carefully, as it might be due to the loading condition with pure moments. The intradiscal pressure data showed that the implant allows the disc to take part of the load, which is consistent with the natural biomechanics of the disc. However, the absolute values cannot be compared directly to the in vivo conditions because no preload could be simulated. Additionally, the hydrostatic pressure can be determined accurately only in a nondegenerated disc. Load on the Pedicle Screws To evaluate the efficacy of the polyaxial pedicle screws and the ability to fixate the TOPS device to the lumbar spinal vertebrae, a comparative test was performed. The examination was performed on a cadaveric spine with the same polyaxial screws. Strain gauges were applied to the same four screws so that the mechanical stress and resulting strains transferred to them from the leading
554
P A R T
V I I I
Lumbar Facet Replacement
n
F I G U R E 72–3. The TOPS system uses four standard hydroxyapatite-coated (HA-coated) polyaxial pedicle screws for fixation to the vertebrae.
competitor (Dynesys) and TOPS devices could be monitored while the spine simulator manipulated the spine segments (Fig. 72–6). Results of this study indicate that the load transmission to the pedicle screws is significantly less with the TOPS system than the leading competitor.21 As clinical results of the Dynesys device have indicated a 6% to 8% screw loosening rate at 2- to 3-year clinical follow-ups, it would be expected that the TOPS device will fare as well if not better after long-term clinical implantation.8,22
OPERATIVE TECHNIQUE The patient is positioned prone on the operative frame with either a four-poster or double-roll type configuration to simultaneously recreate the lumbar lordosis as well as to ensure that the abdomen is free and uncompressed. Preoperative biplanar fluoroscopic confirmation of the spinal alignment and the target level incision is obtained. Preinjection of the skin, fascia, and musculature with 0.25% lidocaine with 1:200,000 epinephrine is useful to decrease
CHAPTER 72
TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System
219.1 197.2 175.3 153.4 131.6 109.7 87.82 65.95 44.07 22.2 0.325
Stress von mises N(mm2) n
F I G U R E 72–4. The principal stress generated in the device during maximum anticipated loading is well below the yield stress for the titanium alloy from which it is fabricated.
postoperative pain as well as to decrease intraoperative bleeding. The surgical field is then prepped and draped in the usual sterile fashion. Using a No. 10 scalpel blade, a vertical incision is made down to the level of the lumbodorsal fascia. Subperiosteal dissection is then continued with the use of bovie cautery in combination with periosteal elevators to elevate the dorsal musculoligamentous complex off the spinous processes, laminae, and facets of the vertebrae above and below the target motion segment. As compared to classical spinal arthrodesis, in which exposure of the transverse processes is needed for future bone graft placement, exposure and retraction of the musculature for TOPS placement need only be carried to the lateral aspect of the facet complex. Additionally, particular care should be taken to preserve the capsule and muscular attachments surrounding the superior facet complex above the target level. The surgical exposure is then secured using a selfretaining retractor system. Standard decompression of the index level is then completed via laminectomy and facetectomy techniques. Classically, spinal stenosis occurs at the central canal, lateral recess, and lateral neural foramen from a combination of disc herniation, dorsal uncovertebral joint spurring and lipping, facet osteophytes, and facet subluxation. Depending on the exact pathology of the individual case, the
555
degree of bony, synovium, ligamentum flavum, and disc resection may thus vary accordingly (Fig. 72–7). With specific regard to the TOPS implant, there are three unique points that should be taken into account. First, it is recommended that aggressive resection of the intervertebral disc be avoided and that only herniated or bulging material be removed as needed to decompress the neural elements. Thermal annuloplasty with the bipolar forceps may be desired to stiffen and reinforce the annulus at the point of bulge or herniation. Due to the specific design of the TOPS implant, total or neartotal resection of the spinous process at the lower vertebrae of the index segment is needed to properly seat the four arms and centroid of the device. An implant trial is provided in the system and can be used to estimate the degree of laminar and spinous process removal that will be required (Fig. 72–8). Finally, as the TOPS implant serves to functionally replace the motion restraint of the native facet complex, it is recommended that the facet joint be effectively decoupled. This can be achieved by aggressive resection through the joint itself or by removing the superior articulating processes from the inferior vertebrae. This maneuver is often readily achieved by creating two surgical osteotomies at the level of the pars interarticularis with either an osteotome or powered drill-bit. Once the necessary degree of bony resection has been achieved, adequate neural decompression is confirmed with a Woodson elevator bilaterally over the thecal sac, exiting and traversing nerve roots. Meticulous hemostasis should be obtained at this point with a combination of bipolar cautery, bone wax, Gelfoam with thrombin, and Surgicel as needed. Some of the authors also advocate placement of a collagen-barrier type material (e.g., Duragen, Helostat) above the exposed neural elements to decrease the incidence of scarring and to create a readily accessible surgical plane in case of revision or implant removal. The pedicle screw entry points are then identified at the superior and inferior vertebral levels. Particular attention should be paid to obtain a more lateral-to-medial vector of pedicle cannulation such that more triangulation of the final screws can be achieved. Careful preservation of the superior facet complex is again desired. A cannulated system is provided with the TOPS implant for use via a semipercutaneous technique if desired. In this fashion, a Jamshidi needle can be used to cannulate the pedicles, and then is exchanged for a Kirshner wire once confirmation of the needle trajectory has been obtained on biplanar fluoroscopic guidance. A unique pendulum type guide is provided in the TOPS system that will ensure that the angle of pedicle cannulation will remain within the acceptable range of angles that can be tolerated by the four-arm extensions of the implant (Fig. 72–9). Then, using the provided serial tissue dilator tubes, a working corridor is obtained through the lateral musculoligametous complex. In this fashion, excessive initial stripping and lateral exposure of the muscles can be minimized. A cannulated awl and tap can then be used to prepare the pedicle for instrumentation. It is recommended that a bicortical or near-bicortical triangulating passage be obtained as to massive the pull-out strength of the individual pedicle screw. Further, utilization of the largest diameter screw that can be accepted by the anatomy of the pedicle is recommended for similar reasons. Last, undertapping by 0.5 to 1 mm will also serve to increase the ultimate strength of the screw threads’ purchase.
556
P A R T
V I I I
Lumbar Facet Replacement Right
Left Intact TOPS w/o facets
NZ − ROM −
−10
−8
−6
−4
−2
NZ + ROM +
0
0
2
ROM − [⬚]
4
6
8
10
ROM + [⬚] Lateral bending (A)
Right
Left Intact TOPS w/o facets
NZ − ROM −
−10
−8
−6
−4
−2
0
0
ROM − [⬚]
NZ + ROM +
2
4
Lateral bending (B)
Right
−8
Intact TOPS w/o facets
−6
8
10
Left
NZ − ROM −
−10
6
ROM + [⬚]
−4
−2
0
NZ + ROM +
0
ROM − [⬚]
2
4
6
8
10
ROM + [⬚] Flexion-extension (C)
A, TOPS system almost ideally restores the motion behavior of a segment in left and right lateral bending. B, Left and right axial rotation after facet removal compared to the intact segment. C, In flexion and extension, the range of motion was 55% of that of an intact segment. NZ, neutral zone; ROM, range of motion.
n
F I G U R E 72–5.
Once the desired length of screw has been determined, the pedicles are then instrumented with the standard cannulated tulip-head polyaxial screws that are provided in the TOPS system. Snap-on slotted extension sleeves are available to facilitate final implantation of the TOPS device (Fig. 72–10). The pedicle screws can be placed with or without the sleeves attached according to the surgeon's preference. Whereas the top two screws should be advanced to the end of their passage as low as desired, it is recommended that the bottom two screws be kept slightly more prominent initially. Using a separate four-armed targeting jig (Fig. 72–11), the crossbars are first placed into the tulip heads of the superior screws. Then, using the adjustable sliding inferior arms of the targeting jig, the inferior screws can be advanced to the appropriate depth to ensure that it will readily accept the four-arm geometry of the TOPS implant. It is recommended by the authors that the screws not be “backed-up” if possible to again maximize their purchase and pull-out strength.
At this point, the TOPS device is prepared for implantation. Earlier in the process, the exact profile (regular or low-profile) has been determined using the sizing jigs already described. A small amount of sterile saline is injected through a small port in the gasketed portion of the centroid of the implant to serve as a nonhydraulic lubricant. The device is then loaded on a specialized claw-armed holder (Fig. 72–12). The four arms of the TOPS implant are then passed into the tulip heads of the polyaxial screws. If the screw-extension sleeves were utilized, these can help to facilitate the passage of the arms at this point. Once in place, each of the four crossbar arms are secured using a standard locking nut and then countertorqued to their final tightness (Fig. 72–13). Careful inspection of the implant should confirm that all crossbars are well seated with no evidence of crossthreading or inadequate surface area of the bars within the tulip-head channels. If present, the extension sleeves are then unclipped from the tulip heads to complete the TOPS device implantation
CHAPTER 72
TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System Dynesys flexion-extension test Cadaver 57401432; 4/8/04 Preoad 630 N
2.5
2.5
2.0
2.0
1.5
1.5 Moment (N*m)
Moment (N*m)
TOPS flexion-extension test Cadaver 57401432; 4/8/04 Preoad 630 N
1.0 0.5 0
1.0 0.5 0
−0.5
−0.5
−1.0
−1.0
−1.5
0
2.5
0
2.5
5.0
7.5
5.0
7.5
10.0
−1.5 0
2.5
5.0
7.5
10.0
Time (sec)
10.0
Time (sec)
Screw 3 Screw 4
Screw 1 Screw 2
Dynesys lateral bending test Cadaver 57401432; 4/8/04 Preoad 630 N
TOPS lateral bending test Cadaver 57401432; 4/8/04 Preoad 630 N 3.0
3.0
2.5
2.5
2.0
2.0 Moment (N*m)
Moment (N*m)
557
1.5 1.0 0.5
1.5 1.0 0.5
0
0
−0.5
−0.5 −1.0
−1.0 0
2
4
6
Time (sec) n
8
10
0
Screw 1 Screw 2
2
Screw 3 Screw 4
4
6
8
10
Time (sec)
F I G U R E 72–6. Spine simulator manipulating the spine segments.
procedure. Biplanar fluoroscopic confirmation of the device and screw positions should be obtained prior to final closure to ensure good alignment and angle of the overall segment and construct (Fig. 72–14). Copious antibiotic-impregnated irrigation of the entire wound and implants is completed. A Davol or Jackson-Pratt wound drain should be placed at this time if there are concerns regarding hemostasis or seroma accumulation in the potential surgical dead space. Interrupted O-Vicryl sutures are then used to appose and close the lumbodorsal fascia in a near-watertight fashion. Inverted 2–0 or 3–0 absorbable sutures are then used to appose the subcutaneous layer of the skin. Skin closure can proceed according to the preference of the surgeon. Postoperatively, a rigid clamshell lumbosacral orthosis is generally not recommended due to the motion-preserving nature of the
TOPS implant. However, a lumbar corset or semirigid brace is often useful in the early perioperative period for support and the comfort of the patient. In comparison to spinal fusion patients, patients after TOPS implantation are encouraged to mobilize and ambulate early in their recovery process. CLINICAL OUTCOMES Patients and Methods A total of 29 patients with the diagnosis of moderate-to-severe lumbar spinal stenosis were enrolled in a prospective clinical trial using the TOPS system (Class II). The sites included in this study are São Paulo, Brazil; Istanbul, Turkey; Zreifin, Israel; and Antwerp, Belgium. The average age of the enrolled patients was 64.2 years (range, 52–72 years) (Table 72–1). Inclusion
558
P A R T
V I I I
Lumbar Facet Replacement
n
F I G U R E 72–7. Depending on the exact pathology of the individual case, the degree of bony, synovium, ligamentum flavum, and disc resection may thus vary accordingly.
n
F I G U R E 72–8. An implant trial used to estimate the degree of laminar and spinous process removal that will be required.
CHAPTER 72
TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System
559
n F I G U R E 72–9. The pendulum guide provided in the TOPS system will ensure that the angle of pedicle cannulation will remain within the acceptable range of angles that can be tolerated by the four-arm extensions of the implant.
criteria for this population included patients requiring single-level spinal decompression and fusion surgery between L2 and L5. Radiographic analysis using computed tomography (CT), magnetic resonance imaging (MRI), myelography, or plain x-ray was performed to diagnose and better characterize the exact location of lumbar spinal stenosis. Specifically, these studies were used to confirm the presence of any of the following: thecal sac or cauda equina compression, nerve root impingement by either osseous or nonosseous elements, or hypertrophic facets with canal encroachment. All patients displayed both low back and sciatic pain, with or without claudication. Secondary inclusion criteria for this study included, but did not require, degenerative spondylolisthesis up to Grade 2, advanced facet arthrosis, and radiologic segmental instability (based on dynamic flexion-extension radiographic analysis). Standard anteroposterior, lateral, flexion, and extension radiographs were obtained preoperatively and at 3 weeks, 6 weeks, and 6 months. Sagittal angulation was measured on the lateral radiographs from L1 to S1 using the Cobb method. Flexion, extension, and total range of motion were measured. The radiographs were examined by two independent observers for evidence of hardware failure, loosening, or signs of spinal instability. Clinical outcomes were assessed using Visual Analog Score (VAS), Oswestry Disability Index (ODI), and Zurich Claudication
Questionnaire (ZCQ) questionnaires administered preoperatively and at 6 weeks, 3 months, 6 months, and 1 year following surgery. Results To date, the patients enrolled are at various stages of follow-up ranging from 6 weeks to 1 year (Table 72–2). Of the 29 patients enrolled, 28 study patients were treated at L4-L5 and one patient was treated at L3-L4. Fifteen of the 29 patients had degenerative spondylolisthesis of Grade 1 or 2 (see Table 72–1). The mean surgical duration was 3.1 hours, with a standard deviation of 0.89 hour. The mean blood loss was approximately 200 mL. The average preoperative ODI score was 57%. The ODI scores were 20% and 16% at the 6-month and 1-year time points, respectively (Fig. 72–15). The mean VAS leg score was 88 at baseline, 21 at 6 weeks, 19 at 3 months, 18 at 6 months, and 12 at 1 year (Fig. 72–16). The mean ZCQ score was 57% initially and 26% at 1 year (Fig. 72–17). The radiographic findings were reviewed by an independent panel of radiographic specialists at 6 weeks, 3 months, 6 months, and 1 year after the procedure. These studies revealed that all the motion segments instrumented with the TOPS device were stable and that no device migrations or malfunctions occurred. The preoperative disc heights at the treatment level and adjacent
560
P A R T
n
V I I I
Lumbar Facet Replacement
F I G U R E 72–10.
Snap-on slotted extension sleeves facilitate final implantation of the
TOPS device.
levels were measured and recorded using plain x-rays and confirmed with sagittal CT scans. This was reevaluated at 3 months and 1 year and no subsequent disc height loss was observed. The degree of spondylolisthesis was recorded at all time points and no cases of slip progression were observed. The screw-bone interface was analyzed with thin-slice CT scans. In this independent analysis, none of the 29 patients exhibited any signs of screw loosening at any time point. Global spinal motion was also evaluated using flexion-extension films taken at 3 months and 12 months (Fig. 72–18). This evaluation confirmed that global motion was preserved in all patients (n ¼ 11) through the 1-year time point. Safety analysis revealed no device-related adverse events (AEs) during this study. Non-device–related AEs included three dural tears, four postoperative seromas, and one neurologic deficit. The
non-device–related AEs occurred during the normal course of decompression and were not due to implant insertion. None of the dural tears or seromas developed long-term sequelae. None of the non-device–related AEs required additional surgery. During the study, one patient developed a transient neurologic deficit 1 day following surgery. CT evaluation showed that the position of the TOPS device and pedicle screws were normal. However, it also revealed the presence of a postoperative hematoma that was causing central canal compression. Although the patient's preoperative blood coagulation parameters were normal, the patient reported similar postoperative bleeding complications following previous knee surgery, with a late hemorrhage that required surgical intervention. The patient has been treated conservatively and demonstrated rapid improvement in neurologic function.
CHAPTER 72
TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System
n
561
F I G U R E 72–11. Four-armed targeting jig.
CONCLUSION
n
F I G U R E 72–12.
Claw-armed holder.
The TOPS total facet replacement system is a novel posterior stabilizing device able to preserve motion, restore the biomechanical neutral zone, and prevent abnormal load-sharing at the treated and adjacent spinal motion segments. Within the limited clinical study, it appears that decompression combined with implantation of the TOPS device is able to achieve functional outcomes as favorable as historical control arms for fusion for stenosis and spondylolisthesis. ODI and VAS scores indicate that the device is effective in treating the back pain component of these patients as well as rigid spinal fusion in prior studies. Although the device itself has been tested for over 10 million cycles with no functional failures and minimal wear debris, questions concerning screw purchase and pullout will require far more extensive follow-up and study. As the device appears to have less screw-bone interface stresses than other predicate devices, it would be reasonable to expect a failure rate less than or equivalent to previously published predicate screw pullout rates of 6% to 8%.8, 22 Additional long-term studies will also be required to examine the effect this device has on the incidence of ASD as well. Overall, the TOPS device represents a novel means of treating back pain while still preserving the native motion of the pathologic spinal segment.
n F I G U R E 72–13. Once in place, each of the four cross-bar arms are secured using a standard locking nut and countertorqued to their final tightness.
A
B
n F I G U R E 72–14. Biplanar fluoroscopic confirmation of the device and screw positions should be obtained prior to final closure to ensure good alignment and angle of the overall segment and construct.
CHAPTER 72
TABLE 72–1.
Patient Demographics in TOPS Clinical Trial
Category
9 VAS (Leg) 8
Data
Gender Male Female Age Mean (SD) Min, max Body mass index Mean (SD) Min, max Spondylolisthesis Grade 0 Grade I Grade II
563
TOPS: Total Posterior Facet Replacement and Dynamic Motion Segment Stabilization System
7
12 (41.4%) 17 (58.6%)
6
64.2 (6.10) 52.0, 72.0
5
28.8 (4.94) 18.1, 39.2
4
48.3% 44.8% 6.9%
3 2 1 0 Preop (n = 24)
6 wks (n = 22)
3 mo (n = 22)
6 mo (n = 18)
12 mo (n = 10)
n F I G U R E 72–16. Visual Analog Score (VAS) for the leg at baseline (preop) and at 6 weeks, 3 months, 6 months, and 1 year after surgery.
TABLE 72–2.
Patient Follow-Up Distribution per Study Site Baseline
Theoretical Expected Actual Follow-up (%)
29 29 29
Patient Follow-Up Distribution 6 wks
3 mo
6 mo
12 mo
29 29 25 86
29 28 18 79
25 20 18 90
17 12 11 92
60
60
Oswestry
Zurich (Total) 50
40
40 Percent
50
30
30
20
20
10
10
0
0 Preop (n = 29)
6 wks (n = 25)
3 mo (n = 22)
6 mo (n = 18)
12 mo (n = 11)
n F I G U R E 72–15. Oswestry Disability Index (ODI) scores before surgery (preop) and at 6 weeks, 3 months, 6 months, and 1 year.
Preop (n = 26)
6 wks (n = 19)
3 mo (n = 21)
6 mo (n = 19)
12 mo (n = 10)
n F I G U R E 72–17. Mean Zurich Claudication Questionnaire (ZCQ) score before and after surgery up to 1 year.
564
P A R T
V I I I
Lumbar Facet Replacement
7⬚ Extension n
1.5⬚ Flexion
F I G U R E 72–18. Flexion-extension films taken at 3 months and 12 months.
REFERENCES 1. Abumi K, Panjabi MM, Kramer HM, et al: Biomechanical evaluation of lumbar spinal stability after graded facetectomies. Spine 15 (11):1142–1147, 1990. 2. Blumenthal S, McAfee PC, Guyer RD, et al: A prospective, randomized, multicenter Food and Drug Administration investigational device exemptions study of lumbar total disc replacement with the CHARITE artificial disc versus lumbar fusion, Part I: Evaluation of clinical outcomes. Spine 30(14):1565–1575; discussion E387– E391, 2005. 3. Resnick DK, Chourdhri TF, Dailey AT, et al: Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine, Part 5: Correlation between radiographic and functional outcome. J Neurosurg Spine 2(6):658–661, 2005. 4. Resnick DK, Chourdhri TF, Dailey AT, et al: Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine, Part 7: Intractable low-back pain without stenosis or spondylolisthesis. J Neurosurg Spine 2(6):670–672, 2005. 5. Resnick DK, Chourdhri TF, Dailey AT, et al: Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine, Part 9: Fusion for patients with stenosis and associated spondylolisthesis. J Neurosurg Spine 2(6):679–685, 2005. 6. Resnick DK, Chourdhri TF, Dailey AT, et al: Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine, Part 10: Fusion following decompression in patients with stenosis without spondylolisthesis. J Neurosurg Spine 2(6):686–691, 2005. 7. Fischgrund JS, Mackay M, Herkowitz HN, et al: 1997 Volvo Award winner in clinical studies: Degenerative lumbar spondylolisthesis with spinal stenosis: A prospective, randomized study comparing decompressive laminectomy and arthrodesis with and without spinal instrumentation. Spine 22:2807–2812, 1997. 8. Grob D, Benini A, Junge A, Mannion AF: Clinical experience with the Dynesys Semirigid Fixation System for the lumbar spine: Surgical outcome and patient oriented outcome in 50 cases after an average of 2 years. Spine 30(3):324–331, 2005.
.
9. Huang J, Girardi F, Camissa F: The prevalence of contraindications to total disc replacement in a cohort of lumbar surgical patients. Spine 29(22):2538–2541, 2004. 10. Hunter LY, Braunstein EM, Bailey RW: Radiographic changes following anterior cervical spine fusions. Spine 5:399–401, 1980. 11. Kostuik JP: Complications and surgical revision for failed disc arthroplasty. Spine J 4:289S–291S, 2004. 12. Lee CK: Accelerated degeneration of the segment adjacent to a fusion. Spine 13:375–377, 1988. 13. Lee CK, Langrana NA: Lumbosacral spinal fusion: A biomechanical study. Spine 9:574–581, 1984. 14. Lu WW, Luk KD, Holmes A: Pure shear properties of lumbar spinal joints and the effect of tissue sectioning on load-sharing. Spine 30 (8):E204–E209, 2005. 15. Olsewski JM, Garvey TA, Schendel MJ: Biomechanical analysis of facet and graft loading in a Smith-Robinson type cervical spine model. Spine 19:2540–2544, 1994. 16. Geisler FH: Surgical technique of lumbar artificial disc replacement with the Charite Artificial Disc. Neurosurgery 56(1):46–57; discussion 46–57, 2005. 17. Gamradt SC, Wang JC: Lumbar disc arthroplasty. Spine J 5:95–103, 2005. 18. Yone K, Sahou T, Kawauchi Y, et al: Indication of fusion for lumbar spinal stenosis in elderly patients and its significance. Spine 21:242–248, 1996. 19. Panjabi MM: Clinical spinal instability and low back pain. J Electromyogr Kinesiol 13(4):371–379, 2003. 20. Panjabi MM: The stabilizing system of the spine, Part II: Neutral zone and instability hypothesis. J Spinal Disord 5(4):390–396; discussion 397, 1992. 21. McAfee P: Biomechanics and results of implant testing of the TOPS Facet Replacement Device (Luncheon Symposium presentation— TOPS Device). Presented at the 5th Annual Spine Arthroplasty Society Meeting, New York, New York, May 9, 2005. 22. Stoll TM, Dubois G, Schwarzenbach O: The dynamic neutralization system for the spine: A multi-center study of a novel non-fusion system. Eur Spine J 11(2):170–178, 2002.
CHAPTER
73
Total Facet Arthroplasty System (TFAS) Scott A. Webb and Gordon Neil Holen
K E Y l
l
l
P O I N T S
Total Facet Arthroplasty System (TFAS) is a nonfusion spinal implant developed for the treatment of patients with moderate-to-severe lumbar spinal stenosis. TFAS allows for the replacement of the diseased facets and laminae and is an alternative to traditional spinal fusion. The potential benefits for posterior motion devices include stabilization with controlled kinematics, excision of pain generators, allowance of a more complete lateral recess decompression compared to standard laminectomy, and avoidance of increasing loads at and accelerating degeneration of adjacent levels.
Total Facet Arthroplasty System (TFAS) (Archus Orthopedics Inc., Redmond, WA) is a nonfusion spinal implant developed for the treatment of patients with moderate-to-severe lumbar spinal stenosis. TFAS allows for the replacement of the diseased facets and laminae and is an alternative to traditional spinal fusion. This procedure allows for a wide decompression and maintains, and often restores, intervertebral motion, stability, and sagittal balance in the replaced spinal segment. In contrast to spinal fusion, TFAS also eliminates the need for autologous bone grafting and its well-known associated comorbidities. Currently, TFAS is limited to investigational use only within the U.S. The current clinical trial is a multicenter prospective randomized controlled study comparing the safety and effectiveness of TFAS to spinal fusion for patients with moderate-to-severe lumbar spinal stenosis. A great deal of time and interest has been focused on the development of motion-preserving technology and artificial disc replacement, with little attention paid to the posterior structures. Huang and associates showed that the majority of contraindications for total disc replacement candidates (89%) involve stenosis and facet arthritis.1 The potential benefits for posterior motion devices include stabilization with controlled kinematics, excision of pain generators, allowance of a more complete lateral recess decompression compared to standard laminectomy, and avoidance of increasing loads at and accelerating degeneration of adjacent levels.
BIOMECHANICS The functional spine unit (FSU) comprises the five-joint complex of the intervertebral disc, corresponding facet joints, and ligaments. The FSU has three main roles: (1) to stabilize the vertebrae in relation to each other and protect the nerves through a range of motion and under various loads; (2) to provide proper motion of each segment and of the spine as a whole; and (3) to transfer and share loads within the spinal column. The balance of these three biomechanical elements can be disrupted by disease as well as surgical procedures. The primary role of the facet joints is to both allow and limit motion, thus acting as spinal segment stabilizers. Facet joints also act to protect the lumbar disc from excessive stress. They are complex synovial joints which bear loads in both compression and shear (Fig. 73–1A and B). Facet joint load depends on location and can be increased by degeneration of the intervertebral disc. Facet joint capsules are innervated by Type I, II, and III mechanoreceptors. It is thought that their proprioceptive role can be compensated for by the balance of non-removed capsules in the spine. Work done by Jiang and associates in 1995 showed that the supraspinous and interspinous ligaments were innervated by bundles of Type III receptors suggesting that spinal ligaments are indeed active in monitoring the loading of spinal joints.2 This provides at least static positional awareness for postural control. This offers additional support to the concept that ligaments act as part of the neurologic feedback mechanism for protection and stability of the spine. Thus, the neural input from the facets may indeed be important for proprioception and may modulate muscular action that could control joint stability. Damage to and pain from the facet joints and corresponding capsular ligaments can be independent of surgical intervention and result from trauma, disease, and degeneration. Extensive work has been published on facet joints as pain generators. Ghormley first described this as Facet syndrome in 1933.3 The facet synovial linings and capsules are highly innervated and have free nerve endings within their tissues. Cartilage breakdown can lead to progressive joint
565
566
P A R T
V I I I
Lumbar Facet Replacement
Fibrous capsule
Articular cartilage
Synovial membrane Synovial membrane
A
Load Joint cavity
Articular cartilage
Meniscus Calcified cartilage
Synovial fluid Cancellous bone and marrow
Load
Subchondral cortex
B F I G U R E 7 3 – 1. A, Schematic drawing of an apophyseal joint. (Adapted from Adams MA, Hutton WC: The mechanical function of the lumbar apophyseal joints. Spine 8(3):327–330, 1983.) B, Generic drawing of a diarthrodial (synovial) joint. (Adapted from Mow VC, Hayes WC (eds): Basic Orthopaedic Biomechanics, 2nd ed. Philadelphia, Lippincott-Raven, 1997.)
n
degeneration, inflammation, spur formation, and possible synovial cyst formation. Cyst formation may also lead to stenosis and radiculopathy. A grading system was proposed for facet degeneration by Fujiwara4 (Figs. 73–2 and 73–3). The function of the facet joints is determined by their shape, size, orientation, and location (Fig. 73–4). In the cervical spine the facets are located laterally and are tilted in abduction in the coronal plane. This allows significant freedom in lateral bending, extension, and axial torsion. Cervical facets are subjected to the lowest effective transmitted loads in the spine. Lumbar facets, in contrast, are larger, more centrally located, and oriented in a more sagittal or adducted position. This orientation allows for the main motion of the lumbar segments, which is flexion and extension. Axial rotation and lateral flexion are limited. The joints act as cam-like stops, preventing hyperextension and axial torsion. Lumbar facets are also subjected to the highest magnitude load in the spine.
degenerative diseases of the facets, central or lateral recess stenosis with claudication, and in cases with Grade I degenerative spondylolisthesis. It is currently allowed for single-level implantation at L3-L4 and L4-L5. A U.S. Investigational Device Exemption (IDE) study allows for decompression of up to three levels but limits implantation of the device to a single level. The first TFAS within the U. S. was performed on August 26, 2005, under the FDA IDE study, and TFAS is not yet available outside study centers within the U. S. The European Commission, however, approved the general use of TFAS within the European Union on March 2, 2005. Contraindications in the current U.S. IDE study include a dual-emission x-ray absorptiometry (DEXA) scan with a T score below 2.5, immunosuppressive disorder, metal implant sensitivities, BMI greater than 35, scoliosis with a Cobb angle greater than 25 degrees, and endocrine or metabolic disorders that may affect osteogenesis.
INDICATIONS AND CONTRAINDICATIONS
DESCRIPTION OF DEVICE
TFAS is currently indicated for use as an adjunct to decompressive procedures in place of traditional fusion. It can be used in
The Total Facet Arthroplasty System is a modular implant with a metal-on-metal articulation. The present device is fixated to the
CHAPTER 73
Total Facet Arthroplasty System (TFAS)
Grade I
Grade II
Grade III
Grade IV
567
n
F I G U R E 73–2. Facet cartilage degeneration MRI scale proposed by Grogan. Grade I: Uniformly thick cartilage covers the articular surfaces completely. Grade II: Cartilage covers the entire surface of the articular processes but with erosion of the irregular region evident. Grade III: Cartilage incompletely covers the articular surfaces, with regions of the underlying bone exposed to the joint. Grade IV: Cartilage is absent except for traces on the articular surfaces. (From Fujiwara A, Lim T, An H, et al: The effect of disc degeneration and facet joint arthritis on the segmental flexibility of the lumbar spine. Spine 25(23):3036–3044, 2000.)
pedicles with standard polymethyl-methacrylate (PMMA) cement, although a cementless implant is under development. It comprises two cephalad bearings and two caudal housings or cups, which articulate via a cross-arm assembly (Figs. 73–5 and 73–6). TFAS allows flexion of 13 degrees and extension of 2 degrees. Lateral bending is allowed up to 7.5 degrees and axial rotation up to 2 degrees. The instant axis of rotation is in the posterior third of the vertebral body. Kinematic testing has demonstrated that the TFAS implant restores motion of an unstable FSU to that of an intact FSU. Motion was stabilized in flexion and lateral bending and restored in extension. Motion was limited in axial rotation.
In vivo loading has been performed comparing TFAS to a pedicle screw fusion system. With similar loads applied to both systems, the strains in the actual metal surfaces are lower than the rigidly connected pedicle screw system because of the motion capability of the system. This implies that TFAS will transmit lower loads to the implant-bone interface. Strength testing demonstrates that the construct strength, utilizing cement, is two to three times greater than maximum static in vivo loads that the device will experience (Fig. 73–7). The strength of the fixation in vertebral bone is two times stronger than that of pedicle screws. The integrity of the construct and interconnection mechanisms were tested in vitro to 10 million
n F I G U R E 73–3. Examples of lumbar vertebrae from highly degenerative (left) to normal (right) patients contrasting the large variability in geometry which affects load-bearing and motion.
568
P A R T
V I I I
Lumbar Facet Replacement
Z L3L4_L3Ca M-L
X Y Ce-Ca
L3L4_L3Ce
Alpha
L5S1_FAT2 L5S1_FAT1
n F I G U R E 7 3 – 4. The three-dimensional morphology of the facets is complex but can be well characterized through computed anthropometric analysis of CT-scan reconstructions. Accurate characterization of facet morphology is necessary because curvature, pose, and location of vertebral facet joints play an important role in their load-bearing characteristics.
cycles and there were no implant connection failures or dissociations of the device and PMMA under maximum static loading. The strength of the system is key, and TFAS sets the standard for bone-implant interface strength. Implant wear was assessed through 10 million cycles of testing in a custom-designed wear simulator that allows variable and cross-coupled motion and load application to replicate the activity of daily living conditions of the intact spine. The structural members of the implant are titanium-aluminum-vanadium alloy (TI6Al4V) while the articulating surfaces are cobalt-chromium (CoCr). The amount of debris generated is similar to that generated by other orthopaedic implant-bearing surfaces under physiologic testing conditions. Particle size and distribution also were comparable to current metal-on-metal hips. In animal studies there was no neurotoxic response noted locally or systemically. Load-sharing has been shown to be more physiologic with TFAS when compared to a pedicle fusion construct as demonstrated in tests using strain gauges. Testing was performed with strain gauges placed in the disc space. The disc loads were measured and the TFAS-treated disc segment acted similar to an intact FSU, while the pedicle screw system demonstrated significantly lower non-physiologic loads at the disc space. TFAS has been studied in complete laminectomy-facetectomy models. Studies have shown that TFAS is able to restore segmental stability to that of an intact functional spine unit (Fig. 73–8). Because TFAS is designed to be a total joint replacement, it not only restores the physiologic range or limits of motion but also the quality of motion of the operated segment to that of an intact spinal segment. Quality of motion can be defined as the ability of a device (e.g., posterior stabilizing or spinal arthroplasty) to replicate the characteristic kinematic signature of the intact spine in both its limits as well as profile. The highly nonlinear nature of lumbar spine kinematics, especially in flexion-extension and lateral bending, is a result of the interaction of all the multiple nonlinear subsystems of the spine, such as ligaments, discs, nucleus, muscles, and cartilaginous facet contact. Therefore, the TFAS implants have been designed to physiologically reproduce the biomechanics of the spine after implantation. A design objective of reproducing only the limits of motion rather than the whole motion path during activities of
Archus bone cement Integral cone tip end
Cephalad stem with Tecotex® surface
Archus bone cement Clamp housing with hexatube set screw
Caudal stem (5.75 & 8.5 mm) n
F I G U R E 7 3 – 5.
Caudal bearing
Cross-arms with cephalad bearings
The TFAS construct in relation to the vertebral bodies with individual component labels.
CHAPTER 73
Total Facet Arthroplasty System (TFAS)
569
n
F I G U R E 73–6. Schematic of the TFAS components that are assembled in order to custom-fit the highly variable patient anatomy.
Caudal bearing and stem
Clamp housing with set screw
daily living could result in a non-physiologic loading of the tissues and implants and lead to morphologic changes in the tissues due to disuse or overload during the duty cycle of the device after implantation. Other devices that are meant to dynamically stabilize and preserve motion rather than restore physiologic motion may only grossly reproduce the kinematic behavior of the lumbar spine and not truly reproduce the physiologic motion profile of the spine but rather allow only for similar motion ranges or limits (Fig. 73–9).
Cross-bar with cephalad bearings
Cephalad stem
CLINICAL PRESENTATION Standard preoperative planning, as with any surgical spine procedure, begins with a systematic history and physical examination. Radiographs should include a minimum of anteroposterior, lateral, and flexion-extension radiographs. Magnetic resonance imaging (MRI) or myelography is performed to document underlying stenosis. An attempt should be made to verify that the diameter of the involved pedicles will allow creation of a 6.5-mm diameter drill channel. Otherwise, no special testing is required preoperatively for total facet arthroplasty.
SCIENTIFIC TESTING RESULTS Our initial experience with TFAS was in Romania. We prospectively evaluated 13 consecutive patients who underwent total facet arthroplasty. Our first operation was on May 13, 2005. Our initial series involved eight male and five female patients with an average age of 60.7 years and average BMI of 28.9. Four operations were at L3-L4 and nine at L4-L5. Results were quantified using the Zurich Claudication Questionnaire (ZCQ) for both symptom and function as well as Visual Analog Scale (VAS) measurements. The mean percentage of improvement in ZCQ Symptom Raw Score was 50.2%, ZCQ Function Raw Score was 46%, VAS Leg improvement was 78.2%, and VAS Back was 63.7%.
n
F I G U R E 73–7.
constructs.
OPERATIVE TECHNIQUES Positioning The patient is positioned on a Jackson table in the prone position after general endotracheal intubation. This position allows the chest to be supported, legs extended, and the abdomen free from external compression and will place the patient in good lordotic alignment and allow C-arm visualization. Lordotic positioning is important so that when the implant is inserted, the cephalad bearing is bottomed out on the caudal bearing. This will allow full range of motion postoperatively. If the patient is positioned in a nonlordotic or flexed position and the implant is bottomed
Examples of test apparatus utilized for static and fatigue strength testing of the TFAS
P A R T
570
V I I I
Facet Removal
12
Intact Laminectomy TFAS
Flexion-extension angle (deg)
Lumbar Facet Replacement
8 4
-12
-8
0
-4
4
8
12
-4 -8
The articulating facets for the desired instrumented level are excised. Because TFAS restores stability to the operative level, a full laminectomy may be performed. The facet capsules at adjacent levels should be preserved. The inferior articulating surfaces should be resected at their intersection with the pedicle. A portion of the facet may be left to help in identifying the appropriate entry point of the pedicle stem. This portion of the facet is then subsequently removed. The superior articulating surface should be resected to the level of the lamina. In cases of wide decompressive laminectomy, the entire spinous process and lamina will be removed.
-12
Pedicle Preparation
Applied moment (Nm) n
F I G U R E 73–8. Kinematic signature in flexion-extension from a spine simulator of a cadaveric spine tested intact, surgically destabilized, and then implanted with TFAS. Note that TFAS restored quality of motion to an otherwise unstable FSU (no facets or posterior ligaments) by completely re-establishing the characteristic kinematic signature of the intact spine in both its limits as well as profile. This was accomplished for motion in all planes.
out when implanted, extension and potentially flexion may be restricted by the implant. Exposure After verification of the appropriate level, the spine is exposed in standard fashion. Decompression is performed and associated bone is removed as needed. Care should be taken not to destabilize levels undergoing decompression adjacent to the TFAS level.
Intact or TFAS Stabilizing device
12
The entry point of each involved pedicle is used by identifying the intersection of the midline of the transverse process and the lateral border of the superior facet. A burr or awl is utilized to access the pedicle. Although the pedicle dictates the angulation of the probe, the ideal trajectory is 20 degrees lateral from the sagittal plane. The 4.0 drill bit is then used to create a channel to within 10 mm from the anterior cortex of the vertebral body or the 55 mm marking is reached on the drill bit. Utilize fluoroscopy to verify that the drill bit is centered in the pedicle. Measure the distance from the end of the channel to the exit point of the pedicle to confirm appropriate drill depth and corresponding stem size. Repeat the steps above for the remaining pedicles. Note and record the depth measurement for all four holes (left and right caudal, left and right cephalad). These lengths will be needed in order to select the appropriate stem implants later in the surgical technique. A pedicle probe should be used to verify that there is no breach in the pedicle. If a significant breach is detected, the TFAS should not be implanted in order to avoid cement extravasation and a standard pedicle screw fusion system should be implanted. Smaller breaches can be managed by the insertion of bone graft prior to cement insertion under fluoroscopic visualization.
8
-10
-5
4
0
5
Same ROM
Applied moment (Nm)
Caudal Stem Selection Different quality of motion
-4 -8 -12
Flexion-extension angle (deg) n
F I G U R E 73–9.
Utilizing range of motion (ROM) can be deceiving when evaluating the ability of a motion-restoring spinal implant, because two devices having identical ranges of motion can have drastically different kinematic signatures and thus may not be attaining total proper motion quality.
After obtaining an accurate anteroposterior (AP) C-arm image, the appropriate caudal selector (left or right) is placed into the 4-mm caudal hole. The trial does not need to be fully seated in order for the correct measurement to be obtained, but it should be stable (Fig. 73–10). The lateral edge of the trial should be visually aligned with the spinous processes. An AP C-arm image is taken of the target level with the trial in place. The image displayed on the C-arm screen will show three indicator lines and a ball. The surgeon should determine which line is closest to the ball. Intermediate markings are provided for guidance when the ball is in the middle. The line chosen provides the correct angulation for the caudal stem. When combined with the previously measured length of the pedicle hole, the caudal stem may be selected for both the left and right sides (Fig. 73–11).
CHAPTER 73
n
Total Facet Arthroplasty System (TFAS)
571
F I G U R E 73–10. Caudal alignment trials on the back table (left) and in situ (right) during surgery.
Caudal Cup Selection A direct lateral image should then be taken. The C-arm is positioned such that the cephalad end plate of the inferior vertebral body is visible. With the same caudal selector instrument in place, an image is taken. Three short lines on the implant, which are divergent, appear on the C-arm image. The ridged marker lines correspond to the right side and the smooth lines to the left. Outside the sterile field, the caudal selector template is applied over the image on the C-arm screen, and the lines of the template are laid over the ridged and smooth lines on the C-arm image. The superior end plate of the templated vertebral body is now aligned with the appropriate lines on the template. The line that appears most parallel to the cephalad end plate will be the correct angulation of the cup ( 9, 4, 1, 6, or 11 degrees) (Fig. 73–12).
Using this angle, the correct (left or right) caudal cup is selected, and the same procedure is performed on the contralateral side. Care should be taken to avoid scratching or marring the highly polished cup surfaces. Cephalad Arm Selection Caudal trials corresponding to the correct angulation are placed into the pedicles. Verify that the caudal trial is fully seated, and if necessary, bone should be removed until the caudal trial sits on the bone. The narrow body housing trial is placed into the caudal trials. If the narrow housing trial does not span the distance between the cup trials, the wide housing trial should be used. The 85-degree cephalad trial is placed into the pedicle and the housing trial. The necessary final position is completely seated in
Intraop fluoro
30°
20° Markers for sizing of implants
10°
n F I G U R E 73–11. Caudal trial fluoroscopic image indicating radiopaque features clearly visible . When the indicator (ball) is closest to the most lateral line, a 10-degree caudal stem is selected. When the indicator is closest to the middle line, a 20-degree caudal stem is selected, and when the ball is closest to the most medial line, a 30-degree caudal stem is indicated.
572
P A R T
V I I I
Lumbar Facet Replacement
CAUDAL SELECTOR LATERAL TEMPLATE Superior/ roirepuS Anterior/ roiretnA +11/ 11+
+6/ 6+
+1/ 1+
Posterior/ roiretsoP Align with caudal selector marker
–4/ 4– –9/ 9–
Inferior/ roirefnI
–9° –4° 1° 6° 11°
11°
115°
n F I G U R E 73–12. Caudal trial fluoroscopic image indicating radiopaque features clearly visible. When the alignment lines on the template are parallel with the alignment lines on the trial, the indicator (angled) line on the template, which is parallel to the superior end plate of the inferior vertebral body of the treated FSU, is used to select the proper caudal cup angulation. (From Archus Orthopedics, Inc., Redmond, WA.)
the pedicle, while also being flush with the bottom of the housing, and the cephalad trial must extend at least even with the far edge of the housing. If the medium cephalad trial is too short, the long cephalad trial should be used. If the medium cephalad trial extends significantly outside the housing trial, potentially causing impingement on the level below, the short cephalad trial should be used (Fig. 73–13). If the 85-degree cephalad trial of the correct length (short, medium, long) impinges on the caudal trial housing or otherwise cannot be fully seated, a 75-degree cephalad trial of the same length should be evaluated. With both cephalad trials in the trial housings, the arms should be placed so they are as symmetric as possible. Both trial arms should sit flush on the bottom slots of the housings. Using the lengths previously noted, the appropriate small, medium, or large cephalad arm implants are selected. Housing Implant Selection With the cephalad trials seated in the housing trial, look at the markings on the inside of the trial housing. If the line is in the shaded portion of the trial housing, the 15-degree housing implant is selected. If the line is in the unshaded part of the housing, the 35-degree housing implant should be selected. Selection is completed for both housing implants, and the selected implants are set aside (Fig. 73–14). Upon completion of the housing selection, remove all trials from the implant site. All appropriate implant components except
for the cross-arm and bearings should now have been selected and placed in the selection tray. Final Preparation of the Pedicle Channels Based on dimensions of the isthmus of the pedicle, the appropriate 5.75-mm or 6.5-mm drill bit is chosen and attached to the T-handle. Observe the drill bit depth markings (35–55 mm in 5-mm increments) for the previously measured pedicle depth. Maintaining good alignment with the previously drilled hole to avoid breaching the pedicle, the pedicle is drilled to the appropriate depth. A ball-tipped probe is used to verify the integrity of the pedicle. The drill bit stop is adjusted for the measured depth for each pedicle and drilled in a similar fashion. A 5.75- or 6.5-mm blunt-tipped reamer is also available and, though less aggressive, is less likely to penetrate the pedicle wall or anterior vertebral body. The reamer and drill allow for a 1-mm nominal cement mantle surrounding the implant within the pedicle. CAUDAL IMPLANT ASSEMBLY The appropriate caudal implants are prepared for assembly. Verify that both the caudal stem and cup are for the same side, and slide the cup onto the caudal stem via the dovetail connection. The cup will slide onto the stem only in one direction, from medial to lateral.
CHAPTER 73
Total Facet Arthroplasty System (TFAS)
n F I G U R E 73–13. Superimposed schematic representations of the size and angular permutations of the TFAS cephalad stems.
573
Stem Lengths
85° 75°
35−55 mm (5 mm increments)
Small (35 mm) Medium (45 mm) Large (55 mm)
Standard cephalad stem
Offset 10 cephalad stem
The ratcheting quick-release T-handle is attached to the caudal assembly instrument. The cup and caudal stem are placed into the caudal assembly instrument, and the retention lever is turned flush with the instrument. The screw mechanism is advanced completely to the visual stop, locking the cup and stem together. Assembly is completed in the same fashion for the opposite side. Once fully locked, cup and stem components cannot be disassembled. Cement Mixing and Preparation The cement is then mixed in the provided container. Mixing is continued for 30 to 60 seconds or until the cement appears to be smooth and consistent; 10-mL syringes are then filled by drawing the cement from the cement mixing bowl or vial. The cement cannula is then screwed onto the front of the syringe. Once it reaches a “doughy” state, the cement is ready for delivery (Fig. 73–15). Cementing the Caudal Stem/Cup Assembly
n F I G U R E 73–14. Cephalad trials sitting with housing trials. The upper trial housing line is outside the shaded area so a 35-degree final housing should be used. The lower trial housing line is inside the shaded area, so a 15-degree final housing should be used. A universal design housing has been created that accommodates all variation in angulation so this measurement is no longer required.
The cement cannula is inserted into the bottom of the pedicle channel. Using fluoroscopy, dispense cement slowly into the bottom of the drill channel, verifying penetration into the surrounding cancellous bone. After verifying a bolus of cement has been delivered into the surrounding cancellous bone, begin backfilling the pedicle drill channel in retrograde fashion by slowly withdrawing the cement cannula. Imaging should be used to check carefully for any extravertebral or extrapedicular extrusion of cement or any venous uptake of cement. Extrusion or venous uptake during delivery should be monitored. Excess cement should be removed from around the opening of the pedicle drill hole.
574
P A R T
V I I I
Lumbar Facet Replacement
Cross-Arm Assembly
n F I G U R E 73–15. PMMA (polymethyl-methacrylate) bone cement extruding from the delivery gun. When ready, the cement will flow, will have a dull sheen, and will not drip freely when expressed from the syringe.
The caudal holder (left or right) is attached to the corresponding left or right caudal assembly. A small amount of cement is applied to the distal portion of the stem and around the cone tip. The medial (open side) face of the cup is aligned so that it is parallel to the spinous process, and the caudal stem is inserted slowly until the cup is fully seated. Cement that extrudes from the pedicle should be removed. No cement should be allowed around the dovetail connection of the stem and cup or any part of the articulating surface. A final visual check is again made to ensure that the pedicle has not been breached. If a breach is noticed, the surgeon should remove all cement from the exterior of the breach and verify, after the cement has fully hardened, that no additional cement removal is needed. The contralateral caudal assembly implant is implanted in similar fashion. Remove the caudal holders.
n
The cross-arm selector is placed into the caudal cups, and a corresponding cross-arm size is read directly from the size indicator on the instrument dial. Using the bearing cross-arm assembly chart, select the correct bearings and cross-arm combination from the implant tray to achieve the measured dimension. The cross-arm selector is placed with smooth spheres to articulate with the caudal cups. Once it is locked in place, a measurement can be made to determine cross-arm length (Fig. 73–16). The cross-arm and bearings are assembled by manually pressing one bearing onto the end of the cross-arm. Then, sequentially, transfer the already selected housings, verify one left and one right, and slide them onto the cross-arm. Manually press the second bearing onto the cross-arm. Place the assembled cross-arm housing assembly into the crossarm assembly instrument and attach the T-handle. Turn the T-handle assembly until the handle has been advanced completely to the visual stop. Once fully locked, the cross-arm, housing, and bearing components cannot be disassembled or reused. If the incorrect components are assembled, they must be discarded. Cross-Arm Assembly Insertion and Cementing of Cephalad Arms Attach the cross-arm holder onto the cross-arm assembly and place the assembly into the caudal cups. Cement should be mixed as previously described. Dispense cement slowly into the bottom of the pedicle channel, again verifying penetration into the surrounding cancellous bone using fluoroscopic imaging. After verifying a bolus of cement has been delivered into the surrounding cancellous bone, begin backfilling the pedicle drill channel in retrograde fashion while slowly
F I G U R E 73–16. The cross-arm selector (left) is put in place and engaged with the final caudal implants (right) to determine cross-arm size.
CHAPTER 73
Total Facet Arthroplasty System (TFAS)
575
withdrawing the cement delivery nozzle. Imaging should be used to check for cement extrusion and excess cement should be removed. The appropriate cephalad arm implant is attached to the cephalad stem holder. A small amount of cement is applied to the distal portion of the stem and around the cone tip. The cephalad stem implant is inserted in line with the appropriate housing and slowly pushed into the pedicle until it rests flush against the bottom of the housing. Excess cement is again removed (Fig. 73–17). The same technique is used on the contralateral side to cement the other arm. Alignment Verification and Locking of Cephalad Arms Place the bearing positioner instrument onto the bearings and into the caudal cups to center the bearings in the caudal cups. Apply a set screw to each housing using the initial set screwdriver and provisionally tighten each set screw so that the arm is flush against the bottom of the housing. Slide the countertorque wrench over the housing and engage the arm. Attach the torque adapter to the T-handle and final set screwdriver and tighten the set screws until the torque wrench reaches its slip limit, signifying that the set screws have been tightened to the required torque. Perform the same technique to the other housing and set screw. This completes the assembly of the structure (Fig. 73–18). Allow the cemented components to fully cure and remove all implant holders. Verify that all excess cement has been removed. Irrigate and close in a standard fashion (Fig. 73–19). POSTOPERATIVE CARE Postoperative antibiotics are continued for 24 hours. A lumbosacral orthotic brace is worn for 6 weeks while out of bed. Patients are instructed to discontinue smoking and avoid using NSAIDs (non-steroidal anti-inflammatory drugs). Bending, lifting, twisting, and particularly hyperextension are to be avoided initially. At 3
n
n
F I G U R E 73–18.
Final TFAS construct.
months patients can resume light activities. At 4 months a progressive trunk strengthening and stabilization program is initiated. COMPLICATIONS In our initial European experience, we had no operative complications. There was one death in the follow-up period unrelated to the surgery or device. In our U.S. IDE experience, we have had one device-related complication. This involved a patient who presented for his first postoperative visit on which radiographs revealed a unilateral dislocation of the cup and ball assembly. The patient was asymptomatic with this, but was taken back to surgery and successfully revised.
F I G U R E 73–17. Cephalad stem being inserted into pedicle and housing (left) and after set screw is
used to fix construct.
576
P A R T
V I I I
Lumbar Facet Replacement
Flexion
Extension
n F I G U R E 73–19. Typical TFAS postoperative anteroposterior radiograph (left) along with flexionextension radiographs (right) indicating the full range of motion attained.
CONCLUSION Our initial clinical experience is promising. It demonstrates that posterior element replacement can be successfully used as an adjunct treatment for adult patients in the treatment of lumbar spinal stenosis. The device can restore functional spine unit biomechanics without off-loading the disc and increasing stress at adjacent segments. Long-term outcome studies will be the only way to determine its final efficacy as a stand-alone device and in the future may play a potential role in the development of 360-degree spinal arthroplasty. REFERENCES 1. Huang RC, Lim MR, Girardi FP, et al: the prevalence of contraindications to total disc replacement in a cohort of lumbar surgical patients. Spine 29:2538–2541, 2004. 2. Huang RC, Lim MR, Girardi FP, Cammisa FP: The prevalence of contraindications to total disc replacement in a cohort of lumbar surgical patients. Spine 29(22):2538–2541, 2004. 3. Ghormley RK: Low back pain, with special references to articular facet, with presentation of an operative procedure. JAMA 101:1773–1777, 1933.
4. Fujiwara A, Lim T, An H, et al: The effect of disc degeneration and facet joint arthritis on the segmental flexibility of the lumbar spine. Spine 25(23):3036–3044, 2000.
BIBLIOGRAPHY Bowden AE, Villarraga ML, et al: In Situ Biomechanics of Total Facet Replacement Using Finite Element Analysis. Presented at NASS, Sept. 2006. Larson CR, Qingan Z, et al: Tissue Loading of the Total Facet Arthroplasty System (TFASW). Presented at SAS, May 6, 2005. Mow VC, Lai WM, Hou JS: A triphasic theory for the swelling and deformation behaviors of articular cartilage. J Biomech Eng 113 (3):245–258, 1991. Sjovold SG, Zhe Q, et al: Loading of the Total Facet Arthroplasty System (TFASW) Compared to a Rigid Posterior Instrumentation System. Presented at SAS, May 2005. Webb SA, Presbeana R, Branea R: Preliminary Experience with Total Facet Arthroplasty (TFASW). Presented at SAS, May 2006. Zhu Q, Larson CR, et al: Biomechanical Evaluation of the Total Facet Arthroplasty System (TFASW) Kinematics. Presented at SAS, May 2005.
CHAPTER
74
Anatomic Facet Replacement System (AFRS) Allen Carl, Carlos E. Oliveira, Robert W. Hoy, William Lavelle, Vijay K. Goel, and Bryan W. Cunningham
K E Y l
l
l
l
l
P O I N T S
Fusion and dynamic stabilization may lead to accelerated adjacent level disease due to the reduction or elimination of segmental motion. There is a clinical need for motion preservation devices that address the posterior spine. Anatomically based long bone joint replacement designs have historically resulted in clinical success. Successful arthroplasty systems have utilized precision instrumentation to place implants accurately and repeatably. Building on the clinical success achieved by total hip and total knee arthroplasty systems, the Facet Solutions AFRS uses the general principles of anatomic design and reproducible instrumentation to provide pain relief, normal motion with stability, and optimal implant survival.
The recognition of adjacent level disease as a condition linked to spinal arthrodesis1 has led to the emergence of spinal motion preservation devices as alternatives to fusion. Initially, the focus was on developing disc replacement devices to treat patients with anterior column pathologies. However, more recently, posterior devices have been developed as well. Lumbar spinal stenosis cases in which posterior element neural decompressions lead to instability and thus require fusion2,3 have spawned an interest in motion preservation devices that attempt to replace and mimic the function of the facet joints. In addition, these devices may provide a treatment option for patients with low-grade spondylolisthesis or isolated posterior pain and total disc replacement patients with postoperative facet-related pain. Arthroplasty procedures for many other articulating synovial joints have used implant systems designed to mimic normal healthy anatomy with excellent clinical results for decades.4,5 The Facet Solutions Anatomic Facet Replacement System (AFRS) (Facet Solutions, Inc., Logan, UT) has evolved adhering to the basic concepts that brought success to total hip and knee prostheses. Although their development arose during different eras, their success is due to the same fundamental understanding of materials and biomechanics. Long-term cyclic loading success in long bone joint replacements is felt to be due to meticulous operative intervention while mimicking the normal anatomy with implant design. Preservation of surrounding soft tissues, ligaments, and proper
tissue balancing allows for structural conformity, retained stability, and a uniform distribution of forces leading to optimal implant life. The objective for the Facet Solutions AFRS was to address posterior lumbar spinal pathologies while preserving natural lumbar biomechanics (stability with natural motion), thereby mitigating the adjacent level effects resulting from the reduction or elimination of motion as seen in dynamic stabilization and fusion devices.6 To this end, the development of a facet arthroplasty system comprising anatomic facet implants and precision instrumentation providing for reproducible implant placement was undertaken. INDICATIONS AND CONTRAINDICATIONS The Facet Solutions AFRS implant is intended for use in the treatment of patients with osteoarthritis of the facet joints resulting in subarticular, lateral, or central canal stenosis. In addition, patients with stenosis caused by low-grade spondylolisthesis (Grade I or less) may benefit from this procedure. The implant is specifically designed to preserve motion for patients who are susceptible to adjacent level disorder, commonly caused by spinal arthrodesis. While not included in the clinical trial indications, patients with isolated facet joint pain and no stenosis or nondegenerative patients with recent traumatic facet fractures would also be considered candidates for this surgery. This procedure is contraindicated for patients with the following conditions. (Note: This list is not complete and only includes contraindications that may be of interest to the reader.) 1. Previous fusion attempt(s) with pedicle instrumentation at the operative or adjacent levels 2. Osteoporosis 3. Greater than a Grade I spondylolisthesis at the operative level 4. A recent traumatic pars fracture at the operative level or either adjacent level 5. Trauma to the lumbar spine 6. Metabolic bone disease (osteomalacia, osteogenisis imperfecta) 7. Spondylolisthesis at levels other than the operative level 8. Scoliosis of the lumbar spine (defined as more than 11 degrees of sagittal deformity), as indicated by plain x-ray films, magnetic resonance imaging (MRI), or discography 577
578
n
P A R T
V I I I
Lumbar Facet Replacement
F I G U R E 74–1. Anatomic Facet Replacement System (AFRS).
9. Medically significant obesity as defined by a body mass index (BMI*) greater than 40 kg/m2 10. Known allergy to cobalt-chromium-molybdenum DESCRIPTION OF THE DEVICE The Facet Solutions AFRS is composed of superior and inferior facet implants and utilizes conventional pedicle screw fixation (Fig. 74–1). A crossbar links the left and right inferior facet implants, providing additional construct stability. The conforming articulating surfaces of the facet implants are manufactured from cobalt-chromium-molybdenum, a highly polished, superhard, wear-resistant alloy. This material has a long and successful history of clinical use in total joint replacement systems.7,8
n
F I G U R E 74–2. Kinematics test setup. (Courtesy of Vijay Goel, University of Toledo, Spine Research Center.)
SCIENTIFIC TESTING OUTCOMES
INTACT, DESTABILIZED, AND IMPLANTED ROM 8 Intact Destabilized Implanted 6
ROM (°)
Preclinical testing consisted of a series of implant static strength, fatigue strength, and fixation strength evaluations. This testing demonstrated that the construct's strength far exceeded physiologic loads. Additionally, a cadaveric kinematics study was performed in which the intact, destabilized, and implanted conditions were evaluated for each specimen (Fig. 74–2). This test established that the AFRS restored intact kinematics (Fig. 74–3). A coupled-motion 20 million cycle wear debris biocompatibility assessment was performed in which no adverse histopathologic response was observed (Fig. 74–4). The Investigational Device Exemption (IDE) study submission for this device was approved by the FDA on the first attempt, illustrating the rigorous and thorough nature of the bench-top testing regimen.
4
2
CLINICAL PRESENTATION AND EVALUATION The clinical evaluation of the Facet Solutions AFRS began in São Paulo, Brazil in 2005 where the first implantations were performed by Dr. Carlos Oliveira and Dr. Allen Carl. In 2006 Facet
0 Flex n
*BMI ¼ weight(kg) (height [m])2
Ext
LB
AR
F I G U R E 74–3. Kinematics test results. (Courtesy of Vijay
Goel, University of Toledo, Spine Research Center.)
CHAPTER 74
F I G U R E 7 4 – 4 . There was no adverse histopathologic response to an implantation of wear debris representing 20 million coupled motion cycles. (Courtesy of Bryan Cunningham, Orthopedic Biomechanics Laboratory, Union Memorial Hospital.)
n
Solutions was granted CE mark approval and FDA IDE study approval for the AFRS. The company is currently in the patient enrollment phase for their U.S. study. Although the 1-year follow-up results from the OUS clinical study are encouraging, the small number of patients that have reached that time point limits the data's statistical power. All of the patients will continue to be followed and outcome measures will be recorded per protocol. When more OUS patients reach the later follow-up periods, statistically meaningful results will be reported (Fig. 74–5). OPERATIVE TECHNIQUE The patient will be placed in a prone position with padded prominences and no pressure on the abdominal contents. General anesthetic typical for any posterior spine procedure will be administered. The procedure calls for a standard midline approach with an incision length comparable to a posterior instrumented lumbar fusion. However, the instrumentation also lends itself to a minimally invasive double Wiltse approach. A decompression is performed to relieve any subarticular, lateral, or central canal stenosis. A set of precision instrumentation then allows for repeatable bony preparation and implant trials. Finally, the pedicle screws and facet implants are installed and secured much like a screw and rod system. This technique was designed to build upon what spine surgeons are
n
Anatomic Facet Replacement System (AFRS)
579
F I G U R E 74–5. AFRS lateral standing radiograph.
comfortable with today. The “look and feel” of the procedure is not all that different from a posterior instrumented fusion. As the general population of spine surgeons migrates toward more minimally invasive muscle-splitting approaches, micro-decompressions, and image or computer guidance systems, the Facet Solutions AFRS technique will easily accommodate those advances. POSTOPERATIVE CARE The hospital stay and postoperative care will be similar to what is expected for a posterior lumbar instrumented fusion. The patient requires 6 weeks of soft corset bracing and curtailing of aggressive activity, but walking activity is permissible. COMPLICATIONS AND AVOIDANCE Adhering to the indications for use as defined in the IDE clinical study protocol is the best way to avoid complications and steer clear of high-risk patients. Like any new procedure, there is an absence of long-term clinical data to support broadly defined indications. For the sake of study clarity and safety, the FDA and sponsors tend to err on the conservative side when establishing study indications. As the base of clinical knowledge grows, the indications can be refined to include all the patients who will benefit from this procedure.
580
P A R T
V I I I
Lumbar Facet Replacement
ADVANTAGES/DISADVANTAGES: AFRS Advantages Anatomic by design Easy-to-use and reproducible instrumentation Provides stability with natural motion Conventional pedicle screw fixation Disadvantages Investigational use only
CONCLUSION As improvements in diagnostic capabilities and a more complete understanding of pain generators are achieved, the degeneration of the three-joint complex can be broken down into more discrete stages. Filling the surgeon's armamentarium with technologies that specifically address each stage of the degenerative process is where the future of spinal motion preservation will take us. These therapies will range from nonoperative care to procedures that address multilevel anterior and posterior disorders simultaneously with minimal invasion. The Facet Solutions Anatomic Facet Replacement System has a unique place in this continuum for patients suffering from facet-related, subarticular, lateral, and central stenosis. Building from the clinical success achieved by total hip and total knee arthroplasty systems, the Facet Solutions AFRS uses the general principles of anatomic design and reproducible
instrumentation to provide for pain relief, normal motion with stability, and optimal implant survival. Caution: The Facet Solutions AFRS is an investigational device and is limited by U.S. law to investigational use.
REFERENCES 1. Shono Y, Kanada K, Abumi K, et al: Stability of posterior spinal instrumentation and its effects on adjacent motion segments in the lumbosacral spine. Spine 23(14):1550–1558, 1998. 2. Knaub MA, et al: Lumbar spinal stenosis: Indications for arthrodesis and spinal instrumentation. Instr Course Lect 54:313–319, 2005. 3. Yuan PS, Booth RE Jr, Albert TJ: Nonsurgical and surgical management of lumbar spinal stenosis. Instr Course Lect 54:303–312, 2005. 4. Ramaniraka NA, Rakotomanana LR, Rubin PJ, Leyvraz P, et al: Noncemented total hip arthroplasty: Influence of extramedullary parameters on initial implant stability and on bone-implant interface stresses. Rev Chir Orthop Reparatrice Appar Mot 86(6):590–597, 2000. 5. Kelley MA: Patellofemoral complications following total knee arthroplasty. Instr Course Lect 50:403–407, 2001. 6. Schmoelz W, Huber JF, Nydegger T, et al: Dynamic stabilization of the lumbar spine and its effects on adjacent segments an in vitro experiment. J Spinal Disord Tech 16(4):418–423, 2003. 7. Howie DW, McCalden RW, Nawana NS, et al: The long-term wear of retrieved McKee-Farrar metal-on-metal total hip prostheses. J Arthroplasty 20(3):350–357, 2005. 8. Rieker CB, Schon R, Kottig P, et al: Development and validation of a second-generation metal-on-metal bearing: Laboratory studies and analysis of retrievals. J Arthroplasty 19(8 suppl 3):5–11, 2004.
CHAPTER
75
The Zyre Facet Replacement Device Carl Lauryssen, Scott H. Kitchel, and Jason D. Blain
K E Y l l l
P O I N T S
The Zyre device is used to perform lumbar facet spacer arthroplasty. Implantation of the device is a minimally invasive procedure. The procedure can be performed with or without decompression procedures.
A proliferation of motion preservation technologies has surfaced in recent years with the bulk of the development being focused on devices intended to relieve or supplement the forces seen by the anterior column of the spine. These developments follow the relative successes of the ProDisc (Synthes, West Chester, PA) and CHARITÉ (DePuy Spine, Raynham, MA) intervertebral total disc arthroplasty devices. These devices along with other total disc arthroplasty, nucleus replacement, and dynamic stabilization devices are well covered in other chapters of this book. Relatively few new technologies have focused exclusively on the two posterior elements of the three-joint complex that comprises a spinal functional unit. Although this area of development is still emerging, there appear to be two basic disciplines of thought with respect to addressing the facts with implantable device technology. The predominate focus thus far appears to be on technologies that require the resection of substantial amounts of the facet joint or the entire facet joint in conjunction with a midline incision approach. The facet joints are then replaced with a mechanical functioning means. The Zyre Facet Replacement Device from Spinal Elements, Inc. (Carlsbad, CA), differs from this approach. The Zyre device is an interpositional arthroplasty device that requires no bony resection and can be implanted using less disruptive techniques with small bilateral incisions directly over the facets. The primary functions of the facets are to absorb loads transmitted through the spine and to restrict motion. The facets may be a source of pain generation themselves. Their composition of a synovial-bathed articulation of hyaline cartilage over subchondral bone leaves them susceptible to the same degenerative osteoarthritis mechanism that affects other joints. Degeneration of the facet joints alters normal spinal kinematics. Without the healthy functioning of this motion component, other motion preservation devices are in jeopardy of failure.
Radiographic facet degeneration following CHARITÉ lumbar disc replacement has already been shown.1 The technology behind the Zyre device, interpositional arthroplasty, has been applied in both large and small joints of the appendicular skeleton. As of this writing, the clinical efficacy of the device has yet to be proved; however, the design, mechanical testing, and background technology provide strong evidence that this implant will be capable of providing a solution to maintaining motion in the posterior elements in a simple, elegant manner that makes it attractive as a stand-alone device or as an adjunct to the implantation of other motion preservation technologies. INDICATIONS AND CONTRAINDICATIONS As previously stated, the clinical efficacy of the Zyre device has yet to be proven. Therefore, the indications and contraindications presented represent the authors’ best estimate of application of the device based on preclinical testing that has been performed to date. The indications for use of this device are similar to those for other synovial joints: painful degeneration in the appendicular skeletal system that is unresponsive to conservative treatment. Additionally, we are exploring the device and its use to address neurologic claudication. Any diagnosis must involve a survey of the topography of the facets through computed tomography (CT) scans. Magnetic resonance imaging (MRI) will provide an evaluation of the fluid content within the joint as well as an assessment of the articular cartilage and the joint capsule. A bone scan will also be a key diagnostic tool, demonstrating evidence of uptake in a painful facet joint. Anesthetic blocks of the facet joint may offer the best method of isolating back pain to the facets.2 The use of the Zyre device to address isolated degeneration of the facet joint is an enticing possibility, as the degeneration of each aspect of the three-joint complex of the spine concordantly is a well-documented phenomenon.3 Additionally, this device may be best used as an adjunct to the implantation of other motion preservation devices that address the anterior column. This combined usage would give the practitioner the tools to address each segment of the spinal joint. 581
582
P A R T
V I I I
Lumbar Facet Replacement
Contraindications to the implantation of the Zyre device are similar to those for posterior spinal surgery and those of other motion preservation devices. They include systemic diseases; infection; failed previous procedures at the operative level, including hemilaminectomy and facetectomy; previous fusion at the operative level; ankylosed discs or facets; pars fracture; and other factors that may prevent a positive surgical outcome (obesity, neurologic disorder, chemical abuse, etc.). DESCRIPTION OF THE DEVICE The Zyre Facet Replacement Device is an interpositional spacer that consists of three basic components—a dish-shaped spacer, a polymer cord, and retainers. The spacer is manufactured from cobalt-chromium and is designed to reside within the joint space. Multiple sizes of the spacer are available, varying in diameter and thickness to fit the individual anatomy and pathology. The spacer has a central hole that allows passage of the polymer cord. The cord is made from polyester (PET) fiber and is threaded across the facet joint and through the spacer. The retainers are also made from cobalt-chromium. They slide over the cord on either side of the facet joint and secure to the cord by means of a set screw. A picture of the individual components can be seen in Figure 75–1, and a picture of the device implanted in a spine model is provided in Figure 75–2. SCIENTIFIC TESTING AND CLINICAL OUTCOMES Biomechanical testing was performed on the Zyre Facet Replacement Device with the goal of determining several factors related to the device's performance. The device was implanted in excised cadaveric functional spinal units (FSUs) and loaded both statically and dynamically to determine the range of motion, stability, and endurance of the device. After mechanical testing was completed, the Zyre devices were removed and examined for histologic evidence of wear. Four adult cadaveric excised spines with adequate bone quality and adequate intervertebral discs were obtained for the testing. DEXA (dual-emission x-ray absorptiometry) scans were
n
F I G U R E 7 5 –1.
Device.
Components of the Zyre Facet Replacement
n
F I G U R E 75–2. Construct in spine model.
performed on all specimens as part of the pretest evaluation. The spines were then sectioned into L2-L3 and L4-L5 FSUs, and all nonfunctional soft tissue was removed. This provided eight FSUs for testing. The FSUs were potted in fixtures with the aid of bone cement, and each unit was tested prior to implantation by the application of a 500-N load placed 15 mm anterior to the neutral loading axis to cause the FSU to go into flexion. The units were loaded a total of 20 times each and data related to flexion angle and stiffness were recorded. Following the short-run testing of the undisturbed FSUs, six of the units were implanted with the Zyre Facet Replacement Device, and the short-run testing described above was repeated for an additional 20 cycles. All constructs and controls were then cycled in a similar fashion to 1 million cycles. The FSUs were kept in a saline solution bath that was temperature regulated to 37 C. After the completion of the dynamic cycling, all FSUs were again subjected to the short-cycle loading to get final values for range of motion and construct stiffness. A chart indicating the initial and final data is illustrated in Figure 75–3. The Zyre implants were then removed from the FSU constructs, and the facets and posterior vertebral bodies and discs were sectioned from the FSU, dehydrated, and infiltrated with methyl/methacrylate reagent grade polymer. Once the polymer cured, the specimens were cut into 2-mm sections using a diamond blade saw. Sagittal section cuts were taken across the right side of the specimens, and transverse cuts were made across the left side of the specimens. An image of a typical section can be seen in Figure 75–4. No evidence of wear was seen in any of the sections taken. The interpositional spacer was not able to be removed from one specimen. A picture of this specimen is depicted in Figure 75–5. This image depicts the conformance of the interpositional spacer to the surrounding anatomy. The static and dynamic tests indicated that the Zyre Facet Replacement Device is capable of withstanding cyclic loading in a cadaveric model, and the histologic examination after the testing indicates that wear is not evident.
CHAPTER 75
The Zyre Facet Replacement Device
583
SUMMARY OF STATIC FLEXION TEST, (n = 3)
16.0 Without With With post Control (n = 1) Control post (n = 1)
14.0
12.0
Relevant unit
10.0
8.0
6.0
4.0
2.0
0.0 EBS (Nm/Deg)
ROM (Deg)
NZ (Deg)
EBS (Nm/Deg)
NZ (Deg)
ROM (Deg)
L4-L5
L2-L3 n
FIGURE 75–3. Chart of test data.
OPERATIVE TECHNIQUES
n
F IGURE 75–4.
complex.
Histologic section of normal non-operated facet
The implantation of the Zyre facet system may be performed in line with a minimally invasive technique that involves very little muscle disruption. Each joint of the affected level is addressed with its own incision and exposure. The first incision of about 25 mm may be made over the first facet to be implanted. Soft tissues are then retracted to expose the facet capsule. Care should be taken to maintain as much of the facet capsule as possible. An incision can then be made at the joint line of the capsule. The incision should be long enough to fit the interpositional component of the implant system into the joint space. By placing a trial into the joint space, the facet can be sized for the appropriate corresponding implant. Fluoroscopic verification will reveal if the trial is the appropriate size by comparison to the overall joint space and position relative to the neural foramen. The trial and subsequent implant should cover the articular surfaces without interfering with surrounding structures. The appropriately sized interpositional implant may be place in the articular space. Some forming may be necessary to get the implant to better match the native anatomy. This is accomplished with specialized instruments. If distraction is needed during the implantation of the interpositional spacer, the joint space may be opened by
584
P A R T
V I I I
Lumbar Facet Replacement
COMPLICATIONS As with any implant, the complications for this device will fall into one of two categories—device-related and non-device–related. Non-device–related complications will be similar to those for any posterior lumbar spine procedure and will be related to anesthesia, infection, and underlying medical conditions of the patient. Additional experience with the Zyre device will reveal what, if any, device-related complications will arise. As with any motion preservation device, whether for the spine or extremities, wear, implant migration, and failure of adequate pain relief will be the complications to monitor. CONCLUSION
n
F I G U R E 7 5 – 5.
Histologic section of device with spacer.
distracting on the spinous processes. The spinous processes may be accessed through the incision made over the contralateral facet. Once the interpositional component is placed, a hole is drilled across the facet joint and through the spacer. The polymer cable may then be threaded through the hole in a suture-like fashion. Retainers are placed on the medial and lateral aspect of the joint and secured by a set screw that bears against the cable. The contralateral facet may be addressed in a similar fashion. Radiographic assessment is performed with anteroposterior and lateral fluoroscopic imaging to assure appropriate placement.
There are many exciting new developments in the area of motion preservation for the spine. Many of the technologies remain to be proved, and finding a position in the continuum of spine care may take years of exposure to these devices. The Zyre Facet Replacement Device is a new technology that appears capable of providing a solution to motion preservation in the posterior elements. Its capacity to be a stand-alone device or used in concert with other motion preservation technologies, in addition to its ease of implantation, makes the device seem like a promising innovation for the future of spine surgery. REFERENCES 1. Phillips FM, Piment LF: Facet degeneration after CHARITÉ Disc replacement. Orthop Today 25:45, 2005. 2. Lewinnek GE, Warfield CA: Facet joint degeneration as a cause of low back pain. Clin Orthop Rel Res 213:216–222, 1986. 3. Kirkaldy-Willis WH: Managing Low Back Pain. New York, Churchill Livingstone, 1983.
CHAPTER
76
FENIX Facet Resurfacing Implant Teddy Fagerstrom and Horace Hale
K E Y l l l l l
P O I N T S
Concept of facet joint resurfacing versus replacement Preserves supportive structures of the spine Mimics normal motion of the intact spine Eliminates origin of pain Compatible with any disc replacement techniques
Spinal degeneration and arthritis are well-known causes of several conditions of the spine leading to surgical intervention. The origin of pain in the degenerated lumbar spine is diagnostically a challenge, sometimes making the choice of surgical technique difficult.1,2 Resurfacing techniques have effectively been used in other areas of the body as an alternative to more extensive joint replacement techniques.3 The FENIX Facet Resurfacing Implant (Gerraspine, St. Gallen, Switzerland) is the first resurfacing system for use in the lumbar spine. Osteoarthritis of the spine appears much like osteoarthritis elsewhere in the body. Affecting the synovial facet joint, it can result in osteophytosis, joint space narrowing, subchondral sclerosis, and subchondral cyst formation.4 The prevalence of facet joint pain is considered common and estimated at 15% in younger, injured workers with chronic low back pain and about 40% in older, noninjured, rheumatology patients.5 Several clinical studies have shown that back pain and referred pain in patients can be relieved by anesthetizing one or more of the facet joints. Other studies have shown that facet joint injections and facet nerve blocks may be of equal value as diagnostic tests.2 Total disc replacement has become a popular technique over the last several years. This technique does not address the potential pain arising from the facet joints. Actually, Philips and associates showed in a recent study that 44% of the patients undergoing total disc replacement developed symptomatic pain from the facets and increased facet degeneration. Facet joint arthritis is considered a contraindication to total disc replacement.6,7 Besides being a potential source of low back pain, facet joint osteoarthritis, with osteophytosis, may cause nerve root impingement or compression when extending into the lateral recess of the spinal canal (Fig 76–1).
Facet joints are the only true articulation in the lumbar and lumbosacral spine. In conjunction with the soft tissue supporting structures, they provide axial, rotational, and shear load support to the spine as well as sliding articulation, thus being an important component of the spinal motion segment. Therefore, if normal function of the spine is to be restored and subsequently maintained, the facet joints must be addressed, while taking care to preserve the natural supporting structures, during surgical treatment of debilitating low back pain. The FENIX Facet Resurfacing Implant is manufactured from cobalt-chromium alloy. It is designed to resurface—not replace— the degenerated or otherwise diseased articulating surfaces of the human facet joint and restore its function in subjects with low back pain derived from degenerated facets of the lumbar spine (from L1 to S1), while preserving the supporting soft tissue structures. The cause of the degeneration may be primary or secondary. These implants are designed for single-use only and are fixed by means of the following: 1. A translaminar screw for immediate fixation of the inferior facet component 2. Crossed fins for immediate fixation of the superior facet component 3. Titanium-plasma sprayed contact surfaces for osseous ingrowth and to facilitate and ensure long-term fixation of both the inferior and superior components of the implant
INDICATIONS AND CONTRAINDICATIONS Indications 1. Significant facet disease, including facet arthrosis, clinically restricted or painful extension range of motion (ROM), or facet blocks 2. Single- or multilevel symptomatic lumbar facet syndrome from L1-L2 to L5-S1 confirmed by subject history, physical examination, diagnostic facet injections, and radiographic studies 585
586
P A R T
V I I I
Lumbar Facet Replacement
8. Morbid obesity, having a body mass index (BMI) of 40 or greater* 9. Active infection at the operative site or other systemic site DESCRIPTION OF THE DEVICE The FENIX Facet Resurfacing Implant consists of three primary components: the FENIX superior facet resurfacing component, FENIX inferior facet resurfacing component, and the FENIX locking screw (Fig. 76–2). SCIENTIFIC TESTING RESULTS Three-Dimensional Flexibility Testing L2-S1 n
F I G U R E 76–1. Radiographic image showing marked
osteophytosis in the spine of a patient with osteoarthritis of the lumbar facet joints.
3. Radicular symptoms caused by lateral canal stenosis 4. Chronic low back pain with a pain intensity of a minimum of 4 on a scale of 10 on the Visual Analog Scale (VAS) lasting for more than 6 months 5. Age between 18 and 85 years old; males and females 6. Intact posterior elements of the involved segment 7. Failed at least 6 months of nonoperative care Contraindications 1. 2. 3. 4.
Age is greater than 85 or less than 18 years Osteoporosis or osteomalacia Isthmic spondylolisthesis Prior spine surgery at the involved segment (other than nucleotomy, total disc replacement, or decompression of the medial spinal canal) 5. Congenital conditions altering the posterior anatomy 6. Medication that may reduce bone metabolism 7. Facet joints that are absent or fractured
In flexion-extension, the total range of motion (from 8 Nm extension to 8 Nm flexion) of the spine specimen was slightly increased following implantation of the Gerraspine device (34.3 degrees vs. 30.9 degrees). The neutral zone was also slightly increased (6 degrees vs. 4 degrees) (Fig. 76–3A). In lateral bending, the total range of motion (from 8 Nm right to 8 Nm left) of the spine specimen was slightly increased following implantation of the Gerraspine device (31.2 degrees vs. 29.6 degrees). The neutral zone was similar (Fig. 76–3B). In torsion, the total range of motion (from 8 Nm right to 8 Nm left) of the spine specimen was unchanged following implantation of the Gerraspine device (18.5 degrees vs. 18.4 degrees) (Fig. 76–3C). The neutral zone was unchanged. 1. The L4-L5 segmental range of motion is not affected by the device. 2. The centers of rotation lie in the same range for the intact spine and for the spine after implantation. The progression of centers of rotation through the entire flexion-extension, lateral bending, or rotation motions seems “smoother” with the implant compared to the intact spine. 3. The coupled motions (i.e., lateral bending or torsion in response to a principal motion of flexion-extension) are much cleaner with the implant. With the intact spine, *BMI ¼ weight (in kg) [height (in meters)]2.
C C
D B B
A
A
F I G U R E 76–2. FENIX implant parts from above (left) and below (right). A, superior component; B, inferior component; C, locking screw; D, crossed fins.
n
CHAPTER 76
FENIX Facet Resurfacing Implant
FLEXION-EXTENSION 15 10 5
Angle (degrees)
0 –10
–8
–6
–4
–2
0
2
4
6
8
10
2
4
6
8
10
–5 –10 –15 –20 –25 –30 Moment (Nm) Intact Implant
A
LATERAL BENDING 20 15
Angle (degrees)
10 5 0 –10
–8
–6
–4
–2
0 –5 –10 –15 –20 Moment (Nm) Intact Implant
B n
F I G U R E 76–3. Results of three-dimensional flexibility testing.
587
588
P A R T
V I I I
Lumbar Facet Replacement TORSION
10
Angle (degrees)
5
0 –10
–8
–6
–4
–2
0
2
4
6
8
10
–5
–10
–15 Moment (Nm) Intact Implant
C n
F I G U R E 76–3 Cont'd.
coupled motions in torsion are quite erratic and do not follow the timing or magnitude of the principal motion. With the implant in place, the coupled motions follow the timing of the principal motions and are the same for each successive load application. In general, it seems that the implant cleans up the motion of the segment without having a negative influence on range of motion8 (Fig. 76–4A and B; Table 76–1).
OPERATIVE TECHNIQUE The surgical procedure can be performed by traditional open exposure or by minimally invasive techniques and requires the following steps: S T E P 1 : P R E O P E R A T I V E . Preoperatively, the surgeon must decide which intervertebral levels to treat. This may be done using a variety of diagnostic techniques such as diagnostic facet injections, radiographs, magnetic resonance imaging (MRI), discography, patient history, and physical examination. S T E P 2 : P A T I E N T P O S I T I O N I N G . The patient is positioned in the prone or kneeling position. A table should be used that accommodates both lateral and anteroposterior radiographs. S T E P 3 : E X P O S U R E . Exposure of the facet joints at the affected level can be effected through a classical midline approach or a paramedian approach modified according to Wiltse to reduce surgical trauma to the erector trunci muscles. The surgeon should use the approach that he is trained in and that is appropriate for the patient being treated.
Commercially available surgical instruments are used to perform the exposure down to the level(s) to be treated. They are also used to maintain the exposure via the appropriate retractors. The exposure is complete when the facets of the appropriate level are exposed and sufficient retraction has been done to permit access. S T E P 4 : I D E N T I F Y L O C A T I O N A N D O R I E N T A T I O N . The identification of the target facet is done by imaging control or based on the local anatomy. Beginning on the side of your choosing, remove the capsule. Moving the spinous processes with towel clamps of the involved vertebrae may help to identify the joint space and the orientation of the facet. The yellow ligament is detached by inserting a blunt dissector along the ventral side of the cranial lamina and the medial border of the facet is identified. A simultaneous decompression with removal of the medial border of the facet can be performed to decompress the lateral part of the spinal canal and the entry to the foramen. S T E P 5 : J O I N T P R E P A R A T I O N . With a small curette and rongeur, create a 3- to 4-mm intraarticular space by removing the cartilaginous surfaces of the facet joint. The removal of cartilage should be complete, and the subchondral bone has to be carefully decorticated without weakening the bone. If the narrowness of the joint requires removal of bone, this should be done on the side of the inferior facet and not of the superior facet. With the specially designed inferior facet burr, prepare the inferior facet for placement of the inferior facet implant. Be
CHAPTER 76
n
589
F I G U R E 76–4. Radiographic imaging; anteroposterior (A) and lateral (B) images of specimen.
careful to “lateralize” the apex of the curve of the preparation tool. The resection plane should be chosen in a way that the plane is parallel to the original plane of the facet. The power burr is moved in a caudal to cephalad direction achieving a clean bony surface to receive the prosthesis. The penetration of the canal is avoided by the normal curve of the facet in the ventral part and should be controlled by insertion of the dissector along the medial border of the facet (Fig. 76–5A and B). With the specially designed superior facet burr, prepare the superior facet. Ensure that the flat superior end of the rasp is in contact and in alignment with the previously prepared inferior facet. The rasp is then moved in a cephalad to caudal direction with the power on. This motion effects a clean and uniform resection of the joint space in the shape and dimension of the superior implant. This cephalad to caudal motion is continued until a bleeding bone bed is achieved.
TABLE 76–1.
FENIX Facet Resurfacing Implant
CAUTION: Do not remove more than 2 mm of the medial aspect of the superior facet. S T E P 6 : L A M I N A R P R E P A R A T I O N . With the specially designed aiming device, drill a 3.2-mm hole for the placement of the translaminar screw. The spacer is mounted on the aiming device and is placed centrally in the prepared space of the facet and in contact with the resected inferior facet. Through the drill guide, the hole for the translaminar fixation screw is made. The drill hole exits ideally in the center of the previously prepared inferior facet surface. With the specially designed fin-cutting instrument, prepare the lamina for acceptance of the fins of both the superior and the inferior implants. By hammering upward, this instrument will create a space at the location of the drill hole, in the center of the prepared
Values of Motion-Testing Results Compared to Control Levels L4-L5 Instrumented Level
Frame
Disc Angle (deg)
0 1 2
8.02 6.18 10.84
Frame
Disc Angle (deg)
0 1 2
19.52 17.11 21.69
Disc Angle C
0 1.84 2.82
Ant Disc Hgt
Pst Disc Hgt
64.08 60.78 70.2
38.94 41.38 36.32
Ant Disc Hgt
Pst Disc Hgt
72.65 69.79 75.11
16.89 20.68 13.42
Ant Disc Disp
Pst Disc Disp
0 3.29 6.12
0 2.44 2.63
Avg Disc Hgt
51.51 51.08 53.26
Avg Disc Disp
0 0.43 1.75
Translation (p)
COR X (pix)
COR Y (pix)
0 16.22 31.39
0 34.94 40.73
Translation (p)
COR X (pix)
COR Y (pix)
0 0.17 1.49
0 12.71 12.45
0 2.21 2.82
L5-S1 Intact Level Disc Angle C
0 2.41 2.17
Ant Disc Disp
Pst Disc Disp
0 2.86 2.46
0 3.79 3.47
Avg Disc Hgt
44.77 45.24 44.27
Avg Disc Disp
0 0.47 0.5
0 78.52 56.3
P A R T
590
V I I I
A
Lumbar Facet Replacement
B
D
C
E n
F I G U R E 7 6–5.
Instrumentation for facet surface preparation and implant introduction.
inferior facet, in order to receive the small tower and the fins of the inferior implant. By hammering downward, the trench on the superior facet is made in appropriate size, depth, and position to receive the fins of the superior prosthesis (Fig. 76–5C and D). S T E P 7 : Repeat steps 3 to 6 on the contralateral side. S T E P 8 : I N F E R I O R I M P L A N T I N S E R T I O N . With the screw measurement gauge, determine the correct length of translaminar fixation screw required. Place the inferior facet component in the prepared bed, using the specially designed implant introducer. Distraction of the segment by interspinous spreading facilitates the introduction of the components. By placing the tower of the prosthesis into the bony excavation, the prosthesis is placed automatically in the correct position. Insert the translaminar screw. Tightening the screw will “lag” the inferior facet implant to the lamina thereby ensuring tight contact between the prosthesis and the bony surface. Repeat on the contralateral side (Fig. 76–5E). S T E P 9 : S U P E R I O R I M P L A N T I N S E R T I O N . Place the superior facet component in the prepared bed, using the specially designed implant introducer. Distraction of the segment by interspinous spreading facilitates the introduction of the components.
The superior facet component must be introduced from caudal to cranial. Avoid unnecessary and gross movement of the implant before final positioning in order to maintain the integrity of the cross-fin trenches, which were cut earlier. The prosthesis is pressed against the bony surface by relieving segmental distraction and is held in place by the fins. Repeat on the contralateral side. Following final implant insertion, lateral and anteroposterior radiographs may be taken to assure proper implant placement. S T E P 1 0 : C L O S U R E . The muscle and skin are closed in the usual fashion (Fig. 76–6A and B). POSTOPERATIVE CARE The recommended postoperative regimen for surgeons and their patients is as follows: 1. Mobilization is begun on the first or second postoperative day. 2. A soft brace is worn during daily activities for 3 months to restrict gross motions. 3. The patient is encouraged to walk but instructed to limit activity until advised by the surgeon. Specifically, heavy lifting, repetitive bending, and extension of the spine should be avoided. No physical therapy is performed during the first 2 months. 4. Avoid steroidal drugs for at least 60 days postoperatively.
CHAPTER 76
FENIX Facet Resurfacing Implant
591
n F I G U R E 76–6. A, FENIX implant in place, posterior view. B, FENIX implant in place, posterolateral view, with transparent vertebrae to show positioning of and the translaminar fixation of the superior part of the implant, which is attached to the inferior facet of L4.
B
A
COMPLICATIONS This implant relies on the support of the posterior elements of the vertebrae for immediate and long-term fixation. Therefore, based on this requirement, there will be some complications and contraindications for its use. l Severe osteopenia or systemic bone disease may cause implant loosening through the absence of bony ingrowth into the porous surfaces of the implant. l The posterior elements should be able to receive a 3-mm screw. l For anatomic reasons, isthmic spondylolysis patients should be avoided. l Degenerative and mild (Grades 1–2) dysplastic sponylolisthesis patients should be considered because the joints, interarticular surfaces, and the laminae are usually intact, even when altered. However, great care should be taken in the preparation of the joint surfaces and when preparing the path for the translaminar screw. l MRI diagnostic techniques can be difficult with cobalt-chromium alloys. However, we are hopeful that the new generation of low-carbon, cobalt-chromium alloys will be better suited to these diagnostic techniques. l In the event that the implant should be explanted (loosening, infection, etc.), implant removal should be simple, and all salvage options (e.g., for fusion) will still be open to the surgeon.
ADVANTAGES/DISADVANTAGES: FENIX Advantages Unique, innovative, simple concept Mimics normal motion Resurfacing procedure—minor osseous and no ligamentous resection required Eliminates pain source Stand-alone implant Works with any presently available disc opportunity for complete motion segment repair Adds additional stability and longevity to the diseased spine when used alone or in conjunction with an anteriorly placed implant Easy, straightforward implant insertion Broadens the treatment modalities for spine disease Gives spine surgeons a viable alternative to fusion or a more traumatic procedure A complementary anterior device is easily added A true multilevel device; able to implant from L1 to S1 in the same patient Disadvantages Patient selection sometimes difficult Addresses only pain originating from the facet joints Limited experience with use of resurfacing techniques in the spine
n
F I G U R E 76–7. Chart depicting the different surgical approaches to the facet arthroplasty.
FACET concepts
Articular replacement
AFRS– facet solution
TFAS– Archus
TOPS– Impliant
Articular spacer
Articular resurfacing
Zyre–Spinal Elements
FENIX–Gerraspine
592
P A R T
V I I I
Lumbar Facet Replacement
CONCLUSION Ongoing questions center on the following issues: l Indications l Pain source l Actual efficacy of and need for a facet implant Even the best proven prophylactic, diagnostic, and therapeutic methods must continuously be challenged through research for their effectiveness, efficiency, accessibility, and quality (Fig. 76–7). Our in vitro investigations show the FENIX Facet Resurfacing Implant to be a safe and effective implant that improves the function of an arthritic facet joint without destroying surrounding and needed supporting structures, mimicking the function and motion of a normal facet joint. Such an implant has the potential to enhance the quality of life of patients suffering from debilitating low back pain due to degenerative osteoarthritis. Further work is needed to gather experience and results from clinical use of the FENIX, our intention being to develop a safe and effective implant which in the future can be expanded for use in all areas of the human spine.
REFERENCES 1. Bogduk N, Long DM: The anatomy of the so-called “articular nerves” and their relationship to facet denervation in the treatment of lowback pain. J Neurosurg 51(2):172–177, 1979. 2. Marks RC, Houston T, Thulbourne T: Facet joint injection and facet nerve block: A randomised comparison in 86 patients with chronic low back pain. Pain 49(3):325–328, 1992. 3. Roberts P, Grigoris P: Removal of acetabular bone in resurfacing arthroplasty of the hip. J Bone Joint Surg Br 88(6):839, 2006. 4. Richardson ML: Approaches to differential diagnosis in musculoskeletal imaging. University of Washington Medical School, version 2.0. October 2001 web posting. 5. Schwarzer AC, Aprill CN, Derby R, et al: Clinical features of patients with pain stemming from the lumbar zygapophysial joints: Is the lumbar facet syndrome a clinical entity? Spine 19(10):1132– 1137, 1994. 6. Cammisa FP: The prevalence of contraindications in a cohort of lumbar surgical patients. Spine 29(22):2538–2541, 2004. 7. Philips FM, Diaz R, Pimenta L: The fate of the facet joints after lumbar total disc replacement: A two-year clinical and MRI study. Global Symposium on Intervertebral Disc Replacement and Non Fusion Technology. New York, New York, 2005. 8. Ferguson S: Summary of results of flexibility with the FENIX facet implant: Report from biomechanical testing. MEM Research Center: 2004.
CHAPTER
77
Cervical Disc Replacement Combined with Cervical Laminoplasty Seok Woo Kim* and Paul C. McAfee
K E Y l
l
l
l
P O I N T S
Cervical laminoplasty techniques were developed in Japan, and they are a time-honored method of increasing the cross-sectional area of the spinal canal without requiring spinal fusion. They are the original motion preservation method for the cervical spine. Cervical spinal radiculopathy can be treated either anteriorly or posteriorly, whereas multilevel cervical spinal stenosis is preferred to be treated by posterior decompressive techniques. Cervical disk replacement can be combined with multilevel cervical laminoplasty in myeloradiculopathy patients with multilevel stenosis and one- or two-level herniated discs. Posterior laminoplasty is performed first to relieve myelopathy, and if the patient has persistent or recurrent symptoms of radiculopathy, anterior cervical disk replacement could be added. Of the several methods of expanding the spinal canal with laminoplasty, the two methods that are most widely used are (1) the hinge door from the junction of the lamina and facet and (2) the midline French door, splitting the spinous processes in the midline. Both types are meticulous surgical techniques that decompress the spinal cord and roots and frequently require spacers or plates to keep the neural arch open.
Recently, cervical laminoplasty has received increased attention for the treatment of multilevel cervical compressive myelopathy. Laminoplasty was developed to widen the spinal canal dimensions without permanently removing the dorsal elements of the cervical spine. The retained dorsal elements should prevent scar formation on the dura and potentially reduce the incidence of postoperative instability. Cervical motion is theoretically preserved. Although the clinical relevance of the range of motion (ROM) after laminoplasty is controversial, several authors have noted the benefit of maintaining the cervical ROM. This improved ROM has the potential to decrease axial neck pain and prevent adjacent segment cervical disease. Although most cervical myelopathy can be addressed surgically with an isolated anterior or posterior approach, some patients present with more complex cervical spine disease, such as myelopathy with superimposed radiculopathy. The problem with anterior
*Seok Woo Kim should be credited as the first spinal surgeon in the world to innovate and to combine cervical disc replacement and cervical laminoplasty in the successful treatment of patients.
cervical decompression and fusion (ACDF) for the treatment of acute disc herniation could be that there is reduction of motion and there could be significant morbidity if autologous bone graft harvesting is done. Consequently, there has been more recent emphasis on cervical disc arthroplasty to maintain motion, reduce adjacent segment deterioration, allow for an adequate decompression, and avoid the use of iliac bone grafts. Cervical disc replacement combined with laminoplasty could be successful. Anterior total disc replacement (TDR) combined with posterior multilevel laminoplasty could preserve ROM after laminoplasty for multilevel cervical compressive myelopathy in addition to correcting radiculopathy in patients presenting with myeloradiculopathy due to multilevel stenosis and a herniated disc at one or two levels. This chapter will focus on the possible merits of the cervical disc replacement combined with laminoplasty, especially from the perspective of the maintenance of ROM, by first reviewing classical laminoplasty methods used to treat complex compressive myelopathy and radiculopathy in the cervical spine. DEVELOPMENT OF LAMINOPLASTY—HISTORICAL BACKGROUND OF OPERATIVE TREATMENT IN CERVICAL SPONDYLOTIC MYELOPATHY Degenerative cervical spondylosis and ossification of the posterior longitudinal ligament (OPLL), rheumatoid arthritis, and trauma remain common causes of compression of the spinal cord that can result in cervical spondylotic myelopathy (CSM). Patients suffer from clumsiness of the hands, difficulty walking, impaired balance and coordination, and sensory complaints of numbness or tingling in the hands and feet. Cervical spondylosis, which most commonly occurs after age 50, is the most common cause of cervical myelopathy. It evolves from degeneration of the disc that leads to reduced disc height and bulging of the disc posteriorly into the spinal canal. The bulging disc may then calcify. This calcified disc, along with marginal osteophyte formation and uncovertebral spurring, plays an important role for narrowing the spinal canal. The resultant foraminal and spinal canal stenoses produce radiculopathy and myelopathy, respectively. 595
596
P A R T
I X
Hybrid Nonfusion Techniques
Operative treatment is usually required to decompress the neural elements, restore lordosis, and stabilize the spine to prevent additional degeneration at the affected level. Surgery for cervical myelopathy has been performed by both posterior (laminectomy with or without fusion, laminoplasty) and anterior (corpectomy, multilevel discectomy) approaches. The decision to use either an anterior or posterior approach depends on many factors, including the number of vertebral segments involved in the disease process, cervical alignment, the source of spinal cord compression, presence of instability on dynamic radiographs, and surgeon's preference for the various surgical techniques. Other clinical factors to consider include axial neck symptoms, presence of congenital spinal stenosis, medical comorbidities, previous surgery, and the presence of radiculopathy. Anterior decompression with stabilization obtained by anterior arthrodesis allows direct removal of the compressive sources. The procedure is accomplished through multilevel discectomies and fusions with segmental instrumentation or through cervical corpectomy with vertebral reconstruction using autograft, allograft, or titanium mesh cages with plate fixation. The disadvantages include the risks of an extensive anterior cervical approach, the need for graft healing, and the potential problems at adjacent levels. Emery and associates reviewed 108 cases of cervical spondylotic myelopathy which had been managed with anterior decompression and arthrodesis via partial or subtotal corpectomy.1 In their report, 46% of patients with preoperative gait abnormality had a normal gait at follow-up, and 62% presenting with a preoperative motor deficit had full recovery. However, 15% of the patients developed pseudarthrosis after surgery. It has been well documented that the rate of pseudarthrosis increases with each segmental level added to an anterior decompression and fusion.2 Additional complications of anterior cervical decompression and fusion surgery include graft fracture, dislodgement or settling, and surgery-related complications including dysphagia, recurrent laryngeal nerve dysfunction, esophageal injury, carotid vessel injury, and late adjacent segment degeneration. Historically, laminectomy has been regarded as the standard procedure to decompress the spinal cord for cervical myelopathy secondary to multisegmental spondylosis, ossification of the posterior longitudinal ligament (OPLL), with or without developmental spinal canal stenosis. However, segmental instability, kyphosis, perineural adhesions, and late neurologic deterioration occurred and became well-known complications after laminectomy. Pal and Cooper tested load transmission in the anterior (vertebral bodies) and posterior (facet joints and articular processes) cervical columns in cadavers.3 They demonstrated that 36% of load transmission was through the anterior columns, whereas 64% was through the posterior columns. Therefore, the posterior neural arch is responsible for most of the load transmission in the cervical spine, and significant loss of integrity of these posterior columns can result in instability, causing the weight-bearing axis to shift anteriorly. Kyphosis progresses subsequent to this loss of sagittal balance and places the cervical musculature at a mechanical disadvantage, requiring constant contractions to maintain upright head posture. This progression causes most of the weight to be borne by the discs and anterior vertebral bodies, which leads to further degeneration and spondylosis. Some authors advocate to
augment with fusion after laminectomy because of the frequency of instability and loss of normal cervical lordosis. To avoid these problems, surgeons developed a different strategy in which decompression of the spinal cord could be achieved while preserving posterior arch, which was thought to contribute to postoperative cervical alignment and stability. Oyama and associates introduced the expansive Z-shaped laminoplasty in which the posterior wall of the spinal canal was preserved by Z-plasty of the thinned laminae.4 Hirabayashi developed “expansive open-door laminoplasty” in 1978.5 Since then, many modifications of the procedure have been reported, such as “double-door laminoplasty,” which was introduced by Kurokawa and co-workers in 1982.6 In this procedure, spinal canal enlargement is achieved by splitting the spinous process, and graft bones are inserted between the split spinous process. This procedure has shown several new features, including symmetric configuration of reconstructed posterior elements that theoretically allow symmetric expansion of the spinal cord, closed-ring configuration of the neural arch for each segment through insertion of bone graft, and wide access to the spinal canal to perform additional procedures. Because of these advantages, this procedure is widely accepted and a variety of modifications have been developed. In 1992, Nakano and associates developed a hydroxyapatite (HA) spacer instead of an autogenous bone graft to decrease the operation time, decrease blood loss, and avoid postoperative pain around the donor site.7 Over time various modifications and supplementary procedures were devised for posterior decompression of the spinal canal while preserving stability and mobility of the cervical spine: reattachment of the spinous process and extensor musculature, dome-like decompression on the C2 lamina, use of thread wire saw (T-saw) to minimize the bony loss at the time of sagittal cutting of laminae, and modification of the spinous process spacer. With refinements of the surgical technique and the development of more stable fixation devices, the reproducibility of the procedure has been improved. All of the laminoplasty variants effectively expand the cross-sectional area of the spinal canal while preserving alignment, stability, and motion. Surgeons have demonstrated equivalent neurologic outcomes utilizing the various laminoplasty methods. Many authors have shown that laminoplasty can be effectively utilized in a patient to multilevel cervical spondylotic myelopathy with similar or superior clinical outcomes as compared to multilevel anterior surgery or laminectomy combined with fusion.8 In addition, laminoplasty can be combined with posterior foraminotomies to relieve compression of the nerve roots and with fusion at one or two levels if instability such as spondylolisthesis is present. TECHNIQUES OF LAMINOPLASTY Nearly all of the laminoplasty variants effectively expand the crosssectional area of the spinal canal while preserving alignment, stability, and motion. The following techniques are most commonly used: l Z-plasty l Hirabayashi laminoplasty (open-door laminoplasty) l French door laminoplasty (double-door laminoplasty) l Kurokawa modification l Tomita modification
CHAPTER 77
Cervical Disc Replacement Combined with Cervical Laminoplasty
597
Novel techniques designed to reposition the laminae and open the spinal canal continue to evolve. There are two basic types of procedures. The first is the “open-door laminoplasty,” whereby one side of the lamina is hinged open, along with the ligamentum flavum on the same side. The other procedures involve creating a midline opening with the left and right hemilaminae hinging on both sides. Newer techniques, such as the use of hydroxyapatite (HA) or ceramic spacers and titanium miniplates, have been proposed to decrease the surgical time and improve the safety of the procedure. Z-Plasty This technique was first described by Oyama and associates.4 In the Z-plasty, troughs are first drilled into the laminae at the junction of the lateral mass. Then the laminae are thinned. After this, a Z-shaped cut in the laminae is done with a high-speed drill. After the Z cut, the thinned sections of laminae may then be separated and the canal opened or expanded. The laminae may then be secured or wired to maintain the expanded canal. Naito and associates reported Z-plasty had longer operation time and more blood loss, and it has subsequently disappeared.9 Hirabayashi Laminoplasty Hirabayashi and collegues described the expansive open-door laminoplasty.5, 10–12 In this technique, a trough is first drilled in the laminae at the junction of the lateral mass. The cut is taken down to dura. The ligament and any remaining thinned bone should be removed with a Kerrison rongeur. After drilling the complete trough on the open-door side, a second trough is drilled on the closed side. After the troughs have been cut, the ligamentum flavum between ventral and caudal vertebrae and remaining structures should be removed with a Kerrison rongeur. After cutting on the open side, the block of laminae is rotated toward the closed (hinged) side This effectively opens and expands the spinal canal. The laminae may be held open with sutures passing around or through the spinous processes and the facet capsule on the hinged side or held open with titanium miniplates. Hirabayashi advocated stay sutures within the spinous process and paraspinal muscles to avoid closure of the opened laminae. Even with the added sutures, there were several cases of a spring-back phenomenon (i.e., the door closed), in which patients showed both progressive neurologic and radiologic deterioration. Since then, in order to prevent the spring-back phenomenon, many techniques were developed to maintain canal patency. Recently, a novel technique augmented by titanium miniplates to maintain patency of the open-door laminoplasty was introduced by O'Brien and associates13 (Figs. 77–1, 77–2). They reported a significant improvement in sagittal canal diameter and canal area with no hardware failure during their follow-up. French Door Laminoplasty (Double-Door Laminoplasty, Midline Opening Laminoplasty) In this type of laminoplasty, the spinous processes are split in the midline; therefore, the door is opened in the midline and creates a
n F I G U R E 7 7 – 1. Anteroposterior radiograph of a 63-year-old female patient after Hirabayashi open-door laminoplasty augmented by titanium miniplates.
symmetric opening in the canal, unlike Hirabayashi type open-door laminoplasty in which the canal is opened on one side and hinged on the other and thus creates an asymmetric expansion of the canal. Troughs are then cut into the laminae at the junction of the lateral masses. The laminae are only thinned and are not cut completely through. After the completion of the trough cuts, the spinous processes are split in the midline. Then the spinous processes are spread in the midline and held open, just like French doors. They may be held open with small bone grafts or other grafts. Ceramic or hydroxyapatite (HA) spacers have also been reported as a substitute for autogenous bone graft (Figs. 77–3, 77–4). The Kurokawa modification means that the dorsal aspect of the spinous processes is removed and used as grafts.6 The spinous processes are cut in the midline. The spinous processes are split in the midline and held open with bone grafts that are wired in place. The main disadvantages of this procedure are the technical difficulty and high risk of the procedure associated with sagittal splitting of the spinous processes with a burr in such close proximity to the spinal cord. In the Tomita modification, the spinous processes are split with a wire saw (threadwire saw; T-saw) to avoid the use of a burr while splitting the midline. The polyethylene sleeve enclosing the T-saw is passed cranially along the epidural space, and then the sleeve is withdrawn over the saw. The saw is pulled tight to cut the midline of the inner wall of the lamina arch by using a reciprocating saw motion (Figs. 77–5 to 77–7). An important safety consideration is that cervical kyphosis places the cord at risk during the splitting of the spinous processes.
598
P A R T
I X
Hybrid Nonfusion Techniques
n
F I G U R E 7 7 – 4.
Polyethylene sleeve enclosing the T-saw.
The effects of increasing amounts of facet resection with multilevel cervical open-door laminoplasty and laminectomy on cervical spine stability were tested in nine cadaveric specimens. In that study, resection of 25% or more of the facet bilaterally resulted in a high increase in cervical motion; thus, the authors recommended a
concurrent arthrodesis should accompany a laminectomy with more than 25% bilateral facectectomy. On the other hand, cervical laminoplasty was not significantly different from the intact control except for a marginal increase in axial torsion. Baisden and associates14 observed the effects on structural properties between multilevel laminectomy and laminoplasty using an in vivo goat model over a 6-month period. No significant radiographic differences in cervical alignment were observed initially. However, at 16 weeks after surgery, the laminectomy group demonstrated a significant decrease in normal cervical lordosis. This decrease progressed throughout the 6-month period, whereas the laminoplasty group did not significantly change its cervical alignment during the study period. Unfortunately, few in vitro biomechanical studies evaluate the ability of laminoplasty to maintain stability of the cervical spine. More biomechanical studies need to be evaluated on a long-term basis analyzing stability and changes in structural alignment parameters.
n F I G U R E 7 7 – 3. Hydroxyapatite (HA) spacer as a substitute for autogenous bone graft.
spinous process.
n F I G U R E 7 7 – 2. Lateral radiograph of a 63-year-old female patient after Hirabayashi open-door laminoplasty augmented by titanium miniplates.
Biomechanical Studies of Laminoplasty
n
F I G U R E 7 7 – 5.
Threadwire saw (T-saw) used to split the
CHAPTER 77
Cervical Disc Replacement Combined with Cervical Laminoplasty
599
laminoplasty with unilateral foraminotomy on the open-door side. Overall, the majority of studies reported some neurologic improvement. French Door Laminoplasty and Its Modifications The mean recovery rate after French door laminoplasty is approximately 50% to 70%.19 The recovery rate is not likely related to the specific surgical procedure. Clinical outcome is more likely related to preoperative neurologic status. Short-Term Results Almost all of the studies based on short-term follow-up have shown satisfactory results after any types of laminoplasty. F I G U R E 7 7 – 6. Exposed dura after splitting the spinous processes with T-saw (modified French door laminoplasty).
n
CLINICAL OUTCOMES Many studies have evaluated the clinical outcomes after various laminoplasty procedures. Most studies report clinical outcomes based on JOA (Japanese Orthopaedic Association) scores to demonstrate some improvement in preoperative neurologic deficits. Hirabayashi Open-Door Laminoplasty The mean recovery rate after Hirabayashi expansive laminoplasty is approximately 50% to 60%.5,10–12,15,16 Hirabayashi and Satomi reported a 54% neurologic improvement rate over an average 3-year follow-up.12 Satomi and associates17 found recovery rates of 56% for OPLL and 61% for cervical spondylotic myelopathy over an average 8-year follow-up for 33 patients with OPLL and 18 patients with cervical spondylotic myelopathy. Herkowitz18 reported a 90% neurologic recovery rate after Hirabayashi
F I G U R E 7 7 – 7. Postoperative photograph showing modified French door laminoplasty using hydroxyapatite (HA) spacers.
n
Long-Term Results A few studies have evaluated the long-term clinical outcome after laminoplasty. Miyazaki and associates20 reported that improved neurologic status was maintained at a mean of 12 years after surgery. Kawaguchi and associates21 observed the long-term outcome over 10 years of 126 patients after en bloc laminoplasty. They found that neurologic recovery rates were maintained to 55.1% at the last follow-up. Most of the patients improved their JOA scores after surgery and maintained them at the final follow-up, although 20 patients has a worsened JOA score during followup. Postoperative ROM decreased to 25.1% of preoperative motion, with 61% of patients showing some reduction in ROM. Kyphotic changes were noted in eight patients, who showed poor neurologic recovery rates. Seichi and associates22 also examined the average 12-year follow-up of 25 patients who underwent a double-door laminoplasty. They observed that the average preoperative JOA score was improved after surgery, and the average score increased by 5-year follow-up. However, 10 years after surgery, the average score had decreased slightly until the final follow-up. Late neurologic deterioration was noted in four patients at an average of 8 years. Another long-term study by Chiba and associates23,24 examined the average 14-year follow-up of 80 patients who underwent expansive open-door laminoplasty. They reported that the average JOA score and recovery rate improved significantly until 3 years after surgery and remained at an acceptable level in both cervical spondylosis and OPLL patients with slight deterioration after 5 years. Segmental motor palsy developed in 8 patients. Late neurologic deterioration, mainly lower extremity motor score decline, was noted in 8 CSM and 16 OPLL patients. Overall, cervical range of motion decreased by 36%. Cervical lordosis decreased gradually in both patient groups. Such changes in alignments did not affect surgical results in CSM patients, and OPLL patients with preoperative kyphosis had lower recovery rates than those with straight and lordotic alignments. OPLL progression was detected in 66% of patients but did not affect clinical results.
600
P A R T
I X
Hybrid Nonfusion Techniques
POSTOPERATIVE COMPLICATIONS Most studies report similar problems after laminoplasty, which include postoperative axial neck pain, reduced range of motion, motor nerve root palsy (canal stenosis), and the loss of cervical lordosis (development of some degree of kyphosis). In addition to these complications, the issue of range of motion will be discussed later at the end of this chapter. Axial Neck Pain Sani and associates25 reported that the overall incidence of neck pain ranged from 6% to 60% without apparent dependence on the specific variation of laminoplasty using meta-analysis in 71 retrospective studies. Hosono and associates26 also reported that the prevalence of postoperative axial symptoms was significantly higher (60%) after laminoplasty than after anterior fusion (19%). In a study by Wada and associates27 comparing subtotal corpectomy and laminoplasty, axial pain was observed in 15% of the corpectomy group and in 40% of the laminoplasty group. The etiology of axial neck pain remains unclear, but it might be related to posterior soft tissue and muscle dissection or facet joint disruption. Motor Nerve Root Palsy Segmental motor nerve paralysis is common after laminoplasty. Motor paralysis most commonly involves the C5 nerve root, possibly caused by the result of a tethering effect on the nerve root at or near the apex of lordosis induced by excessive posterior cord migration after decompression. Sani and associates25 reported the incidence of C5 root palsy as 8%, and most cases resolved spontaneously. The motor palsy usually recovers to normal or near normal within 12 months after surgery. Loss of Cervical Lordosis (Development of Kyphosis) The range of worsening spinal alignment, not necessarily kyphosis, has been reported to range from 22% to 53%.27,30 However, with the use of modern instrumentation techniques, the incidence of development of kyphosis has been reported to range from 2% to 4%, with 0% in the hardware augmentation group.13 Although many studies have evaluated the preoperative and postoperative cervical alignment using various methods, there are few reports on the correlation between loss of normal lordosis and potential long-term neurologic deterioration. DISCUSSION Laminoplasty versus Laminectomy and Fusion In a retrospective matched cohort analysis of cervical laminoplasty versus cervical laminectomy with fusion for the treatment of multilevel cervical myelopathy, laminoplasty was associated with a higher improvement in function and a lower frequency of complications. Although most studies that used laminectomy and fusion for the treatment of multilevel cervical myelopathy have reported favorable neurologic outcomes, complications seem to occur in
12% to 40% of patients. These complications include screw loosening, broken hardware, pseudarthrosis, adjacent segment degeneration, progression of myelopathy, development of cervical kyphosis, deep wound infection, epidural hematoma, seroma, and donor site discomfort. Importantly, no laminoplasty technique is effective for the restoration of lordosis in an already kyphotic spine, because the first goal of laminoplasty is maintaining the expanded spinal canal and cervical alignment. On the other hand, laminectomy should decompress the cervical canal; therefore, any type of fusion must be followed until bony fusion has been documented radiographically. In addition, the fusion procedure is likely to have a higher rate of adjacent segment degeneration due to the lost motion through several spinal motion segments. Laminoplasty versus Multilevel Anterior Corpectomy and Discectomy In a retrospective matched cohort analysis of multilevel cervical corpectomy and fusion versus laminoplasty for the treatment of multilevel cervical myelopathy, the percentage of patients with subjective improvement in symptoms was similar between the two groups; however, complications in the corpectomy group were more frequent than in the laminoplasty group.29 Complications were progression of myelopathy, nonunion, screw malposition, persistent dysphagia, and persistent dysphonia. Yonenobu and associates24 also reported similar clinical results (neurologic outcomes) after multilevel anterior decompression and fusion compared with laminoplasty; however, complications were more frequent in the subtotal corpectomy and strut graft group than in the laminoplasty group. Overall, multilevel anterior cervical discectomies and corpectomies have been shown to have more frequent complications. To reduce these complications, combined anterior and posterior surgery might be an alternative treatment to address the nonunion and instrumentation-related problems, but greater surgical risk could arise. Range of Motion in Laminoplasty In almost all the reviewed studies, cervical ROM has been reported to decrease 17% to 50%, with an average of approximately 50% after the Hirabayashi laminoplasty. Cervical ROM was decreased greatly (70% to 80%) when laminoplasty was augmented with fusion. The clinical relevance of this decreased ROM is controversial. Some authors emphasized the importance of limiting cervical ROM after laminoplasty. Kimura and associates31 reported that the 40% reduction in cervical ROM after laminoplasty and this decreased ROM were thought to contribute to limiting dynamic factors, which are some of the possible reasons for the development of myelopathy. Yoshida and associates32 documented that the decreased ROM may prevent late deterioration caused by progression of OPLL. In contrast, other authors emphasized the importance of preservation of cervical ROM after laminoplasty. They thought that axial neck pain was decreased and adjacent segment degeneration was prevented by preserving cervical ROM. Morimoto and associates28 reported that 83% of “normal range of cervical motion” was preserved at 3-year follow-up using their modification of
CHAPTER 77
the Kurokawa laminoplasty and noted the importance of preserving cervical ROM. Hirabayashi and Bohlman10 proposed that the restrictive effects of cervical laminoplasty produce spinal stability without the complete restriction of motion that is obtained by fusion techniques. In long-term follow-up studies, there is a trend to loss of cervical ROM after laminoplasty. Seichi and co-workers22 observed that a significant rate of spontaneous facet fusion occurred in patients followed for at least 10 years after a Kurokawa laminoplasty. They therefore recommended use of hydroxyapatite spacers instead of autogenous bone graft and reduced postoperative neck immobilization to prevent these effects. One of the most confusing aspects of the laminoplasty might be related to the cervical ROM. The gain from this changed ROM after laminoplasty appears to lie between solid fusion with elimination of cervical motion and laminectomy alone with complete preservation of normal cervical range of motion. And it is clear that this wide range of cervical motion was retained after surgery while maintaining cervical stability regardless of the type of laminoplasty, patient age, and postoperative immobilization, although there has been inconsistent emphasis in the literature regarding maintenance of preoperative ROM and its effect on progressive myelopathy. Also, the range of remaining ROM must lie between the ROM from solid fusion and the ROM from laminectomy alone. Therefore, if the patient is doing well without any loss of cervical stability and any undesirable complications after laminoplasty, it is not necessary to try to reduce the remaining motion from the surgery, because nobody knows whether this motion could be of a beneficial nature, such as decreasing neck pain and preventing adjacent segment level disease. Range of Motion in Cervical Arthroplasty There have been more recent interests on new emerging technologies, such as cervical disc arthroplasty, to maintain motion, avoid deformity, reduce adjacent segment deterioration, and allow for an adequate decompression without having to use bone graft. Although there are no long-term data available yet, most of cervical arthroplasty prostheses available on the market have shown satisfactory clinical and radiographic outcomes, including maintaining ROM at the treated level and overall cervical spine (C2C7; to check the global cervical motion) without any remarkable device-related complications. Goffin and associates33observed that motion was maintained at the treated segment, and this motion was similar to preoperative motion in almost 90% of the group at the 2-year follow-up after implantation of the Bryan cervical disc (Medtronic Sofamor Danek, Memphis, TN). Duggal and associates34 demonstrated quantitatively the functional motion-preserving capabilities of cervical arthroplasty (Bryan device). They found that the mean sagittal ROM at the treated disc space was 7.8 degrees on follow-up assessments, and this did not differ significantly from the 10.1 degrees measured before surgery. Moreover, no significant changes in ROM were noted at individual segments either adjacent to or distant from the operative site, and global cervical sagittal motion distributed across all levels (C2-C7) increased moderately in the late follow-up evaluations. Bertagnoli and associates35reported that ROM at level of implants was 12 degrees after implantation of
Cervical Disc Replacement Combined with Cervical Laminoplasty
601
ProDisc-C cervical artificial disc. McAfee and associates36 demonstrated that ROM was maintained at the treated level in their biomechanical study using the porous-coated motion (PCM) device (Cervitech, Inc., Rockaway, NJ) and reported that ROM was corroborated with dynamic fluoroscopy in 23 patients (32 implants) who underwent PCM cervical artificial disc replacement surgery. Cervical Disc Replacement Combined with Cervical Laminoplasty Recently, the capability of maintaining motion of the treated level makes it possible to extend the indications of cervical artificial disc. Sekhon37 reported the use of an artificial disc at one or two levels in the management of cervical spondylotic myelopathy. In his study, a total of seven patients who underwent Bryan artificial disc prosthesis preserved motion at the final follow-up, and improvement in lordosis was achieved in 29% of cases. McAfee also reported the use of PCM cervical arthroplasties in the management of radiculopathy, myelopathy, and especially Klippel-Feil syndrome to preserve the motion. It is known that cervical motion is important to assist in chewing in this class of Klippel-Feil patients. Therefore, the proven advantages of the foregoing procedures, which can preserve and maintain ROM at the treated level, might have a major impact on the primary indications for cervical arthroplasty and more widespread use of implantations. Case Report A herniated nucleus pulposus occurred in a 51-year-old male patient with classic symptoms of cervical spondylotic myelopathy due to cervical stenosis, and the patient underwent posterior cervical laminoplasty (Figs. 77–8 and 77–9). He was suffering from hand clumsiness (both hands), gait disturbance, and difficulty in fine finger movement bilaterally. The senior author (SWK) performed a posterior cervical modified French door midline splitting laminoplasty using a T-saw and hydroxyapatite spacers to preserve the motion. After surgery, the patient was dramatically improved neurologically. He could walk independently without aids and moved all his extremities unassisted. However, several months later, he was starting to complain of a tingling sensation and pain in his right hand; herniated nucleus pulposus was noted at the C5C6 level on a new MRI. As he was completely happy from the previous laminoplasty surgery, especially in terms of preservation of motion, even after such a major surgery like multilevel laminoplasty, he wanted to have a surgery which afforded him the opportunity to maintain the motion at the herniated disc level. Therefore, an anterior decompression and PCM artificial disc implantation were performed at C5-C6. He was completely satisfied with the results after cervical artificial disc replacement. At 1-year follow-up, he has returned to work and is unrestricted in performance of the activities of daily living without any limitation of neck motion (Figs. 77–10 and 77–11). Although there has been much debate on the beneficial or harmful effects of the motion after laminoplasty, we have realized how motion is important for the patient's life quality (satisfaction). It is hoped that by combining posterior cervical laminoplasty with anterior cervical artificial disc replacement surgery, the
602
P A R T
I X
Hybrid Nonfusion Techniques
n F I G U R E 7 7 – 1 0. PCM artificial disc implantation in a patient who underwent previous multilevel posterior laminoplasty (flexionlateral radiograph).
n F I G U R E 7 7 – 8. Anteroposterior view of a 51-year-old male patient after posterior laminoplasty (modified French door laminoplasty using hydroxyapatite spacers).
n F I G U R E 7 7 – 9. Lateral radiographs of a 51-year-old male patient after posterior laminoplasty (modified French door laminoplasty using hydroxyapatite spacers).
historically successful outcomes attained by posterior laminoplasty can be married to the known advantages of maintenance of normal motion attained by cervical artificial disc replacement. Currently, there is a tremendous interest in the maintenance of the biomechanical properties of the cervical spine that focus on the preservation of the motion segment.
n F I G U R E 7 7 – 1 1. PCM artificial disc implantation in a patient who underwent previous posterior laminoplasty (extension-lateral radiograph).
CHAPTER 77
The patient's anticipation for a better postoperative outcome and avoidance of the known problems associated with fusion suggest that the new procedures such as cervical artificial disc replacement and laminoplasty may play an important role in the management of cervical spondylotic myelopathy combined with radiculopathy. A larger series of patients with long-term followup data is needed before recommending this technique routinely. REFERENCES 1. Emery SE, Bohlman HH, Bolesta MJ, Jones PK: Anterior cervical decompression and arthrodesis for the treatment of cervical myelopathy:two to seven-year follow-up. J Bone Joint Surg Am 80:941–951, 1998. 2. Farey ID, McAfee PC, Davis RF, Long DM: Pseudoarthrosis of the cervical spine after anterior arthrodesis: treatment by posterior nerveroot decompression, stabilization, and arthrodesis. J Bone Joint Surg Am 72:1171–1177, 1990. 3. Pal PP, Cooper HH: The vertical stability of the cervical spine. Spine 13:447–449, 1988. 4. Oyama M, Hattori S, Moriwaki N: A new method of posterior decompression(in Japanese). Centr Jpn J Orthop Traumatol Surg 16:792–794, 1973. 5. Hirabayashi K: Expansive open-door laminoplasty for cervical spondylotic myelopathy(in Japanese). Operation 32:1159–1163, 1978. 6. Kurokawa T, Tsuyama N, Tanaka H: Enlargement of spinal canal by the sagittal splitting of the spinous process. Bessatsu Seikeigeka 2:234–240, 1982. 7. Nakano K, Harata S, Suetsuna F, et al: Spinous process-splitting laminoplasty using hydroxyapatite spinous process spacer. Spine 17:41–43, 1992. 8. Heller JG, Edwards CC, Murakami H, et al: Laminoplasty versus laminectomy and fusion for multilevel cervical myelopathy: an independent matched cohort analysis. Spine 26:1330–1336, 2001. 9. Natio M, Ogata K, Kurose S, Oyama M: Canal-expansive laminoplasty in 83 patients with cervical spondylotic myelopathy. Int Orthop 18:347–351, 1994. 10. Hirabayashi K, Bohlman KH: Controversy: multilevel cervical spondylosis. Spine 20:1732–1734, 1995. 11. Hirabayashi S, Kumano K: Contact of hydroxyapatite spacers with split spinous processes in double-door laminoplasty for cervical myelopathy. J Orthop Sci 4:264–268, 1999. 12. Hirabayashi K, Satomi K: Operative procedure and results of expansive open-door laminoplasty. Spine 13:870–876, 1988. 13. O'Brien MF, Peterson D, Casey AT, et al: A novel technique for laminoplasty augmentation of spinal canal area using titanium miniplate stabilization. A computerized morphometric analysis. Spine 21:474–483, 1996. 14. Baisden J, Voo LM, Cusick JF, Pintar FA, Yoganandan N: Evaluation of cervical laminectomy and laminoplasty: a longitudinal study in the goat model. Spine 24:1283–1289, 1999. 15. Kawaguchi Y, Matsui H, Ishihara H: Surgical outcomes of cervical expansive laminoplasty in patients with diabetes mellitus. Spine 25:551–555, 2000. 16. Kimura K, Oh-Hama M, Shingu H: Cervical myelopathy treated by canal-expansive laminoplasty: computed tomographic and myelographic findings. J Bone Joint Surg Am 66:914–920, 1995.
Cervical Disc Replacement Combined with Cervical Laminoplasty
603
17. Satomi K, Nishu Y, Kohno T, Hirabayashi K: Long-term follow-up studies of open-door expansive laminoplasty for cervical stenotic myelopathy. Spine 19:507–510, 1994. 18. Herkowitz H: A comparison of anterior cervical fusion, cervical laminectomy, and cervical laminoplasty for the surgical management of multiple level spondylotic radiculopathy. Spine 13:774–780, 1988. 19. Takayasu M, Takagi T, Nishizawa T, et al: Bilateral open-door cervical expansive laminoplasty with hydroxyapatite spacers and titanium screws. J Neurosurg(Spine 1) 96:22–28, 2001. 20. Miyazaki K, Korohuji E, Ono S: Extensive simultaneous multisegmental laminectomy and posterior decompression with posterolateral fusion. J Jpn Spine Res Soc 5:167, 1994. 21. Kawaguchi Y, Knamori M, Ishihara H, et al: Minimum 10-year follow-up after en bloc cervical laminoplasty. Clin Orthop 411:129–139, 2003. 22. Seichi A, Takeshita K, Ohishi I, et al: Long-term results of doubledoor laminoplasty for cervical stenotic myelopathy. Spine 479–487, 2001. 23. Chiba K, Ogawa Y, Ishii K, et al: Long-term results of expansive open-door laminoplasty for cervical myelopathy-average 14-year follow-up study. Spine 31(26):2998–3005, 2006. 24. Chiba K, Toyama Y, Watanabe M, et al: Impact of longitudinal distance of the cervical spine on the results of expansive open-door laminoplasty. Spine 25:2893–2898, 2000. 25. Sani S, Ratliff JK, Cooper PR: A critical review of cervical laminoplasty. Neurosurg Ouart 14:5–16, 2004. 26. Hosono N, Yonenobu K, Ono K: Neck and shoulder pain after laminoplasty: a noticeable complication. Spine 21:1969–1973, 1996. 27. Wada E, Suzuki S, Kanazawa A, et al: Subtotal corpectomy versus laminoplasty for multilevel cervical spondylotic myelopathy: a longterm follow-up study over 10 years. Spine 26:1443–1448, 2001. 28. Morimoto T, Matsuyama T, Hirabayashi H, Sakaki T, Yabuno T: Expansive laminoplasty for multilevel cervical OPLL. J Spinal Disord 10:296–298, 1997. 29. Edwards CC, Heller JG, Murakami H: Corpectomy versus laminoplasty for multilevel cervical myelopathy: an independent matched cohort analysis. Spine 27:1168–1175, 2002. 30. Yonenobu K, Hosono N, Iwasaki M, Asano M, Ono K: Laminoplasty versus subtotal corpectomy: a comparative study of results in multisegmental cervical spondylotic myelopathy. Spine 17:1281–1284, 1992. 31. Kimura I, Shingu H, Nasu Y: Long-term follow-up of cervical spondylotic myelopathy treated by canal-expansive laminoplasty. J Bone Joint Surg Br 77:956–961, 1995. 32. Yoshida M, Otani K, Shibasaki K, et al: Expansive laminoplasty with reattachment of spinous process and extensor musculature for cervical myelopathy. Spine 17:491–497, 1992. 33. Goffin J, Van Calenberg F, Van Loon J, et al: Intermediate follow-up after treatment of degenerative disc disease with the Bryan cervical disc prosthesis: single-level and bi-level. Spine 28:2673–2678, 2003. 34. Duggal N, Pickett GE, Mitsis DK, et al: Early clinical and biomechanical results following cervical arthroplasty. Neurosurg Focus 17:62–68, 2004. 35. Bertagnoli R, Yue JJ, Pfeiffer F, et al: Early results after ProDisc-C cervical disc replacement. J Neurosurg Spine 2:403–410, 2005. 36. McAfee PC: The indications for lumbar and cervical disc replacement. Spine J 4:177s–181s, 2006. 37. Sekhon LH: Cervical arthroplasty in the management of spondylotic myelopathy: 18-month results. Neurosurg Focus 17:55–62, 2004.
CHAPTER
78
Hybrid Nonfusion Techniques Rudolf Bertagnoli
K E Y l
l
l
l
l
P O I N T S
The use of motion sparing and/or fusion technologies should be assigned based on the specific pathology or pathologies in a patient. Patients may require a combinaton of motion sparing or fusion technologies concurrently or during different temporal surgical times. The anterior and posterior spinal compartments should be carefully and individually evaluated. Select patients with prior fusions may be candidates for motion sparing surgery. Contraindications to motion sparing technology should be reviewed prior to any hybrid surgery technique.
Classical fusion technologies display a wide range of disadvantages such as obliteration of normal anatomy, elimination of mobility, and increased stiffness. Each consequence can increase the possibility of other long-term complications (“fusion diseases”), such as facet hypertrophy, facet arthritis, spinal stenosis, osteophyte formation, posterior muscular debilitation, and adjacent-level disc degeneration.1–6 In order to avoid (circumvent) these problems, nonfusion motion-preserving techniques (disc arthroplasty and posterior dynamic stabilization) have been developed that offer the possibility of achieving intersegmental stabilization while preserving/restoring physiologic mobility. In cases in which total disc replacement (TDR) technologies fail to produce the expected result, cannot address the complete disease process, or are contraindicated, hybrid techniques may be the resultant method of choice.
Single-Level Hybrid Constructs Single-Stage Motion-Preserving Hybrids: Types 1, 2a, 2b (see Table 78–1)
Because the motion segments consist of three moving areas, disc arthroplasty technologies (partial and total disc arthroplasty) may need to be combined with elements that achieve posterior stabilization, such as dynamic pedicle screw devices, interspinous devices, or facet replacements. Multistage Motion-Preserving Hybrids (Types 2a, 2b)
Patients who had a prior surgery in which a motion-preserving device was implanted can also receive an additional implantation at a later date. This is defined as “multistage” (Table 78–1, types 2a, 2b) and allows the combination of primary anterior or primary posterior technologies with secondary posterior or secondary anterior technologies. Multilevel Hybrids: Types 3, 4, 5a to 5c (see Table 78–2) The combination of anterior motion-preserving technologies (nucleus arthroplasty, total disc arthroplasty) with posterior motion-preserving technologies (interspinous devices, pedicle screw systems, facet replacement) at more than one level is termed a multilevel hybrid. With these combinations, a three-dimensional, motion-preserving, and biomechanically stable reconstruction of the affected motion segments can be achieved, and a physiologic range of motion can be reestablished. These multilevel motion-preserving hybrids can be classified in three subgroups: single stage type 3; fusion (single stage type 4); and previous fusion (multistage type 5a to c).
CLASSIFICATION OF HYBRID CONSTRUCTS In principle, hybrid constructs can be subgrouped into two categories: single-level hybrid constructs and multilevel hybrid constructs (Tables 78–1, 78–2), which may be single-stage and multistage. Multilevel hybrids can be subgrouped into pure motion-sparing technologies and a mixture of fusion and motionsparing technologies.
604
Motion-Preserving Hybrid: Single Stage Type 3
Anterior (nucleus arthroplasty, total disc arthroplasty) and posterior (interspinous devices, pedicle screw systems, facet replacement) technologies can be implanted in different levels and can be combined in a single stage. The dynamic treatment of all affected segments with motionpreserving technologies can be achieved with this type of hybrid.
CHAPTER 78
TABLE 78–1.
Single-Level Hybrid Constructs
Type
Timing of Surgical Procedures
Anterior
Posterior
1 2a 2b
Single stage Multistage Multistage
Yes Previous surgery Yes
Yes Yes Previous surgery
Motion-Preserving Hybrid: Fusion (Single Stage Type 4)
In some cases, it makes sense to apply a hybrid construct in which a motion-preserving device and a fusion technique are combined. If different pathologies are found within multiple motion segments, it is not advantageous to reconstruct only the most affected segment with motion-preserving technologies. This is the case if, for example, one segment shows severe spondyloarthritis, spondylolysis, and/or spondylolisthesis, the second segment has a reduced disc height of at least 50%, and the third segment shows a large central disc herniation (in conditions with no severe posterior
TABLE 78–2.
Multilevel Hybrid Constructs Type of procedure
Type
Timing of Surgical Procedures
Level X to Z
Level Y to Z
3
Single stage
Motion-preserving technology (anterior, posterior, or combinations)
4
Single stage
Motion-preserving technology (anterior, posterior, or combinations) Fusion procedure (any kind)
5a
Multistage
Previous surgery: fusion procedure (any kind)
5b
Multistage
5c
Multistage
Previous surgery: motion preservation (any kind) Previous surgery: motion preservation (any kind)
Motion-preserving technology (anterior, posterior, or combinations) Motion-preserving technology (anterior, posterior, or combinations) Fusion procedure (any kind) Motion preservation (any kind)
Hybrid Nonfusion Techniques
605
element disease at all levels). With the application of a fusion device in the lower or middle area, a mechanically stable construct can be achieved and should be taken into consideration. In these cases, single-stage surgery is preferred. Motion-Preserving Hybrid: Previous Fusion (Multistage Type 5a to c)
For symptomatic adjacent level instability, a surgery with implantation of a motion-preserving technology subsequent to a prior fusion surgery may be beneficial to the patient. The hybrid is then a combination of a fusion device with a motion-preserving technology implanted in different, consecutive sessions. Anterior and posterior technologies or combinations of single-level hybrids can be used in these cases. INDICATIONS AND CONTRAINDICATIONS Indications Indications for hybrid constructs are similar to those for the anterior or posterior motion-sparing technologies. The reconstruction of a motion segment should be analyzed on a three-dimensional basis, taking biomechanical and morphologic changes into consideration. Therefore, the degree of degeneration and the degree of mechanical insufficiency must be considered by the surgeon before deciding which combinations of technologies to use. The scenarios in which hybrid constructs might be beneficial for the patient typically involve multilevel-diseased spines in which all affected levels are symptomatic. Thus, multilevel degenerative disc disease (DDD) with or without degenerative spondylolisthesis, degenerative scoliosis, combinations of isthmic or hyperplastic spondylolisthesis with DDD-affected adjacent levels, and progressively degenerated motion segments above or below a fusion (TZS, transitional zone syndrome) are typical indications for hybrid constructs. Contraindications In general, the same contraindications that are valid for fusion and nonfusion technologies can be considered for hybrid constructs as well. Major contraindications are osteopenia (T-score < 1.5), osteoporosis, and other severe bone pathologies; infectious diseases; and severe psychosocial factors. These disorders significantly reduce the load-bearing capacities of the ventral bodies and their end plates. In fusion-only reconstructions, however, this is of less concern. Additionally, acute spinal fractures, spine tumor, and discitis should be excluded as well.
CASE STUDIES Posterior Dynamic Stabilization and Lumbar TDR Hybrid Construct Type 1: Dynesys System plus ProDisc-L Prosthesis A 68-year-old male patient underwent a previous fusion surgery at level L4-L5 (posterior lumbar interbody fusion) with two titanium cages (Fig. 78–1A, B). At 6 months after the surgery he developed continuous low back pain due to a failure of bony ingrowth. In a revision surgery with anterior explantation of the cages, a ProDisc-L prosthesis (Synthes, Inc., Paoli, PA) was implanted (Fig. 78–1C, D). In the same session, a posterior stabilization procedure with a modified Wiltse approach was used to implant a dynamic instrumentation system (Dynesys; Zimmer Spine, Inc., Warsaw, IN). The patient reported significant pain reduction a few days after the surgery with maintainance of relief at the 1-year follow-up.
606
P A R T
Hybrid Nonfusion Techniques
I X
F I G U R E 78–1. Preoperative radiographs: anteroposterior (A), lateral (B). Postoperative radiographs: anteroposterior (C), lateral (D).
n
Hybrid Construct Type 2a: ProDisc-L Prosthesis plus Coflex System A 44-year-old female patient underwent ProDisc-L surgery in level L4-L5. The prosthesis was placed too far posterior, impinging the existing nerve root and creating a segmental scoliosis with facet compression pain (Fig. 78–2A, B). Posterior repositioning of the ProDisc and implantation of a coflex (Paradigm Spine, LLC, New York, NY) interspinous implant were performed at level L4-L5 (Fig. 78–2C, D) to unload the facets. Patient reported improvement after surgery.
CHAPTER 78
Hybrid Nonfusion Techniques
607
F I G U R E 78–2. Preoperative radiographs: anteroposterior (A), lateral (B). Postoperative radiographs: anteroposterior (C), lateral (D).
n
Hybrid Construct Type 3: Dynesys System plus ProDisc-L Prosthesis (Three-Level Procedure) A 42-year-old male patient underwent laminectomy at L2-L3 and L4-L5 and a disectomy at L3-L4 with progressive low back pain (Fig. 78–3A, B). Discographically, L2-L3 and L4-L5 disc levels have been negative, and the L3-L4 disc was highly positive. In a single-stage procedure, first implantation of ProDisc-L prosthesis was done at level L3-L4, and then Dynesys was implanted at levels L2-L3 and L4-L5 (Fig. 78–3C, D).
608
n
P A R T
I X
Hybrid Nonfusion Techniques
F I G U R E 78–3. Preoperative radiographs: anteroposterior (A), lateral (B). Postoperative radiographs: anteroposterior (C), lateral (D).
CHAPTER 78
Hybrid Nonfusion Techniques
609
Hybrid Construct Type 5c A 46-old-female patient received a ProDisc-L at L5-S1 to relieve her low back pain (Fig. 78–4A, B). Three years later she developed a relevant vertical segmental instability at L4-L5 with back and leg pain. Our treatment choice was to perform a posterior dynamic stabilization with the DSS system (Abbott Spine PLC, Austin, TX) (Fig. 78–4C, D), because a posterior procedure is less risky than an anterior one.
n
F I G U R E 78–4. Frontal (A) and anterolateral (B) positioning of the ProDisc-L. C, D, Three years later,
implantation of DSS.
610
P A R T
I X
Hybrid Nonfusion Techniques
Interspinous Implants with Nucleus Replacement Hybrid Construct Type 2a: Nucleus Replacement with NeuDisc plus coflex System A 44-year-old female patient underwent nucleus replacement surgery with NeuDisc (Replication Medical, Inc., Cranbury, NJ) at level L4-L5 (Fig. 78–5A). Two years after the surgery, the patient developed a facet hypercompression pain syndrome at the same level that could be confirmed with fluoroscopically guided facet injections (Fig. 78–5B, C). Implantation of a coflex interspinous device (Paradigm Spine, LLC, New York, NY) decompressed the facet joint by maintenance of mobility for dynamic stabilization at level L4-L5 was carried out (Fig. 78–5D, E). Significant pain reduction was reported by the patient.
n F I G U R E 78–5. A, Magnetic resonance images after NeuDisc implantation. B, C, Magnetic resonance images after 2 years with facet hypercompression pain. Anteroposterior (D) and lateral (E) radiographs after coflex implantation.
CHAPTER 78
As a general rule for any surgical procedure, proper adherence to accepted indications and contraindications is crucial for achieving maximum success of the surgery and good postoperative results.
Hybrid Nonfusion Techniques
611
future, controlled single- and multicenter studies are required to confirm the first promising results from the use of hybrid constructs. It is important that the casual use of these constructs is strictly avoided.
CONCLUSION Treatment of only one moving zone of the three-column segment might be in some situations an insufficient solution. In these situations, the use of motion-sparing technologies in all three moving parts could be beneficial in complete reconstruction of the segment by maintaining motion. Especially in multilevel degenerative disc disease in which not all the segments have degenerated to the same degree or in which the pain generators are located in the front and back, very individualized selections of treatment options might be the best solution (multilevel motion-sparing hybrid). In contrast to these cases, highly degenerated or affected areas might need a fusion procedure that can now be combined with motion-sparing technology at another level. Based on these facts, there has been a dramatic expansion of indications toward patients with multilevel or multicolumn degenerative disease. Early experience with the expansion of indications by using combinations of motion-preserving technologies also revealed promising results. It is very important, however, to verify the results scientifically while performing very stringent patient selection for application of these hybrid combinations. This level of concern is a basic necessity as there are only limited nonrandomized data available for some of these new techniques, and therefore the likelihood for complications can increase drastically when combining them in one patient. Short-, intermediate-, and longterm complications of these combinations are yet unknown, and therefore a careful follow-up of patients is also necessary. In the
REFERENCES 1. Bertagnoli R: Disc surgery in motion. SpineLine 6:23–28, 2004. 2. Bertagnoli R, Kumar S: Indications for full prosthetic disc arthroplasty: A correlation of clinical outcome against a variety of indications. Eur Spine J 11(2):S131–136, 2002. 3. Hilibrand AS, Robbins M: Adjacent segement degeneration and adjacent segment disease: The consequences of spinal fusion? Spine J 4 (6):190S–194S, 2004. 4. Malter AD, McNeney B, Loeser JD, Deyo RA: 5-year reoperation rates after different types of lumbar spine surgery. Spine 23(7):814–820, 1998. 5. Bono CM, Lee CK: The influence of subdiagnosis on radiographic and clinical outcomes after lumbar fusion for degenerative disc disorders: An analysis of the literature from two decades. Spine 30(2):227–234, 2005. 6. Bertagnoli R: Review of modern treatment options for degenerative disc disease. In Kaech DL, Jinkins JR (ed): Spinal Restabilization Procedures: Diagnostic and Therapeutic Aspects of Intervertebral Fusion Cages, Artificial Discs and Mobile Implants. Amsterdam, Elsevier, 2002, pp 365–375. 7. Bertagnoli R, Tropiano P, Zigler J, et al: Hybrid constructs. Orthop Clin North Am 36:379–388, 2005.
CHAPTER
79
Dynamic Pedicle-Screw Stabilization with Nucleus Replacement Rolando Garcı´a and Brett A. Osborn
K E Y l
l
l
P O I N T S
The primary function of a nuclear replacement is to help resist axial compressive forces working in conjunction with the annulus. The application of hybrid technologies to the lumbar spine theoretically serves to counteract the destabilizing consequences of degeneration, trauma, and standard lumbar procedures. While there may be theoretical advantages, surgeons must resist the temptation to combine these technologies purely based on intuition.
The ideal spinal arthroplasty procedure aims to restore normal prepathology anatomy and biomechanics. Although certain individuals are appropriately treated with individual systems, other situations may require more than one technology to achieve this goal. The combined application of these technologies, however, may carry a potentially amplified risk of complication. An unanswered challenge, therefore, is defining the indications for hybrid technologies in the context of the associated risks. With thousands of nuclear replacement devices and posterior dynamic stabilization devices having been implanted, the major technical complications of these systems individually are now well established.1–5 A major concern with the utilization of nuclear replacement devices is the potential for migration and subsidence,6 but the concern of posterior dynamic stabilization is loosening of instrumentation and subsequent need for decompression or fusion.4 The application of hybrid technologies may theoretically reduce the incidence of complications of motion preservation technologies applied in isolation and potentially extend their inherent limitations. In a medical environment increasingly scrutinized for cost efficiency, it will be particularly challenging to implement hybrid protocols that will no doubt substantially increase cost.7 It will be particularly interesting to observe the coverage policies that will be established regarding the combination of implant technologies. Of course, there is precedent for optimism; currently, although the value of bone morphogenetic protein as an adjunct to posterior instrumentation in place of iliac crest bone is not firmly established, their combined use is gaining widespread popularity.8,9 612
One can see why when considering that the surgeon’s first responsibility is to offer and do what is best for a patient, regardless of effort or cost. In the same vein, it will be interesting to watch how different arthroplasty technologies will be combined at least in part as a result of the surgeon’s preference for certain manufacturers. Utilizing the metaphor that we must learn how to walk before we can run, most surgeons will likely develop considerable experience with individual systems before venturing into hybrid applications. The goal of this chapter is not to propose a manner in which to treat certain disorders with combined arthroplasty techniques but rather to help the reader question and then evaluate the merit and limitations of hybrid techniques. ANATOMIC, BIOMECHANICAL, AND PATHOLOGIC CONSIDERATIONS The intervertebral motion segment includes not only the disc and the facet joints but also a number of secondary structures such as the supraspinous ligament, interspinous ligament, as well as the origin and insertion of a variety of paraspinal muscles. Considering the complexity of this motion model, it should be no surprise that treating global segment instability with either an anterior or posterior approach can sometimes lead to failure. Furthermore, the effect of degeneration on the function of these anatomic structures must be understood in order to apply the proper corrective measures. In Newtonian terms, in order to properly stabilize a motion segment we must accurately define the instability forces and then apply equal but opposing forces. Failure to properly counteract the destabilizing forces may actually accentuate instability rather than reduce it. A spinal segment lacking axial stability, as is often seen following discectomy, will not be properly served by a system that applies compressive forces, such as a posterior dynamic stabilization. Similarly, a segment lacking posterior flexural control, as can be seen after laminectomy, would not be properly corrected simply by insertion of a nuclear replacement.
CHAPTER 79
The primary function of a nuclear replacement is to help resist axial compressive forces working in conjunction with the annulus. Biomechanical studies have shown that the axial forces on the disc are substantial. Wilke and associates reported that intradiscal pressure can be as high as 2,300 kPa with heavy lifting.10 In addition, studies have shown that the degree of load shared by the nucleus is proportional to its cross-sectional area.11 Although the disc’s primary role is that of shock absorption, the facet’s primary function is that of rotational control. Furthermore, the function of the facets is influenced by their position and degree of segmental degeneration. Early biomechanical testing by Adams and Hutton has shown that 15% of the total axial compression load is carried by the facets when the segment is in the neutral position and that this load can increase to as much as 40% when the segment is in extension.12 Destabilizing or instability forces can take many forms including traumatic, developmental, degenerative, and iatrogenic. Furthermore, destabilizing forces can be subclassified as either anterior or posterior. Prior to applying hybrid technologies of any sort, one must possess a sound knowledge of the anatomic and biomechanical properties of the spine and familiarity with the application of motion-preservation technologies in isolation. Specifically the facet joint orientation and its favored ranges of motion are crucial. The lumbar facet joints, for example, allow for flexion and extension and resist rotation. The intervertebral discs allow for polyaxial movement but primarily serve to resist compressive forces (axial loading). The facet joint architecture often is altered during posterior decompressive procedures. Similarly, during discectomies, the intervertebral disc (both annulus and nucleus) is violated. Partial removal of the posterior elements while gaining access to the spinal canal potentially destabilizes the vertebral unit. The altered motion dynamics may, via accelerated facet degeneration, lead to the development of refractory back pain in the postoperative period. Also, the ipsilateral discectomy potentially destabilizes the vertebral unit. The combined motion alterations, albeit subtle, may contribute to the high rate of recurrent disc herniation and refractory low back pain. Alluding to this is Barr’s published conclusion that “the thesis that every patient should have a spine fusion done at the time of laminectomy is tenable.”13 The application of hybrid technologies to the lumbar spine theoretically serves to counteract the destabilizing consequences of degeneration, trauma, and standard lumbar procedures. In order to select the proper combination of spinal arthroplasty systems, we must understand the limitations of each of the individual arthroplasty systems. In a recently published (2006) study by McAfee and co-workers, the authors concluded that implantation of a total disc replacement (TDR) does not restore rotational stability. Rotational motion is 120% to 140% after one-level TDR and 240% after two-level TDR.14 By not significantly disrupting the annular architecture, nuclear replacement following microdiscectomy should not alter rotational stability to the same degree as TDR, especially when supplemented by posterior dynamic systems. In fact, nuclear replacement may work in concert with the existing annulus and longitudinal ligmanents to restore segmental stability and motion.15
Dynamic Pedicle-Screw Stabilization with Nucleus Replacement
613
Although rotational instability is a major concern following TDR, sagittal motion behavior or misbehavior is not.16 In contrast, vertical translation is disproportionately magnified following discectomy and fenestrations, but axial hypermobility is not.17 The degree of disc and facet degeneration will play an increasingly important role in the selection process of the surgical approach. In a biomechanical and imaging study, Fujiwara and associates evaluated the effect of disc and facet degeneration on lumbar segmental motion.18 The authors found that while facet degeneration leads to increased axial motion, axial motion is most affected by disc degeneration. Tanaka and colleagues have reported that moderate disc degeneration is associated with greater segmental motion, and severe disc degeneration is associated with reduced segmental motion or stiffening. This is likely secondary to the interference between adjacent osseous structures.19 RATIONALE FOR HYBRID STABILIZATION In simple terms, the goal of hybrid stabilization is to reduce the limitations of the individual systems while broadening the indications or applications. In order for arthroplasty techniques to be successful, their relationship must be symbiotic and not parasitic. For example, the combination of interspinous stabilization with posterior pediclebased dynamic stabilization would be counterproductive because most interspinous devices work as flexion spacers, which would often limit the motion and function in the case of posterior dynamic stabilization. There are two basic questions that must be addressed before we can support the combination of nuclear replacement and posterior dynamic stabilization: 1. Can posterior dynamic stabilization decrease the rate of migration of nuclear replacement while augmenting stability? 2. Can nuclear replacement decrease the rate of screw loosening while providing anterior column support? Clinical trials addressing hybrid technologies have yet to be published. The aforementioned questions accordingly remain unanswered but essential to paradigm development as it applies to utilization of motion preservation technologies. REVIEW OF THE LITERATURE The history of spinal arthroplasty is the topic of another chapter in this book. From Fernstrom’s original work in the 1960s to Graf's work on ligamentoplasty, there is extensive literature on the individual use of nuclear replacement (NR) and posterior dynamic stabilization (PDS).20,21 However, there are little data available on the application of hybrid arthroplasty techniques. Although there have been anecdotal reports of the combined use of nuclear replacement with posterior dynamic stabilization, the authors were unable to find any peer-reviewed publications on this topic. There are no Class I or Class II studies on the use of nuclear replacement with concomitant posterior dynamic stabilization. Furthermore, it will prove particularly challenging to prove the value of such combination techniques for two primary reasons:
614
P A R T
I X
Hybrid Nonfusion Techniques
First, defining control group criteria will be difficult. Secondary is the foreseen difficulty of two competing instrumentation manufacturers combining their products (and efforts) in an Investigational Device Exemption (IDE) study. The authors believe that the hybrid use of NR and PDS would be best compared with a posterolateral fusion with pedicle screw instrumentation, utilizing newer less invasive techniques and bone morphogenetic protein. A comparison of a NR device in combination with a PDS system with a traditional and open ALIF (anterior lumbar interbody fusion) and PLIF (posterior lumbar interbody fusion) with pedicle screw instrumentation and iliac crest bone graft would hardly seem fair or appropriate. A major point that can be drawn from publications on the individual use of NR and PDS is the most common limitations and complications of these devices.1–5 In general, the major complications with nuclear replacement are device migration and subsidence.1,2 On the other hand, the major complications with posterior dynamic stabilization are screw loosening and persistent back pain.3–5 The major complication reported with the use of the PDN (prosthetic disc nucleus) has been device migration. Shim and associates reported on 46 patients followed for a minimum of 6 months. The authors reported four cases of device migration and one case of infection.22 In a prospective multicenter study by Stoll and associates, the authors reported on 83 consecutive patients who underwent flexible stabilization with the Dynesys system (Zimmer Spine, Inc., Warsaw, IN). Although the authors reported favorable clinical results, the authors reported a high rate of additional surgery: three required implant removal and conversion to fusion for persistent pain, seven required additional surgery for transitional level degeneration, one required a laminectomy at the operated level, and one required screw removal because of loosening. In addition, seven patients demonstrated signs of screw loosening.3 In a retrospective study by Grob and associates, the authors reported on 31 patients after an average of 2 years following Dynesys stabilization. The authors reported that during the follow-up period 42% of the patients underwent additional decompression. The authors also reported that 67% of the patients reported improvement in back pain, but 33% of the patients reported either no improvement or worsening of their back symptoms at follow-up. Similarly, 35% of the patients reported either no improvement or worsening of their leg symptoms.4 In a study by Schnake and associates, the authors reported on 26 patients with a minimum follow-up of 2 years who underwent Dynesys stabilization for degenerative spondylolisthesis. Again, although the authors reported good clinical results, they also reported an implant failure rate of 17%.23 INDICATIONS Once we have established the intellectual and clinical merit of combining nuclear replacement and posterior dynamic stabilization, we must define the procedure's indications. It would seem logical to look at the indications for nuclear replacement and posterior dynamic stabilization and combine these. However, the combined use may limit some indications while expanding other
applications. Furthermore, this chapter deals with the primary and initial combined application of these two techniques. An additional set of indications would be the use of one of the two technologies to “salvage” the motion segment after the failure of one of the individual systems. Indications for this hybrid procedure in the treatment of herniated nucleus pulposus (HNP) can be classified into two categories: primary and revision (secondary). Primary Indications Foraminal HNP Necessitating Vigorous Unilateral Medial Facetectomy
Treatment options for patients with a large foraminal HNP are many. However, most would agree that a discectomy would be the preferred procedure if the patient did not report significant back pain. However, most clinicians would be troubled by the need of some facet resection as well as by the removal of a substantial amount of disc material. In this circumstance, a TLIF (transforaminal lumbar interbody fusion), PLIF, or XLIF (extreme lumbar interbody fusion) would be strong considerations but all at the expense of motion. The isolated application of either NR or PDS in this clinical scenario would be technically feasible but not recommended. On the other hand, the combined application of NR and PDS is particularly appealing because the NR would help to restore disc biomechanics, and the PDS would help to counteract the destabilizing forces from the partial facet violation. HNP with Spondylolysis without Listhesis
The treatment options of a patient with a HNP at the level of a spondylolysis without spondylolisthesis include (1) limited discectomy either endoscopic or open, (2) discectomy and pars repair, and (3) and perhaps the most common, discectomy and fusion. Although all three of these options would address the problem, they all suffer from significant limitations. Discectomy alone may lead to progressive disc collapse and listhesis may ultimately lead to arthrodesis and segmental motion loss. Discectomy and pars repair may lead to bony nonunion or progressive disc collapse and spondylolisthesis. Fusion would, of course, address the problem but at substantial collateral sacrifice. The combined application of NR and PDS would be theoretically attractive because the NR would help to restore anterior column kinematics, and the PDS would help to augment posterior instability, decreasing the potential for progressive spondylolisthesis. HNP at the Apex of a Scoliotic Curve
Most surgeons would address this clinical scenario with either a limited decompression or with a decompression and fusion of the curve. Certainly the extent of the decompression and the magnitude of the deformity would be major determinants of the treatment choice. Nevertheless, a decompression at the apex of the curve, particularly in the presence of any rotatory listhesis, would likely lead to deformity progression and instability. The drawbacks of a multilevel fusion for a moderate deformity otherwise asymptomatic except for the stenotic level are obvious. In this particular
CHAPTER 79
presentation, the hybrid use of NR and PDS would work synergistically by restoring disc height and behavior, while the PDS could be used to help correct asymmetric disc height loss and provide supplemental rotational control.
Dynamic Pedicle-Screw Stabilization with Nucleus Replacement
615
posterior motion. As alluded to above, however, this option should only be considered in patients with a relative paucity of low back pain (relative to radicular pain). HNP Adjacent to a Fusion
HNP and Contralateral Facet (Synovial) Cyst
Two major options are currently used to address this situation: (1) a discectomy and decompression with excision of the facet cyst; (2) a decompression and fusion. The major determinants for the addition of the fusion to the decompression would be the presence of significant disc degeneration and back pain. In this case, the isolated use of NR would be at risk of failure given the need for bilateral decompression and some facet resection. The isolated use of PDS in this case would fail to address the anterior column destabilizing forces potentially leading to persistent mechanical back pain and screw loosening. The joined use of NR and PDS would potentially recalibrate the motion segment after discectomy and facet cyst excision by addressing both anterior column insufficiency as well as by augmenting the loss of posterior motion control. Large Central HNP
The treatment of a young patient with a large central disc herniation is particularly controversial. Although most patients do well with minimally invasive discectomy, surgeons grow particularly wary of removing substantial amounts of disc material. Although TDR is an attractive alternative to fusion in this patient subgroup, it suffers from two major disadvantages: (1) the risks associated with anterior surgery and (2) the limited ability to decompress the canal in a patient with a migrated, extruded, and potentially sequestered fragment. In this patient population, the application of a NR by itself would be of concern if the posterior longitudinal ligament is deficient and there is a large annular defect. The use of PDS in this application would not serve to properly address the anterior column insufficiency. In this application, the combined use of NR and PDS would allow for restoration of anterior column support with the NR while the PDS would help motion, which would in turn reduce the likelihood of device migration.
Most surgeons would consider an HNP adjacent to a fusion an indication for extending the fusion. However, others may argue that a minimally disruptive decompression carries little risk of subsequent instability and dysfunction. Newer technologies such as TDR or PDS are attractive choices for this application but experience in this application is limited. In this particular scenario, few, if any, would advocate the use of NR alone, but the combination of NR and PDS would offer substantial biomechanical advantages over PDS alone. “Salvage” Procedures
As stated above, NR and PDS systems may be hybridized (during revision surgery) when the individual systems, implanted in isolation, have failed. For example, in the event that a nuclear replacement migrates from the disc space, the surgeon may consider supplementation with a posterior dynamic stabilization system during reoperation. Similarly, NR systems may be implanted during a reoperative procedure targeting an extruded disc in a level previously stabilized by a PDS system. CONTRAINDICATIONS Contraindications can also be subclassified into two categories: relative and absolute. Absolute Contraindications l l l l l l l
Revision (Secondary)
l
Bilateral laminectomies with medial facetectomy Markedly collapsed/attenuated disc space Total or subtotal facetectomy Segmental instability on dynamic x-rays Grade II or greater spondylolisthesis Infection Osteoporosis Rheumatoid or inflammatory arthritis
Recurrent HNP
Currently the two main choices in patients with recurrent disc herniation are either a revision discectomy or a discectomy and fusion. Once again, these ends of the treatment spectrum seem sometimes to be either insufficient or overtreatment. The obvious concern with a revision discectomy is the potential for mechanical insufficiency and debilitating discogenic pain. On the other hand, a discectomy and fusion, particularly in a patient without significant back pain, seems overly aggressive. Although some would opt for a nuclear replacement device, the risk of device migration or failure would seem significant given the substantial annular defect and the violation of posterior bony structures usually necessary for safe removal of a recurrent HNP. The combination of NR and PDS would seem to address both the need for anterior column reconstitution as well as the posterior column deficiency while allowing for
Relative Contraindication l
Bilateral laminectomies (without significant facet disruption)
HYBRID PERMUTATIONS There are myriad nuclear replacement and posterior dynamic stabilization systems. Nuclear replacement options are listed in Table 79–1. Posterior dynamic stabilization systems are given in Table 79–2. CONCLUSION The field of spine arthroplasty is no longer in its infancy but rather in its adolescence, a time of rapid and rebellious growth. The early
616
P A R T
TABLE 79–1.
I X
Hybrid Nonfusion Techniques
Nuclear Replacement Options
Implant Name
Manufacturer
Biomet DASCOR NeuDisc NUBAC PDN Nucore SINUX
Regain Disc Dynamics Replication Medical Pioneer Surgical Technology Raymedica Protein Polymer Technologies Sinitec
TABLE 79–2. Posterior Dynamic Stabilization Systems Implant Name
Manufacturer
Dynesys Graf ligament SoftFlex Cosmic TFAS
Zimmer Spine SEM Sarl Globus Medical Ulrich GmbH Arcus Othopedics
promising experience with application of arthroplasty techniques such as nuclear replacement and posterior dynamic stabilization in isolation will undoubtedly lead to the combination of such technologies. Although there may be theoretical advantages, surgeons must resist the temptation to combine these technologies purely based on intuition. As the art and science of arthroplasty evolve and a myriad of motion preservation options becomes available, it will become increasingly challenging for surgeons to tailor the treatment to each patient. Interestingly, the future will likely show an increase in fusion of arthroplasty techniques in an effort to avoid fusion. REFERENCES 1. Ray CD: The PDN prosthetic disc nucleus device. Eur Spine J 11(2): S137–S142, 2002. 2. Bertagnoli R, Karg A, Voigt S: Lumbar partial disc replacement. Orthop Clin North Am 36(3):341–347, 2005. 3. Stoll TM, Dubois G, Schwarzenbach O: The dynamic neutralization system for the spine: A multi-center study of a novel non-fusion system. Eur Spine J 11(2):S170–S178, 2002. 4. Grob D, Benini A, Junge A, Mannion A: Clinical experience with the Dynesys semi-rigid fixation for the lumbar spine. Spine 30:324–331, 2005. 5. Gardner A, Pande KC: Graf ligamentoplasty: A 7 year follow up. Eur Spine J 11(2):S157–S163, 2002.
6. Bertagnoli R, Schonmayr R: Surgical and clinical results with the PDN prosthetic disc nucleus device. Eur Spine J 11(2): S143–S148, 2002. 7. Ray CD: The artificial disc: Introduction, history, and socioeconomics. In Weinstein JN (ed): Clinical Efficacy and Outcome in the Diagnosis and Treatment of Low Back Pain. New York, Raven, 1992, pp 205–225. 8. Vaccaro AR, Patel T, Fischgrund J, et al: A pilot study evaluating the safety and efficacy of OP-1 Putty (rhBMP-7) as a replacement of iliac crest bone autograft in posterolateral lumbar arthrodesis for degenerative spondylolisthesis. Spine 29(17):1885–1892, 2004. 9. Boden SD, Kang J, Sandhu H, Heller JG: Use of human bone morphogenetic protein-2 to achieve posterolateral lumbar spine fusion in humans: A prospective, randomized clinical pilot trial: 2002 Volvo Award in Clinical Studies. Spine 27(23):2662–2673, 2002. 10. Wilke HJ, Neel P, Caimi M, et al: New in vivo measurements of pressures in the intervertebral disc in daily life. Spine 24:755–762, 1999. 11. Nachemson A: Lumbar intradiscal pressure. In Jayson MIV (ed): The Lumbar Spine and Back Pain. London, Churchill Livingstone, 1987, p 191. 12. Adams MA, Hutton WC: The mechanical function of the lumbar apophyseal joints. Spine 8:327–330, 1983. 13. Barr JS: Ruptured intervertebral disc and sciatic pain. J Bone Joint Surg 29:210–214, 1947. 14. McAfee PC, Cunningham PW, Hayes V, et al: Biomechanical analysis of rotational motions after disc arthroplasty: Implications for patients with adult deformities. Spine 31(19):S152–160, 2006. 15. Ray CD, Hale JE, Norton BK: Prosthetic disk nucleus partial disk replacement: Pathobiological and biomechanical rationale for design and function. In Kim DH, Cammisa FP, Fessler RG (eds): Dynamic Reconstruction of the Spine. Thieme, 2006. 16. Cunningham BW, Gordon JD, Dmitriev EV, et al: Biomechanical evaluation of total disc arthroplasty: An in vitro human cadaveric model. Spine 28(20):S110–117, 2003. 17. Lu WW, Luk KD, Ruan DK, et al: Stability of the whole lumbar spine after multi-level fenestration and discectomy. Spine 24(13): 1277–1282, 1999. 18. Fujiwara A, Lim TH, An HS, et al: The effect of disc degeneration and facet joint osteoarthritis on the segmental flexibility of the lumbar spine. Spine 25(23):3036–3044, 2000. 19. Tanaka N, An HS, Lim TH, et al: The relationship between disc degeneration and the flexibility of the lumbar spine. Spine J 1(1): 47–56, 2001. 20. Fernstrom U: Arthroplasty with intercorporeal endoprosthesis in herniated disc and painful disc. Acta Chir Scan Suppl 357:154–159, 1966. 21. Graf H: Lumbar instability: Surgical treatment without fusion. Rachis 412:123–137, 1992. 22. Shim CS, Lee SH, Park CW, et al: Partial disc replacement with the PDN prosthetic disc nucleus device: Early clinical results. J Spinal Disord Tech 16(4):324–330, 2003. 23. Schnake KJ, Schaeren S, Jeanneret B: Dynamic stabilization in addition to decompression for lumbar spinal stenosis with degenerative spondylolisthesis. Spine 31(4):442–449, 2006.
CHAPTER
80
Simultaneous Lumbar Fusion and Total Disc Replacement Scott L. Blumenthal, Fred H. Geisler, Thomas F. Roush, and Donna D. Ohnmeiss
K E Y l
l
l
l
l
P O I N T S
In patients with two-level symptomatic disc degeneration, total disc replacement (TDR) may not be indicated at both levels. Combining total disc replacement at one level with fusion at another level may be a viable treatment alternative for two-level painful disc degeneration. Biomechanical studies lend support that combined TDR and fusion functions similar to a single-level fusion. There is relatively little data available on the use of hybrids at this time, but preliminary results are promising and suggest that if there is no contraindication for TDR at one level, this procedure yields acceptable results. The combination of TDR with fusion at separate levels may have a protective effect on the adjacent segment compared to a two-level fusion.
Total disc replacement (TDR) has yielded favorable results in patients with disc degeneration.1 One of the keys to successful outcome, however, is careful patient selection. This becomes more critical in patients with two-level degeneration. Even though one of the levels may be appropriate for TDR, the other may not be. Theoretically, the potential benefit for TDR may be greater in patients with two-level degeneration because a fusion (which accelerates degeneration at the adjacent segment) may compound the already present degeneration process. In a common case such as this, TDR may play a more important role. However, the enthusiasm to allow motion should not obviate strict adherence to selection criteria. In this chapter, we will review the biomechanical and clinical use of the “hybrid” spine procedure in which TDR is performed at one level and fusion at the adjacent segment during one operative setting (Fig. 80–1). Additionally, a brief review of the clinical experience of performing TDR adjacent to a previously fused segment will be provided. RATIONALE There are many patients with two-level symptomatic disc degeneration who have failed nonoperative management and are considered surgical candidates. Although one may be tempted to
perform two-level TDR for potential motion preservation at both painful levels, one must rigorously adhere to the appropriate disc arthroplasty screening regimen. In some of these patients with two-level problems, one level may be suitable for TDR, but the other level may not be. There may be osteophytes or significant facet joint changes, or the disc space may be too collapsed for the prosthesis. In addition, the anatomic environment during a revision operation in the case of prior fusion may be such that accurate placement of the prosthesis may be difficult or improbable and would otherwise compromise the prosthesis’ functionality. In such cases, TDR may be undertaken at the level appropriate for it and fusion at the level not appropriate for TDR. The options for fusion are the same as for any patient with respect to graft type, instrumentation, and the combination of a posterior fusion procedure. Huang and associates reported a relationship between the amount of motion at the level implanted with TDR and degeneration of the adjacent segment.2 In that study, they found that the segment above a TDR allowing greater motion was less likely to be degenerated than a disc above a TDR that allowed less motion. Similarly, in a study of dynamic posterior stabilization, it was reported that fewer segments adjacent to those treated with Graf ligamentoplasty were degenerated as compared to segments above a fusion at a minimum of 5-year follow-up.3 These studies lend clinical support for the concept that using dynamic stabilization at a lumbar level reduces the incidence of adjacent segment degeneration compared to the changes seen above immobile or fused segments. BIOMECHANICS The biomechanical concept for the potential benefit of combining TDR with fusion rather than performing a two-level fusion is that the TDR in at least one level will reduce stress on the adjacent segment. Previous biomechanical studies have shown that fusion increases the intradiscal pressure at levels adjacent to a simulated fusion.4–6 The pressure increased further when an additional 617
618
P A R T
I X
Hybrid Nonfusion Techniques
produced motion similar to the intact specimen.11 However, when the L5-S1 level was replaced with a simulated fusion, the total range of motion was decreased. Additionally, they reported that the L4-L5 level with the TDR did not increase motion to compensate for the reduction of motion at L5-S1. Grauer and associates used a finite element model to evaluate the effect of two-level TDR to a hybrid TDR and fusion combination on the implanted and adjacent segment.12 They found that the two-level TDR produced motion and facet loading values very different from those predicted from a combined fusion and TDR hybrid model. For the motion, the magnitude of change from the intact model was the same, but the pattern of change at the various segments were opposites. The potential clinical implication of the changes in motion patterns has yet to be established. CLINICAL EXPERIENCE
A
B
F I G U R E 80–1. A, Preoperative MRI showing disc degeneration at the L4-L5 and L5-S1 levels. B, Postoperative radiograph showing a CHARITE´ at L4-L5 and a STALIF at the L5-S1 level. n
segment was added to the simulated fusion.6 This finding may be suggestive that multilevel fusions have a greater potential for accelerated deterioration of adjacent segments. There has been little investigation on the effect of fusion on bone mineral density (BMD). In an early study using canines, it was suggested that the stress shielding caused by posterior instrumentation was related to reduced BMD.7 In a clinical study of patients undergoing combined anterior/posterior instrumented lumbar fusion at two levels, it was found that the BMD of the vertebral body above fusion decreased significantly as early as 3 months and remained so at 6 months.8 The authors contributed the decreased BMD to altered biomechanics caused by the twolevel combined instrumented fusion. Another study reported that BMD increased during long-term follow-up in patients undergoing single-level posterior fusion; however, their study involved only seven patients.9 The combined TDR and fusion may help prevent the potential for reduced BMD following fusion procedures. Serhan and associates performed simulated testing with various combinations of one- and two-level CHARITÉ (DePuy Spine Inc., Raynham, MA) and fusions to investigate the motion patterns of the operated and adjacent segments.10 They found that single-level TDR preserved physiologic motion, and with two-level TDR, the motion was altered only at the L4-L5 level. With fusion, the motion was significantly altered with one- or two-level surgery. When testing combined TDR and fusion, they found that fusion at L5-S1 combined with TDR at L4-L5 produced results similar to those for a single-level L5-S1 fusion. This is supportive of the use of hybrid surgery, that is, combining fusion at the lower level with TDR at the upper level, in patients with two-level symptomatic disc degeneration. In a cadaveric model simulating the treatment of two-level disc degeneration, Rivera and associates found that two-level TDR
There are a few reports of the use of TDR combined with fusion at different lumbar levels. Some of these reports deal with procedures performed at the same operative setting, and others are on the results of TDR when performed at the level adjacent to a previous fusion. The second group is primarily using TDR to treat adjacent-segment deterioration following an earlier fusion. In the area of “true hybrids,” that is, combining fusion with TDR at different levels, the indications for such may be a patient with two-level symptomatic disc degeneration, but in whom TDR is not indicated at one of the discs. Rather than fusing both levels, TDR may be used at the level appropriate for this procedure. Possible reasons for not doing TDR at one level are significant facet changes or severely collapsed disc spaces. There have been few reports on this application. In a preliminary report of our own experience with 17 patients,13 there was approximately a 50% reduction in Oswestry Disability Index (ODI) and VAS pain scores following the combined fusion-TDR procedure (Fig. 80–2). These results were very similar to those from the multicenter Investigational Device Exemption (IDE) trial involving only single-level TDR. Geisler and associates also reported favorable outcomes for this hybrid procedure in a series of 36 patients from
8 Pre-op 6 mo post-op 6 4 2 0
n
TDR + fusion
TDR above prior fusion
Blumenthal et al.; Spine, 2005
F I G U R E 80–2. VAS (visual analog scale assessing pain) data from our own experience with hybrid TDR and fusion.13 In both the TDR þ fusion group (performed at the same operative setting) and in the TDR above a previous fusion, the preoperative and 6-month postoperative VAS scores were very similar to the values reported from the FDA IDE trial.1
CHAPTER 80
10 centers.14 Aunoble and associates reported on a series of 45 patients undergoing fusion at L5-S1 and TDR with CHARITÉ at L4-L5 with a mean follow-up of 16 months.15 They noted no pseudarthrosis at L5-S1. The motion at L4-L5 was 8.4 degrees. ODI scores improved 29.6%. The VAS pain scores improved 39.1%, and SF-36 mental and physical components improved significantly. There have been two reports published dealing with the use of TDR next to a previously fused segment for the treatment of adjacent-segment deterioration. Bertagnoli and associates prospectively assessed the efficacy of treating a degenerated level adjacent to remote prior fusion with ProDisc (Synthes, West Chester, PA) arthroplasty in 20 patients with 2-year follow-up.16 The authors noted statistical improvements in VAS pain scores and ODI scores both at 3 and 24 months after arthroplasty. The patient satisfaction rate was 86% at 24 months. The authors suggested that patients should be screened carefully for evidence of facet joint impingement/degeneration, hardware-induced pain, and nonunion at prior fusion levels before undergoing disc replacement surgery, but that performing a disc arthroplasty adjacent to prior fusion was both safe and efficacious. Kim and co-workers also reported their experience with TDR at a level adjacent to a previous fusion.17 This was a series of five patients with at least 6-month follow-up. They found a decrease in ODI scores from 64 to 24. The authors were encouraged by these early results. In our clinic's experience with TDR at the level next to a previous fusion, despite heterogeneity in fusion constructs, there was an improvement in the VAS pain score of approximately 50%, which is similar to that reported for single-level TDR in patients with no previous fusion (see Fig. 80–2).13 In addition, the range of motion was similar to that for single-level TDR. SUMMARY Combining TDR with fusion, either concurrent or remote, appears to be a viable treatment option. The impending research in this area will undoubtedly play a critical role in the routine application of such hybrid technologies. Of particular interest will be the longevity of the prostheses, on which more demand may be placed from the adjacent-segment immobility. Many years must pass before such changes, if present, become evident. In the meantime, theoretical benefits based on current knowledge must be relied upon. At this time, it seems plausible that the incorporation of TDR may help avoid accelerated adjacent disc degeneration in patients with multilevel disc disease. Although the current clinical experience with hybrid TDR and fusion procedures is relatively limited, the results appear promising. The range of motion and clinical outcomes are comparable to those reported for single-level TDR in patients with no previous fusion.
Simultaneous Lumbar Fusion and Total Disc Replacement
619
REFERENCES 1. Blumenthal S, McAfee PC, Guyer RD, et al: A prospective, randomized, multicenter Food and Drug Administration investigational device exemptions study of lumbar total disc replacement with the CHARITÉ artificial disc versus lumbar fusion, Part I: Evaluation of clinical outcomes. Spine 30:1565–1575, 2005. 2. Huang RC, Girardi FP, Cammisa FP Jr, et al: Long-term flexionextension range of motion of the Prodisc total disc replacement. J Spinal Disord Tech 16:435–440, 2003. 3. Kanayama M, Hashimoto T, Shigenobu K, et al: Adjacent-segment morbidity after Graf ligamentoplasty compared with posterolateral lumbar fusion. J Neurosurg 95:5–10, 2001. 4. Cunningham BW, Kotani Y, McNulty PS, et al: The effect of spinal destabilization and instrumentation on lumbar intradiscal pressure: an in vitro biomechanical analysis. Spine 22:2655–2663, 1997. 5. Rao RD, David KS, Wang M: Biomechanical changes at adjacent segments following anterior lumbar interbody fusion using tapered cages. Spine 30:2772–2776, 2005. 6. Weinhoffer SL, Guyer RD, Herbert M, et al: Intradiscal pressure measurements above an instrumented fusion: A cadaveric study. Spine 20:526–531, 1995. 7. McAfee PC, Farey ID, Sutterlin CE, et al: The effect of spinal implant rigidity on vertebral bone density: A canine model. Spine 16:S190–197, 1991. 8. Bogdanffy GM, Ohnmeiss DD, Guyer RD: Early changes in bone mineral density above a combined anteroposterior L4-S1 lumbar spinal fusion: A clinical investigation. Spine 20:1674–1678, 1995. 9. Singh K, An HS, Samartzis D, et al: A prospective cohort analysis of adjacent vertebral body bone mineral density in lumbar surgery patients with or without instrumented posterolateral fusion: A 9- to 12-year follow-up. Spine 30:1750–1755, 2005. 10. Serhan H, Malcolmson G, Teng E, et al: Hybrid Testing for Adjacent- and Other-Level Effects Following Arthroplasty with the CHARITÉ Artificial Disc vs. Simulated Fusion. International Meeting on Advanced Spine Technologies. Athens, Greece, 2006. 11. Rivera Y, Mehbod A, Garvey T, et al: Two level disc disease: Two level disc arthroplasty versus a hybrid model. International Meeting on Advanced Spine Technologies. Athens, Greece, 2006. 12. Grauer JN, Biyani A, Faizan A, et al: Biomechanics of two-level CHARITÉ artificial disc placement in comparison to fusion plus single-level disc placement combination. Spine J 6:659–666, 2006. 13. Lhamby J, Guyer R, Zigler J, et al: Patients Undergoing Total Disc Replacement with Spinal Fusion at Different Lumbar Levels. International Society for the Study of the Lumbar Spine. Bergen, Norway, 2006. 14. Geisler F, Banco R, Cappuccino A, et al: Lumbar Total Disc Replacement Combined with Fusion at an Adjacent Level: Early Results from a Multi-Center Retrospective Review of a Hybrid Procedure for Multi-Level Degenerative Disease. International Meeting on Advanced Spine Technologies. Athens, Greece, 2006. 15. Aunoble S, Huec J-CL, Gornet M, et al: Hybrid surgery for DDD: Fusion L5-S1 and disc prosthesis L4-L5. Meeting, Spine Arthroplasty Society. Montreal, Canada, 2006. 16. Bertagnoli R, Yue JJ, Fenk-Mayer A, et al: Treatment of symptomatic adjacent-segment degeneration after lumbar fusion with total disc arthroplasty by using the ProDisc prosthesis: A prospective study with 2-year minimum follow-up. J Neurosurg Spine 4:91–97, 2006. 17. Kim WJ, Lee SH, Kim SS, et al: Treatment of juxtafusional degeneration with artificial disc replacement (ADR): Preliminary results of an ongoing prospective study. J Spinal Disord Tech 16:390–397, 2003.
CHAPTER
81
Repair and Reconstruction of the Annulus Fibrosus with the Inclose Surgical Mesh System Joseph C. Cauthen and Steven L. Griffith
K E Y l
l
l
l
l
P O I N T S
Reconstruction of the annulus fibrosus after lumbar discectomy is an important consideration, because current procedures without reconstruction might have unsatisfactory clinical outcomes requiring repeat surgery. Repair of the annulus fibrosus may reduce subsequent postoperative disc herniation and other secondary operations. Nucleus fragment removal with subsequent annular repair, rather than a more extensive nucleus removal without repair, is projected to reduce postoperative back pain that ultimately requires arthrodesis due to instability. Biomechanical, animal, and early human clinical studies of a surgical mesh implant (Inclose Surgical Mesh System) are favorable. Repair and reconstruction of the annulus with a surgical mesh system, such as Inclose, appears to be a fast, simple, and safe method to close a portal of re-herniation.
There is increasing awareness that postoperative clinical outcomes are less than satisfactory in many cases after discectomy surgery using current standard procedures that do not specifically address the opening in the annulus fibrosus.1,2 Recurrent disc herniation requiring additional surgery remains a common problem.3 Further, expulsion of prosthetic nucleus replacement devices has been a perceived negative factor in their development. Recent interest in improving these technical and clinical outcomes has focused on reconstruction and repair of the lumbar disc annulus. In this chapter, preliminary results describing the use of a surgical mesh placed below the aperture in the tissue surface will be presented and include animal and biomechanical studies. Preliminary human trial results will be described. The Inclose Surgical Mesh System (Fig. 81–1; Anulex Technologies Inc., Minnetonka, MN) is a mesh that supports soft tissue and can be used after disc decompression surgery. It can be inserted into the disc through a tear or incision in the annulus fibrosus. After appropriate positioning inside the intervertebral disc, it is deployed into its final configuration, expanding the implant to act as a barrier and eventually a tissue scaffold, thus containing intradiscal contents that remain after disc decompression.
INDICATIONS AND CONTRAINDICATIONS The Inclose Surgical Mesh System in the U.S. has a general indication “. . .to support soft tissue where weakness exists, or for the repair of hernias requiring the addition of a reinforcing, or bridging material, such as the repair of groin hernias.” In the European Union, the device's CE registration includes the repair of “the annulus fibrosus of the intervertebral disc.” The mesh is contraindicated where tissue may be contaminated or infected. And the mesh should not be used in infants or children in whom future growth may be compromised by its use. For complete indications, contraindications, potential adverse events, warnings and precautions, the product's Instruction for Use should be consulted. This chapter addresses a specific use for this system for repair of the annulus fibrosus. Additional information is being assembled to establish long-term outcomes.
DESCRIPTION OF THE DEVICE The Inclose Surgical Mesh System is composed of a braided mesh cylinder that is biocompatible and expandable. The basic material is polyethylene-terephthalate (PET). The nonexpanded 3.5-mm cylindrical implant is placed in the recipient’s disc site beneath the surface of the annulus by means of a disposable delivery tool. Once in place, the delivery tool is used to expand the implant, which conforms to the available evacuated nucleus cavity. In the cephalad-caudal direction, the disc space is defined by the end plates. It is recommended that at least 6 mm of disc height be available for adequate mesh deployment to avoid overconstraining mesh deployment and expansion. An integral latching mechanism holds the two ends of the cylinder together, forming an implant that acts as a barrier that is larger than the opening in the annulus. In a typical application, the existing pathologic annular tear or fissure is used as the point of insertion for the mesh implant. Alternatively, an incision into the annulus overlying a contained subannular nuclear fragment can be used. The expansion of the mesh into its configuration and latching of the cylinder constitutes 623
624
P A R T
X
Annular Repair
Mesh delivery tool
Surgical mesh Pledget T-anchor Anchor bands Mesh
Suture
Anchor band
Anchor band delivery tools n
F I G U R E 81–1. The Inclose Surgical Mesh System. The mesh is supplied sterile and premounted on a disposable delivery tool. Anchor bands are used to secure the expanded mesh to available surrounding tissue and below the aperture in the annulus.
a barrier to further displacement of intradiscal tissue or material. After satisfactory nuclear fragment removal and nerve root decompression, the mesh implant is further stabilized by tethering it to the surrounding tissue with suture anchor bands. SCIENTIFIC TESTING OUTCOMES A biomechanical study was conducted by Cunningham and associates using the device.4 This study evaluated segmental spinal mechanics after placement of Inclose surgical mesh and after cyclic loading. In this acute in vitro study, there was no measurable effect on the spinal mechanics compared to nonrepaired discs (i.e., stiffness, range of motion, or neutral zone) as a result of the placement of the device. This finding is meaningful because the reconstruction and repair of the annulus, in contrast to other intradiscal reparative
n
technologies such as nucleus pulposus replacements or total artificial disc prostheses, should ultimately not affect normal physiology. This study further demonstrated that the device remained in position below the annulus tissue surface after complex cyclic loading. No evidence of mechanical failure (fraying, unraveling, etc.) of the mesh was seen, and no significant particulate debris was noted. The device was further evaluated in a chronic animal model. The lumbar discs of Nubian cross-bred goats were implanted with smaller prototype mesh implants via a lateral retroperitoneal approach.5 This study demonstrated that the mesh barrier could be safely placed in proximity to the disc annulus. And when the mesh is properly positioned and affixed, radiographic evidence suggested the device remained in position during a 12-week observation period. Incorporation of the device into the annulus tissue was noted without deleterious effects on the surrounding tissues (Fig. 81–2).
F I G U R E 81–2. Histologic appearance of Inclose mesh. In an undecalcified histologic section from a goat model, the mesh can be seen adjacent to the inner annulus with extracellular matrix in and around the mesh fibers. No adverse tissue reactions were noted.
CHAPTER 81
Repair and Reconstruction of the Annulus Fibrosus with the Inclose Surgical Mesh System
625
CLINICAL PRESENTATION AND EVALUATION
OPERATIVE TECHNIQUES
Indications for the use of this device include signs and symptoms of displacement of the intervertebral disc (specifically nucleus pulposus tissue) corroborated by definitive radiographic images. In the ideal candidate, radicular symptoms are predominant and can include gluteal or leg pain, leg weakness or numbness, diminished reflexes, and positive straight-leg raising test. Leg pain should be qualitatively greater than the patient’s back pain complaint, thus suggesting nerve root compression or irritation. The pain level is typically intractable and must be concordant with the suspected site of nerve involvement. Additionally, failure of nonoperative measures, including strengthening exercise, epidural steroid injections, and avoidance of lumbar exertional stress, is a prerequisite before surgical intervention. Radiographic correlation to the symptom complex should be confirmed by magnetic resonance imaging (MRI), myelogram, or computed tomography (CT) (Fig. 81–3). Application of this particular device should be approached cautiously in patients who do not meet the above-mentioned criteria and in those patients who have normal neurologic findings, nonlocalizing pain, infection, significant comorbidities, or a target disc space that will not anatomically accommodate the surgical mesh as discussed next.
The use of this mesh implant is adjunctive to standard discectomy or microdiscectomy procedures, and, therefore, no special surgical techniques are needed. The implant has reportedly6 been used in more than 47 patients in both standard loupe-assisted and microscopeassisted discectomy as well as through endoscopic portal systems such as METRx (Medtronic Sofamor Danek, Memphis, TN). Under general anesthesia with the patient in the prone position with abdominal compressive forces minimized, the incision is limited to that required to accommodate the preferred open approach, microscopic approach, or minimal access endoscopic approach through a tubular retractor. The surgeon's preferred technique for posterior lumbar laminotomy and nerve decompression can be used. The ligamentum flavum is excised. Closure of epidural veins is achieved using microbipolar forceps allowing isolation and retraction of the dura and exposure of the disc annulus. With the nerve root retracted, a tear or fissure in the annulus fibrosus and herniated disc material is identified. Extradural and intradiscal fragments are removed with exploration of four quadrants of the epidural space. In the case of contained herniations, the annulus fibrosus is opened equidistant between the end plates using a vertical slit followed by evacuation of loose
n
F I G U R E 81–3. Exemplary magnetic resonance image of two-level disc herniation appropriate for Inclose mesh repair. In this clinical example, two lumbar disc herniations are seen at L4-L5 (a contained herniation) and L5-S1 (a sequestration). Adequate disc space for mesh deployment can also be appreciated.
626
P A R T
X
Annular Repair
nuclear fragments in the intradiscal space. After satisfactory disc removal and nerve root decompression has been completed, attention can turn to repairing the defect in the annulus fibrosus through which the herniation occurred. Proof of sufficient disc space height to accommodate the expanded mesh implant can be confirmed by probing the disc with an instrument of known dimensions. A minimum 3-mm opening in the annulus fibrosus is required for insertion of the unexpanded mesh implant. A minimum of 6 mm between the end plates is recommended to allow the mesh to expand and act as a flexible barrier (rather than a load-bearing implant); a tissue space constrained less than 6 mm does not allow the mesh to take the most appropriate shape or to retain its flexibility (tight space constraint, Fig. 81–4). Further qualitative assessment of the integrity of annulus fibrosus should be made to determine whether the annulus is sufficiently competent for annular reconstruction (i.e., the mesh will be attached to a firm tissue base) or if there might be attachment of the mesh to a patulent annulus. In the latter case or in the case of very narrow disc spaces, consideration should be given to an alternative annular repair technique. Tightening of a patulous annulus may best be achieved by reapproximation of the tissue layers with other reparative techniques. The Inclose surgical mesh is inserted through the fissure in the annulus. Visual depth confirmation is achieved by means of a yellow marker band on the shaft of the instrument. Maintaining the position of this yellow marker band at or near the tissue surface ensures anticipated placement of the deployed mesh below the surface. Deployment is best accomplished by both pulling the finger tabs on the delivery tool and readjusting the position of this yellow indicator band. An audible click is heard when the mesh is fully deployed and it has been latched. Upward pressure on the delivery tool is recommended to further seat the mesh on the inner annular wall.
Constrained
After the Inclose mesh is deployed below the tissue surface, the delivery tool is held firmly to assure close proximity to the annulus while medial and lateral suture anchor bands are used to tether the implant to the remaining firm annulus fibrosus. Suture anchor bands preloaded in disposable delivery needles are inserted into surrounding annulus tissue and ultimately through the mesh deployed below this tissue surface. A tissue depth indicator located on the shaft of the insertion needle tool protects overpenetration of the distal needle tip. While holding the tissue depth indicator on the tissue surface, the suture anchors are deployed by pulling the finger tabs up. This causes an anchor that is attached to a suture to be placed on the underside of the mesh; simultaneously a circular pledget located on the outer tissue surface and attached to the anchoring suture is cinched down, securing the tissue and the mesh implant. POSTOPERATIVE CARE There is no special perioperative or postoperative care required as a result of the use of this implant. After routine closure, the patient is discharged on the same day as surgery with no modification to standard postoperative precautions. Avoidance of lifting, stretching, or combining compression and twisting are recommended for a period of 12 weeks to allow for tissue healing in and around the implant. COMPLICATIONS AND AVOIDANCE Potential technical or perioperative complications as a result of the use of this mesh device can result if the anteroposterior dimensions of the disc are smaller than the length of the tools. For adequate placement of this device, a minimum depth of 34 mm from the tissue surface is required. Visual marks on the mesh's delivery
Constraint removed
9 mm Yes
6 mm Yes
4 mm No
n
F I G U R E 81–4. Inclose deployed shape comparison. On the left, Inclose mesh was deployed with the noted constraints, and on the right, the same implant is seen after removal of the constraints. With constraints 6 mm or greater, the mesh is adequately deployed in its intended shape and retains its flexibility. As noted in the 4-mm constraint, the mesh becomes overly compressed and can lose its flexibility.
CHAPTER 81
Repair and Reconstruction of the Annulus Fibrosus with the Inclose Surgical Mesh System
tool and a circular tissue stop on the anchor band delivery tools aid in avoiding overpenetration through the anterior side of the disc. Furthermore, probing the disc after decompression prior to mesh insertion with an instrument of known dimensions also can be used. Analysis of poor clinical outcomes and complications following laminotomy and partial discectomy is multifactorial. Because this device is adjunctive to the discectomy performed rather than an entirely new procedure, it is difficult to distinguish specific complications of the implant. When using the implant, careful patient selection and meticulous discectomy techniques are critical to prevent untoward or unsatisfactory outcomes. Patients whose symptoms are complex and cannot be corroborated with preoperative or intraoperative imaging studies should be considered a high risk for poor outcomes. In spite of the intended goal of avoiding repeat surgery, there is certainly the potential for complications requiring additional treatment in cases in which the patient’s pain is unresolved or returns. Because this implant is obviously a local barrier, if the implant is not properly placed spanning the opening in the annulus or the herniation, then the implant may not function as intended. The placement of this implant is through an in-line approach, and for instance, a far lateral herniation may not be adequately approached from a standard paramedial, interlaminar approach. As previously described, a patient whose disc space has narrowed below 6 mm should be approached cautiously. In its current configuration, Inclose may not expand effectively into the subannular space because of the constraints of the end plates. Furthermore, intraoperative assessment of the integrity of the annulus is warranted to achieve the best possible technical outcomes for maintaining the position of the mesh. Effective fixation and ultimately the positional stability of the implant may be compromised if these anatomic limitations are not appreciated. As with any other application of a surgical mesh that supports soft tissue, there is the potential for inadequate attachment or fixation of the implant because of lack of overlying tissue (< 2mm) or generalized tissue incompetency.7 It should come as no surprise that the environment of the soft tissue of the intervertebral disc annulus is not substantially different from other soft tissues in the body. If the annulus soft tissue cannot support fixation of
ADVANTAGES/DISADVANTAGES: INCLOSE SURGICAL MESH SYSTEM Advantages Rather than leave tears, holes, or fissures in the annulus fibrosus unclosed after discectomy, the Inclose Surgical Mesh System can be used to support the soft tissue of the annulus. Placement of the Inclose mesh implant does not add significant surgical time and can be performed quicker than manually suturing the tissue. Placement of the Inclose mesh below the tissue surface and expansion of the mesh greater than the annular tear, fissure, or incision facilitate positional stability. Disadvantages An incompetent, patulent annulus fibrosus or a degenerative, narrowed disc space is not adequate for this type of annular repair using a flexible barrier implant.
627
the mesh or the opening is too large relative to the dimensions of the mesh, then there is the potential for re-extrusion of the implant, particularly in light of intradiscal pressures. In cases in which a secondary surgery is required, the device can be removed if needed. Furthermore, should additional intradiscal decompression be required, the implant would have to be retrieved prior to disc excision. Depending on the timing of any repeat surgery, re-exploration would not be unlike other repeat discectomies in which the surgeon would have to contend with fibrotic scar. CONCLUSION Lumbar discectomy or microdiscectomy is considered to be a successful operation both in terms of technical ease and clinical outcomes. But the wide variety of approaches and techniques used can be debated with respect to their advantages and disadvantages. Simply removing the offending nucleus pulposus fragment without an extensive intradiscal decompression is favored by many surgeons in attempts to maintain disc height. Other surgeons rely on a more aggressive, yet subtotal, disc decompression in attempts to decrease the likelihood that reherniation might occur. Carragee and associates8 studied this technical aspect prospectively in 86 patients. Two-year follow-up suggested that a “limited discectomy” may result in better patient outcomes, but the reherniation rate in his study was twice as high as in those patients who had subtotal discectomy. These authors ultimately suggested that an effective barrier may be clinically useful. Description of advantages and disadvantages of a particular device or technique implies comparisons. In the case of annular repair with a barrier, the Inclose Surgical Mesh System is one of the first implants to be used clinically for repairing the soft tissue of the intervertebral disc. Therefore, benefits of this implant are most appreciated in comparison to the current surgical technique of leaving the annulus fibrosus unclosed at the conclusion of the discectomy procedure. The potential advantages of repairing the annulus following discectomy have been suggested previously9,10 and most extensively studied by Cauthen.11 In his study of 254 patients with tedious manual suture closure of the annular fissure with or without a fascial autograft patch, he showed a greater than 50% reduction in disc reherniation rates over a 2-year period. In contrast, Ahlgren and associates12 used studies in animal models to suggest that the healing response of the annulus is not influenced by directly repairing the fissure with sutures or with a muscle-fascial overlay graft. The ability to achieve closure of the annulus fibrosus with Inclose mesh is a significant advance over the previously described suture methods. Because of the engineering in the delivery instruments, the implant does not add significant time to the procedure. The device can typically be delivered and secured in place in 5 minutes or less. Expansion of the mesh device larger than the opening in the annulus and below the tissue surface further facilitates its function as an effective barrier. Data from biomechanical testing,1 laboratory testing in animal implantation trials,2 and initial clinical experience13 suggest that a surgical mesh barrier such as Inclose placed into the intervertebral disc affixed with suture anchor bands has a reasonable chance
628
P A R T
X
Annular Repair
of providing a safe, simple, and fast solution to the problem of recurrent disc herniation in the human spine. As in any pioneering effort, long-term conclusions are not yet possible due to the limited length of follow-up, but early indications strongly suggest that this device is stable and may provide a mechanical barrier to secondary disc herniation. Additional potential benefits of annular repair techniques such as this include the reduction of chemical pain mediators escaping through the open annulus following conventional discectomy, thus reducing secondary effects of irritation of the nerve root ganglion, epidural fibrosis, or the potential for ingrowth of nerve fibers that are mediators of postoperative discogenic pain.14–16 REFERENCES 1. Asch HL, Lewis PF, Moreland DB, et al: Prospective multiple outcomes study of outpatient lumbar microdiscectomy: Should 75 to 80% success rates be the norm? J Neurosurg 96(1):34–44, 2002. 2. Fritsch W, et al: The failed back surgery syndrome—Reasons, intraoperative findings, and long-term results: A report of 182 operative treatments. Spine 21:626–633, 1996. 3. Atlas S, et al: Long-term outcomes of surgical and nonsurgical management of sciatica to a lumbar herniation: 10 year results from the Maine Lumbar Spine Study. Spine 30:927–935, 2005. 4. Cunningham BC, Hu N, Beatson H, McAfee P: Acute in vitro stability of an annular repair device after multi-directional cyclic fatigue evaluated in human specimens. Presented at the Global Symposium on Motion Preservation Technology (SAS), New York, NY, 2005. 5. Peppelman W, Davis R, Sherman J, et al: Feasibility results of a novel anular repair device in a goat model. Presented at the World Spine III—An Interdisciplinary Congress on Spine Care, Rio de Janeiro, Brazil, 2005.
6. Sherman J, Bajares G, Cauthen J, et al: Evaluation of a mesh device to repair the anulus fibrosus. Presented at the 13th Annual International Meeting of Advanced Spinal Techniques (IMAST), Athens, Greece, 2006. 7. Luijendijk RW, Hop WCJ, van den Tol P, et al: A comparison of suture repair with mesh repair for incisional hernia. N Engl J Med 343(6):392–398, 2000. 8. Carragee EJ, Spinnickie AO, Alamin TF, et al: A prospective controlled study of limited versus subtotal posterior discectomy: Shortterm outcomes in patients with herniated lumbar intervertebral discs and larger posterior anular defect. Spine 31(6):653–657, 2006. 9. Yasargil MG: Microsurgical operation of herniated lumbar disc. Adv Neurosurg 4:81, 1977. 10. Lehmann T, Titus MK: Refinements in technique for open lumbar discectomy. Proceedings of the International Society for the Study of the Lumbar Spine (ISSLS), June 1997. 11. Cauthen JC: Microsurgical annular reconstruction (annuloplasty) following lumbar microdiscectomy. In Guyer RD (ed): Spinal Arthroplasty: A New Era in Spine Care. St. Louis, Quality Medical Publishing, 2005, pp 156–177. 12. Ahlgren BD, Lui W, Herkowitz, HN: Effect of annular repair on the healing strength of the intervertebral disc—A sheep model. Spine 25 (17):2165–2170, 2000. 13. Bajares G, Perz-Oliva A: A pilot study evaluating a novel device for annular repair following spinal discectomy. Presented at the Global Symposium on Motion Preservation Technology (SAS), New York, NY, 2005. 14. Olmarker K: Neovascularization and neoinnervation of subcutaneously placed nucleus pulposus and the inhibitory effects of certain drugs. Spine 30(13):1501–1504, 2005. 15. Nygaard OP, Mellgren SI, Osterud B: The inflammatory properties of contained and noncontained lumbar disc herniations. Spine 22:2484–2488, 1997. 16. Coppes MH, Marani E, Thomeer RT, et al: Innervation of “painful” lumbar discs. Spine 22:2342–2349, 1999.
CHAPTER
82
The Intrinsic Therapeutics Barricaid Device Jacob Einhorn, Oscar Yeh, and Greg Lambrecht
K E Y l l
l l l
P O I N T S
Annular repair with the Barricaid may reduce reherniation risk. Disc height has been better maintained in prospective, controlled clinical trials. The Barricaid covers the entire posterior annulus. The Barricaid requires limited removal of nucleus. Prospective clinical trials are currently under way.
l
THE CLINICAL PROBLEM Over 275,000 lumbar discectomy procedures were performed in the United States in 2005,1 with an additional 260,000 projected outside the United States.2 These patients generally present with radicular symptoms or low back pain that is unresponsive to conservative therapy and magnetic resonance imaging (MRI) or computed tomography (CT) confirmation of a nuclear herniation or extrusion as the immediate cause of their symptoms. Although lumbar discectomy is frequently successful in immediately reducing symptoms of sciatica and radiculopathy, over time many patients suffer a relapse in symptoms, requiring reoperation. One prospective, multicenter study of lumbar discectomy patients reported a 20% rate of reoperation within the first 5 years,3 and another reported a 13% rate within the first 2 years.4 The discectomy procedure itself, despite its short-term efficacy, is not benign to the disc—it is an ablative, not a restorative procedure. Additional nucleus from within the disc space is frequently removed, and the weakness or defect in the annulus, through which the nucleus bulged, protruded, or extruded, remains, or is even enlarged, following discectomy. Both create very real risks for the patient: 5 l Reherniation: Rogers reported a 21% reherniation rate with a fragmentectomy. Carragee and associates6 reported a 9% reherniation rate in a study of 187 patients done with limited nuclear removal. Asch and colleagues4 report a 13% reoperation rate for recurrence of herniations in a study of 212 patients in a multicenter study. l Loss of disc height: Almost all discectomy patients lose disc height after surgery,7 a finding that is not surprising given that material has been removed from their disc, and the hole
l
in the annulus has remained or has been enlarged. Laboratory studies have shown that disc height loss leads to overstressing and degeneration of the annulus,8,9 while clinical studies have shown that disc height loss of greater than 25% after discectomy is correlated with poor clinical outcomes.10,11 Increased back pain: As described earlier, most discectomy patients lose disc height, and those with the most disc height loss have significantly worse clinical outcomes. A large component of this degradation of outcomes is due to increased back pain, which has been reported in greater than 12% of discectomy patients.10,12 Recurrence of symptoms: Although discectomy is quite effective in the short term, only about 70% of primary discectomy patients report improvement in their predominant symptoms several years following surgery.4,10
ANNULAR REPAIR Patient Characteristics and Surgical Technique Several prospectively enrolled studies have correlated patient- as well as technique-related risk factors with poor outcomes following discectomy: l Size of defect: Not surprisingly, the size of the defect found at the time of primary discectomy has a huge effect on the reherniation rate. Carragee and associates reported an overall reherniation rate of 9% in 187 patients. However, in patients with massive annular defects (defined as larger than a Penfield-1 probe [6.5 mm]), the reherniation rate was 27%, but in patients with slit defects the reherniation rate was only 1%. This would indicate that if larger defects could be effectively reduced in size, reherniation rates could fall dramatically in this at-risk group. l Amount of nucleus removed: Aggressive nuclear removal can also impact the risk of reherniation. In a follow-up study, Carragee and associates compared a group of patients in whom nucleus was aggressively removed with an historical cohort from their previous study in whom only the offending 629
P A R T
630
X
Annular Repair
fragments of nucleus were removed.13 They found that more limited nucleus removal resulted in a doubled risk of reherniation (18% vs. 9% within the first 2 years), while aggressive nucleus removal resulted in significantly higher levels of back pain, worse clinical outcomes (per Oswestry Disability Index), prolonged delay of return to work, and higher levels of required pain medication at 1 year. These data highlight the dilemma a surgeon faces when performing a discectomy: On the one hand, aggressive nuclear removal can reduce reherniations but has been correlated with worse clinical outcomes and increased disc height loss. On the other hand, limited nuclear removal, which results in greater maintenance of disc height and better clinical outcomes, leads to a significantly higher risk of symptomatic reherniation. In short, for a given herniation type, making the annular defect smaller could reduce reherniation rates. A device that closed and reinforced the annular defect could allow the surgeon to retain as much native nucleus as possible without an increased risk of reherniation. Such a device would provide surgeons and patients with a new option that results in both the best long-term clinical outcome and a reduced risk of reherniation. Challenges The goal of achieving a lasting repair or closure of defects in the annulus is not an easy one. This is in large part due to constraints imposed by the harsh intradiscal environment. These constraints include high pressures, large ranges of motion, lack of natural healing, a wide variety of defect sizes and location, and the risk of weakness on the contralateral side not visible at the time of surgery. l High intradiscal pressures: The highest pressures in the human body have been measured in the lower lumbar discs, with pressures as high as 2.3 MPa (334 psi) having been recorded in vivo during activities of daily living.14 These high pressures can cause intradiscal devices to be expulsed from the disc space at reported frequencies up to 38% in some studies.15 Any device intended to close defects in the annulus must remain in place when challenged with these high intradiscal pressures. l High ranges of motion: Even though axial rotation and lateral bending in the lumbar discs are relatively limited, the range of motion of the disc in flexion-extension can be as high as 24 degrees.8 Because the center of rotation of the disc is typically anterior of the posterior annulus, the strain range of the posterior annulus during these high ranges of motion can often be on the order of 100% (e.g., for a posterior disc height of 4 mm, during flexion-extension this height may go down to 2.5 mm and up to 6.5 mm, for a total strain range of 4 mm). Designing a device that can maintain its integrity while sealing defects that also go through this strain range is extremely challenging. l Poor primary annular healing: Because of the extremely high intradiscal pressures, the high range of motion, and the avascular nature of the disc, the annulus does not heal well. Heggeness and associates16 reported that posterior extravasation
l
l
of discography dye is significantly more likely (P ¼ 0.025) in a postdiscectomy disc than in an unoperated disc. Ahlgren and associates17 reported that attempts at direct repair of annular incisions (slit, cruciate, or box) did not significantly affect the strength of the annulus after discectomy as judged by the disc's ability to maintain intradiscal pressure. A successful annular repair device will likely not be able to rely on primary healing of the annulus itself. Variety of defect locations: The location of the naturally occurring annular defect may range from perfectly central to far lateral. Ninomiya and Muro demonstrated that over 80% of defects are primarily or entirely between the facets.18 As surgical exposure is limited, the ideal annular repair device would protect aspects of the annulus that are not immediately in front of the surgeon's field of view. Risk of contralateral herniations: The main goal of annular repair is to contain nuclear material within the disc space following a primary discectomy. As discussed previously, a recurrent herniation occurs in a significant number of these patients. However, this recurrence is not always through the same annular defect, or even on the same side of the disc. Contralateral herniations have been reported in prospective studies to account for up to 37% of recurrent herniations that occur at the operated level.19 An annular repair device that can protect the contralateral side of the disc, and not just the side of primary access, from recurrence would be ideal.
Product Requirements Based on the clinical problems and challenges described here, as well as the ever-present clinical desire to maintain function and reduce operative trauma, the ideal annular repair device would have the following characteristics: l Prevent reherniation l Preserve disc height and function l Cover defects of a variety of sizes and locations l Protect as much of the posterior annulus as possible l Require minimal or no removal of nucleus l Easily used with current surgical approach and techniques l Have minimal impact on operative time l Endure the high ranges of motion and pressure without failure or migration DESCRIPTION OF THE DEVICE The Intrinsic Therapeutics (Woburn, MA) Barricaid annular repair device (Fig. 82–1) is designed for use as an adjunct to lumbar discectomy and is intended to close annular defects and support the weakened annular soft tissue. The Barricaid implant serves as an internal patch that is positioned between the annulus and the nucleus and covers substantially all of the posterior annulus (Fig. 82–2). Figure 82–3 shows a herniated disc with nucleus material extruding out of the disc (left), and the defect repaired with the Barricaid device (right). The Barricaid consists of a flexible nitinol support frame surrounded by an e-PTFE sleeve, which rests along the posterior
CHAPTER 82
n
F I G U R E 82–1. Intrinsic Therapeutics Barricaid annular repair
device.
annulus after insertion into the disc. The frame has anterior projections that aid with insertion and serve to stabilize the device in its final position. These projections are connected by a PTFE link and stabilizing ring. The materials used in the Barricaid were chosen for their unique mechanical properties, excellent biocompatibility, and long clinical implant history. Nitinol is a nickel-titanium alloy known mainly for its shape-memory and superelasticity, and is widely used in implantable medical devices such as stents and bone anchors. Nitinol provides the Barricaid with the strength to resist the extreme pressures within the disc and the flexibility to respond to motions of the treated level. PTFE (also known commonly by the trade names Teflon and Gore-Tex) is an inert polymer that is widely used in implantable medical devices such as abdominal hernia meshes, stent grafts, and dural substitutes. PTFE augments the Barricaid's ability to seal against extrusion of nucleus and increases compliance to the surrounding tissues. OPERATIVE TECHNIQUE The Barricaid annular repair device is implanted via a standard open posterior discectomy approach, with or without the
n
F I G U R E 82–2. Rendering of the Barricaid in its implanted position between the annulus and nucleus, spanning the defect, and covering substantially all of the posterior annulus.
The Intrinsic Therapeutics Barricaid Device
631
assistance of a microscope. The Barricaid can be implanted through a defect as small as a 5 mm 5 mm cruciate cut, and can be used to repair defects that are up to 6 mm high and 10 mm wide. As part of a limited discectomy, the offending fragment of nucleus is removed prior to implantation. No additional intradiscal nucleus removal is required to place the Barricaid. Indeed, one of the benefits of the Barricaid is that it enables the retention of intradiscal nucleus without an elevated risk of recurrent herniation. Following removal of the nuclear fragment and prior to insertion of the implant, two measurements are made using custom tools supplied by Intrinsic Therapeutics. The first is the thickness of the posterior annulus, to ensure that the implant is placed along the border between the annulus and the nucleus; the second is the internal width of the disc, to ensure that the appropriate size Barricaid is implanted to maximize the coverage and protection of the posterior annulus. Once these measurements are made, the implant is folded into the delivery instrument and implanted in the appropriate location. The entire process of measuring the annular thickness and disc width and implanting the device takes approximately 5 to 10 minutes. No changes to postoperative care are required and the patient can return to normal activities per the advice of the treating physician.
PRECLINICAL TESTING Expulsion from the Disc Space An extremely aggressive mechanical environment exists in the lumbar intervertebral disc. In vivo intradiscal pressures have been reported to be as high as 23 atm (338 psi).14 The Barricaid serves as a patch that is positioned along the inside of the posterior of the damaged or weakened annulus. The pressure of the nucleus acts to push the Barricaid against the posterior annulus and hold it in place. This intradiscal pressure may also act to expel the Barricaid from the disc space, which represents the device's primary mode of failure in terms of safety. We test the device's resistance to this failure mode in two ways: (1) on the benchtop in a simulated disc model and (2) in a challenging fresh-frozen human cadaver model, which tests for expulsion resistance following a series of loaded and unloaded motions designed to challenge the stability of the implant. All cadaver and benchtop expulsion testing is done with an annular defect that is 6 mm tall by 10 mm wide, and all testing is conducted at 37 C. Intradiscal pressure is monitored throughout all testing, which is performed to 23 atm (338 psi). When testing intradiscal implants for resistance to expulsion, no test should be considered a “success” unless the appropriate amount of pressure is achieved, and this pressure is accomplished in a disc oriented such that the pressure is acting to expel the implant. The efficacy of the device in preventing extrusions of nuclear material has also been assessed by performing cadaveric expulsion testing in self-matched specimens. Each specimen was tested in two rounds—first with the device implanted and then with the device removed. A measure of efficacy is obtained by comparing the pressures at which nuclear extrusions occurred in these two scenarios.
P A R T
n
X
Annular Repair
F I G U R E 82–3. Posterior view of a herniated disc (A) and a disc whose annulus has been repaired with the Barricaid device (B).
Cadaver Testing
Loads and moments on the spine have largely been inferred, and no specific combination of loads and specimen orientation reliably produce nuclear extrusions in an in vitro biomechanics model. We have gone to great lengths to develop a cadaver model that can reliably be used to create nuclear extrusions,20 and we use this model while continuously monitoring intradiscal pressure to challenge the resistance of the Barricaid device against expulsion from the disc. The Barricaid implant has been implanted in over 150 freshfrozen cadaver specimens. Over 60 of these have been dynamically tested, with an axial preload of 500 N (to simulate body weight) and unconstrained applied moments of 10 Nm. These motions and loads are applied for several hundred cycles in various orientations—flexion-extension, lateral bending, and flexion combined with each of right and left lateral bending. After this motion testing, the disc is flexed such that the intradiscal pressure is pushing the nucleus and Barricaid posteriorly, and the particular combination of flexion angle and load vector is identified such that intradiscal pressure is maximized. This worst-case orientation is then used to compress the disc until at least 23 atm (338 psi) of pressure is reached, typically around 3 to 5 kN (675 to 1125 lbf). For a test to be considered a success, this pressure must be reached, and the Barricaid implant must remain in position. If a specimen cannot generate this high pressure, or if the specimen fails prior to reaching this pressure (e.g., through an end plate fracture), the device has not been appropriately challenged, and this data point cannot be counted as a success. All the Barricaid implants survived this aggressive testing without incidence of expulsion from the disc space. To perform efficacy testing, nuclear fragments from an adjacent level were inserted into the disc space from an anterior approach.21 Use of these fragments increases intradiscal pressure, and the
mobility of the fragments fully challenges the device. The fragments were placed anterior to the device after being stained to differentiate the fragments from native nucleus and soaked in radiopaque dye for visualization under fluoroscopy. Expulsion testing, as described earlier, was performed, and the pressure at which extrusion of nuclear material occurred was recorded. After this initial round of testing, the implant was carefully removed. If nuclear material had extruded, an equivalent volume of nuclear material was then inserted back into the disc space. Expulsion testing was then repeated and acted as a control for the first round of testing. In a group of eight specimens, the Barricaid device increased extrusion pressure by at least three times (P < 0.0001; Fig. 82–4). During the initial round of testing (i.e., with the Barricaid implanted), extrusions did not occur in all specimens. Testing was stopped at 25 atm of intradiscal pressure in order to preserve the specimens for further testing, so actual extrusion pressures for the implant group were even higher.
30 25 Pressure (atm)
632
20 15 10 5 0
n
Barricaid
Control
F I G U R E 82–4. Intradiscal pressure at which herniation
occurred with and without the Barricaid.
CHAPTER 82
Benchtop Testing
The benchtop expulsion testing is done following fatigue testing of the Barricaid implant, under the assumption that devices that are challenged by fatigue testing would be theoretically weaker than a virgin device would be against expulsion. All devices are first implanted into a plastic disc model. There are two types of fatigue testing to which implants are subjected in vivo: pressure- (or load-) driven fatigue due to cyclic changes in intradiscal pressure, and displacement-driven fatigue due to cyclic motions of the vertebral bodies.
Testing Implant Fatigue Intradiscal Pressure Changes
As the intradiscal pressure increases and decreases as a result of various activities of daily living, the Barricaid implant may be cyclically deformed and could fatigue as a result. Such fatigue could theoretically result in fragmentation of the Barricaid or decreased resistance to expulsion from the disc space, both of which would constitute device failures. Resistance of the Barricaid implant against fatigue due to this cyclic pressurization is tested as part of our “Cyclic Pressure Fatigue Test.” Intradiscal pressures recorded in living humans in vivo have been reported for a number of different activities of daily life.14,22 These activities include four general categories: sitting/standing, walking/jogging, climbing stairs, and lifting objects with various techniques. We tested the Barricaid implant in cyclic pressure fatigue in the greatest pressure ranges in each of these four categories, with the following exception: lifting 20 kg with good technique was substituted for climbing stairs to yield an even more rigorous loading protocol. A loading protocol totaling 10 million cycles was formulated based on the activity frequency estimates provided in other published references, representing the upper end of the range of expected annual cycles for these normal and extreme activities. All tested devices passed this test; no devices fragmented, and all devices survived benchtop expulsion testing, as described earlier, following fatigue testing without extrusion from the disc model.
Vertebral Body Movement The Barricaid implant is not intended to support the adjacent vertebrae, or keep them separated from each other in any way. That is, the Barricaid is not intended to support load along the axis of the spine or to support the vertebral bodies in any other way. On the contrary, it is designed to be quite flexible and to move in concert with the surrounding soft tissue and accommodate the natural range of motion of the bony anatomy. As the vertebral bodies move during activities of daily living, the Barricaid implant may be cyclically deformed and could fatigue as a result. Such fatigue could result in fragmentation of the Barricaid or decreased resistance to expulsion from the disc space, both of which would constitute device failures. Resistance of the Barricaid against fatigue due to this cyclic deformation is tested as part of our “Motion Fatigue Test.”
The Intrinsic Therapeutics Barricaid Device
633
Intrinsic Therapeutics has adopted the philosophy of developing benchtop tests from observations of the implant's deformations and motions during aggressive cadaveric testing, and confirming (or modifying) these deformations based on clinical observations when possible. For this benchtop test, Barricaid implants were observed fluoroscopically in fresh-frozen human cadaver specimens while undergoing motions intended to simulate the extreme ranges of motion during daily living activities. The greatest range of implant deformation was seen in lateral bending. This implant deformation was used to simulate a range of activities that occur a combined 400 times per day as estimated from various literature sources.23,24 These extreme ranges of motion are run for 10 million cycles, or nearly 70 years of simulated activities. The cadaver specimen group consisted of a population of 23 specimens tested with unconstrained flexion-extension and lateral bending moments of 10 Nm and an axial preload of 500 N simulating the static load of the upper body. Fluorographs were taken at the extreme of each motion, as well as in the neutral position, and analyzed digitally to determine the maximum and minimum implant height (the difference between these being referred to as “implant compression”). These implant compression levels were then compared to implant compression seen in the functional x-rays (i.e., flexion-extension and lateral bending to the patient's limit) of 14 clinical patients implanted with the Barricaid device. Analysis was also performed to investigate the difference in implant compression between standing and lying down. We did this by testing cadaver specimens with and without the compressive preload, which was meant to simulate the weight of the upper body, in the neutral configuration. The mean of the maximum mesh compression from all of these cycles was used as the basis for this displacement-controlled test. Testing to these compression levels was accomplished to 10 million cycles without failure; no devices fragmented, and all devices survived benchtop expulsion testing as described earlier for fatigue testing.
Summary In summary, mechanical resilience of the Barricaid implant was determined using benchtop and cadaver testing to simulate the following scenarios: 1. Fatigue of the implant due to end plate motion (displacementcontrolled, 10 million cycles) 2. Fatigue of the implant due to intradiscal pressure changes of the nucleus pulposus (pressure-controlled, 10 million cycles) 3. Resistance to expulsion from intradiscal pressure of the simulated nucleus in a benchtop model (pressure-controlled “expulsion testing,” 338 psi) after completion of implant fatigue (motion or pressure) 4. Stability and expulsion resistance in a cadaver model with worst-case disc motion, orientation, intradiscal pressure, and annulus defect size (i.e., physiologic motion simulation, followed by >338 psi of intradiscal pressure with a 10-mm defect in the annulus)
634
P A R T
X
Annular Repair
Further, the biocompatibility of the Barricaid was assessed per the requirements of EN/ISO 10993 standards. The results indicate that the Barricaid is biocompatible. CLINICAL DATA Initial Postmarketing Study The patients reported here were treated in five European centers and were divided into two cohorts. One group (“treatment”) consists of 39 primary discectomy patients treated since March 2006 who were implanted with the Barricaid. The second group (“control”) consists of 108 primary discectomy patients treated since January 2003 who were not implanted with the device. All patients were enrolled prospectively, and enrollment in both cohorts is currently ongoing. Exclusion criteria for this initial postmarketing study include prior spine surgery and multilevel herniations. Inclusion criteria include severe sciatica or sciatica unresponsive to conservative therapy for 6 weeks, and MRI or CT confirmation of lumbar herniation as the cause. Follow-ups occur at 6 weeks, 3 months, 6 months, 12 months, and then annually through 5 years after surgery. Data gathered at each follow-up include Visual Analog Scale (VAS) (back and both legs), Oswestry Disability Index (ODI), and a custom clinical function score. Functional (standing) x-rays and CT scans are taken at each follow-up to assess stability of the implant over time, as well as disc height maintenance. Recurrent herniations (as evidenced by a confirmatory MR image) and reoperations are tracked throughout the follow-up period. Symptomatic reherniations were diagnosed first clinically, through a recurrence of leg pain and neurologic symptoms, and then confirmed by MRI. Both defect size and amount of disc material removed were measured at the time of surgery. To account for patient variability, the amount of disc material removed was normalized against the total volume of the intervertebral space using a previously reported CT algorithm.25 See Figure 82–5 for a representation of a CTbased model of the disc space. Mean follow-up on the control group has been 2.2 years, and on the implanted group 8.0 months. The volume of nucleus removed ranged from 0.4 to 7.0 cm3 (3.6 to 95.9% when normalized for disc volume), and the defect area ranged from 20 to 84 mm2. No patients have been lost to follow-up in the implanted group, and only one patient has been lost to follow-up in the control group at 1 year.
n
F I G U R E 82–5. This CT-based model of the disc space is used to normalize the volume of nucleus removed during surgery.
treatment group is required to see if the decrease in rate of reherniation remains over time. Disc Height Maintenance Control patients lost an average of 21% of their disc height by 6 months following the surgery and maintained that level at 1 year. Percent disc height loss at 1 year was positively correlated with both preoperative disc height and normalized nuclear volume removed when regressed with defect size (both P < 0.001). That is, for a given disc height, the more nucleus removed at the time of primary discectomy, the greater the disc height loss at 1 year following the operation. At 6 months, Barricaid patients lost only 7.5% of their preoperative disc height versus 21% in nonimplanted patients (P < 0.0002, Fig. 82–7). While less nucleus was removed from Barricaid patients (P < 0.000001), a multivariate analysis of disc Kaplan-Meier Survivorship (failure = symptomatic reherniation confirmed by MRI) 100
Symptomatic Recurrent Herniation Percentage
There have been no recurrent herniations in any of the 39 treatment group patients. The symptomatic reherniation rate in the control group has been 4% (4 of 108 patients) within the first 3 months and 9% (10 of 108 patients) by the 1-year follow-up (Fig. 82–6). Reherniation rate among the control patients (based on survivorship analysis) was positively correlated on a univariate basis with defect height (P ¼ 0.0024) and defect width (P ¼ 0.0104), and negatively correlated with nuclear volume removed (P ¼ 0.0466). That is, the greater the defect width or defect height created or found at the time of primary discectomy, the greater the risk of reherniation at a given time point. Further follow-up on the
95 90 85 80
Discectomy patients implanted with Barricaid Discectomy patients (controls)
75 0
2
4
6
8
10
12
14
Time (mos) n F I G U R E 82–6. Kaplan-Meier survivorship curve comparing Barricaid patients to control discectomy patients (failure defined as symptomatic reherniation confirmed by MRI).
CHAPTER 82
Percentage of disc height loss
Disc Height Loss at 6 Months 30 25 20 15 10 5 0 Barricaid n
Control
F I G U R E 82–7. Comparison of disc height loss between
Barricaid patients and control discectomy patients at 6 months.
height loss as a function of various parameters that may affect it (e.g., gender, body mass index, age, defect size, nuclear volume removed, implantation of Barricaid) demonstrated that Barricaid implantation alone accounts for 13% greater maintenance of disc height at 6 months. As an anecdotal check of this analysis, a subset of the 108 total patients in this abstract was selected with comparable amounts of normalized nucleus removed (P ¼ 0.97). At 6 months, Barricaid patients lost significantly less disc height (7.5% compared to 21% in the control group). Further follow-up on the treatment group is required to see if the greater disc height maintenance is retained over time. Summary After a successful initial pilot study, 39 patients have been implanted with the Barricaid annular repair device as part of this ongoing follow-up clinical study. Implantation has been shown to be safe and easy. The device is performing its function of retaining nuclear material within the disc as evidenced both by reherniation rate and disc height maintenance. Recurrent herniation following primary discectomy is a welldocumented clinical problem, reported in up to 21% of the general primary discectomy patient population. To date, there have been no recurrent herniations among patients implanted with the Barricaid at the time of a primary discectomy procedure. This is in contrast to recurrent herniation rates of 4% at 3 months and 16% at 1 year in an equivalent patient population not treated with the Barricaid. Loss of disc height has been correlated with disc degeneration in both the clinic and the laboratory. Early data indicate that the Barricaid may be effective in maintaining preoperative disc height. In our prospective clinical study, the patients treated with the Barricaid have lost just 7.5% of their preoperative disc height at 6 months, compared to 21% for the control patients. These early data provide a strong indication that an annular closure device may be effective in delaying disc degeneration and improving pain outcomes for discectomy patients. Further follow-up is ongoing to determine if the decrease in recurrent herniation rate and improvement in disc height maintenance provided by the Barricaid annular repair device remain over time.
The Intrinsic Therapeutics Barricaid Device
635
INDICATIONS AND CONTRAINDICATIONS The indications for the postmarketing study were narrower than clinically necessary in order to have as clean a data set as possible. Many of the contraindications for the postmarketing study (e.g., single level affected, prior spine surgery) will likely go away as further data are gathered. Indications and contraindications for the Barricaid annular repair device are generally in line with those for a standard discectomy procedure. Contraindications specific to the Barricaid include annular defects wider than 10 mm; annular defects taller than 6 mm; foraminal, extraforaminal, or anterior herniations; and disc height in the target level less than 5 mm. SUMMARY The need for annular repair is clear. The annular weakness or defect that is present in every discectomy patient presents a very real dilemma to the surgeons performing the more than 500,000 discectomies that are done each year: remove only the nuclear fragment causing the patient's immediate symptoms and run a high risk of a recurrent herniation, or remove as much nuclear material as possible at the time of the primary discectomy and risk disc collapse and significantly worse clinical outcomes in general. The Barricaid annular repair device has been designed to meet the product requirements outlined here: l Prevent reherniation: In a prospective clinical study, no reherniations have occurred among patients implanted with the Barricaid. l Preserve disc function: In a prospective controlled clinical study, patients implanted with the Barricaid preserved significantly more disc height at 6 months than the unimplanted control group. l Cover defects of a variety of sizes and locations: The Barricaid can cover defects as tall as 6 mm and as wide as 10 mm and can be inserted through a defect as small as a 5 mm 5 mm cruciate cut. l Protect as much of the posterior annulus as possible: The Barricaid is sized intraoperatively to cover substantially all of the posterior annulus. l Require minimal or no removal of nucleus: With the Barricaid device, only the fragment of nucleus that is causing the patient's immediate symptoms needs to be removed. l Easily used with current surgical approach and techniques: The Barricaid is implanted through a standard posterior discectomy access. l Minimal impact on operative time: Use of the Barricaid device adds roughly 5 to 10 minutes to the overall operative time. l Endure the high ranges of motion and pressure without failure or migration: Rigorous cadaver and benchtop testing has demonstrated the robustness of the Barricaid design under the aggressive loading environment of the lumbar disc. The early clinical results described here have shown the Barricaid annular repair device to be an extremely promising solution to the problems associated with the annular defects with which discectomy patients currently leave the operating room.
636
P A R T
X
Annular Repair
REFERENCES 1. Thompson Healthcare, Solucient InpatientView: 2005 estimated data for ICD-9 80.51, 2007. 2. Viscogliosi AG, Viscogliosi MR, Viscogliosi JJ: Spine Arthroplasty: Market Potential and Technology Update. Viscogliosi Bros, LLC. p. 89, 2001. 3. Atlas SJ, Keller RB, Chang Y, et al: Surgical and nonsurgical management of sciatica secondary to a lumbar disc herniation: Five-year outcomes from the Maine lumbar spine study. Spine 26(10): 1179–1187, 2001. 4. Asch HL, Lewis PJ, Moreland DB, et al: Prospective multiple outcomes study of outpatient lumbar microdiscectomy: Should 75 to 80% success rates be the norm? J Neurosurg: Spine 96:34–44, 2002. 5. Rogers, LA: Experience with limited versus extensive disc removal in patients undergoing microsurgical operations for ruptured lumbar discs. Neurosurgery 22(1):82–85, 1988. 6. Carragee EJ, Han MY, Suen PW, Kim D: Clinical outcomes after lumbar discectomy for sciatica: The effects of fragment type and anular competence. J Bone Joint Surg Am, 85-A(1):102–108, 2003. 7. Hanley EN Jr., Shapiro DE: The development of low-back pain after excision of a lumbar disc. J Bone Joint Surg Am, 71-A(5):719–721, 1989. 8. White AA, Panjabi MM: Clinical Biomechanics of the Spine. 2nd edition, Philadelphia, Lippincott Williams & Wilkins, 1990. 9. McMillan DW, McNally DS, Garbutt G, Adams MA: Stress distributions inside intervertebral discs. J Bone Joint Surg Am, 78-B(6): 81–87, 1996. 10. Yorimitsu E, Chiba K, Toyama Y, Hirabayashi K: Long-term outcomes of standard discectomy for lumbar disc herniation: A followup study of more than 10 years. Spine 26(6):652–657, 2001. 11. Mochida J, Toh E, Nomura T, Nishimura K: The risks and benefits of percutaneous nucleotomy for lumbar disc herniation: A 10-year longitudinal study. J Bone Joint Surg Am 83-B(4):501–505, 2001. 12. Hanley EN Jr., Shapiro DE: The development of low-back pain after excision of a lumbar disc. J Bone Joint Surg Am, 71-A(5):719–721, 1989. 13. Carragee EJ, Spinnickie AO, Alamin TF, Paragioudakis S: A prospective controlled study of limited versus subtotal posterior
14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.
discectomy: Short-term outcomes in patients with herniated lumbar intervertebral discs and large posterior anular defect. Spine 31(6): 653–657, 2006. Wilke HJ, Neef P, Caimi M, et al: New in vivo measurements of pressures in the intervertebral disc of daily life. Spine 24(8): 755–762, 1999. Klara PM, Ray CD: Artificial nucleus replacement: Clinical experience. Spine 27(12):1374–1377, 2002. Heggeness MH, Watters WC 3rd, Gray PM Jr: Discography of lumbar discs after surgical treatment for disc herniation. Spine 22: 1606–1609, 1997. Ahlgren BD, Lui W, Herkowitz HN, et al: Effect of anular repair on the healing strength of the intervertebral disc. Spine 25(17): 2165–2170, 2000. Ninomiya M, Muro T: Pathoanatomy of lumbar disc herniation as demonstrated by computed tomography/discography. Spine 17(11): 1316–1322, 1992. Cinotti G, Gumina S, Giannicola G, Postacchini F: Contralateral recurrent lumbar disc herniation: Results of discectomy compared with those in primary herniation. Spine 24(8):800–806, 1999. Yeh O, Chow S, Small M, et al: Novel approach to closing anular defects: a biomechanics study. Poster presented at North American Spine Society, Philadelphia. September 2005. Brinckmann P, Porter RW: A laboratory model of lumbar disc protrusion. Fissure and fragment. Spine 19(2):228–235, 1994. Sato K, Kikuchi S, Yonezawa T: In vivo intradiscal pressure measurement in healthy individuals and in patients with ongoing back problems. Spine 24(23):2468–2474, 1999. Morlock M, Schneider E, Bluhm A, et al: Duration and frequency of everyday activities in total hip patients. J Biomech 34:873–881, 2001. Kaynak H, Kaynak D, Oztura I: Does frequency of nocturnal urination reflect the severity of sleep-disordered breathing? J Sleep Res 13(2):173–176, 2004. Kamaric E, Yeh O, Velagic A, et al: Disc Collapse Following Discectomy: A Novel Approach to Measuring Disc Height. Poster presented at Spine Arthroplasty Society, New York. May 2005.
CHAPTER
83
Animal Models for Human Disc Degeneration Kern Singh, Koichi Masuda, and Howard S. An
K E Y l
l
l
l
P O I N T S
The availability of an experimental animal model that consistently reproduces intervertebral disc degeneration would facilitate the investigations of the disease. Animal models of human disc degeneration can be subdivided into two main categories: naturally occurring and experimentally induced. Naturally occurring animal models have the drawback of the basis for the high rate of disc degeneration not being known. Although the interventions in artificial animal models are known, the true relationship between these interventions and the actual events leading to disc degeneration in humans is not.
Intervertebral disc degeneration is a major cause of functional disability in humans.1 The macroscopic features characterizing disc degeneration include the formation of tears within the annulus substance and progressive fraying and dehydration of the nucleus pulposus (NP).2–4 Despite the significant impairment associated with this disease, a clear understanding of the basic mechanisms of disease pathogenesis and specific therapeutic agents is still lacking. Unquestionably, disc degeneration is a multifactorial process influenced by genetics, lifestyle conditions (including obesity, occupation, smoking, and alcohol consumption),1–6 biomechanical loading and activities, and other health factors (diabetes and aging). The availability of an experimental animal model that consistently reproduces the disease would facilitate the investigations of intervertebral disc degeneration. This chapter provides an overview of the reported animal models that have been used to evaluate human disc degeneration. INTERVERTEBRAL DISC ANATOMY The intervertebral disc acts as a load-bearing structure with two distinct components: the NP and the annulus fibrosus (AF). Each component has very distinct biomechanical properties. The NP, rich in proteoglycan, acts as an internal semifluid mass, whereas the AF, rich in collagen, acts as a laminar fibrous container.5 The hydrostatic properties of the disc arise from its high water content. The NP, when palpated in a young adult, acts as a viscid fluid under applied pressure, but also exhibits considerable elastic
rebound, assuming its original physical state upon release.3 Histologic analysis has shown that the NP is derived from embryonic notochordal tissue. The NP from young individuals is composed of loose, delicate fibrous strands embedded in a gelatinous matrix.7 The transition from notochordal to chondrocytic cells in NP varies among species, and therefore, the investigator should factor the age of the animals into consideration when performing experiments.8 This process of transition from notochordal cells to chondrocytic NP cells involves mitochondrial capase,9 apoptosis of notochordal cells, and migration of chondrocytes from the cartilaginous end plate toward the NP.9–11 The major function of the NP is to resist and redistribute compressive forces within the spine; the major function of the AF is to withstand tension. The unique combination of biochemical and biomechanical properties of the AF and NP allows the intervertebral disc to absorb and disperse the normal loading forces experienced by the spine.3,6 The presumption is that when one of these two units, either the AF or the NP, is structurally compromised, degenerative changes will ensue because of the alteration in mechanical force distribution across the functional spinal unit. ANIMAL MODELS—REQUIREMENTS AND SELECTION Animal models are essential in making the transition from scientific concepts to clinical applications. Ethical issues are always of concern. The Animal Welfare Act and the Public Health Service Animal Welfare Policy require that an Institutional Animal Care and Use Committee review and approve each protocol. The use of appropriate technologies to eliminate or reduce pain and euthanize is required. The usage of the minimum number of animals from which significant conclusions can be inferred is not only ethically necessary but also cost-effective. In addition, the species selected should be carefully chosen with serious consideration given to all applicable federal regulations, public health service policy, and institutional policies. The degree of disc degeneration in an animal model should be controllable and selectable to aid the researcher in proving the hypothesis. The validation of the reproducibility of an animal model allows results from different scientific researchers to be 639
640
P A R T
X I
Cell-Based Biologic Scientific Methods and Future Clinical Applications
compared. Furthermore, for validation, the interobserver variability of outcome measures should be fully studied. If a surgical technique or other environmental change is used, the procedure should be standardized in detail to increase the transferability of models to other groups. CATEGORIES OF ANIMAL MODELS FOR DISC DEGENERATION Animal models of human disc degeneration (Table 83–1) can be subdivided into two main categories: naturally occurring and experimentally induced. Naturally Occurring Disc Changes Sand Rat
The sand rat (Psammomys obesus) is indigenous to eastern Mediterranean deserts. Silberberg12 and Adler13 first described the spontaneous appearance of disc changes in the rat as similar to those that occur in humans. Cysts, tears, and occasionally bony bars were seen more frequently in the AF of 18- to 30-monthold animals. According to Silberberg, the low water and high salt diet of the sand rat may somehow be related to an altered metabolism potentially leading to the internal derangement of the NP. Moskowitz and associates14 further studied the sand rat by analyzing the lumbar as well as the thoracic spine. The authors fed a group of sand rats a normal diet and included radiographic and histologic assessment of the intervertebral disc as well as measurements of serum glucose and insulin. Roentgenographically observed disc degeneration increased with age, leading to 50% of the animals being affected by 18 months of age. The authors believed that the observation of degeneration occurring earlier than 3 months lent support to causative factors other than aging. Ziv15 and associates compared hydration, fixed charge density (FCD), and varying osmotic pressures in young, old, and diabetic sand rats. The discs of the young diabetic sand rats demonstrated decreased hydration, FCD, and ability to resist compression under osmotic pressures (assessed by hydration after equilibration in vitro with either 15 or 25 g/dL polyethylene glycol) when compared to young healthy sand rats. The authors noted that the diabetic intervertebral disc was more similar in nature to the older sand rat intervertebral disc. Gruber and associates16 recently published a large cross-sectional and prospective analysis of Psammomys obesus. They analyzed 158 animals in a cross-sectional study and 22 animals in a longitudinal study (12 months of age). The authors noted that naturally occurring radiographic signs of degeneration were evident by 2 months; wedging, narrowing, irregular disc margins, and end plate calcification were the most common degenerative changes in older animals. The sand rat was also recently used as a small animal model for autologous disc cell implantation by Gruber and associates.17 Cells were harvested from a lumbar intervertebral disc, expanded in monolayer tissue culture, labeled with agents that allow subsequent immunolocalization of these cells, and implanted in a second disc site of the donor animal. The cells were either engrafted in a
bioresorbable carrier tested for compatibility or injected into the recipient disc. Data from 15 animals were obtained up to 33 weeks. Immunocytologic identification of engrafted cells showed integration into the disc with normal matrix surrounding the cells 8 months after engraftment. Engrafted cells exhibited either a spindle-shaped morphology in the AF or a rounded chondrocyte-like morphology in the NP. Pintail Mouse
Other animal models have also been developed with characteristics that resemble the spontaneous onset of disc degeneration observed in humans. Berry18 noted that in the pintail mouse (Anas acuta), the NP in the subcervical discs underwent changes similar in appearance to those in humans (loss of mucopolysaccharides). This change was much more dramatic in mutant homozygotes (pBr strain mutated by exposure to methylcholanthrene), leading the authors to conclude that disc degeneration in humans may be genetically linked. Chinese Hamster
The Chinese hamster (Cricetulus griseus) was used by Silberberg and Gerritsen19 because of its tendency to develop diabetes. The hamster was evaluated in order to determine a correlation between hyperostotic spondylosis and diabetes. Histologic evaluations revealed a higher incidence of spondylosis (60% versus 39%), but a lower incidence of disc herniations (9% versus 30%) in animals with diabetes. Both spondylosis and disc degeneration increased with age. Rabbit
Kim and associates conducted a biochemical and radiologic comparison of four disc injury models to produce disc degeneration in rabbits.20 In the first experiment, seven New Zealand white rabbits (1 year old, 3.5–4.5 kg body weight) were used to test four different disc injury models: intradiscal injection of camptothecin (an apoptotic agent) using a 23-gauge needle at L2-L3, nucleus aspiration using a 21-gauge needle at L3-L4, three annulus punctures using a 21-gauge needle at L4-L5, and one annulus puncture using an 18-gauge needle at L5-L6. Lumbar spinal magnetic resonance images were assessed using four grades of disc degeneration. In the second experiment, the 21-gauge three-puncture and the 18-gauge one-puncture models, thought most effective in producing disc degeneration in the first experiment, were again used in a second study. Six rabbits were killed 8 weeks later, the water and sulfated glycosaminoglycan contents being measured as in the first experiment. The authors found that in the first experiment, the 21gauge three-puncture and the 18-gauge one-puncture models produced the most consistent disc degeneration in the rabbit lumbar spine. When these two models were again studied in the second experiment, the 21-gauge three-puncture technique was superior in producing disc degeneration over a shorter period of time. Dog
Canine models, including those using basset hounds, beagles, and dachshunds, have been extensively described in the literature.21–24 Several authors have noted that the disc degeneration
CHAPTER 83
TABLE 83–1.
641
Animal Models for Human Disc Degeneration
Summary of the Animal Models of Disc Degeneration
Animal Model
Method of Analysis
Onset of Degeneration
Author
Year
Observation on natural diet—high salt, low water Observation on natural diet—high salt, low water Histologic analysis of IVD, measurements of serum glucose/insulin Varied hydration, fixed charge density and osmotic pressures (hydration after equilibration in vitro with either 15 or 25 g/dL polyethylene glycol) Biomechanical pressure evaluation of the IVD; peak force and force decay (1 versus 500 s) were measured in response to a flexionproducing step displacement Observational prospective study of natural IVD degeneration Autologous disc cell transplantation; cells were harvested from a lumbar IVD, expanded in monolayer tissue culture, and labeled with agents to allow immunolocalization pBr strain mutated by exposure to methylcholanthrene Comparison of hyperostotic spondylosis and diabetes
6 months 1.5–2.5 years 3–30 months 6 weeks
Silberberg12 Adler13 Moskovitz14 Ziv15
1979 1983 1990 1992
3–18 months
Ziran80
1994
2–12 months 33 weeks
Gruber16 Gruber17
2002 2002
2–4 months 1–3 months
Berry18 Silberberg19
1961 1976
Comparison of collagen content Characterization of degenerated disc disease in multiple breeds
0–3 months 0–124 months 2 months 0–15.7 years
Goggin22 Ghosh23
1970 1976
Cole21 Lauerman25
1986 1992
0–6 months 1–6 months 3–12 months 1–3 months
Goff 26 Yamada27 Higuchi28 Cassidy29
1957 1962 1983 1988
0–4 weeks
Hutton30
1999
7–28 days 0–56 days 0–2 wk
Lotz31 Iatridis32 Ching33
1998 1999 2002
1–28 days
Kroeber34
2002
60–90 days
Hadjipavlou35,36
1998
6–12 months 3–6 months
Miyamoto37 Sullivan38
1991 1971
3–9 months
Phillips39
2002
0–3 months 0–3 months
Keyes40 Smith41
1932 1951
0–3 months
Haimovichi42
1970
Naturally Occurring
Sand rat (Psammomys obesus)
Pin tail mouse Chinese hamster (Cricetulus griseus) Beagle/greyhound Multiple canine breeds Beagle Baboon (Papio cynocephalus anubis) Experimentally Induced
Evaluation of PG content in lumbar IVD Roentgenographic evaluation Method of Injury
Surgically/Physically Induced Posture Change Wistar rat Mouse Wistar rat
Forelimb amputation—midhumeral Midhumeral amputations—effects of experimental posture Light/electron microscopic evaluation of NP in bipedal rats Observation of lumbosacral herniations in bipedal rats
Sprague-Dawley rat
Tail suspension (simulated weightlessness)
Mouse Sprague-Dawley rat
Mouse tail exposed to external compression Rat tail compression using an Ilizarov-type apparatus Compressive stress via pins inserted in the 6th and 7th caudal vertebrae; cyclic loading of 0.5, 1.5, or 2.5 Hz was applied for 1 hour each day from the 3rd to 17th day; angular compliance, angular laxity, and inter-pin distance were measured Axial dynamic compression loading
Tail suspension
Axial Loading
New Zealand white rabbit
Torsional Injury New Zealand white rabbit
Facetectomy and 30-degree lumbar torsion
Mouse New Zealand white rabbit
Removal of cervical spinous process and paravertebral muscles Resection of the inferior articular process at one level Resection of the contralateral inferior articular process at an adjacent level
Resection of Spinal Process or Facet Joint
Adjacent Level Fusion New Zealand white rabbit
Lumbar fusion L5-L6 Annulotomy
New Zealand white rabbit
Anterior annulotomy using a scalpel Ventral annulotomy (transverse, mid-disc, 4 mm long and through the AF into the NP) Ventral annulotomy similar to Smith and intramuscular ACTH injection every other day
(Continued)
642
P A R T
TABLE 83–1.
X I
Cell-Based Biologic Scientific Methods and Future Clinical Applications
Summary of the Animal Models of Disc Degeneration—Cont'd
Animal Model
Merino sheep Domestic pig (Sus scrofa) Gottingen pig
Method of Analysis
Onset of Degeneration
Author
Year
Full-thickness stab through the anterior part of the AF into the NP (2 mm in length); the NP and anterior AF of the intact IVD and the central region of the lesioned IVD were removed and analyzed; biochemical measures included analysis of PG, water and hyaluronic acid concentration as well as PG monomer size and total PG content Topographic analysis of anterior full-thickness annulotomies Puncture with 16G–21G needle Slow degeneration compared to stab model; disc height, MRI and histologic analysis Puncture with 16G needle, MRI analysis Left anterolateral annulotomy parallel and adjacent to the inferior end plate with a controlled depth of 5 mm; the inner third of the AF and the NP were left intact closely reproducing the rim Lear lesion Stab lesion with a No. 11 scalpel midline anteriorly (depth 13 mm)
1–200 days
Lipson43
1981
0–3 months 12 weeks
1986 2002–2004
12 weeks 4–12 months
Urayama44 Aota45 Muehleman,46 Masuda47 Kom pel48 Osti49
0–2 weeks
Kaapa50–52
1994
53
2003 1990
0–12 months
Pfeiffer
1994
3 months
Holm54
2004
200k cycles
Wada55
1992
2–26 weeks 6–20 weeks 2–52 weeks 8 weeks
Bradford81,82 Kahanovitz69 Spencer61 Nitobe65
1983, 1984 1985 1985 1988
Pig Rabbit Rabbit
Injection of chymopapain Injection of chymopapain Injection of collagenase and chymopapain Application of chymodiactin to the lumbar spine IVDs of canines and postinjection biomechanical analysis Injection of collagenase and chymopapain Injection of collagenase Injection of chondroitinase ABC
4 weeks 2–4 weeks 1–12 weeks
1986 1987 1990–1999
Canine
Injection of chondroitinase ABC
1–21 days
Sheep Rat Rat
Injection of chondroitinase ABC Injection of trypsin, collagenase, chymopapain, and hyaluronidase Effects of chondroitinase ABC on rat tail IVDs
1–4 weeks 1 week 2 weeks
Zook56 Olmarker64 Kato,57 Eurell83 Takahashi,59 Sumida60 Takahashi,58 Ono63 Sasaki62 Takenaka84 Norcross74
16 weeks
Anderson75
2003
8 weeks 8 weeks
Oda77 Iwahashi76
2004 2002
1 month
Sahlman85
2001
Hamrick79
2003
Analyzed lumbar IVDs after a more extensive annulotomy, by a window in the AF anterolaterally, which was then plugged by fibrin glue followed by injection of chymopapain End Plate Injury
Domestic pig
Analyzed lumbar IVDs after perforation of end plate into the NP Muscle Stimulation
Rabbit Chemically/ Genetically Induced Canine Canine Canine Canine
Repetitive flexion-extension of the C-spine (trapezius stimulation, 200k cycles) Chemonucleolysis
1997–1998 2001 1987 2003
Injection of Fibronectin Rabbit
Injection of fibronectin fragments, radiographically, histologically, and biochemically analyzed
Rabbit Rat
Smoke inhalation via control box Nicotine-induced (110 ng/mL)
Smoking
Genetic Knockout COL2a1 knockout mice
GDF-8 knockout mice
Analysis of skeletal tissues from mice with an inactivated allele of the Col2a1 gene for type II collagen (“heterozygous knockout”); radiographic analysis; conventional, quantitative, and polarized light microscopy; immunohistochemistry for the major extracellular components, and in situ hybridization for procollagens alpha-1 (I) and alpha-1 (II) Analysis of skeletal tissues from GDF-8 (myostatin) knockout mice
AF, annulus fibrosus; G, gauge; IVD, intervertebral disc; MRI, magnetic resonance imaging; NP, nucleus pulposus; PG, proteoglycan.
CHAPTER 83
in these models differs from human intervertebral disc degeneration. Typically, by 1 year of age in these canine models, the NP becomes occupied by chondrocytes with a matrix containing an increased collagen content and decreased proteoglycan and water content. Unlike the human disc, this matrix may become calcified after disc herniations. Primates
Primate models, including the baboon (Papio cynocephalus anubis), have also been used. Lauerman and associates25 radiographically demonstrated a significant correlation between age and disc degeneration grade (Spearman correlation coefficient of 0.726, P < 0.0001) as well as kyphosis angle (coefficient of 0.6333, P < 0.0001).
Experimentally Induced Disc Changes Experimentally produced disc changes have been extensively described in the literature. These artificially induced degenerative changes are invaluable in creating clinically applicable and reproducible animal models of degeneration. Surgically or Physically Induced Animal Models Postural Change
Goff and Landmesser,26 as well as Yamada,27 developed a method for studying disorders of posture and changes in bone dimensions, gait, and behavioral characteristics of bipedal animals. The authors performed neonatal midhumeral surgical bilateral amputations in the Wistar rat and DBA strains of mice. Postural changes resulted in the reduction of the usual lumbar kyphosis and gradual degenerative changes in the AF and NP at 3 to 12 months. The authors also noted that disc herniations developed in some of the bipedal rats. Higuchi and associates28 used electron microscopy to further analyze changes in the NP of bipedal mice. The authors noted that over time the discs in bipedal mice degenerated in a similar but much more rapid fashion than their quadruped counterparts. Cassidy29 noted similar findings in bipedal Wistar rats with radiographic evidence of anterior vertebral body wedging, degeneration of the intervertebral discs, and lumbosacral herniations of the NP. Tail Suspension
Hutton and co-workers examined the effects of simulated weightlessness on the intervertebral disc by using a rat-tail suspension model.30 The authors tail-suspended 32 Sprague-Dawley rats for either 2 or 4 weeks and biochemically evaluated the lumbar discs using enzyme-linked immunosorbent assays. At 4 weeks, the authors noted a significant decrease in proteoglycan content (35%). No appreciable difference was found at 2 weeks. There were no statistically significant differences between the two groups in type I or II collagen content at either time point. Axial Loading
In a mouse model, Lotz and associates studied the biomechanical effects of static compressive loading on tail intervertebral discs.31 Mouse-tail discs were loaded in vivo with an external compression
Animal Models for Human Disc Degeneration
643
device. The expression of type II collagen was suppressed at all levels of stress, whereas the expression of aggrecan decreased at the highest stress level in proportion to decreased nuclear cellularity. The authors concluded that compressive loading results in a dose-dependent apoptotic response and a downregulation of collagen II and aggrecan gene expression. Several authors have described rat-tail compression models.32,33 Iatridis and associates originally described the rat-tail compression model using an Ilizarov-type apparatus to apply chronic compression.32 The authors noted that chronically applied compression had effects similar to immobilization but induced the changes earlier and in larger magnitudes. In addition to an increase in proteoglycan content of the intervertebral disc, biomechanically, there was an increase in disc thickness, angular laxity, and axial and angular compliance. Ching and associates also studied the effects of cyclical compression on the rat-tail intervertebral disc.33 Sixty SpragueDawley rats were subjected to daily compressive stress via pins inserted in the sixth and seventh caudal vertebrae over a 2-week loading period. Animals were randomly divided into a sham group (pin insertion, no loading), a static loading group, or cyclic loading group (0.5, 1.5, or 2.5 Hz). Loading was applied for 1 hour each day from the third to seventeenth day following pin insertion. The angular compliance, angular laxity, and inter-pin distance were measured in vivo at days 0, 3, 10, and 17. The authors noted that, in general, cyclical loading resulted in less marked changes than static loading, but cyclical loading at certain frequencies (0.5 and 2.5 Hz) produced severe degenerative changes. More recently, Kroeber and colleagues34 reported a new rabbit disc degeneration model (New Zealand white), which applies a controlled and quantified axial mechanical loading. After 14 and 28 days of axial loading, the discs demonstrated a significant decrease in disc height. Histologically, disorganization of the AF occurred, and the number of dead cells increased significantly in the AF and cartilage end plate. These changes were not reversible after 28 days of unloading. Torsional Injury Model
Hadjipavlou and associates described a rabbit model involving a torsional injury that leads to accelerated disc degeneration.35,36 Sixty-five New Zealand rabbits underwent a surgical facetectomy and a 30-degree torsional lumbar injury. The authors noted that within 60 to 90 days the rabbits that received the torsional strain exhibited clear signs of disc changes, including thinning, increased phospholipase A2, and decreased NP volume. The control group (surgical facetectomy without the torsional strain) did not exhibit these findings, which suggests the role of torsional strain as a possible mechanism of disc degeneration. Resection of Spinal Process or Facet Joint
Miyamoto and associates described an easily reproducible cervical spondylosis model in the rat. The authors noted that cervical disc degeneration was accelerated with detachment of the posterior paravertebral muscles from the vertebrae and resection of the spinous processes along with the supraspinous and interspinous ligaments. Pathologic changes that occurred as a result of the instability included proliferation of cartilaginous tissue and fissures in the AF, shrinkage of the NP, disc herniation, and osteophytic formation.37
644
P A R T
X I
Cell-Based Biologic Scientific Methods and Future Clinical Applications
Facetectomy studies have been performed by Sullivan and associates38 in the lumbar spine of immature white rabbits. The authors resected the inferior articular process on one side at a selected vertebral level and on the opposite side at the adjacent level. The disc height was decreased at the surgical level in 50% of the discs at 6 months and 74% at 12 months. The discs in the transverse plane at 9 to 12 months showed thinning of the posterior AF, circumferential slits in the peripheral AF, and an increased area and decreased organization of the NP. The facet joints opposite the facetectomy began to show degeneration at 6 months. The authors concluded that the posterior facet joint protects the intervertebral disc against rotational stresses. Adjacent-Level Lumbar Fusion
Phillips and associates described a noninjury rabbit model of disc degeneration. Disc degeneration was created at levels proximal (L4-L5) and caudal (L7-S1) to a simulated lumbar fusion and was studied for up to 9 months after arthrodesis.39 Loss of the normal parallel arrangement of collagen bundles within the annular lamellae was observed in intervertebral discs adjacent to the fusion at 3 months. At 9 months of follow-up, the disc had been replaced by disorganized fibrous tissue. An initial cellular proliferative response was subsequently followed by a loss of chondrocytes and notochordal cells in the NP. Degeneration was accompanied by a decrease in the monomer size of proteoglycans. Annulotomy
Annulotomies have been performed both anteriorly and posteriorly in a variety of methods. The annular puncture model of the rabbit intervertebral disc has become a popular and easily reproducible model of disc degeneration. Although Lipson was the first to biochemically analyze the effects of annulotomies, the earliest work in this area was performed by Keyes and Compere in 1930.40 Both authors described artificially induced injuries to the AF (radiographically and visually) that were noted to be similar to human disc degeneration. Smith and Walmsley41 performed a similar study, prompted by the superficial healing seen after posterior annulotomies in prior studies. Fifty-five immature adult rabbits were grossly and microscopically examined at times ranging from 1 to 25 months after ventral annulotomy. The ventral annulotomy was transverse, mid-disc, 4 mm long, and through the AF into the NP. The authors found that healing occurred in the outermost portion of the lesion and progressive degeneration occurred throughout the disc even though the lesion was only in the anterior part of the AF. In 1970, Haimovici used a similar rabbit annular laceration model to describe the negligible effects of adrenocorticotropic hormone injections repeated every other day.42 Lipson and associates were the first to describe the biochemical effects of full-thickness stab injuries to the anterior part of the AF (ventral annulotomy, 2 mm in length, and through the AF into the NP).43 From each animal, the NP and the anterior AF of the intact disc and the central region of the lesioned disc were removed and analyzed. Biochemical measures included analysis of proteoglycan, water, and hyaluronic acid concentration as well as proteoglycan monomer size and total proteoglycan content. The authors noted that rabbit discs were similar to human discs in some characteristics (water concentration and proteoglycan size)
and different in other characteristics (lower hyaluronic acid concentration and percentage of aggregated proteoglycan). Urayama44 focused on the topographic distribution of postannulotomy changes within the AF after annulotomies of rabbit discs as described by Lipson,43 Relative to the control discs, the greatest changes in the AF were in its inner layer; they were more pronounced for light microscopic measures than for electron microscopic measures. Our research group has recently developed a new rabbit model of mild, reproducible disc degeneration by an annulus needle puncture. The newly developed annulus needle puncture procedure, utilizing 16- to 21-gauge (G) needles with controlled depth, resulted in a slower decrease in disc height than the classical stab procedure. The gauge size significantly affected the degree of disc degeneration by magnetic resonance imaging (MRI) grading and the histologic score of disc degeneration throughout the experimental period. At all three time points, the histologic scores were significantly higher in the 16-G and 18-G groups than in the control group. The radiographic, histologic, and MRI results suggested that the use of 16-G and 18-G needles for puncture produces the most predictable, slowly progressive disc narrowing. This new model may be more useful in studying changes in the biomechanical and biochemical properties of disc degeneration progression and, due to its milder degeneration, is better suited to test biologic therapy.45–47 Kompel and associates showed more detailed findings in the MRI using a 16-G needle puncture. The MRI analysis of eight rabbits revealed that 12 weeks after the anterior annular puncture, the mean NP signal intensity (T2-weighted images) of stabbed discs had significantly decreased to approximately 60% of their respective preoperative values (P < 0.05). By week 24, the mean NP signal intensity had significantly decreased further to 34% to 54% of the preoperative values.48 Annulotomy models have also been extensively evaluated in larger animal models as well. Annulotomy-induced lesions, similar to those discussed previously, were performed in sheep models. Osti and associates49 studied only partial thickness anterior annulotomies. In 21 adult sheep, a cut was made in the left anterolateral AF of three randomly selected lumbar discs. The cut was parallel and adjacent to the inferior end plate and had a controlled depth of 5 mm. This left the inner third of the AF and the NP intact and closely reproduced the rim Lear lesion. Using 2-year-old merino sheep, the authors noted that, although the outermost AF showed the ability to heal, the defect induced by the cut led initially to deformation and bulging of the collagen bundles and eventually to inner extension of the tear and complete failure. These findings suggested that discrete tears of the outer AF might have a role in the formation of concentric clefts and in accelerating the development of radiating clefts. Peripheral tears of the AF, therefore, may play an important role in the degeneration of the intervertebral joint complex. Overall, the authors concluded that disc degeneration occurred in an outside-in manner with deterioration occurring first in the AF followed by that in the NP. Porcine annulotomy models have also been reported in the literature. Kaapa and associates50–52 studied domestic pigs (Sus scrofa). Their iatrogenically induced lesion was an anterior midline stab with a No. 11 blade scalpel to a depth of 13 mm. The
CHAPTER 83
concentration of total collagen (hydroxyproline [Hyp]), the activities of the two key enzymes in collagen biosynthesis (prolyl-4-hydroxylase [PH] and galactosylhydroxylysyl glucosyltransferase [GGT]) and the concentration of mature collagen cross-links (hydroxypyridinium [HP]) were determined. Considerable morphologic changes were apparent, particularly in the NP, which became small, fibrous, and yellowish. The annular lamellar structure was partially destroyed and had been replaced by granulation tissue in the region of the injury. Pfeiffer and associates53 analyzed lumbar discs of Gottingen minipigs after more extensive annulotomies, for example, by a window in the AF anterolaterally, which was then plugged by fibrin glue followed by injection of chymopapain. The authors noted intervertebral disc degeneration in all groups by 3 weeks with an associated decrease in intradiscal pressure.
Animal Models for Human Disc Degeneration
645
the rabbit.75 Fibronectin fragments or control substances were injected, and the discs were examined radiographically, histologically, and biochemically, and the gene expression was measured at the 2-, 4-, 8-, 12-, and 16-week time points. A progressive loss of the normal architecture of the NP and AF was observed over the 16-week study period. Similar biochemical results were shown with decreasing proteoglycan content and downregulation of aggrecan mRNA. Smoking
Wada and associates evaluated possible overuse injury and its impact on cervical spondylosis. The authors provided a flexionextension moment to young rabbit spines through electrical stimulation of the trapezius muscle. After 200,000 cycles, the authors noted severe delamination of the AF with an associated osteophyte formation at the same level. No severe degeneration of the NP occurred despite the repetitive loading.55
Iwahashi and associates studied the effects of nicotine on vascular buds in rabbits for elucidating the mechanism of nicotine-induced vertebral disc degeneration.76 The authors used a pump filled with a dilute nicotine solution placed subcutaneously for 8 weeks. The model maintained nicotine blood concentrations at approximately 110 ng/mL. After 8 weeks of nicotine exposure, necrosis and hyalinization of the nucleus pulposus were noted in all the rabbits. The AF demonstrated a pattern of overlapping laminae with clefts and separation. Furthermore, the authors noted necrotic changes in the endothelial cells and narrowing of the vascular lumen. The authors concluded that the reduction in density of the vascular buds and narrowing of the lumen resulted in a decreased oxygen tension, which leads to the decreased synthesis of proteoglycan and collagen, thereby facilitating disc degeneration. Similar effects of nicotine on rat intervertebral disc degeneration was noted by Oda and associates.77 The authors exposed 8week-old rats to indirect tobacco smoke inhalation resulting in blood concentrations twice that of human smokers. After 8 weeks of exposure, the rats were sacrificed and the chondrocytes were noted to have a disordered AF layer that was larger than normal. Furthermore, the interleukin 1(IL-1)b levels were significantly higher than the nonsmoking control group. The authors concluded that nicotine exposure resulted in an increase in local production and release of inflammatory cytokines and resultant decomposition of chondrocyte activity.
Chemically or Genetically Induced Animal Models
Knockout Mouse
Chemonucleolysis
Human twin studies suggest that particular genes may play a role in disc degeneration.78 Sahlman and associates evaluated the heterozygous inactivation (“knockout”) of the Col2a1 gene and its role in growth development and disc degeneration.85 Skeletal tissues of mice with an inactivated allele of the Col2a1 gene for type II collagen (heterozygous knockout) were studied. The tissues were studied using radiograph analyses; conventional, quantitative, and polarized light microscopy; immunohistochemistry for the major extracellular components; and in situ hybridization for procollagens a1 and a2. The authors found that the gene-deficient mice had shorter limbs, skulls, and spines, as well as thicker and more irregular vertebral end plates. The mice also had a lower concentration of glycosaminoglycans in the AF, in the end plates, and in the vertebral bone than the controls. The authors concluded that gene-deficient mice (heterozygous knockout of Col2a1) showed early skeletal manifestations and late degenerative changes resembling human disc degeneration. Because these mice were deficient in Col2a1 it is unclear whether the degenerative changes were from improper development
End Plate Injury
Holm and associates developed an animal model of disc degeneration induced by an injury to the end plate.54 The L4 cranial end plate of domestic pigs was perforated into the NP at the center using a 3.5-mm drill bit from the lateral cortex at the midheight of the vertebral body. After 3 months, biochemical analysis of the L3-L4 disc showed a reduction in the water content in the outer AF of the degenerated disc. Morphologically, the loss of the gel-like nature of the NP as well as the delamination of the AF layer was seen. A T2-weighted MR image of the spine showed the loss of MR signal in the NP of the degenerated disc. Muscle Stimulation
Various animal models have been evaluated after the application of chemonucleolytic agents.32,53,56–73 Canine, murine, porcine, and rabbit models have been extensively described in the literature. Kahanovitz and colleagues reported one of the earlier studies regarding the application of chymodiactin to the lumbar spine of canines.69 Biomechenically, the authors noted a significant decrease in torsional stiffness, as well as loss of anterior and medial-lateral shear stiffness of the L4-L5 interspace. More recently, Norcross described the effects of chondroitinase ABC on the intervertebral disc of the rat tail.74 The findings were similar to the changes observed in degenerative disc disease: reduced intervertebral disc height, diminished proteoglycan content, loss of NP cells, and increased stiffness of the disc. Fibronectin Fragment Injection
Anderson and associates have shown that an injection of fibronectin fragments induced the degeneration of intervertebral discs in
646
P A R T
X I
Cell-Based Biologic Scientific Methods and Future Clinical Applications
during growth or due to Col2a1 deficiency as an adult. Similar degenerative changes were also observed in the lumbar spine of mice lacking the GDF-8 (myostatin) gene.79 Hamrick and colleagues noted that there was a loss of proteoglycan staining in the hyaline end plates and inner AF of the knockout mice. Results from this study suggest that increased muscle mass in mice lacking myostatin is associated with increased bone mass as well as degenerative changes in the intervertebral disc. CONCLUSION Many techniques have been applied in order to develop a successful experimental animal model of human disc degeneration. However, no one particular model currently parallels the complex nature of human disc degeneration. Naturally occurring animal models have the drawback that the basis for the high rate of disc degeneration is not known. The availability of animals is based on the rate of occurrence, so a predictable experiment is difficult. Although the interventions in artificial animal models are known, the true relationship of these changes to the actual events leading to disc degeneration in humans is not. The careful comparison of data obtained from an animal model with those from human pathologic specimens will shed light on the detailed mechanisms of disc degeneration. With recent progress in biomechanics, cell biology, and molecular biology, an easily reproducible and valid animal model may help unlock the complex cascade of events surrounding human disc degeneration. Only then might it be possible to offer insight into the prevalent and disabling condition of back pain. Although there is no perfect animal model to study the degeneration and regeneration of the intervertebral disc, when choosing a proper model to prove a therapeutic hypothesis, careful consideration should be given to balancing the humane treatment and judicious use of animals with the potential benefits to humans.
10. 11.
12. 13. 14. 15. 16.
17. 18. 19. 20. 21. 22. 23.
REFERENCES 1. Ferguson SJ, Steffen T: Biomechanics of the aging spine. Eur Spine J 12(2):S97–S103, 2003. 2. Boos N, Weissbach S, Rohrbach H, et al: Classification of agerelated changes in lumbar intervertebral discs: 2002 Volvo Award in basic science. Spine 27(23):2631–2644, 2002. 3. Lotz JC, Hsieh AH, Walsh AL, et al: Mechanobiology of the intervertebral disc. Biochem Soc Trans 30(Pt 6):853–858, 2002. 4. Ariga K, Miyamoto S, Nakase T, et al: The relationship between apoptosis of endplate chondrocytes and aging and degeneration of the intervertebral disc. Spine 26(22):2414–2420, 2001. 5. Gruber HE, Hanley EN Jr: Ultrastructure of the human intervertebral disc during aging and degeneration: comparison of surgical and control specimens. Spine 27(8):798–805, 2002. 6. Roughley PJ, Alini M, Antoniou J: The role of proteoglycans in aging, degeneration and repair of the intervertebral disc. Biochem Soc Trans 30(Pt 6):869–874, 2002. 7. Oegema TR Jr: Biochemistry of the intervertebral disc. Clin Sports Med 12(3):419–439, 1993. 8. Hunter CJ, Matyas JR, Duncan NA: The functional significance of cell clusters in the notochordal nucleus pulposus: Survival and signaling in the canine intervertebral disc. Spine 29(10):1099–1104, 2004. 9. Kim KW, Ha KY, Park JB, et al: Expressions of membrane-type I matrix metalloproteinase, Ki-67 protein, and type II collagen by chondrocytes migrating from cartilage endplate into nucleus pulposus
24. 25. 26. 27. 28. 29.
30. 31.
in rat intervertebral discs: A cartilage endplate-fracture model using an intervertebral disc organ culture. Spine 30(12):1373–1378, 2005. Kim KW, Kim YS, Ha KY, et al: An autocrine or paracrine Fas-mediated counterattack: A potential mechanism for apoptosis of notochordal cells in intact rat nucleus pulposus. Spine 30(11):1247–1251, 2005. Kim KW, Lim TH, Kim JG, et al: The origin of chondrocytes in the nucleus pulposus and histologic findings associated with the transition of a notochordal nucleus pulposus to a fibrocartilaginous nucleus pulposus in intact rabbit intervertebral discs. Spine 28(10):982–990, 2003. Silberberg R, Aufdermaur M, Adler JH: Degeneration of the intervertebral disks and spondylosis in aging sand rats. Arch Pathol Lab Med 103(5):231–235, 1979. Adler JH, Schoenbaum M, Silberberg R: Early onset of disk degeneration and spondylosis in sand rats (Psammomys obesus). Vet Pathol 20(1):13–22, 1983. Moskowitz RW, Ziv I, Denko CW, et al: Spondylosis in sand rats: A model of intervertebral disc degeneration and hyperostosis. J Orthop Res 8(3):401–411, 1990. Ziv I, Moskowitz RW, Kraise I, et al: Physicochemical properties of the aging and diabetic sand rat intervertebral disc. J Orthop Res 10(2):205–210, 1992. Gruber HE, Johnson T, Norton HJ, Hanley EN Jr: The sand rat model for disc degeneration: Radiologic characterization of agerelated changes: Cross-sectional and prospective analyses. Spine 27(3):230–234, 2002. Gruber HE, Johnson TL, Leslie K, et al: Autologous intervertebral disc cell implantation: A model using Psammomys obesus, the sand rat. Spine 27(15):1626–1633, 2002. Berry R: Genetically controlled degeneration of the nucleus pulposus in the mouse. J Bone Joint Surg 43B:387–393, 1961. Silberberg R, Gerritsen G: Aging changes in intervertebral discs and spondylosis in Chinese hamsters. Diabetes 25(6):477–483, 1976. Kim KS, Yoon ST, Li J, et al: Disc degeneration in the rabbit: A biochemical and radiological comparison between four disc injury models. Spine 30(1):33–37, 2005. Cole TC, Ghosh P, Taylor TK: Variations of the proteoglycans of the canine intervertebral disc with ageing. Biochim Biophys Acta 880(2–3):209–219, 1986. Goggin JE, Li AS, Franti CE: Canine intervertebral disk disease: Characterization by age, sex, breed, and anatomic site of involvement. Am J Vet Res 31(9):1687–1692, 1970. Ghosh P, Taylor TK, Braund KG, Larsen LH: The collagenous and non-collagenous protein of the canine intervertebral disc and their variation with age, spinal level and breed. Gerontology 22(3):124– 134, 1976. Gillett NA, Gerlach R, Cassidy JJ, Brown SA: Age-related changes in the beagle spine. Acta Orthop Scand 59(5):503–507, 1988. Lauerman WC, Platenberg RC, Cain JE, Deeney VF: Age-related disk degeneration: Preliminary report of a naturally occurring baboon model. J Spinal Disord 5(2):170–174, 1992. Goff C, Landmesser W: Bipedal rats and mice: Laboratory animals for orthopaedic research. J Bone Joint Surg 39A:616–622, 1957. Yamada K: The dynamics of experimental posture: Experimental study of intervertebral disk herniation in bipedal animals. Clin Orthop 25:20–31, 1962. Higuchi M, Abe K, Kaneda K: Changes in the nucleus pulposus of the intervertebral disc in bipedal mice. A light and electron microscopic study. Clin Orthop 1983(175):251–257, 1962. Cassidy JD, Yong-Hing K, Kirkaldy-Willis WH, Wilkinson AA: A study of the effects of bipedism and upright posture on the lumbosacral spine and paravertebral muscles of the Wistar rat. Spine 13(3):301–308, 1988. Hutton WC, Yoon ST, Elmer WA, et al: Effect of tail suspension (or simulated weightlessness) on the lumbar intervertebral disc: Study of proteoglycans and collagen. Spine 27(12):1286–1290, 2002. Lotz J, Colliou O, Chin J, et al: Compression-induced degeneration of the intervertebral disc: An in vivo mouse model and finite element study. Spine 23(23):2493–2506, 1998.
CHAPTER 83
32. Iatridis JC, Mente PL, Stokes IA, et al: Compression-induced changes in intervertebral disc properties in a rat tail model. Spine 24(10):996–1002, 1999. 33. Ching CT, Chow DH, Yao FY, Holmes AD: The effect of cyclic compression on the mechanical properties of the inter-vertebral disc: An in vivo study in a rat tail model. Clin Biomech (Bristol, Avon) 18(3):182–189, 2003. 34. Kroeber MW, Unglaub F, Wang H, et al: New in vivo animal model to create intervertebral disc degeneration and to investigate the effects of therapeutic strategies to stimulate disc regeneration. Spine 27(23):2684–2690, 2002. 35. Hadjipavlou AG, Simmons JW, Yang JP, et al: Torsional injury resulting in disc degeneration in the rabbit, II: Associative changes in dorsal root ganglion and spinal cord neurotransmitter production. J Spinal Disord 11(4):318–321, 1998. 36. Hadjipavlou AG, Simmons JW, Yang JP, et al: Torsional injury resulting in disc degeneration, I: An in vivo rabbit model. J Spinal Disord 11(4):312–317, 1998. 37. Miyamoto S, Yonenobu K, Ono K: Experimental cervical spondylosis in the mouse. Spine 16(10):S495–500, 1991. 38. Sullivan J, Farfan H, Kahn D: Pathologic changes with intervertebral joint rotational instability in the rabbit. Can J Surg 14:71–79, 1971. 39. Phillips FM, Reuben J, Wetzel FT: Intervertebral disc degeneration adjacent to a lumbar fusion: An experimental rabbit model. J Bone Joint Surg Br 84(2):289–294, 2002. 40. Keyes D, Compere E: The normal and pathological physiology of the nucleus pulposus of the intervertebral disc: An anatomical, clinical, and experimental study. J Bone Joint Surg 14:897–938, 1932. 41. Smith J, Walmsley R: Experimental incision of the intervertebral disc. J Bone Joint Surg 33B:612–625, 1951. 42. Haimovici EH: Experimental disc lesion in rabbits: The effect of repeated ACTH administration. Acta Orthop Scand 41(5):505– 521, 1970. 43. Lipson SJ, Muir H: Experimental intervertebral disc degeneration: Morphologic and proteoglycan changes over time. Arthritis Rheum 24(1):12–21, 1981. 44. Urayama S: Histological and ultrastructural study of degeneration of the lumbar intervertebral disc in the rabbit following nucleotomy, with special reference to the topographical distribution pattern of the degeneration. Nippon Seikeigeka Gakkai Zasshi 60:649–662, 1986. 45. Aota Y, Masuda K, Nguyen C, et al: Radiological and MRI analyses of a novel rabbit model: A mild, progressive disc degeneration. Ortho Res Soc Trans 27:118, 2002. 46. Muehleman C, Masuda K, Imai Y, et al: Histological assessment of a novel needle puncture model: A mild, progressive intervertebral disc degeneration. Ortho Res Soc Trans 28:86, 2003. 47. Masuda K, Aota Y, Muehleman C, et al: A novel rabbit model of mild, reproducible disc degeneration by an annulus needle puncture: Correlation between the degree of disc injury and radiological and histological appearances of disc degeneration. Spine 30(1):5–14, 2005. 48. Kompel J, Sobajima S, Clarke C, et al: MRI and histological analysis of a rabbit model of disc degeneration. Ortho Res Soc Trans 28:253, 2003. 49. Osti OL, Vernon-Roberts B, Fraser RD: 1990 Volvo Award in experimental studies: Anulus tears and intervertebral disc degeneration. An experimental study using an animal model. Spine 15(8):762–767, 1990. 50. Kaapa E, Holm S, Han X, et al: Collagens in the injured porcine intervertebral disc. J Orthop Res 12(1):93–102, 1994. 51. Kaapa E, Zhang LQ, Muona P, et al: Expression of type I, III, and VI collagen mRNAs in experimentally injured porcine intervertebral disc. Connect Tissue Res 30(3):203–214, 1994. 52. Kaapa E, Gronblad M, Holm S, et al: Neural elements in the normal and experimentally injured porcine intervertebral disk. Eur Spine J 3(3):137–142, 1994. 53. Pfeiffer M, Griss P, Franke P: Degeneration model of the porcine lumbar motion segment: Effects of various intradiscal procedures. Eur Spine J 3:8–16, 1994. 54. Holm S, Holm AK, Ekstrom L, et al: Experimental disc degeneration due to endplate injury. J Spinal Disord Tech 17(1):64–71, 2004.
Animal Models for Human Disc Degeneration
647
55. Wada E, Ebara S, Saito S, Ono K: Experimental spondylosis in the rabbit spine: Overuse could accelerate the spondylosis. Spine 17(3): S1–S6, 1992. 56. Zook BC, Kobrine AI: Effects of collagenase and chymopapain on spinal nerves and intervertebral discs of cynomolgus monkeys. J Neurosurg 64(3):474–483, 1986. 57. Kato F, Iwata H, Mimatsu K, Miura T: Experimental chemonucleolysis with chondroitinase ABC. Clin Orthop 1990(253):301–308, 1986. 58. Takahashi T, Nakayama M, Chimura S, et al: Treatment of canine intervertebral disc displacement with chondroitinase ABC. Spine 22 (13):1435–1439; discussion 1446–1447, 1997. 59. Takahashi T, Kurihara H, Nakajima S, et al: Chemonucleolytic effects of chondroitinase ABC on normal rabbit intervertebral discs. Course of action up to 10 days postinjection and minimum effective dose. Spine 21(21):2405–2411, 1996. 60. Sumida K, Sato K, Aoki M, et al: Serial changes in the rate of proteoglycan synthesis after chemonucleolysis of rabbit intervertebral discs. Spine 24(11):1066–1070, 1999. 61. Spencer DL, Miller JA, Schultz AB: The effects of chemonucleolysis on the mechanical properties of the canine lumbar disc. Spine 10 (6):555–561, 1985. 62. Sasaki M, Takahashi T, Miyahara K, Hirosea T: Effects of chondroitinase ABC on intradiscal pressure in sheep: An in vivo study. Spine 26(5):463–468, 2001. 63. Ono A, Harata S, Takagaki K, Endo M: Proteoglycans in the nucleus pulposus of canine intervertebral discs after chondroitinase ABC treatment. J Spinal Disord 11(3):253–260, 1998. 64. Olmarker K, Rydevik B, Dahlin LB, et al: Effects of epidural and intrathecal application of collagenase in the lumbar spine: An experimental study in rabbits. Spine 12(5):477–482, 1987. 65. Nitobe T, Harata S, Okamoto Y, et al: Degradation and biosynthesis of proteoglycans in the nucleus pulposus of canine intervertebral disc after chymopapain treatment. Spine 13(11):1332–1339, 1988. 66. Miyabayashi T, Lord PF, Dubielzig RR, et al: Chemonucleolysis with collagenase: A radiographic and pathologic study in dogs. Vet Surg 21(3):189–194, 1992. 67. Lu DS, Shono Y, Oda I, et al: Effects of chondroitinase ABC and chymopapain on spinal motion segment biomechanics: An in vivo biomechanical, radiologic, and histologic canine study. Spine 22(16):1828–1834; discussion 1834–1835, 1997. 68. Kudo T, Sumi A, Hashimoto A: Experimental chemonucleolysis with chymopapain in canine intervertebral disks. J Vet Med Sci 55 (2):211–215, 1993. 69. Kahanovitz N, Arnoczky SP, Kummer F: The comparative biomechanical, histologic, and radiographic analysis of canine lumbar discs treated by surgical excision or chemonucleolysis. Spine 10 (2):178–183, 1985. 70. Ishikawa H, Nohara Y, Miyauti S: Action of chondroitinase ABC on epidurally transplanted nucleus pulposus in the rabbit. Spine 24 (11):1071–1076, 1999. 71. Henderson N, Stanescu V, Cauchoix J: Nucleolysis of the rabbit intervertebral disc using chondroitinase ABC. Spine 16(2):203–208, 1991. 72. Henderson N, Stanescu V, Cauchoix J: [Nucleolytic action of chondroitinase ABC on the lumbar disc of the rabbit]. C R Acad Sci III 307(7):403–406, 1988. 73. Fry TR, Eurell JC, Johnson AL, et al: Radiographic and histologic effects of chondroitinase ABC on normal canine lumbar intervertebral disc. Spine 16(7):816–819, 1991. 74. Norcross JP, Lester GE, Weinhold P, Dahners LE: An in vivo model of degenerative disc disease. J Orthop Res 21(1):183–188, 2003. 75. Anderson GD, Li X, Tannoury T, et al: A fibronectin fragment stimulates intervertebral disc degeneration in vivo. Spine 28(20):2338– 2345, 2003. 76. Iwahashi M, Matsuzaki H, Tokuhashi Y, et al: Mechanism of intervertebral disc degeneration caused by nicotine in rabbits to explicate intervertebral disc disorders caused by smoking. Spine 27 (13):1396–1401, 2002.
648
P A R T
X I
Cell-Based Biologic Scientific Methods and Future Clinical Applications
77. Oda H, Matsuzaki H, Tokuhashi Y, et al: Degeneration of intervertebral discs due to smoking: Experimental assessment in a ratsmoking model. J Orthop Sci 9(2):135–141, 2004. 78. Sambrook PN, MacGregor AJ, Spector TD: Genetic influences on cervical and lumbar disc degeneration: a magnetic resonance imaging study in twins. Arthritis Rheum 42(2):366–372, 1999. 79. Hamrick MW, Pennington C, Byron CD: Bone architecture and disc degeneration in the lumbar spine of mice lacking GDF-8 (myostatin). J Orthop Res 21(6):1025–1032, 2003. 80. Ziran BH, Pineda S, Pokharna H, et al: Biomechanical, radiologic, and histopathologic correlations in the pathogenesis of experimental intervertebral disc disease. Spine 19(19):2159–2163, 1994. 81. Bradford DS, Cooper KM, Oegema TR Jr: Chymopapain, chemonucleolysis, and nucleus pulposus regeneration. J Bone Joint Surg Am 65(9):1220–1231, 1983.
82. Bradford DS, Oegema TR Jr, Cooper KM, et al: Chymopapain, chemonucleolysis, and nucleus pulposus regeneration: A biochemical and biomechanical study. Spine 9(2):135–147, 1984. 83. Eurell JA, Brown MD, Ramos M: The effects of chondroitinase ABC on the rabbit intervertebral disc: A roentgenographic and histologic study. Clin Orthop 256:238–243, 1990. 84. Takenaka Y, Revel M, Kahan A, Amor B: Experimental model of disc herniations in rats for study of nucleolytic drugs. Spine 12 (6):556–560, 1987. 85. Sahlman J, Inkinen R, Hirvonen T, et al: Premature vertebral endplate ossification and mild disc degeneration in mice after inactivation of one allele belonging to the Col2a1 gene for Type II collagen. Spine 26 (23):2558–2565, 2001.
CHAPTER
84
Growth Factors for Intervertebral Disc Regeneration Koichi Masuda and Howard S. An
K E Y l
l
l
l
l
P O I N T S
In vitro, growth factors, such as bone morphogenetic protein-7 and growth and differentiation factor-5, enhanced cell proliferation and the synthesis and accumulation of proteoglycan and collagen by bovine and human nucleus pulposus and annulus fibrosus cells cultured in alginate beads. In vivo, a single injection of a growth factor into the nucleus pulposus resulted in improved radiographic findings and MRI and histologic grades of intervertebral discs in the rabbit annular-puncture disc degeneration model. The findings of the biomechanical study provided evidence that an injection of a growth factor restored the biomechanical properties of degenerated discs in the rabbit annular-puncture disc degeneration model. The biomechanical and biochemical data in these studies suggested that an injection of a growth factor induced biochemical changes by anabolic stimulation, which may have resulted in the biomechanical restoration of the intervertebral disc. Although further efficacy studies with larger animals and studies using pain as a primary end point for a therapeutic approach for disc degeneration are needed, the preliminary evidence in these rabbit studies established that an injection of a growth factor may be clinically applicable as a therapeutic approach to repair the degenerated intervertebral disc.
MAINTENANCE OF MATRIX HOMEOSTASIS IN THE INTERVERTEBRAL DISC: GROWTH FACTORS AND OTHER BIOLOGIC MOLECULES Biologically, disc cells actively regulate disc tissue homeostasis by maintaining a balance between anabolism and catabolism. The activity of disc cells is modulated by a variety of substances, including cytokines, growth factors, enzymes, and enzyme inhibitors, in a paracrine or autocrine fashion.1,2 Recent therapeutic strategies for disc degeneration have included attempts to upregulate important extracellular matrix components (i.e., aggrecan)3,4 or to downregulate proinflammatory cytokines (i.e., interleukin 1 [IL-1] or tumor necrosis factor-a [TNF-a])5–12 and matrix-degrading enzymes (i.e., matrix metalloproteinases
[MMPs] and members of a disintegrin-like and metalloprotease with thrombospondin motifs (ADAMTS) family (aggrecanases).13–15 It may be possible that a combination of both strategies could be most efficacious. As the name implies, growth factors play an important role in the development of the spine. Interestingly, several growth factors have been found in normal and degenerated intervertebral disc (IVD) tissues, suggesting that IVD cells are capable of expressing and producing growth factors. These factors include insulin-like growth factor-1 (IGF-1),16,17 basic fibroblast growth factor (bFGF),18–21 bone morphogenetic-2 (BMP-2),22 BMP4,22,23 growth differentiation factor-5 (GDF-5),23 platelet-derived growth factor (PDGF),24 and transforming growth factor-b (TGF-b).17,25–29 Taken together, the autocrine and paracrine production of growth factors is considered to be a major regulatory mechanism in IVD tissues.16 For delivering therapeutic biologic agents into IVD tissue, several methods of administration can be considered. The direct injection of a protein is relatively simple and practical; however, the efficiency, duration of action, and possibility of adverse effects are not currently known. To regenerate or repair a degenerated IVD, the injection of protein anabolic factors would be a simple and practical approach. However, several issues, such as half-life, solubility of the protein, a proper carrier, and a presence of inhibitors, etc., need to be taken into consideration. In the past, the halflife of an injected protein was considered to be very short, in the order of minutes.30 However, recent research by our group, in collaboration with Stryker Biotech (Hopkinton, MA), revealed that the half-life of injected radiolabeled osteogenic protein-1 (OP-1, otherwise known as BMP-7) was greater than the 1 month observation period; this result may support our contention that a simple protein injection may have therapeutic efficacy. The preclinical development of an injection of an anabolic factor is one of the most advanced in the biologic treatment of disc diseases. In fact, the first investigational new drug study to test the safety and efficacy of OP-1 has been initiated. 649
P A R T
650
X I
Cell-Based Biologic Scientific Methods and Future Clinical Applications
IN VITRO EVIDENCE FOR THE POSSIBLE THERAPEUTIC ROLE OF GROWTH FACTORS IN DEGENERATIVE DISC DISEASE A disruption of disc tissue homeostasis may be achieved by stimulating matrix synthesis with cytokines or growth factors, resulting in a shift of cellular metabolism to the anabolic state.1 Others have shown that the rate of proteoglycan (PG) synthesis by IVD cells increases several-fold by the addition of TGF-b and epidermal growth factor (EGF).31,32 IGF-1 was found to stimulate matrix synthesis and proliferation in vitro.16 Using human IVD cells in three-dimensional culture, Gruber and associates first demonstrated that, after 4 days’ exposure, TGFb stimulated cell proliferation by annulus fibrosus (AF) cells.32 These authors also reported that IGF-1 and PDGF significantly reduced the percentage of apoptotic AF cells induced by serum depletion during culture.33 More recently, we have shown that OP-1 enhanced the PG metabolism of IVD cells.34,35 OP-1, a member of the TGF-b superfamily, was initially found to exert potent effects on osteocyte and chondrocyte differentiation and metabolism.36,37 In chondrocytes, OP-1 stimulated the synthesis of PGs and type II collagen.38,39 A series of studies by our laboratory has demonstrated that OP-1 strongly stimulated the production and formation of extracellular matrix molecules, including PGs and collagen, by rabbit IVD cells,34 as well as by human IVD cells in vitro (Fig. 84–1).35 After depletion of the extracellular matrix following exposure of IVD cells to IL-1, OP-1 was also found to be effective in the replenishment of a matrix rich in PGs and collagens.40 A similar result with OP-1 was reported when the matrix was first depleted by in vitro chemonucleolysis using chondroitinase ABC.41 Another BMP, BMP-2, stimulated matrix production and cell proliferation by rat IVD cells.42 Using human IVD cells, Kim and associates reported that BMP-2 facilitated the expression of the
chondrogenic phenotype, increased PG synthesis, and upregulated the expression of aggrecan, collagen type I, and collagen type II mRNA, compared to untreated control levels.43 GDF-5, another member of the BMP family, was originally found to be a factor responsible for skeletal alterations in brachypodism mice.44 Recently, an analysis of GDF-5-deficient mice revealed the presence of disc degeneration, as well as a loss of PGs from the IVD.45 GDF-5 was also shown to stimulate PG and type II collagen expression in mouse IVD cells.45 Furthermore, we have shown that recombinant human GDF-5 (rhGDF-5) enhanced cell proliferation and matrix synthesis and accumulation by both bovine nucleus pulposus (NP) and AF cells. The response to rhGDF-5 was greater by NP cells than by AF cells (Fig. 84–2).3 The application of autologous growth factors may be advantageous for the clinical treatment of degenerative disc disease. Wehling and colleagues showed that the combination of autologous IL-1 receptor antagonist (IL-1ra)/IGF-1/PDGF proteins reduced the percentage of apoptosis and the production of IL-1 and IL-6 by IVD cells.46 Platelet-rich plasma (PRP), which is a fraction of plasma that can be produced by centrifugal separation of whole blood in the operating room, contains multiple growth factors concentrated at high levels.47–50 In porcine NP and AF cells cultured in alginate beads, PRP was found to be an effective stimulator of cell proliferation, PG and collagen synthesis, as well as PG accumulation (Fig. 84–3).51 IN VIVO STUDIES FOR THE DEVELOPMENT OF GROWTH FACTOR INJECTION THERAPY IN DEGENERATIVE DISC DISEASE Walsh and associates have reported the in vivo effects of a single injection of several growth factors, including bFGF, GDF-5,
* P 0.05). Bilateral uncovertebral joint resection combined with disc arthroplasty indicated significantly greater segmental motion than the intact spine (P < 0.05). Flexion-extension testing demonstrated a significant increase in range of motion for all groups with discectomy and uncovertebral joint resection when compared to the intact condition or those stabilized with the PCM arthroplasty (P < 0.05). PCM arthroplasty reconstruction following discectomy restored the flexion-extension neutral zone and range of motion to the intact condition (P > 0.05). Unilateral and bilateral uncovertebral joint resection demonstrated increased neutral zone values compared to all other treatments (P < 0.05). Neutral zone values for the unilateral and bilateral uncovertebral joint resection were significantly greater than all other treatments groups (P < 0.05); however, cervical arthroplasty restored neutral zone levels to the intact condition (P > 0.05). DISCUSSION Because of the resection of ligaments and portions of the annulus fibrosis, the preparation for a TDR always destabilizes both the
Spinal Deformity and Motion-Sparing Technology
cervical and the lumbar spinal segments to some extent.8,9 However, implanting an unconstrained cervical TDR restores the rotational stability back to the level of the intact condition, whereas inserting an unconstrained lumbar TDR remains rotationally more unstable. This study characterized four significant biomechanical differences, highlighting four biomechanical and anatomic factors for superior inherent rotational stability in cervical TDR versus lumbar TDR: 1. Cervical TDR has a larger arc of influence (153.6 degrees versus 98 degrees). 2. Cervical TDR had a greater facet/end plate area ratio (0.519 versus 0.283). 3. The tensile load of the major ligaments of the lumbar and cervical spines was quantitated by White and Panjabi. The rank order of ligament tensile strength from highest to lowest for the lumbar spine is ALL ! PLL ! LF ! CL. For the cervical spine the rank order is quite different—CL ! LF ! ALL ! PLL. This difference has important ramifications for arthroplasty stability because the ALL is always resected and the PLL is released or stretched. Whereas the ligamentum flavum (LF) and the capsular ligaments
AXIAL ROTATION–RANGE OF MOTION (ROM) One-level reconstruction
Two-level reconstruction
300
250
∼
L3-L4 L4-L5 L5-S1
** #
**
% Intact
200
Construct 6 **
150 *
* *
100
50
0 Discectomy
CHARITÉ
CHARITÉ + pedicle screws
701
CHARITÉ
CHARITÉ + CHARITÉ FRA + pedicle FRA + pedicle pedicle A screws screws screws ´ . Peak range of n F I G U R E 89–5. A, Rotational biomechanical testing of one-level and two-level CHARITE motion comparing the intact spine to seven methods of reconstruction highlighting changes in segmental motion at the L3-L4, L4-L5, and L5-S1 levels. The seven series of three bar graphs (n ¼ 7) represent the peak range of motion (ROM) at three levels, L3–4, L4–5, and L5-S1, normalized to the intact lumbar cadaveric segment before treatment, with intact percentage designated as 100% on the graph. From left to right the seven treatments were (1) destabilized (discectomy preparation for lumbar total disc replacement) at L4-L5; (2) one-level CHARITE´ at L4-L5; (3) CHARITE´ and pedicle screws at L4-L5; (4) two-level CHARITE´ at L4-L5 and L5-S1; (5) CHARITE´ and pedicle screws at two levels; (6) CHARITE´ þ FRA (femoral ring allograft) spacer and pedicle screws at two levels; and (7) FRA and pedicle screws at two levels. Notice that with one-level CHARITE´ at L4-L5 that the rotational stiffness is 160% (SD 26%), meaning that the acute rotational stability is not restored to the level of the intact segment.
702
P A R T
X I I
Controversies
n F I G U R E 89–5. Cont’d. B, An AP radiograph of construct number 6 is shown in the fully rotated position on the spine simulator. The three levels are intact L3-L4, CHARITE´ at L4-L5, and 360 instrumentation at L5-S1 (femoral ring allograft þ pedicle screws). The AP radiograph of the specimen looks very similar to the rotational deformity of the clinically failed CHARITE´ at L4-L5 shown in Figure 89–2. Notice the rotation of the spinous processes as well as the obliquity of the implant components. We can attest that the CHARITE´ was placed in midcenter of L4-L5 in the laboratory specimen. If the intact disc is taken as 100%, the CHARITE´ demonstrated 290%, whereas the 360-degree instrumentation demonstrated 85% rotation. In absolute degrees, the total rotational deformity of the hybrid construct was 10.86 degrees compared to an intact (unoperated) specimen of 3.31 degrees.
(CL) are, of course, preserved with arthroplasty. The ratio of CL/ALL tensile load for the cervical spine is 204/111 n ¼ 1.93. The CL/ALL tensile load ratio is much smaller for the lumbar spine at 222/450 n ¼ 0.493. 4. Depending on the amount of disc height distraction and lateral decompression with the cervical TDR, the uncovertebral joints contribute to rotational stability. We found with unilateral uncovertebral joint resection, however, that even an unconstrained cervical TDR restored the neutral zone of the intact condition. The capsular ligaments and lateral masses clearly are the primary stabilizers of rotational stability in the cervical spine. The implications in the cervical spine for TDR clearly show that even
for a radical discectomy and uncovertebral resection a TDR can safely restore the stability of the cervical spine back to the physiologic neutral zone. However, in the lumbar spine the annulus fibrosis, ALL, and PLL are the three most important primary stabilizers in rotation and all three are compromised in lumbar TDR. Even with normal preparation of the lumbar disc for TDR, the CHARITÉ does not return the rotational stability back to the physiologic neutral zone but results in 120% to 140% rotation compared to the intact condition. The multilevel TDR lumbar implantation has an additive destabilizing effect, as two-level annulus lumbar resection results in 240% to 260% rotation (with intact being normalized to 100%). The findings in the present biomechanical investigation highlight a variety of important trends at the operative and adjacent
CHAPTER 89
A
Spinal Deformity and Motion-Sparing Technology
703
B
F I G U R E 89–6. A and B, Uncovertebral joint restrains rotation. This is the benchtop rotational testing of stability to an axially rotated cervical segment with an unconstrained prosthesis implanted at C5-C6. The arrows designate the right uncovertebral joint. Notice that with axial rotation to the right side that the uncovertebral joint closes down, serving as a rotational bony block. An exact number or percentage of the contribution of the uncovertebral joint toward rotational stability is not possible—the contribution of the uncovertebral joint is extremely variable, depending on the amount of the uncus resected during the lateral extent of the cervical neural decompression, and the amount of disc space height restored during artificial disc implantation. Notice also that in the cervical spine coupled motions are much more important. As the specimen is rotated toward the right, the two vertebrae demonstrate right lateral bending with the vertebral end plates laterally bending toward each other. This also tends to close down the uncovertebral joint and also serves as a constraint toward cervical rotation.
n
levels. Axial rotation loading produced the greatest differences in segmental range of motion, particularly under two-level reconstructions. There were no differences observed at the L3-L4 level for any treatment group—single or two level. When compared to the intact condition, segmental motion at the L4-L5 level markedly increased from 160 26% to 263 65% with the implantation of the second CHARITÈ at L5-S1. Moreover, the addition of pedicle screws and FRA at L5-S1 further increased the motion at L4-L5 to 292 71%, which was significantly greater than all other groups except the two-level CHARITÉ (P < 0.05). The most interesting finding under this loading modality was that observed with the addition of a second CHARITÈ at L5-S1. While segmental motion at L4-L5 markedly increased, motion at L5-S1 significantly increased to 263 65% over the intact condition, which was different from all other treatments (P < 0.05). This suggests that a disc replacement adjacent to a 360-degree instrumentation can become unstable. These benchtop biomechanical findings from the current study corroborate the clinical work of Sariali and associates10 who investigated the in vivo kinematic response to single- and two-level CHARITÈ reconstructions. A comparative retrospective study of motion in axial rotation at L4-L5 level was carried out on 17 patients with the CHARITÈ device versus six healthy volunteers. Five patients had one prostheses at the L4-L5 level and 12 had two prosthesis at L4L5 and L5-S1 levels. The follow-up ranged from 10.8 to 14.3 years (average 12.4 1, median 12.6). Eleven (65%) patients had a normal mobility in torsion, identical to those of the volunteers and of the literature, whereas six (35%) had an abnormal increased mobility. According to Sariali and associates, if only one disc was replaced, mobility in torsion was identical to that of the volunteers. In the case of two replaced discs, 50% (6 of 12) of the patients had an abnormal
increased mobility. Moreover, in the subgroup of increased mobility, the coupling was different with increased flexion (10 degrees) (P < 0.001). According to Sariali and associates, the implantation of one CHARITÈ device appears to restore kinematics close to that of the healthy volunteers. However, two adjacent prostheses produce a condition of abnormal kinematics in 50% of the cases. There should be a disclaimer—a major shortcoming of this type of acute instability found in biomechanical testing cannot be emphasized enough. The major limitation of all phases of this investigation is the lack of consideration of the contribution of muscular forces, progressive biologic healing, heterotopic ossification, fibrosis, adhesions, and other processes that would be expected to contribute to long-term rotational stability following clinical surgical recovery. What About a More Constrained Prosthesis Design? We did not study the effect of having a more constrained prosthesis design. Perhaps the effect might be more stress internal to the prosthesis with a higher incidence of lateral or scoliotic articulation of the components rather than more obvious obliquity of the vertebral bodies. For example, notice that the same anatomic and biomechanical forces are still relevant to the motion segment. This constrained device required revision due to asymmetric scoliotic collapse and unintended metal-on-metal contact of a nonarticular part of the prosthesis. The spinal malalignment is not as obvious by Cobb measurement, but the premature failure of the prosthesis is just as clinically significant. The removal of an ingrown keeled prosthesis due to premature biomaterial failure less than 12 months postoperatively is a supreme surgical challenge and one of the most difficult procedures in spinal surgery. The use of a
704
P A R T
X I I
Controversies
more constrained prosthesis in the peripheral skeleton does not obviate problems, as a more constrained prosthesis increases the stresses at the metal-bone interface, increases the incidence of loosening, and can reduce the range of prosthetic motion. CONCLUSIONS 1. There are four main rotational stabilizing considerations in the functional spinal unit, and all four favor cervical versus lumbar rotational integrity. 2. Compared to the intact nonoperated motion segment, the preparation for both cervical and lumbar disc replacements compromises the axial rotational stability of both the cervical and lumbar spine—resection of the ALL, stretching or “release” of the PLL, increasing the disc space height which “unlocks” the posterior facet joints, and removal of portions of the uncovertebral joints in the cervical spine. 3. With an unconstrained prosthesis, the uncovertebral joint can be resected and the rotational stability of the intact cervical segment can be restored, but not with bilateral uncovertebral joint resection. 4. Implantation of an unconstrained lumbar disc replacement fails to restore the acute rotational stability of the lumbar spine. However, with an unconstrained prosthesis in the cervical spine, the rotational stability is restored within the neutral zone of the intact condition. With multilevel or successive lumbar artificial disc replacement, the instability is additive—what is lost in resecting the annulus, ALL, and PLL is not compensated for totally in the acute postoperative condition.
REFERENCES 1. McAfee PC, Cunningham BW, Lee GA, et al: Revision strategies for salvaging or improving failed cylindrical cages. Spine 24:2147–2153, 1999. 2. McAfee PC, Cunningham BW, Hayes V, et al: Biomechanical analysis of rotational motions after disc arthroplasty: Implications for patients with adult deformities. Spine 31(19):S152–S160, 2006. 3. Blumenthal SL, McAfee PC, Guyer RD, et al: A prospective, randomized, multi-center FDA IDE Study of lumbar total disc replacement with the CHARITÉ™ Artificial Disc vs. lumbar fusion: Part I—Evaluation of clinical outcomes. Spine 2005 (in press). 4. McAfee PC, Cunningham BW, Holtsapple G, et al: A prospective, randomized, multi-center FDA IDE Study of lumbar total disc replacement with the CHARITÉ™ Artificial Disc vs. lumbar fusion: Part II—Evaluation of radiographic outcomes and correlation of surgical technique accuracy with clinical outcomes. Spine 2005 (in press). 5. Buttner-Janz K, Hochshuler SH, McAfee PC: The Artificial Disk. New York, Springer-Verlag, 2003. 6. McAfee PC: Artificial disc prosthesis: The link SB CHARITÉ. In Kaech DL, Jinkins JR (eds): Spinal Rehabilitation Procedures. St. Louis, Elsevier, 2002, pp 299–310. 7. McAfee PC, Cunningham BW, Orbegoso CM, et al: Analysis of porous ingrowth in intervertebral disc prostheses: A nonhuman primate model. Spine 28:332–340, 2003. 8. McAfee PC, Geisler FH, Saiedy SS, et al: Revisibility of the CHARITÉ Artificial Disc: Analysis of 688 patients enrolled in the US IDE Study of the Charite Artificial Disc. Spine 31(11):1217–1226, 2006. 9. McAfee PC, Geisler FH, Scott-Young M (eds): Roundtables in Spine Surgery: Complications and Revision Strategies in Lumbar Spine Arthroplasty. St. Louis, Quality Medical Publishing, 2005. 10. Sariali EH, Lemaire JP, Pascal-Mousselard H, et al: In vivo study of the kinematics in axial rotation of the lumbar spine after total intervertebral disc replacement: Long-term results: A 10-14 year followup evaluation. Eur Spine J 21:1–10, 2006.
CHAPTER
90
Can Lumbar Disc Replacement Be Used Adjacent to a Scoliotic Deformity? Thierry Marnay, Patrick Tropiano, James J. Yue, and Geneste Guilhaume
K E Y l
l l
l
P O I N T S
Lumbar disc replacement adjacent to pre-existing scoliosis is currently being evaluated. Scoliosis has recently been classified into four types. Use of semiconstrained ProDisc implant adjacent to scoliotic deformity or within scoliotic deformity may be achievable in select patients. Extensive preoperative evaluation is mandatory in this expanded indication patient group.
Although the use of total disc replacement (TDR) has been shown to be at least equally effective and in some studies superior to fusion of the spine,1–6 the use of total disc replacement in the patient with a scoliotic deformity has not been examined in great detail. Most, if not all, prospective randomized studies have excluded patients with greater than 11 degrees of scoliosis in the frontal plane.4,6 In general and in broad terms, scoliosis in the pediatric population can be defined as being idiopathic, congenital, neuromuscular, and syndrome-related.7 In the adult patient, scoliosis has been recently categorized by Aebi into four types8: Type 1: Primary degenerative or de novo scoliosis resulting from disc degeneration or facet joint arthritis with or without signs of spinal stenosis. Type 2: Idiopathic adolescent scoliosis of the thoracic or lumbar spine which progresses in adult life and is usually combined with secondary degeneration or imbalance. The patient who has had fusion of an adolescent curve and now has progression of a pre-existing curve or the formation of a new adjacent curve are categorized as 3a. Type 3: Patients with secondary adult curves are subcategorized into Types 3a and 3b. Type 3a: In the context of an oblique pelvis, for instance, due to a leg-length discrepancy or hip disease or as a secondary curve in idiopathic, neuromuscular, and congenital scoliosis, or asymmetric anomalies at the lumbosacral junction. Type 3b: In the context of a metabolic bone disease (mostly osteoporosis) combined with asymmetric arthritic disease or vertebral fractures.
In Chapter 89, Paul McAfee discussed the formation of sagittal and frontal plane deformities at the level of total disc replacement in patients who had no pre-existing scoliosis. This chapter will evaluate the use of ADR surgery within a pre-existing curve in a degenerative/de novo scoliosis (Aebi Type I) or in a patient who has undergone previous spinal fusion for idiopathic adolescent scoliosis and has developed adjacent level changes to the prior curve. The solutions proposed as a surgical treatment for scoliosis has always been limited to a fusion with an instrumentation that tries to reduce the curve and maintain the reduction until the fusion has healed. With the capacity of the total disc replacement (ProDisc as a semiconstrained implant) to reduce a part of the deformity, stabilize the space, and maintain the motion, we present here a new approach to the scoliosis surgical treatment. CLINICAL BACKGROUND AND METHODS Historically, the senior author (TM) designed the ProDisc5 implant (Synthes, Paoli, PA) to be utilized not in patients with primary disc degeneration in a neutral spine. Rather, the ProDisc implant was designed to be utilized in those patients who had had prior scoliotic surgery and developed adjacent level degeneration. The ProDisc implant is a keeled implant and can resist sheer forces due to the fixed ball-and-socket design. The keel permits not only strong primary fixation but also permits for stabilization of reduction of deformity in the frontal plane in the presence of a lateral spondylolisthesis. The senior author (TM) began a clinical trial in 2000, evaluating using the ProDisc semiconstrained lumbar total disc replacement in the treatment of adult patients either within a scoliotic deformity or adjacent to a scoliotic curve.9 For evaluation purposes the patients were classified into Type 1 (TDR in the curve itself) and Type 2 (TDR below the curve) (Table 90–1). Type 1 patients were stratified into those patients with idiopathic scoliosis, (Type 1aI) and those patients with a 705
706
P A R T
TABLE 90–1. Spinal Curve
X I I
Controversies 100
Classification Total Disc Replacement in Scoliotic
Type 1 Type 2
80
Type 1: Total disc replacement in the curve, 18 patients* Type 1a: Idiopathic scoliosis, 9 patients Type 1b: Degenerative scoliosis, 9 patients Type 2: Total disc replacement below the curve, 24 patients{ Type 2a: Already instrumented, 23 patients Type 2b: Not instrumented, 1 patient
62
60 56 40
32
20 *Six men, 12 women. { Three men, 21 women.
0 Pre-op n
22
26
3 mo
18
10
12
6 mo
12 mo
8 4 24 mo
F I G U R E 90–2. Oswestry Disability Index scores for Types 1
and 2.
degenerative curve (Type 1b). Type 2 patients were stratified into those patients that had prior scoliosis surgery (Type 2a) and those who had not had prior surgery (Type 2b). Average age of patients was 52 years old (range 32–69), the weight 64 kg (50–83 kg), and a height 167 cm (155–185 cm). Minimum follow-up period was 24 months. A total of 42 patients were enrolled in the study. Twelve patients had one-level surgery, 16 patients had bilevel surgery, 10 patients had trilevel surgery, and 4 had four-level surgery. RESULTS Figures 90–1 and 90–2 show clinical outcome scores for Visual Analog Score (VAS) and Oswestry measurements. All groups showed excellent improvement with maintenance of this improvement at 2 years. The patients who needed a TDR in the curve (idiopathic or degenerative) presented a leg pain more important than the group who needed a surgery below the curve. It seemed that the radicular pain coming from the curve dislocation was less tolerated than the roots compression in the degenerated discs below the scoliosis. However, the results were the same after the surgery, regardless of the anatomic situation before
8 7
X
6
X X
5 4 3
X
2 1
X X
X X
3 mo
6 mo
0 Pre-op n
1 VAS back 1 VAS leg 2 VAS back 2 VAS leg
X X X 12 mo
24 mo
F I G U R E 90–1. Visual Analog Scale (VAS) scores (from 0 to 10)
for Type 1a, Type 1b, Type 2a, and Type 2b patient groups. There is a difference in the preoperative signs, with a lower level of leg (neurologic) pain in the group with a need of total disc replacement below the curve.
surgery. Figure 90–3 illustrates pre- and postoperative radiographs of a type 1a patient with a pre-existing idiopathic scoliosis. Figures 90–4 and 90–5 illustrate the pre- and postoperative imaging studies of a Type 1b patient. The patient had a degenerative scoliosis and underwent 3 level ProDisc surgery with reduction of the L3-L4 lateral spondylolisthesis. The patient's 4-year follow-up VAS and Oswestry scrores were 1 and 5, respectively. Figures 90–6 to 90–9 illustrate the pre- and postoperative radiographs of Type 2b patients.
DISCUSSION The usage and indications of total disc replacement surgery continue to evolve. One of these expanded indications that is currently being evaluated is in the patient with either de novo degenerative scoliosis or who has had prior scoliosis surgery and now is suffering from adjacent level degeneration. As presented in other chapters in this text, great care must be taken to evaluate the patient for TDR surgery. The patient with scoliosis is no exception and must be even more carefully assessed for facet degeneration and segmental instability, osteoporosis, and other exclusionary criteria. As the age of the patient is on average 52 (10 years older than the patients with a current degenerative disc disease6) with a majority of females, the osteodensitometry limit for surgery is 1. The quality of the outcomes has to be unlighted, especially in the Type 2a, with the poor results of the extension of the fusion to the pelvis below a scoliosis with the complete ankylosis of the spine, and the mechanical complications of the pelvic fixation procedure. The capacity of the total disc replacement in scoliosis seems for the future to be able to modify the indications in scoliosis surgery, and could change our surgical behaviors—not to extend the fusion below L4, not to operate too early on the lumbar curves, decide to include hybrid constructs with fusions in the thoracolumbar area, and disc replacements below, as we have seen the capabilities we have to extend the TDR to L2-L3. The senior author's early study results appear to show positive value in the usage of TDR surgery in a select group of scoliotic surgery.
Type 1a
n
F I G U R E 90–3. Preoperative and postoperative radiographs of Type 1a patient. The delordotic curve of the lumbar spine is associated with a retroversion of the pelvis. The capability of adaptation of the patient is limited with the stiff kyphotic curve of the thoracic spine. In the frontal plane, there are also no possibilities to balance the occipital axis preoperatively. We can see in postoperative films the capacity for the lumbar spine to reestablish a lordosis curve and so reduce the version angle. The design of the prothesis with an anchorage with a keel and a ball-and-socket as a joint that neutralizes the shear forces allows those changes.
n
F I G U R E 90–4. Preoperative radiographs of Type 1b patient. We can see the asymmetry of the frontal inclination of the L4-L5 disc and the dislocation of L3-L4.
708
P A R T
X I I
Controversies
n
F I G U R E 90–5. Postoperative radiographs of Type 1b patient. The quality of the realignment of the spine in the frontal view shows the capacity of reduction of curves with the release of the space and the semiconstrained prosthesis.
Type 2a
n
F I G U R E 90–6. Preoperative radiographs of Type 2a patient. The kyphosis of the two lowest levels creates a retroversion of the pelvis and a sagittal unbalance.
CHAPTER 90
Can Lumbar Disc Replacement Be Used Adjacent to a Scoliotic Deformity?
5º
29º
No motion ROM = 24º
n
F I G U R E 90–7. Postoperative radiographs of Type 2a patient. The 24 degrees of range of motion (ROM) and especially the restitution of a lumbar lordosis re-create the conditions for a better everyday life.
n
F I G U R E 90–8. Preoperative radiographs of Type 2a patient. The degeneration of the discs below an instrumented idiopathic curve may be solved through a total disc replacement on the degenerated levels. The alternative of an extension of the fusion to the pelvis has never been a satisfactory procedure, with a high rate of pseudarthrosis and mechanical complications on the pelvic fixation, and the fair outcomes of the total spine fusions to the pelvis.
709
710
P A R T
X I I
Controversies
n
F I G U R E 90–9. Postoperative radiographs of Type 2a patient. The ball-and-socket joint allows a frontal inclination of the implant without any risk of lateral dislocation, and is perfectly adapted to this type of pathologic anatomy.
REFERENCES 1. Bertagnoli R, Yue JJ, Kershaw T, et al: Lumbar total disc arthroplasty utilizing the ProDisc prosthesis in smokers versus nonsmokers: A prospective study with 2-year minimum follow-up. Spine 31:992– 997, 2006. 2. Bertagnoli R, Yue JJ, Shah RV, et al: The treatment of disabling multilevel lumbar discogenic low back pain with total disc arthroplasty utilizing the ProDisc prosthesis: A prospective study with 2year minimum follow-up. Spine 30:2192–2199, 2005. 3. Bertagnoli R, Yue JJ, Shah RV, et al: The treatment of disabling single-level lumbar discogenic low back pain with total disc arthroplasty utilizing the Prodisc prosthesis: A prospective study with 2-year minimum follow-up. Spine 30:2230–2236, 2005. 4. Blumenthal S, McAfee PC, Guyer RD, et al: A prospective, randomized, multicenter Food and Drug Administration investigational device exemptions study of lumbar total disc replacement with the
5. 6.
7. 8. 9.
CHARITÉ artificial disc versus lumbar fusion, Part I: Evaluation of clinical outcomes. [erratum appears in Spine 30(20):2356, 2005]. Spine 30:1565–1575; discussion E1387–1591, 2005. Tropiano P, Huang RC, Girardi FP: Lumbar total disc replacement, 7 to 11 years of follow up. J Bone Joint Surg Am 87:490–496, 2005. Zigler J, Delamarter R, Spivak JM, et al: Results of the prospective, randomized, multicenter Food and Drug Administration investigational device exemption study of the ProDisc-L total disc replacement versus circumferential fusion for the treatment of 1-level degenerative disc disease. Spine 32:1155–1162; discussion 1163, 2007. Newton PO, Wenger DR: Pediatric spinal deformity. Orthopaedic Knowlege Update. Spine 2:361–376, 2002. Aebi M: The adult scoliosis. Eur Spine J 14:925–948, 2005. Marnay T, Tropiano P, Geneste G, Blondel B: Scoliosis and TDR Classification of Indications and Preliminary Results. Presented at Spine Arthroplasty Society, Montreal, Canada, 2006.
CHAPTER
91
Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis Charles H. Rivard, Christine Coillard, Souad Rhalmi, Marco Be´rard, Robert T. Chomiak, and Gary L. Lowery
K E Y l
l l l
l
P O I N T S
The Orthobiom is intended to treat pediatric patients suffering from scoliosis. The Orthobiom Spinal System is a pedicle screw system. The Orthobiom allows stabilization and correction of the curve. The primary difference between the Orthobiom and other pedicle screw systems is the incorporation of a mobile connector into the subject device. The Orthobiom has not yet been implanted in humans but has been implanted in minipigs.
Scoliosis is defined by the Scoliosis Research Society as a structural curve of the spine greater than 10 degrees.1 The lateral spinal curvature along with the spinal rotation causes the trunk and the rib asymmetry (rib hump). Scoliosis is typically associated with subnormal thoracic kyphosis; hence, structural scoliosis is a three-dimensional deformity with significant cosmetic implications. Idiopathic scoliosis is the most common type of scoliosis. Onset of scoliosis may be in infancy (age 0–3 years) or childhood (age 4–10 years). The vast majority of patients who present with idiopathic scoliosis are diagnosed during adolescence, during the growth spurt of puberty. The etiology of adolescent idiopathic scoliosis is multifactorial and includes genetic, hormonal, neuromuscular, and growth factors. The course of treatment for patients in different scoliosis age groups varies considerably and depends on a variety of factors including the extent of the curve at the time of diagnosis and during follow-up, the patient’s stage of remaining bone growth, the amount of pain and deformity associated with the condition, and the patient’s willingness and ability to withstand surgery should it be deemed necessary. Treatment options for the various types of scoliosis are categorized into observation, conservative, and surgical, depending on the individual case. Treatment selection is based on the measurement of the Cobb angle. Conservative treatment, such as bracing and casting, aim at preventing or slowing the curve progression and require flexibility and growth potential to be effective. As a consequence, congenital and adult scoliosis are usually nonresponsive to bracing. If the curve is at high risk of progression and
is of significant magnitude, surgical correction may be indicated. For all patients with scoliosis, the goals of treatment are the same, which is to alleviate symptoms and to prevent curve progression. The Orthobiom Spinal System (Paradigm Spine, GmbH, Wurmlingen, Germany) is a posterior pedicle screw system indicated for the treatment of pediatric scoliosis by (1) correction, (2) stabilization, (3) adjustment, and (4) fixation of the scoliotic spine. The Orthobiom Spinal System fuses the apical level to better stabilize and maintain the curve similar to the currently cleared scoliosisspecific pedicle screw systems while allowing the possibility of longitudinal adjustment. DESCRIPTION OF THE DEVICE The Orthobiom Spinal System is a pedicle screw scoliosis system consisting of pedicle screws, rods, fixed and mobile connectors, and cross connectors designed to correct and stabilize the scoliotic spine (Fig. 91–1). The pedicle screws, rods, and fixed connectors behave identically to other pedicle screw systems such as the ISOLA (DePuy Spine, Inc., Raynham, MA), CD Horizon (Medtronic Sofamor Danek, Memphis, TN), and TSRH Spinal Systems (Medtronic Sofamor Danek, Memphis, TN), (with the exception that the polyaxial screw head does not lock). These components rigidly fix the spine and allow for fusion at one or more levels. The mobile connector is similar to the fixed connector except it allows the connector to slide axially on the rod at levels superior and inferior to the fusion site. Therefore, the mobile connector still allows correction of the spine in flexion/extension and lateral bending, but do not axially fix the spine at these levels. This characteristic allows the levels above and below the fused level(s) to potentially continue natural spinal growth (adjustment without repeat operations) but maintain stabilization of the corrected scoliotic curve. The Orthobiom Spinal System is made of three different materials: 1. Nitrogen-strengthened steel (BioDur 108): rods, pedicle screws, and cross connector 711
P A R T
712
X I I
Controversies
Rod
Mobile connector
Cross connector
7. Patient has a previous history of disease of bone metabolism. 8. Patient has a previous history of allergy to any component of the Orthobiom Spinal System. 9. Patient has a previous history of severe allergy or anaphylaxis. 10. Patient has a previous history of malignancy. 11. Patient is pregnant. SCIENTIFIC TESTING
The indications for the Orthobiom are stated as follows: “The Orthobiom Spinal System is a posterior pedicle screw system indicated to treat pediatric scoliosis by (1) correction, (2) stabilization, (3) adjustment and (4) fixation of the scoliotic spine. Also, the Orthobiom Spinal System should be used with bone graft around the fixed connectors only.”*
Because Orthobiom is a dynamic system, normal load-controlled corpectomy testing as described in ASTM F1717 could not be performed. Each screw allows for some degree of rotation and movement and some of the interconnections are not fixed; therefore, the test blocks touch under any amount of load. Since acquisition of Orthobiom by Paradigm Spine, LLC (New York, NY), new test setups and data were generated. Paradigm is committed to characterizing each potential failure mode and each loading scenario; therefore, multiple static and dynamic tests have been designed. Testing is under way, with some of the tests completed while others are still in the preparation stage. Testing that has been completed is described here, but ongoing tests and those in the design phase will not be reported. There are two primary types of components of the Orthobiom Spinal System: the fixed segments and the mobile segments. In order to adequately correct and maintain the scoliotic spine, high bending strength is required initially for the construct to sustain flexion/extension forces caused by daily activities, and medial/lateral forces induced by the realignment of the scoliotic spinal segment. Under this scenario, the construct will mainly apply (or resist) three-point bending forces to (or from) the spine. This variation in the load-sharing/load-bearing requirements imposed to the new construct implies that minimum axial and torsional forces are withstood by the longitudinal elements. There is a potential for wear because some of the components of the Orthobiom articulate or telescope. Therefore, wear resistance is a requisite at the various sliding interfaces of the mobile connectors as well as in the links between those connectors and the pedicle screws.
CONTRAINDICATIONS
Component Testing
Fixed connector
n
F I G U R E 91–1. Orthobiom Spinal System.
2. Standard stainless steel (316LS-type): fixed and mobile connectors 3. UHMWPE (ultra-high-molecular-weight polyethylene): roller inside mobile connector. The Orthobiom Spinal System can be described by its five main components: rod, pedicle screws, fixed connectors, mobile connectors, and cross connector. INDICATIONS
The following are the specific exclusion criteria for the Orthobiom: 1. Patient had a previous spinal surgery with fusion. 2. Patient’s Risser sign is greater than 3 or there is a nonflexible curve. 3. Patient has a neuromuscular disorder. 4. Patient has a pathologic spinal malformation, such as spondylolisthesis. 5. Patient has a previous history of systemic infection or infection at the site of surgery. 6. Patient has previous history of osteopenia, osteoporosis, or osteomalacia to a degree that spinal instrumentation would be contraindicated. *Paradigm Spine, LLC.
Static Torsion Bending (Screw and Rod)
Static torsion bending testing was performed on the screw/rod interconnection in accordance with ASTM F1798. The static test was performed due to quasi-static correction moment. The purpose of this test was to evaluate the rod/clamp interface under a torsional load. The rod was stabilized while the distal end of the screw was loaded, which placed a torsional load on the interconnection of the fixed connector. The average load on the screw was 5.7 Nm of torque. Static Torsion Bending (Cross Connector and Rod)
Static torsion binding testing was performed on the screw/rod interconnection to evaluate torsion moments on the clamp. This test was performed in accordance with ASTM F1798. Only static
CHAPTER 91
testing was performed owing to the quasi-static correction moment. A load was placed on the opposite side of the cross connector, which placed a torsional load on the rod interconnection of the cross connector. The mean load measured on the connector/rod interconnection was 6 Nm of torque. Axial Loading of Fixed Connector
The static strength of this interconnection in the longitudinal direction was measured because the fixed connector needs to secure the rod in position against frictional forces and gravity to provide stability for fusion. An axial load was placed on the fixed connector while the rod was fixed. This test is currently ongoing. Mobile Connector Against Rod
One potential cause for wear generation is between the rod and the mobile connector. As the patient moves, flexes, extends, and bends side to side, the mobile connectors will slide on the rods. Therefore, in order to mimic the worst-case loading scenario that the Orthobiom Spinal System and mobile connector would experience, a lateral load (72 N) was placed on the rod and connector. The mobile connector was then articulated up and down on the rod 5 mm (2.5 mm) with the simultaneous lateral load. This loading scenario was performed out to 10 million cycles, and all wear debris was collected and analyzed. The results of the mobile connector wear testing demonstrated very little wear of the mobile connector (UHMWPE roller). After 10 million cycles, only 0.001 mg of wear was seen per 1 million cycles. This correlates to only 10 mg of wear. Cross Connector Wear
As the patient moves, micromotion on the cross connector parts will take place while quasi-static correction forces will be applied. Wear is assumed to occur between sleeve and bolt, and between sleeve/bolt and clamp/locking screw. The objectives of the test are to quantify the amount of metal-on-metal wear generated as a function of the motion cycles and to evaluate the bearing surfaces for changes. A characterization of the size distribution and shape of the particles generated will be performed. The specimens are mounted on two rods and subjected to cyclic linear motion. During testing, a constant torsional preload of 3 Nm is applied to the rods. After the cross connector was attached to the rods, the rods were moved axially 2 mm (1 mm). This distance was thought to represent a worst-case scenario because the cross connector is only to be applied between the fixed pedicle screws and, therefore, will see only minimal micromotion. The tests are carried out in newborn calf serum. The testing will run out to 10 million cycles at a frequency below 5 Hz. Wear indicators such as weight loss, wear tracks, scratching, and burnishing were measured and analyzed. Visual surface characterization was performed at the end of the test. Frictional forces were monitored throughout the duration of the test and any changes reported. This testing is still ongoing.
Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis
713
Construct Testing Static Axial Compression Bending
A modified ASTM F1717 test was used to measure the static and dynamic axial compression of the Orthobiom VI Spinal System because it includes mobile connectors which cannot transfer longitudinal forces. Once the screws reach their highest range of motion and the rods and screw/rod connection are being loaded, the gauge length of the test setup is representative of what is described in ASTM F1717, which is 76 mm. All other parameters are identical to that described in ASTM F1717. The smallest rod and screw were both used to represent worst-case setup. This test is still under development. Any differences from ASTM F1717 will be justified, and the loading scenario will have a physiologic justification. Dynamic Axial Compression Bending
The dynamic test was performed with the same parameters out to 5 million cycles. This test is expected to have at least two samples run-out at 100 N for 5 million cycles. This resulting load will be the same as several other pedicle screw systems out to 5 million cycles. This test was only run out to 5 million cycles because it is evaluating the effect of the fixed connectors on the rods. The fixed connectors hold the rod in place so fusion can occur, allowing a comparison to other systems on the strength of the Orthobiom at the fusion level. This load is much higher than the anticipated loads expected by a child. Therefore, the Orthobiom VI will withstand the physiologic loading by the patient. Static Corpectomy Lateral Bending
This test was developed to study the effects of lateral bending on the construct. The experimental setup used for the flexion/lateral bending tests of the implant assembly is based on the principles proposed by ASTM F1717 for testing spinal constructs in a corpectomy model. Three PE (polyethylene) blocks, instead of two, simulating the vertebrae are used for setting up a three-point bending test. The implant is assembled using three UHMWPE blocks to simulate the vertebrae. The use of UHMWPE eliminates the effects of the variability of native bone properties. The supports of the PE blocks are positioned so that a horizontal distance of 76 mm between the screw axes is achieved. The fixed and mobile connectors are attached to the polyaxial pins of the pedicle screws by self-breaking nuts, which are tightened until breakage of the nut occurs. Per implant assembly tested, four mobile connectors, two fixed connectors, and one cross-link are used. The two fixed connectors attached to the middle PE block prevent the rods from moving along the rod axis. At a distance of 2 to 3 mm from the fixed connectors, one cross-link is attached to both rods. The position of the cross-link is adapted so that the distance of the rod axes (which is variable due to the polyaxial pins of the pedicle screws) is 18 mm. The outer PE blocks are rigidly attached to the test frame, but a linear bearing is used to enable the middle block to move in the horizontal direction.
714
P A R T
X I I
Controversies
All connectors, screws, and rods were the smallest in size, representing the worst-case scenario. The maximum load that was measured before any component failed was 1,000 N and the yield load was 800 N. The method of failure was yielding of the rods. Dynamic Corpectomy Lateral Bending
The same test set-up was used and tested out to 10 million cycles. During the dynamic testing a sinusoidal compression load is applied at a rate of 5 Hz. The test is carried out at room temperature (20 C) in ambient air. This test setup mimics the physiologic loading of the device when the patient bends side to side. With this test setup, the Orthobiom ran out to 10 million cycles with a load of 250 N. This is much higher than any expected lateral load on the spine by a child. The method of failure was screw and rod breakage. Another failure observed was the screw twisting out of UHMWPE block under loads above the fatigue strength. This result demonstrates how the device is able to correct and maintain the scoliotic spine. Static Corpectomy Flexion/Extension
A similar test setup was used as presented in the static and dynamic corpectomy lateral bending test. Instead of the lateral three-point bending load placed on the construct, an anterior load was placed on the construct. All connectors, screws, and rods were the smallest in size, representing the worst-case scenario. This mimics the patient bending forward. Testing the three-level corpectomy model in a flexion/extension scenario resulted in a maximum load of 1,200 N with a yield load of 900 N. Dynamic Corpectomy Flexion/Extension
Dynamic testing was performed using the same test setup. This test was performed to evaluate how the device behaves when a cyclic load was placed on it, mimicking a patient bending over. The test was run out to 10 million cycles. The results demonstrated that when a flexion/extension load is placed on the Orthobiom, it can run out to 10 million cycles at a load of 250 N. The method of failure was rod breakage. Static Torsion
A static torsion test was performed in close accordance with ASTM F1717. Owing to the technical characteristics of the Orthobiom Spinal System, the standard was modified slightly. A static load was applied to the construct and the failure load and yield load were measured. This test is still under development. FUNCTIONAL ANIMAL STUDY Minipigs The Orthobiom IV was studied in a minipig model to evaluate the effect on the intervertebral joints by allowing intervertebral micromotion as compared to the effect of rigid fixation on the biologic changes in the intervertebral joints incorporated within the instrumented (nonmobile) segments. Twenty minipigs with scoliosis-
induced spines were studied according to the following distribution: Rigid fixation (n ¼ 6) Mobile fixation with Orthobiom (n ¼ 11) Control group, no surgery (n ¼ 3) Minipigs in the rigid fixation group were evaluated at 12 and 18 months. Those in the Orthobiom group were evaluated at 6, 12, 18, and 24 months postoperatively. The Orthobiom group included 2 minipigs with implants removed at 12 months, then were followed for an additional 12 months. Evaluations included anteroposterior (AP) and lateral radiographs (Fig. 91–2), CT scans on select minipigs, and histologic examination. The results demonstrated that the mobile fixation permitted by the Orthobiom preserves the intervertebral joints while providing the stability and correction necessary for a scoliotic spine. Rabbit Particulate Study A particulate injection study was performed in the lumbar spine of New Zealand rabbits to study the local and systemic effects of UHMWPE.2 Macroscopic and histopathologic evaluations were conducted on the nerve structures and lymph nodes. The purpose of the study was to determine if UHMWPE is a suitable material for components of the Orthobiom Spinal System. Medical grade UHMWPE powder was commercially obtained in a mix of the following sizes: 90% 60- to 250-mm; 5% below 60 mm; and 5% above 250 mm (Ticona, NJ). This particle size distribution was selected because it is representative of the observed wear particles collected following wear testing of the rollers. Eighteen New Zealand white female rabbits weighing 2.5 to 3 kg were allocated into the following three groups: UHMWPE with Orthobiom (n ¼ 12) Control group, no treatment (n ¼ 3) Sham procedure, without wear particle injection (n ¼ 3) The quantity of the particles implanted per site ranged from 10 to 15 mg, which is equivalent to 5.5 103 – 8 103 particles per site. Each of the 15 rabbits operated on survived the operation with no remarkable events or complications. Four test rabbits, one control rabbit, and one sham rabbit were used per evaluation period at 1 week, 4 weeks, and 12 weeks. The biologic response of the test rabbit spinal cords (UHMWPE implantation) was evaluated macroscopically in comparison with the spinal cords of the control and sham rabbits. The spinal cord appearance of the control rabbits, at 12 weeks after observation, was identical to the spinal cord of the control rabbits at 1 and 4 weeks after observation as well as to the spinal cord appearance of the sham rabbits at 1, 4, and 12 weeks after surgery. No adverse reactions such as necrosis developed secondary to particle implantation. Inflammation was observed in the UHMWPE group at 12 weeks but was limited to the connective tissue close to the dura mater and had subsided when compared to 1 week and 4 weeks after implantation. This is considered a normal, acute immunologic response to foreign material. The reactions to UHMWPE particles in the spinal cord tissue from test rabbits were evaluated histologically in comparison with
CHAPTER 91
A
Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis
B
715
C
D n
F I G U R E 91–2. Posteroanterior radiographs of a minipig instrumented with a mobile fixation, Orthobiom Spinal System. A, Immediately after surgery; Cobb angle of 20 degrees. B, High magnification of A showing thoracolumbar region. Note the disc thickness. C, At 12 months follow-up: high magnification of the thoracolumbar region. Note the disc thickness that is similar to B; Cobb angle of 15 degrees. D, Immediately after implant retrieval, minipig survival, 12 months follow-up.
the spinal cords of the control and sham rabbits. UHMWPE particles were found in the histologic sections of the implantation sites. Polarized microscopy was used to show the particles present in all the test rabbits. The histology analysis showed normal spinal cord tissue with normal cells and dura mater and normal nerve roots. At 4 weeks and 12 weeks after implantation, the mass of particles at the implantation sites were more adhered to the dura mater than at 1 week after implantation. The embedded mass of particles, with a vascularized lattice of connective tissue and without any appearance of necrosis or swelling, is therefore an evident sign of the material tolerance by the living tissue. The dura mater vascularization and the spinal cord appearance adjacent to the implantation sites appear identical to that of the control and sham rabbits. These results demonstrate that UHMWPE is a suitable material for the Orthobiom System.
OPERATIVE TECHNIQUE Positioning The patient is lying supine with the hips and the knees extended. Both arms should be flexed on each side of the head.
Incision The skin and the soft cutaneous tissue will be opened, going down to the supraspinous ligament by a midline incision usually centered on the proximal and distal spinous process. Orthobiom Spinal System requires a different approach around the spinous process. Because fusion is not wanted, it is important to not go subperiosteal. It means that the surgeon must open the supraspinous ligament approximately 2 mm lateral from the tip of the spinous process; it is very easy to find a plane of soft tissue between the muscle and the insertion of the small muscle at the level of the spinous process. It is easy to develop a complete dissection plane to displace laterally the mass of the two muscles without touching the periosteum at the level of the spinous process. What is important is also not to touch the periosteum at the level of the laminae or the articular facet. Screw Insertion This instrumentation will be used with pedicular screws in the lumbar and thoracic regions. The pedicular screw link incorporates a ball and socket that permits easy insertion of the rod and connector. In the lumbar region, screws of 5 or 6 mm can be used.
716
P A R T
X I I
Controversies
A hole is perforated with a 3.2-mm drill bit for the 5-mm screw and 4.5-mm for the 6-mm screw. The entry point of the pedicular screw should be at the intersection of the longitudinal transverse process midline and the interarticular facet line. The superior articular facet of the inferior vertebrae must not be altered; it must be preserved to ensure some movement and normal articulation during adult life. After the point of entry has been chosen, we will aim at a 150-degree medial angle toward the corpus of the vertebrae. In the thoracic region, screws of 4.5 or 5 mm in diameter can be used. The entry point will be at the intersection of the vertical midline of the inferior articular process of the superior vertebra and a line tangent to the lower point of the same inferior articular process. When the hole has been made in the lumbar or thoracic region, its depth is measured after probing the pedicle on the medial side, the inferior side, the lateral side, and the upper side. This probing of the pedicle hole is very important to ensure that bone is present at the medial and inferior portion of the hole to prevent any damage to the dura mater and for the lateral and upper part of the pedicle to ensure a good grip on the bone. On the concave side of the deformation, it is recommended to use a segmental fixation at every level or at least at every possible level. On the convex side of the deformation, fewer pedicular screws are required and are normally 60% of the number placed on the concave side. As a first step, the pedicular screws are only partially inserted. Once this is complete, AP and lateral radiographs are taken to define the placement and the angle of the screws. If each screw angle is correct, the screws can be inserted to the normal position. If a screw does not have a proper position or angle, it must be corrected and then inserted to the normal position. Curve Correction As with any other scoliosis instrumentation the rod is bent and cut to the proper length to correct the three-dimensional deformation. The mobile and fixed connectors are then slid onto the rod in the proper sequence. All connectors are positioned on the corresponding pedicular screw link. The derotation maneuver of the rod is performed to induce the translation of the spine and then the two fixed connectors are tightened to stabilize the rods and prevent derotation. Completion of Surgery A wake-up test is performed along with evoked potentials. The achieved correction is evaluated through an x-ray examination (AP and lateral) and it can be determined if readjustment of the instrumentation is required. As with any other fusion instrumentation, the surgeon needs to close the muscle aponeurosis, the subcutaneous tissue, and the skin with different separated stitches, and therefore, closure is done in multiple layers. Normal dressing is then performed. POSTOPERATIVE CARE The surgeon must inform the patient regarding the level of activity that he/she can do. This list of activities gives only a general
progression of the allowed postoperative activities for the patient. As for any other surgeries, general postoperative instructions unrelated to the Orthobiom Spinal System should also be provided by the surgeon. COMPLICATIONS The risks associated with the use of the Orthobiom Spinal System include those related to any surgery, those related to any instrumentation of the spine for scoliosis correction, and finally, the risks directly associated with the use of the Orthobiom Spinal System. Risks Associated with any Type of Surgery There is a risk of death associated with any surgery. There are also risks any time a person receives a general anesthestic. These risks include reactions to the anesthetic, blood clots that may move through the blood vessels and damage major organs, pneumonia, abnormal heartbeats, or heart attack. Anesthesia is planned before the surgery and is controlled during the surgery by a qualified anesthetist. At the end of the surgery the wakening of the patient progresses under a close supervision from the hospital staff members who follow a protocol specific to each medical center. The second major risk is related to the blood transfusion, which is controlled by autologous blood donation before surgery. The third major risk is associated with infection that is minimized using prophylactic antibiotic therapy during and following the surgery. The patient may also develop a hematoma, a wound that will not heal properly, bleeding, or a vessel tear at a surgical opening. The close postoperative follow-up during the hospitalization period helps to detect and react properly to these events. Risks Associated with any Instrumentation of the Spine The major risk associated with any instrumentation of the spine is temporary or permanent neurologic deficits for the patient. This is controlled by recording the sensory and motor evoked potentials during the surgery and by doing a wake-up test at the end of the instrumentation procedure. There are also risks related to material breakage, malfunction, or vertebral fracture that could result in a loss of scoliosis curvature correction or injuries to adjacent tissues. The nonfusion of vertebral segments also limits the efficacy of the conventional implants. However, nonfusion does not represent a failure for the Orthobiom Spinal System, because this characteristic is anticipated. In all cases, these possibilities are closely followed by periodic clinical and radiologic examinations. If such an event occurs, a revision of the system may be planned. Additional complications may occur as a result of having a foreign object implanted. This could cause infection and allergic reactions. Risks Associated with the Orthobiom Spinal System The risks associated with the Orthobiom Spinal System include those already described. Two different features specific to the implant itself could also be sources of risks. First, because some parts of the Orthobiom Spinal System are made of polyethylene, a material that has not been previously used for a spinal corrective
CHAPTER 91
implant, biocompatibility could be a concern. However, this material has been used extensively for many other types of orthopaedic implants3–5 for many years and the results of the cytotoxicity tests done by the sponsor before this study showed no negative effects on tissues surrounding the spine. Second, some components of the Orthobiom Spinal System are free to move to a certain extent. This feature adds some degree of risk, because the implant could be exposed to greater stresses, which could cause breakage of certain parts. However, results from exhaustive finite element analysis and mechanical tests support the strength of the implant over a long period of time because only micromotion is anticipated. The mobility of the implant could also affect the spinal system in different ways. It could reduce its ability to maintain the spinal correction or it could increase the loading on the pedicles and induce changes on their bony structure. The close follow-up of the patients recruited for this study allows the clinician to quickly detect the problem. The fusion solution could then be applied and represents a second surgery comparable to a system revision for a conventional implant. On the other hand, fusion could occur, even if not induced during the surgery. This event should not change the result of the spinal correction, but the patient will experience the loss of some potential benefits associated with the use of the Orthobiom Spinal System. Risk is then associated with the uncertainty of the Orthobiom Spinal System effectiveness. The mobility of its components has been designed to take into account the mobility of the spine and to overcome the problems associated with the nonfusion of the spine. There still is a possibility that this available mobility limits the quality of the correction or its progression in time. Based on the same principles, it is unclear whether at maturity, when the Orthobiom Spinal System is removed, the complete spinal correction will persist. However, the neuromuscular system should have adapted and a new equilibrium should have developed. Furthermore, the growth of the patient has ended. If scoliosis curves progress at maturity, a conventional instrumentation with fusion could be carried out. The risks associated with the use of the Orthobiom Spinal System will be minimized by giving intensive training to the investigator/surgeon responsible for performing the surgery with the instrumentation. This consists of theoretical training sessions and instrumentation on animal or cadaver models. In any case, the consequences of the described adverse events will be controlled by a close follow-up of the patients with clinical and radiologic evaluations and appropriate recording of the patient’s condition. DISCUSSION Arthrodesis is an effective means of eliminating disease and providing a permanent structural stability for disease processes that involve mobile joints in both the extremities and the axial skeleton. However, being essentially irreversible and at best nonphysiologic, it is also accused of many undesirable long-term side effects. During the last couple of decades, there has been a definite trend in the field of orthopaedic surgery toward preservation of the mobile joint for a more physiologic functional restoration. Controlling the spine without an arthrodesis is not new. It has a much longer history than spinal arthrodesis techniques. External spinal immobilization by plaster of Paris cast and braces before the instrumentation era and spinal instrumentation without
Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis
717
fusion,6 rod-long-fuse-short method for the treatment of unstable fractures,7 and use of subcutaneous rods and growing rods for pediatric deformities all fall into this category.8 Despite the definite theoretical advantages they offer, such as permitting spinal growth before definite fusion, reduction of the fusion level, or eliminating the need of arthrodesis altogether, these procedures are not very popular except for external immobilization. This is due to their limited indications, uncertainty of the result, and the high rate of mechanical and biologic failure.9,10 Because instrumentation without fusion procedures rely heavily on the spontaneous acquisition of spinal stability either by healing of the destabilizing lesion or anatomic alteration by growth, the spinal implant will have to withstand stress much longer than those used with fusion, making them more vulnerable to mechanical failures. Spontaneous fusion is the most common and important reason for biologic failure that compromises the result of the instrumentation without fusion, as the goal of the surgery is preservation of the motion segments. It also greatly compromises the ability of the vertebral column to reacquire intrinsic stability by limiting the growth-determined alteration of the vertebral shape and also by alteration of the force acting on the unfused segments. Degenerative changes in the intervertebral disc, facet joint articular cartilage, and intervertebral joint ankylosis are biologic failures that affect the long-term result mainly after the removal of the instrument and cause pain, stiffness, or hypermobility and precocious degenerative changes. They occur mainly in the instrumented area but also may occur in the motion segments neighboring the instrumentation due to the stress concentration caused by abrupt change in mobility. Disturbance in the intervertebral disc and facet joint articular cartilage nutrient transport mechanism by rigid fixation seriously compromises the biologic function of the cartilage/disc cells and results in degenerative changes, which may lead to total disappearance of the articular cartilage and bony fusion of the joint in extreme instances. Our experimental results show that the viability of the joint cartilage and the intervertebral disc may be maintained, not only by allowing a normal full range of motion but also by micromotion of the intervertebral joint just sufficient for loading/unloading of the avascular connective tissues. In the rigid fixation group (Fig. 91–3), one animal was sacrificed at 12 months and five animals at 18 months after surgery. Despite significant growth of the animals from 325.14 kg to 62.714.8 kg, the length of the instrumented segment measured by the distance between the uppermost and the lowermost screws remained unchanged. Initial scoliosis of 315 degrees created by rod rotation was maintained at 278 degrees at the time of euthanasia with no significant change in the curve magnitude (P ¼ 0.37, paired t-test), therefore showing an acceptable maintenance of fixation (Table 91–1). In the mobile fixation (Fig. 91–4), one animal was sacrificed at 6 months, three animals at 12 months, three at 18 months, and two at 24 months. In the observation period, the animals demonstrated significant growth from 29.9 to 67.2 kg. The length of the instrumented segment changed from 25.32.0 cm to 30.0 1.5 cm showing a growth of 4.71.4 cm (P ¼ 0.0004, paired t-test). Despite three minor fixation failures detected on the final
718
P A R T
X I I
Controversies
L3-L4 n n
F I G U R E 91–4. Minipig instrumented with a mobile fixation Orthobiom System, disc L3-L4. Nucleus pulposus and annulus fibrosus are well preserved.
x-ray examinations before euthanasia (two pigs with one broken screw, one pig with two broken screws), the experimental scoliotic curve of 19 4 degrees was maintained at 17 5 degrees, showing a reasonable maintenance of fixation with no statistically significant alteration of the curve magnitude (P ¼0.21, paired t-test) (Table 91–2). In fact, with the immunohistochemistry, we realized that the presence of collagen X in the rigid fixation group was far more predominant than in the mobile fixation group (Orthobiom). However, the comparison of the samples between the mobile fixation (Orthobiom) group and the rigid fixation (Colorado and TSRH) group revealed that the presence of collagen X was statistically far more significant in the intervertebral discs of the rigid fixation (Colorado and TSRH). For the rigid fixation, the collagen X represents 92.84% of the field and the background up to 6.42% of the field with the remainder of the total percentage belonging to other types of collagen. For the mobile fixation (Orthobiom) samples, the normal collagen represented 62.96% of the field and for the mean the collagen X represented 7.84% of the background 14.81%; the remainder of the total percentage belongs to other types of collagen. The ANOVA (analysis of variance) statistical analysis for the comparison between the Orthobiom and Colorado system had a P value of 0.00094.
Failure to grow and heal as desired forms the third category of biologic failures and would eventually necessitate a definite restoration of spinal stability by fusion. This may be due to spontaneous fusion, degeneration of the intervertebral and the articular cartilage, and inappropriate mechanical forces hindering the growth in the desired direction.9 Though it is difficult to conclude at this moment, this may be controlled by appropriate combination of rigid and mobile fixations in the same construct. This experiment of comparing the biologic response of the unfused intervertebral joint to the rigidity of spinal fixation was part of our attempt to develop an instrumentation system to be used in nonfusion treatment of pediatric spinal deformities. In younger patients with large deformities, unlike in adults and patients near the end of growth, the bone growth has to be considered in the treatment of the spinal deformities. The remaining growth in the spinal column and the extremities in young children act both as a negative and a positive factor in the treatment of spinal deformities. It is negative in the sense that a lengthy fusion of the spine would result in trunk shortening compared to the extremities, and that fear of such shortening often leads to an inappropriately short fusion, resulting in failure of the deformity control. The positive aspect is that the growth, with an appropriate force application system, may be exploited to restore the intrinsic mechanical stability of the vertebral column, reversing the destabilizing anatomic alteration in the vertebral body. Our goal was to
F I G U R E 91–3. Minipig at 12 months follow-up with a rigid fixation (Colorado). A disc from the instrumented segment shows loss of the nucleus pulposus and degenerative changes in the annulus fibrosus and the end plate cartilage.
TABLE 91–1.
TABLE 91–2.
Results of Group 1: Rigid Fixation
Measurements
Immediate Postoperative
18 Months
P*
Measurement
Scoliosis (degrees) Instrumented section (mm){ Minipig growth (kg)
31 5 276 25
27 8 283 29
0.37 0.10
62.7 14.8
—
Scoliosis (degrees) Instrumented section (mm){ Minipig growth (kg)
32 5.14
*Paired t-test. { Corrected for magnification, using the length of the longitudinal members.
Results of Group 2: Mobile Fixation Immediate Postoperative
18 Months
19 4 253 20
175 30015
29.9
67.2
*Paired t-test. { Corrected for magnification, using the length of the longitudinal members.
P* 0.21 0.0004 —
CHAPTER 91
avoid the negative effects of lengthy fusion while taking advantage of the positive effects of the remaining growth to restore stability. For an adequate function, the instrumentation system needed to offer a reliable fixation of the vertebral column to exert the desired mechanical force and at same time allow some motion in the incorporated intervertebral joints to maintain the viability and translation on the rods for longitudinal growth. This experiment was performed to determine the balance point in the rigidity/flexibility of the implant that permits both the control of the vertebral column and the maintenance of the joint viability, and to test the hypothesis that micromotion of the joint is sufficient to maintain the viability of the intervertebral disc and the articular cartilage. The concept of the micromotion of the joint was particularly important because mechanical fixation failures are the most common mode of failure of the instrumentation without fusion. This could only be overcome by narrowing the difference in the flexibility between the vertebral column and the implant. The difference may be reduced either by making the spinal column stiffer by increasing the fixation rigidity or by reducing the stiffness of the spinal implant construct through modification of the components. However, because the prime goal of the instrumentation surgery is maintenance of the corrective force on the spine, the balance had to be on the side of making the spine significantly stiffer than the uninstrumented spine, leaving little other alternative than controlled micromotion. The implant-bone interface forms the grip on the vertebral body and determines the strength of the vertebral stabilization effected through the instrumentation. Failure here, though heavily influenced by the overall stiffness of the implant construct, may be also influenced by the biomechanical characteristics of the anchoring member, namely the pedicle screws. The characteristic of the longitudinal member determines the overall gross stiffness of the implant construct, but not necessarily the intersegmental motion, as the lever arm of the intersegmental motion is very small compared to the length of the entire implant. A failure here is mainly due to the difference in stiffness of the implant and the vertebral column. A smaller diameter (5-mm) rod was initially created using a titanium alloy to permit 15% elastic deformation of the rods under physiologic loads to reduce the stiffness of the implant. The rods are now made of polished stainless steel, which provides minimal wear debris at the rod-polyethylene roller junction. The connecting mechanism that forms the implant construct determines the stability between the implant members and is the main site determining the intersegmental and the translational
Orthobiom: A Nonfusion Treatment for Pediatric Scoliosis
719
motion. Two types of connecting mechanisms were designed, one called the fixed connector, which locks the interface and offers control of the rods (e.g., prevent slippage and rotation), and the mobile connectors which allow 6 degrees of freedom to allow unconstrained linking between the anchoring and the longitudinal member. By combining two types of connection mechanisms on the same rod, the desired mechanical forces could be exerted on the motion segments incorporated within the instrumented segments. CONCLUSION Pediatric scoliosis can severely affect a large population of children if not treated properly and immediately. The best-case scenario is to treat and arrest the progression of the curve with bracing. However, if progression of the curve continues, other options should be considered, such as the Orthobiom Spinal System. REFERENCES 1. Rogala EJ, Drummond DS, Gurr J: Scoliosis: Incidence and natural history: A prospective epidemiological study. J Bone Joint Surg Am 60(2):173–176, 1978. 2. Rivard CH, Rhalmi S, Coillard C: In vivo biocompatibility testing of peek polymer for a spinal implant system: A study in rabbits. J Biomed Materials Res 62(4):488–498, 2002. 3. McGloughlin TM, Kavanagh AG: Wear of ultra-high molecular weight polyethylene (UHMWPE) in total knee prostheses: A review of key influences. Proc Inst Mech Eng [H] 214(4):349–359, 2000. 4. Edidin AA, Kurtz SM: Influence of mechanical behavior on the wear of 4 clinically relevant polymeric biomaterials in a hip simulator. J Arthroplasty 15(3):321–331, 2000. 5. Bavaresco VP, de Carvalho Zavaglia CA, de Carvalho Reis M, Malmonge SM: Devices for use as an artificial articular surface in joint prostheses or in the repair of osteochondral defects. Artif Organs 24(3):202–205, 2000. 6. Soucacos PN, Zacharis K, Gelalis J, et al: Assessment of curve progression in idiopathic scoliosis. Eur Spine J 7(4):270–277, 1998. 7. Kane WJ: Scoliosis prevalence: A call for a statement of terms. Clin Orthop 126:43–46, 1977. 8. Lonstein JE, Carlson JM: The prediction of curve progression in untreated idiopathic scoliosis during growth. J Bone Joint Surg Am 66(7):1061–1071, 1984. 9. Cobb JR: Outline for the study of scoliosis. Instructional Course Lectures, American Academy of Orthopaedic Surgeons, Vol 5. Ann Arbor, MI, JW Edwards, 1948. 10. Labelle H, Dansereau J: Orthotic treatment of pediatric spinal disorders and diseases. In Spine: State of the Art Reviews, Vol 4, No 1. Philadelphia, Hanley & Belfus, 1990.
CHAPTER
92
Considerations for Spinal Arthroplasty in Elderly and Osteoporotic Patients James J. Yue
K E Y l
l
l
l
P O I N T S
Comprehensive history and physical examination are mandatory prior to considering motion-sparing technology in elderly patients and in those with osteoporotic bone. Lumbar total disc replacement results in an overall lordosis of the lumbar spine. Therefore, circumferential spinal stenosis should be considered a relative contraindication to lumbar disc replacement. Dynamic posterior stabilization procedures require careful analysis of coronal and sagittal instability prior to implementation. The adjunctive use of bone augmentation such as PMMA (polymethyl/ methacrylate) or Cortoss may permit limited use of motion-sparing technologies in the osteoporotic patient.
The use of motion-sparing technologies such as total disc arthroplasty, nucleus replacement, and posterior dynamic stabilization has been shown to be safe and effective in patients who are 18 to 60 years old.1–4 Except for interspinous spacers, the use of nonfusion technologies has not been evaluated to any great degree in the elderly.5–8 A number of considerations such as, but not limited to, the presence, degree, and location of spinal stenosis, degenerative facet disease, bone density, and vascular status should be carefully evaluated in all patients, especially those older than 60, before proceeding with a nonfusion surgical technique. Because of the increase in potential medical complications, this patient population requires careful general medical evaluation prior to undergoing any surgical procedure.
720
is subarticular/foraminal or central canal stenosis, compression of the neural elements may occur and lead to neurologic deficit (Fig. 92–2).10,11 Physical examination should assesss for the presence of abdominal pulsatile masses indicative of an abdominal aortic aneurysm. Pulses should be evaluated for vascular insufficiency. Unilateral low back pain on extension and lateral bending should prompt an evaluation of facet integrity as well as sacroiliac (SI) joint disease. “Hip” pain should prompt careful assessment of intra-articular femoral-acetabular disease. Radiographically, all patients should undergo plain x-rays, MRI, computed tomography (CT), and bone density evaluation. Standing radiographs should be assessed for scoliotic deformity, dynamic degenerative spondylolisthesis, aortic calcification, and osteopenia. MRI evaluation should include careful evaluation for other “red flag” findings such as infection and tumor. In addition, significant facet disease, facet cysts, and osteoporotic compression fractures should be assessed. One of the most significant contraindications, as already mentioned, is circumferential spinal stenosis. The degree and location of foraminal stenosis should be evaluated carefully. CT scans should be obtained to assess for facet degeneration, pars defects, and SI joint disease. Lastly, but equally as important, a DEXA (dualemission x-ray absorptiometry) scan should be obtained in all patients. A T-score less than 1.0 should be considered a contraindication in this patient population.
CONSIDERATIONS FOR DISC ARTHROPLASTY
POSTERIOR DYNAMIC LUMBAR STABILIZATION
Although most patients over 60 years old are not ideal candidates for total disc replacement (TDR), TDR can be performed successfully in a select subset of carefully chosen patients in this age group.9 As with all patients being considered for TDR, a detailed history and physical examination are required. Patients with a history of neurogenic claudication secondary to circumferential stenosis as demonstrated on magnetic resonance imaging (MRI) are not ideal candidates for lumbar TDR (Fig. 92–1). The overall effect of a lumbar TDR is to place the spine in a lordotic position. If there
The indications for posterior dynamic stabilization of the lumbar spine are evolving and are currently undergoing formal randomized testing.12–16 The majority of these devices are pedicle screw based and, therefore, depend on adequate bone quality to assure appropriate bone fixation. Few of these devices, other than the total posterior arthroplasty systems such as the Total Facet Arthroplasty System (TFAS) (Archus Orthopedics, Redmond, WA) or the TOPS device (Impliant Spine, Princeton, NJ), are able to adequately control for sagittal or coronal instability.17,18 Therefore,
CHAPTER 92
n
Considerations for Spinal Arthroplasty in Elderly and Osteoporotic Patients
721
F I G U R E 92–1. Circumferential lumbar spinal stenosis.
careful assessment for excessive spinal motion (e.g., greater than Grade 1 dynamic spondylolisthesis) in either the coronal or sagittal plane should be made. Patients with excessive spinal motion should be considered for fusion. Of all of the motion-sparing devices available to date, the interspinous spacer devices have been examined most critically in the elderly population.5–8 Advantages include limited spinal exposure, minimal blood loss, and the potential for the use of local anesthesia. The device is highly dependent on the structural integrity of the two immediately adjacent spinous processes. Patients with severe spinal stenosis and concomitant cauda equine syndrome should not be considered for this type of intervention. These patients should undergo formal decompression with or without fusion. Patients with poor bone quality defined as T-scores less than 2.5 should also be considered for alternative therapies.
CONSIDERATIONS FOR THE OSTEOPOROTIC PATIENT Subsidence has been reported to occur in 9% of patients when Tscores are less than 1.59 (Fig. 92–3). Vertebral bony augmentation using adjunctive bone fillers such as PMMA or Cortoss (Orthovita, Malvern, PA) is currently being evaluated under controlled studies. Cortoss is a bioactive glass ceramic polymer composite that has a high radiodensity, low exothermic reaction, and low viscosity19 (Fig. 92–4). The polymer induces a local alkaline environment that induces the deposition of calcium phosphate and bone deposition. Yue and associates have presented non-randomized data evaluating the use of Cortoss to augment vertebral body strength in osteoporotic patients undergoing lumbar total disc replacement surgery.20 They evaluated 33 patients with an average T-score of 2.5 (range 1.5 to 4.4). Thirty-three
722
P A R T
X I I
Controversies
A
B
C
D
F I G U R E 92–2. Decrease in central and foraminal area with lordosis of spine (A and C). Increase in central and foraminal area with flexion of spine (B and D). (From Richards JC, Majumdar S, Lindsey DP, et al: The treatment mechanism of an interspinous process implant for lumbar neurogenic intermittent claudication. Spine 30:744–749, 2005; Siddiqui M, Karadimas E, Nicol M, et al: Influence of X STOP on neural foramina and spinal canal area in spinal stenosis. Spine 31:2958–2962, 2006.)
n
percent were smokers. Subsidence rates decreased from 9% to 0% after the use of Cortoss placed anteriorly after implantation of the TDR device. No adverse events occurred in these patients over an average follow-up period of 35 months. SUMMARY Careful clinical evaluation including a comprehensive history and physical examination and radiographic analysis is essential prior
to considering the elderly or osteoporotic patient. Spinal stenosis and neurogenic claudication should be considered relative contraindications to lumbar TDR surgery. Patients with risk factors for osteoporosis should be carefully screened to ensure adequate bone strength. The use of adjunctive bone fillers such as PMMA and Cortoss may increase the relative bone strength and permit the limited use of motion-sparing technologies in this patient population. Further studies are warranted prior to utilizing these adjunctive technologies in all osteoporotic patients.
CHAPTER 92
n
n
Considerations for Spinal Arthroplasty in Elderly and Osteoporotic Patients
F I G U R E 92–3. Example of subsidence.
F I G U R E 92–4. Use of Cortoss in multilevel total disc replacement surgery. (Photograph courtesy of
R. Bertagnoli.)
723
724
P A R T
X I I
Controversies
REFERENCES 1. Bertagnoli R, Yue JJ, Fenk-Mayer A, et al: Treatment of symptomatic adjacent-segment degeneration after lumbar fusion with total disc arthroplasty by using the Prodisc prosthesis: A prospective study with 2-year minimum follow-up. J Neurosurg Spine 4:91–97, 2006. 2. Bertagnoli R, Yue JJ, Kershaw T, et al: Lumbar total disc arthroplasty utilizing the ProDisc prosthesis in smokers versus nonsmokers: A prospective study with 2-year minimum follow-up. Spine 31:992–997, 2006. 3. Bertagnoli R, Yue JJ, Shah RV, et al: The treatment of disabling single-level lumbar discogenic low back pain with total disc arthroplasty utilizing the Prodisc prosthesis: A prospective study with 2-year minimum follow-up. Spine 30:2230–2236, 2005. 4. Bertagnoli R, Yue JJ, Shah RV, et al: The treatment of disabling multilevel lumbar discogenic low back pain with total disc arthroplasty utilizing the ProDisc prosthesis: A prospective study with 2-year minimum follow-up. Spine 30:2192–2199, 2005. 5. Anderson PA, Tribus CB, Kitchel SH: Treatment of neurogenic claudication by interspinous decompression: Application of the X STOP device in patients with lumbar degenerative spondylolisthesis. J Neurosurg Spine 4:463–471, 2006. 6. Chiu JC: Interspinous process decompression (IPD) system (XSTOP) for the treatment of lumbar spinal stenosis. Surg Technol Int 15:265–275, 2006. 7. Hsu KY, Zucherman JF, Hartjen CA, et al: Quality of life of lumbar stenosis-treated patients in whom the X STOP interspinous device was implanted. J Neurosurg Spine 5:500–507, 2006. 8. Kondrashov DG, Hannibal M, Hsu KY, et al: Interspinous process decompression with the X-STOP device for lumbar spinal stenosis: A 4-year follow-up study. J Spinal Disord Tech 19:323–327, 2006. 9. Bertagnoli R, Yue JJ, Nanieva R, et al: Lumbar total disc arthroplasty in patients older than 60 years of age: A prospective study of the ProDisc prosthesis with 2-year minimum follow-up period. J Neurosurg Spine 4:85–90, 2006.
10. Richards JC, Majumdar S, Lindsey DP, et al: The treatment mechanism of an interspinous process implant for lumbar neurogenic intermittent claudication. Spine 30:744–749, 2005. 11. Siddiqui M, Karadimas E, Nicol M, et al: Influence of X Stop on neural foramina and spinal canal area in spinal stenosis. Spine 31:2958–2962, 2006. 12. Beastall J, Karadimas E, Siddiqui M, et al: The Dynesys lumbar spinal stabilization system: A preliminary report on positional magnetic resonance imaging findings. Spine 32:685–690, 2007. 13. Cakir B, Richter M, Huch K, et al: Dynamic stabilization of the lumbar spine. Orthopedics 29:716–722, 2006. 14. Nockels RP: Dynamic stabilization in the surgical management of painful lumbar spinal disorders. Spine 30:S68–S72, 2005. 15. Schwarzenbach O, Berlemann U, Stoll TM, et al: Posterior dynamic stabilization systems: DYNESYS. Orthop Clin North Am 36:363– 372, 2005. 16. Yue JJ, Timm JP, Panjabi M, et al: Clinical application of the Panjabi Neutral Zone Hypothesis: The Stabilimax NZTM Posterior Lumbar Dynamic Stabilization System. Neurosurg Focus 22:E12, 2007. 17. Wilke HJ, Schmidt H, Werner K, et al: Biomechanical evaluation of a new total posterior-element replacement system. Spine 31:2790–2796; discussion 2797, 2006. 18. Zhu Q, Larson CR, Sjovold SG, et al: Biomechanical evaluation of the Total Facet Arthroplasty System: 3-dimensional kinematics. Spine 32:55–62, 2007. 19. Gheduzzi S, Webb JJ, Miles AW: Mechanical characterisation of three percutaneous vertebroplasty biomaterials. J Mater Sci Mater Med 17:421–426, 2006. 20. Yue JJ, Bertagnoli R, Lee R, Kirk J: A prospective, non-randomized analysis of the adjunctive use of Cortoss vertebroplasty in lumbar disc arthroplasty utilizing the Prodisc prosthesis. Presented at the Spine Arthroplasty Society Meeting, New York, NY, 2006.
CHAPTER
93
Multilevel Lumbar Disc Arthroplasty Rudolf Bertagnoli
K E Y l
l
l
l
P O I N T S
For the generation of multilevel disc degeneration, a variety of intrinsic and extrinsic factors are discussed. The treatment of the consequently arising pain scenarios requires at some stage spinal arthroplasty when it is favorable to use motion-sparing devices instead of fusion procedures. If the disc (and not posterior structures) is identified as the pain generator, the use of total disc replacement appears to be beneficial. The use of multilevel motion-preserving devices appears to be superior to multilevel fusion procedures because the overall segmental stability is maintained. These procedures require critical patient selection and great surgical experience and precision. Multilevel fusion procedures will remain the option in treating conditions unrelated to degenerative disc disease such as scoliosis and long degenerations.
Degenerative multilevel lumbar disc disease (MLDD) is one of the most frequent reasons for chronic and disabling low back pain. This pain has an extensive impact on the social, economic, and psychological situation of the affected individual. The patient’s history often includes repetitive external trauma. In most cases these patients received discectomies for disabling radicular symptoms and have also developed low back pain at the affected as well as the adjacent levels. Several risk factors also have to be taken into account when looking for the reasons of MLDD: Heredity factors (discussed in the literature1–4), smoking habits, obesity, and poor physical conditioning all influence the pathogenesis of MLDD. Typically, all nonsurgical options should be exhausted and disability and pain should still be present before operative intervention is considered. Different forms of spinal arthrodesis such as posterior, posterolateral fusion, anterior interbody fusion with or without instrumentation, and combined anterior/posterior fusion for the treatment of single- and bisegmental chronic discogenic low back pain have been investigated and experiences have been described.5 Factors such as age, workers’ compensation, and smoking status were considered. In prospective studies, direct correlation in patient satisfaction rates and fusion success could not be demonstrated.6 There are also some disadvantages of spinal arthrodesis which are not negligible, for example, adjacent-level degeneration (36.1% at 10 years) of discs and facet joints as well as symptomatic
pseudarthrosis, graft site morbidity, and pain due to reaction to instrumentation.7 PATIENT SELECTION The right indication, proper patient selection, and an optimally placed device are the key factors for successful treatment in total disc arthroplasty (TDA) in general and especially in multilevel TDA. Therefore, a thorough preoperative assessment is essential in which the commonly used diagnostic steps and tools for evaluating degenerative disc disease (DDD) should be applied to identify exactly the pain generator. The factors evaluated are age, gender, pain history and presence, prior conservative treatment, prior surgeries (spinal and abdominal), application for workers’ compensation, and social and psychological factors. Furthermore, measurements and scores, such as the Oswestry Disability Index (ODI), SF-36, and Visual Analog Scale (VAS), and clinical investigation are employed for assessment. Noninvasive radiographic evaluation (e.g., x-ray, magnetic resonance imaging [MRI], computed tomography [CT] studies) is mandatory to understand pathologic changes and static considerations. Of great importance, however, is also the direct detection of the pain generators because in multilevel diseased situations often the pathologically changed areas are not the pain generators. In these situations only the invasive methods (e.g., nerve root blocks, facet blocks, discograms, myelography, technetium bone scan) can identify the pain generators. Discograms are very helpful in determining painful discs, and fluoroscopic or CT-guided facet or root blocks can help to detect pain generators in that area. Additionally, osteodensiometries to understand the bone quality should be performed on a regular basis on patients over age 40 to reduce the risk of undetected subsidence in multilevel cases. Three-dimensional (3-D) angiography-CT can help in preoperative planning of the anterior approach. Videofluoroscopy (moving radiograph) is useful for understanding the motion pattern of spinal areas. In summary, only a multimodal investigational pattern can help in finding the objective diagnostic status. 725
726
P A R T
X I I
Controversies
INDICATIONS If in general the disc can be identified as the main pain generator in patients with low back pain, this is a good indication for TDA. If the posterior structures (e.g., facet joints) are the main cause for low back pain, TDA is not the choice of treatment. Also, patients with spinal deformities such as degenerative scoliosis are considered to be a critical group to be treated with TDA. According to the Clinical Success Related Classification (CSRC),8 the prime indication for TDA is monosegmental lumbar symptomatic DDD with or without herniation or large central herniations. The patient should not have had previous surgery at the affected level. Good indications are seen in failed disc surgery syndromes without laminectomy or severe facet alterations as well as bisegmental lumbar symptomatic DDD. Expanded indications are both fusion adjacent-level instabilities, cases with more than two-level DDD, degenerative scoliosis up to 25 degrees Cobb angle, and degenerative, self-reducible spondylolisthesis without mature canal stenosis (M I-II). CONTRAINDICATIONS Contraindications for TDA in general are bone disorders such as osteoporosis, osteopathies, severe canal stenosis, hypertrophic spondyarthrosis, isolated radiculopathy, spinal deformities (spondylolisthesis), infections, tumors, and predominantly psychosocial factors. Significant facet joint arthritis, severe end plate irregularities, hemangiomas, metal allergies, pregnancy, and body mass index greater than 35 are also contraindicated. DESCRIPTION OF THE DEVICE In the clinical application, unconstrained prostheses have been discovered as unsuitable in the use for multilevel disc replacement. In contrast, prosthesis with a semiconstrained kinematics in terms of controlling flexion and extension and left/right lateral bending in the given radius of the ball-and-socket concept proved to be suitable for the multilevel disc replacement in the coupled motion of the lumbar spine. It does not restrict randomized movements (Fig. 93–1). In our clinical experience, the use of ProDisc-L (Synthes, Inc., Paoli, PA) (see Fig. 93–1) has been demonstrated to be a successful device in treatment of DDD patients. OPERATIVE TECHNIQUES For the surgical approach, the patient is positioned in the da Vinci position on a radiolucent imaging table with arms and legs abducted. The surgeon is working between the patient’s legs to get a better view of both sides of the disc and to have a better perspective view when applying the prosthesis perpendicular into the prepared disc space. To determine the level, the obliquity of the diseased disc, and an orthograde postion of the patient fluoroscopic imaging in the lateral and anteroposterior (AP) positions is conducted before skin incision. In multilevel procedures a paramedian left side longitudinal incision is recommended. A selfretaining retractor system such as the SynFrame (Synthes, Inc.) anterior spinal retractor is very useful for generation of a secure and stable retroperitoneal approach of the anterior aspect of the
n
F I G U R E 93–1. Modular design of the ProDisc-L implant.
spine. The affected levels are approached segment by segment starting from the lowest level. As an exemption, the upper levels (L1-L2 and L2-L3) should be exposed first due to their cephalad orientation of the discal corridor line in the cranial portion of the lumbar lordosis. A complete release is essential to allow proper function of the device. Therefore, after a complete discectomy in cases with a disc height below 40% of the normal height, the resection of the posterior longitudinal ligament (PLL) as well as the intraforaminal annulus portion is necessary. A moderate end plate remodeling for a better fit of the prosthesis is often necessary. Generally, the largest possible footprint and the smallest possible disc height should be used to obtain an optimal load transmission onto the vertebrae and to avoid overdistraction of multiple segments. Especially in multilevel procedures, a secure primary fixation is essential to avoid postoperative migration of the devices into an unfavorable kinematics of the coupled motion of the lumbar spine. Keel prosthesis like the ProDisc-L offers large benefits in this regard. A precise positioning of the artificial disc and a precise matching of the centers of rotation are mandatory to allow a harmonic function in the treated areas. The more levels replaced by artificial discs, the greater the importance of these parameters. POSTOPERATIVE CARE Remobilization can begin on the first postoperative day. Especially in multilevel treatment with collapsed discs, a significant increase of overall height leads to significant tension to ligaments, capsules of joints, and muscles. This might be the cause of “distention pain” that may occur after surgery. According to the Straubing Rehabilitation Index (SRI),9 we recommend a dynamic orthesis for 6 weeks and no sports activity during this period. Multilevel patients
CHAPTER 93
should get physiotherapeutic treatment, but not before 6 weeks, and should avoid sports activities for at least 3 months. MATERIALS AND METHODS Prospective data of all multilevel ProDisc-L procedures at our institution were collected for a minimum of 5 years. A complete radiographic assessment including AP, lateral, flexion, and extension imaging was performed before surgery as well as MRI and discography and optional CT scans. To reduce the amount of intraoperative variables (positioning of the patient, the approach or application technique, etc.) all the procedures were performed by the author at a single institution. The data were collected, compiled, and analyzed by independent examiners. Patients were assessed before surgery and after surgery at 3, 6, 12, 24, 36, and 48 months. The results were analyzed before and after surgery. Clinical scores and assessment parameters such as Oswestry Disability Index (ODI), Visual Analog Scale (VAS), patient satisfaction, back pain, radicular pain, medication usage, and complications have been investigated. Radiographically disc heights affected and adjacent levels as well as the range of motion, flexion, and extension were obtained. All adverse outcomes related to the index procedure were recorded. In the study 249 patients (139 male and 110 female patients) with multilevel lumbar ProDisc-L total disc arthroplasty have been included. The median age of all patients was 46 years (23–69 years); 120 patients were treated in two levels, 45 patients in three levels, 1 patient in four levels, and 1 patient in five levels.
Multilevel Lumbar Disc Arthroplasty
727
The VAS score decreased from 7.2 preoperatively to 4.1 at the 1-year follow-up, 3.2 after 2 years, 3.8 after 3 years, and 3.8 after 4 years. Eighty-seven percent of the patients reported to be completely satisfied or satisfied with the surgery 1 year postoperatively and 94% were satisfied 2, 3, and 4 years postoperatively. While all patients complained about continuous back pain preoperatively, 60% could be reported to have no back pain after surgery, 37% had moderate pain, and 3% had less back pain 2 years postoperatively. This improvement did not change at 4 years of follow-up. COMPLICATIONS In eight cases device- or approach-related complications occurred, resulting in an overall complication rate of 3.2%.
Device-Related Complications Five cases of partial implant subsidence (2%) (Fig. 93–2A, B) and one case of polyethylene inlay extrusion (0.4%) have been observed. No other cases of loosening, migration, allergic rejection/reaction, visceral or neurologic injuries were caused by the device components or infection.
Approach-Related Complications CLINICAL OUTCOMES The ODI score was reduced from 49% before surgery to 27% after the 1-year follow-up, 27% after 2 years, 26% after the 3-year and 4-year follow-up periods.
n
F I G U R E 9 3–2 .
subsidence.
One case of subcutaneous hematoma (0.4%) was identified that required reintervention as well as one case of retrograde ejaculation (0.4%). No cases of vascular injury, ureteral, or neurologic injury occurred.
Anteroposterior (A) and lateral (B) postoperative radiographs showing
728
P A R T
X I I
Controversies
CASE STUDIES
CONCLUSION
Case 1 (Two Levels) A 61-year-old man with a multiyear history of low back pain remained resistant to conservative therapy. He had developed vertical segmental instabilities at levels L3-L4 and L4-L5 as well as FDS at level L4-L5 (Fig. 93–3A, B). Surgical treatment was multilevel total disc arthroplasty (two levels) with ProDisc-L at levels L3 to L5. Postoperative radiographs show good maintenance of mobility of the segments (see Fig. 93–3C to E).
The reason for MLDD seems to be a multifactoral combination of genetic, traumatic, social (tobacco use), physical (obesity), and age (senescence) factors.10 Also, two different radiographic appearances of MLDD can be distinguished on MRI: In one, all affected levels seem to be more or less in the same status of degeneration; in another, different levels can be affected in different stages randomly. In patterns with more severe changes at the caudal level, this level (so-called “proximate disc”) obviously is influencing subsequently the adjacent level degeneration.11 Mostly, this proximate disc is situated more caudally. This type of MLDD is termed “proximate MLDD” (Fig. 93–7A, B). We have found extraordinary positive results in patients from this group, which makes this group the ideal group of patients for multilevel lumbar disc arthroplasty. Multilevel ProDisc lumbar surgery appears to ensure instantaneous implant stability and functional mobility, which allows a rapid reintegration in daily life and return to work. This applies for primary surgical procedures as well as for patients with prior posterior decompressive procedures. Multilevel treatment with motion-sparing devices shows a more obvious benefit in maintenance of the segmental stability in the overall function of the patient than patients with multilevel fusion procedures. But these procedures require a high level of technical precision and surgical experience. Careful patient selection is essential for best surgical outcomes. Also, motion-sparing technologies seem to be very beneficial for patients and do have a lot of advantages to multilevel fusion procedures in DDD patients; severe mechanical instabilities (e.g., fractures), tumors, spondylolisthesis, long deformities (e.g., scoliosis, kyphosis, stenosis, long degenerations), severe facet changes, and degenerative scoliosis will still remain for spinal fusion procedures. Therefore, these procedures are not competitive with fusion procedures and allow the surgeon to be more selective in terms of treatment. Long-term data will be necessary to understand the real long-term value of multilevel motion-preserving procedures.
Case 2 (Three Levels) A 42-year-old man was suffering from pseudoradicular lower back pain accompanied by leg pain resisting conservative treatment for years. Surgical treatment of the diagnosed symptomatic disc disease at levels L3-S1 was a three-level total disc replacement with ProDisc-L. X-ray studies showed proper vertebral alignment (Fig. 93–4A to C). Case 3 (Four Levels) A 52-year-old man suffered several years from low back pain. All conservative treatment options failed. Radiographic and MRI revealed vertical segmental instabilities from L1 to S1 and herniations from L1 to L4 (Fig. 93–5A, B). Surgical treatment was multilevel total disc arthroplasty (four levels) with ProDisc-L at levels L1 to L5 and vertebroplasty. Postoperative radiographs showed correct positioning of the devices and a good restoration of disc height (see Fig. 93–5C, D). Case 4 (Five Levels) A 50-year-old man with low back pain for years that failed nonsurgical treatment revealed on radiographic and MRI scans a symptomatic degenerative disc disease. This extended from L1 to S1 with a hyperlordosis and imbalanced lumbar spine. Surgical treatment was a multilevel total disc replacement (five levels) with ProDisc-L prosthesis, which was combined with a vertebroplasty to stabilize the bone (Fig. 93–6A to D). Postoperative imaging showed correct positioning and good restoration of disc heights and a rebalanced spine.
CHAPTER 93
Multilevel Lumbar Disc Arthroplasty
F I G U R E 93–3. A, B, Preoperative radiographs: anteroposterior, lateral (from left to right). C–E, Postoperative radiographs: lateral, extension, flexion (from left to right).
n
729
730
P A R T
n
X I I
Controversies
F I G U R E 93–4. A–D, Pre- and postoperative radiographs: anteroposterior, lateral.
CHAPTER 93
n
Multilevel Lumbar Disc Arthroplasty
F I G U R E 93–5. A–D, Pre- and postoperative radiographs:
anteroposterior, lateral.
731
732
P A R T
X I I
Controversies
n
F I G U R E 93–6. A–D, Pre- and postoperative radiographs:
anteroposterior, lateral.
CHAPTER 93
n
F I G U R E 93–7.
appearance.
Multilevel Lumbar Disc Arthroplasty
A–C, Proximate multilevel lumbar disc disease: example of cascade
733
734
P A R T
X I I
Controversies
REFERENCES 1. Anderson DG, Tannoury C: Molecular pathogenetic factors in symptomatic disc degeneration. Spine J 5(6):260S–266S, 2005. 2. Solovieva S, Lohiniva J, Leino-Arjas P, et al: COL9A3 gene polymorphism and obesity in intervertebral disc degeneration of the lumbar spine: Evidence of gene-environment interaction. Spine 27:2691– 2696, 2002. 3. Sambrook PN, MacGregor AJ, Spector TD: Genetic influences on cervical and lumbar disc degeneration: A magnetic resonance imaging study in twins. Arthritis Rheum 42:1729–1735, 1999. 4. Annunen S, Paassilta P, et al: An allele of COL9A2 associated with intervertebral disc disease. Science 285:409–412, 1999. 5. O’Brien JP: The role of fusion for chronic low back pain. Orthop Clin North Am 14:639–647, 1983. 6. Gibson A, Grant I, Waddel LG: The Cochrane review of surgery for lumbar disc prolapse and degenerative lumbar spondylosis. Spine 24:1820–1832, 1999.
7. Ghiselli GWJ, NN, Hsu WK, et al: Adjacent segment degeneration in the lumbar spine. J Bone Joint Surg Am 86:1497–1503, 2004. 8. Bhatia Bertagnoli R: Indications for total lumbar disc replacement. In Kim DH, Cammisa FP, Fessler RG (eds): Dynamic Reconstruction of the Spine (in process citation). 9. Bertagnoli R, Pfeiffer F: The Straubing Rehabilitation Index (SRI) as a guideline in postop treatment after ProDisc implantation. In process citation. 10. Savitz MH CJ, Yeung AT: The Practice of Minimally Invasive Spinal Technique. 2000, pp 149–165. 11. Bertagnoli R, Yue JJ, Shah RV, et al: The treatment of disabling multilevel lumbar discogenic low back pain with total disc arthroplasty utilizing the ProDisc prosthesis. Spine 30(19):2192–2199, 2005.
CHAPTER
94
Posterior Lumbar Arthroplasty Manoj Krishna
K E Y l
l
l
l
l
P O I N T S
Posterior lumbar arthroplasty has advantages over anterior arthroplasty in being easy to revise, dealing with disc/facet/nerve disease, having an easy approach, and being applicable to the vast majority of patients with lumbar pathology. A paired disc and posterior dynamic stabilizer (PDS) is a concept that is being developed. The instantaneous axis of rotation (IAR) of the disc and PDS needs to be matched for the system to function. The PDS on its own needs to offer shear stability, angular motion, and variable motion and must match the IAR of the disc to deliver the best results. The PDS may significantly reduce the need for posterolateral fusion.
Anterior lumbar arthroplasty is well established as a surgical option in the treatment of discogenic low back pain. Its clinical use has highlighted several problems with this technology including a difficulty in revision, continued facet pain, use of an unfamiliar approach, and its limited use in patients with facet or neural pain. In response to these issues, surgeons started exploring the concept of posterior lumbar arthroplasty. Its main potential advantages are that it uses a familiar approach, deals with all the pain generators in a motion segment, and can be easily revised through an anterior lumbar interbody fusion (ALIF). This chapter explores the concepts and challenges in the development of posterior lumbar arthroplasty. HISTORY The history of anterior lumbar arthroplasty is well documented elsewhere in this book, from its origins in the Charité Hospital in Berlin. The rationale for arthroplasty is to reduce disc degeneration adjacent to a fused lumbar segment and provide more physiologic kinetics and load transmission across the disc. The evidence that it actually does so is emerging from various studies but is by no means conclusive at this stage. If one excludes the PDN and other nucleus replacement devices, there have been few posteriorly inserted arthroplasty devices which replace the disc and allow motion. Globus Medical (Audubon, PA) has developed a posteriorly inserted paired disc replacement that needs to have parallel insertion into the disc space.
A number of dynamic stabilization devices that are pediclebased allow some motion, usually less than 3 degrees. The GRAF ligament system (Neoligaments, Leeds, U.K.), the Dynesys (Zimmer Spine, Inc., Warsaw, IN), Isobar (Scient'x, Aylesbury, U.K.), Agile (Medtronic Sofamor Danek, Memphis, TN), and various PEEK (polyetheretherketone) and metal rods with some give in them are part of this group. Their aim is more stability than motion; any motion provided is designed mainly to prevent breakage of the device under repetitive loading rather than restore normal kinematics to the motion segment. Facet replacement devices form a third group of posteriorly inserted motion devices. The Total Facet Arthroplasty System (TFAS) from Archus Orthopedics Inc. (Redmond, WA) is currently undergoing clinical trials. It is a pedicle-based system that requires the pedicle screws/bolts to be cemented in. This is a concern to some surgeons. Facet Solutions (Logan, UT) has developed the Anatomic Facet Replacement System (AFRS), a pedicle-anchored facet replacement system that is undergoing clinical evaluation. Impliant Spine (Princeton, NJ) has developed the TOPS system, a pedicle-based system that replaces the posterior facets and spinous processes and requires removal of the midline for insertion. This is also undergoing clinical evaluation. The indications for these dynamic stabilization and facet replacement devices are mainly spinal stenosis with instability or following decompression requiring significant facet joint removal. These techniques may obviate the need for fusion and may have advantages in reducing operative time, morbidity, and length of hospital stay and in providing a faster recovery. No system currently exists that combines a posteriorly inserted disc replacement with a coordinated posterior dynamic stabilizer. Disc Motion Technologies (Boca Raton, FL) is developing such a system. It will be a true total disc replacement (TDR) system and inserted through the posterior route, with all its inherent advantages.
ANATOMIC CONSIDERATIONS There are several aspects of the anatomy of the motion segment that influence the design and surgical technique in relation to posterior lumbar arthroplasty (PLA) surgery. 735
736
P A R T
X I I
Controversies
The aim of PLA is to deal with all the pain generators in a motion segment, which comprises the disc anteriorly, the two facet joints posteriorly, and the neural structures. It is important to consider the disc and facets as part of one motion segment. Fujiwara and associates1 studied the radiologic relationship between disc and facet degeneration. They concluded that disc degeneration precedes facet degeneration. Gries and associates2 studied the histologic relationship between disc and facet degeneration. They found that microscopic changes of degeneration could be seen in the disc and facet at an early age and can be quite marked before 40 years of age, but there was no clear correlation between the two. Butler and associates3 found that discs degenerate earlier than facets. Anterior disc arthroplasty has been suggested in patients with no facet degeneration. If the anterior disc replacement does not restore synchronous movement to the facet, either through overdistraction or a mismatched instantaneous axis of rotation (IAR), it is probable that the facet may experience accelerated degeneration. Over a period of time, facet joints will degenerate and likely become pain generators. PLA has a theoretical advantage in being able to deal with the facet joint and the discs at the same time. Tanaka and associates4 studied the relationship between disc degeneration and mobility in cadaveric motion segments. Grades 1 through 4 degeneration was associated with more movement in the segment and Grade 5 with reduced movement. The end plates are thickest at the rim laterally and thinnest centrally. To avoid subsidence problems, devices will need to be placed as laterally as possible. In order to place a device laterally, the lateralmost annulus will need to be removed. This is difficult because it is a solid structure and not routinely accessible to the shavers that go into the space occupied by the nucleus. Placing a device laterally implies that it is more likely to be angled slightly medially, as there is not enough room in the lateral disc to accept a 25-mm long device. If paired bilateral devices are used, then they need to be parallel to each other or must be able to function as one device in a nonparallel insertion. The end plates of the disc in the anteroposterior (AP) view are not parallel to each other but slope laterally toward each other. The extent of this slope varies. It is greater in advanced degeneration and lower levels. This has implications in the design of the disc prosthesis. If one wants to place these devices laterally for maximum end plate strength, then the devices need to be anatomically adapted to the lateral disc space and the implants need to follow the end plate contour. If this is not done, as the device height increases, they will need to be placed more medially. The degree of osteoporosis in the vertebra adjacent to the disc has implications for subsidence. After 55 years of age, a bone density scan needs to be done, and disc replacements avoided in patients with bone density more than 2 standard deviations below the norm. Conjoined nerve roots are seen in approximately 4% of patients. One needs to keep that in mind during surgery and have a technique which excludes a conjoined nerve root before using the cautery to stop bleeding. If a conjoined nerve root is found and retraction is stretching the nerve, then no implant should be placed on that side and one may need to revise the procedure to a fusion.
Epidural veins can cause significant bleeding during surgery, as they cover the disc. Care must be taken to reduce the abdominal pressure. Careful bipolar cautery of the veins needs to be done, but avoid cautery in areas which are close to the nerve (e.g., the foramen or next to the exiting nerve above the disc). The main epidural vein is present just above the lower pedicle. The triangle formed between the traversing nerve and the exiting nerve is the area where the disc can be accessed safely, without needing to retract the traversing nerve root. The course of the exiting nerve above the disc varies. In some cases it can be quite vertical, crossing the disc more medially than normal. Its distance from the upper disc border also varies. In some cases, particularly where the disc height is reduced, it can appear perilously close to the disc and needs careful retraction. BIOMECHANICAL CONSIDERATIONS There are various causes of back pain. We will focus here on the patient with discogenic back pain who has failed conservative measures and is being considered for surgery. These patients typically present with a history of exacerbations and remissions of their symptoms, pain increased on sitting, a catch on extension from the flexed position, a history of instability, and tenderness often localized to the painful segment. Though there are many theories for the cause of axial back pain, clinical and experimental data suggest pain is caused by axial loading of an inflamed disc. This may explain the pain on axial loading and the intermittent flare-ups and remissions. This may also explain the success of interbody stabilization in relieving this pain. Lotz and associates5 studied pathologic disc degeneration via animal studies and suggested a similar theory. The goal of successful surgery for axial low back pain is to address the biomechanics of the painful segment by removing the pain generator (the disc) and restoring axial loading across the disc space. Anterior lumbar arthroplasty, anterior lumbar interbody fusion, and PLIF/TLIF with cages all remove the painful disc and restore loading across the disc. The PLIF/TLIF procedure in addition removes the painful facets and decompresses the neural structures. The goals of posterior lumbar arthroplasty are outlined in Table 94–1. The question arises whether we need to replace just the disc posteriorly or both the disc and facet. Replacing the disc alone would need at least 50% of the facets to be preserved, thus requiring significant dural retraction to insert the discs. There would be a concern that the discs may back out without a posterior tension band. If one were going to replace the disc alone, there would be a good
TABLE 94–1. 1. 2. 3. 4. 5. 6. 7.
Goals of Posterior Lumbar Arthroplasty
Remove painful disc Restore normal loading across the disc Decompress neural structures Remove painful facets if needed Restore motion to the painful disc Allow segment to assume a physiologic position in sitting and standing Match instantaneous axis of rotation of disc and posterior dynamic stabilizing device 8. Preserve normal loading of adjacent discs
CHAPTER 94
argument to do this anteriorly. Adding a posterior dynamic stabilizer allows a generous removal of as much facet as required, a better decompression of the neural structures, and avoidance of the problem of overdistraction of the facet joint capsule (if the facet is removed). My opinion is that the disc and posterior dynamic stabilizer (PDS) should be used in conjunction to start with, and as clinical experience grows, some surgeons may try the posterior disc on its own. If the PDS and the disc are used together, then for any movement to occur, their centers of rotation must be matched (Fig. 94–1). In Figure 94–1 the arc of the posterior dynamic stabilizer (Truedyne) and the posterior prosthetic disc (Truedisc) are matched. Asynchronous movement will result in strains on the adjacent segment. Matching the centers of rotation is not an easy task as the IAR keeps changing during motion and is not in a predictable place in degenerative discs in any case. Rousseau and associates6 highlighted this in their paper on the IAR at the L5-S1 disc. The PDS device will need to provide shear stability throughout its range of movement. The typical lumbar segment moves through an arc in flexion and extension, and the term angular motion is one way of describing it. The disc/PDS combination also needs to restore angular motion to the segment. This is different from devices that allow only rotation around a ball and socket or translation. How much motion needs to be restored to the operated lumbar segment? One has to remember that a diseased disc has less motion than a normal disc in most cases, except in the instability phase of degeneration. The aim is not to restore motion to normal but only to the extent that the disc above or below does not get excessive load transmitted to it or move in an asynchronous manner. Flexion-extension is the main movement to restore, with some lateral bending and rotation. How much motion is then ideal? Finite element analysis (FEA) is probably the best way to address this question.
Posterior Lumbar Arthroplasty
737
n
F I G U R E 94–2. FEM pictures of a posterior lumbar disc prosthesis (TrueDisc) inserted in a parallel and in a nonparallel manner.
The loading on the end plate will be an important consideration in PLA. The contact surface area of a paired disc, placed on either side of the dura, is less than that of an anteriorly placed disc, and so point loading through the end plates will be higher. Careful placement of the discs laterally under the thickest part of the end plate will be important to avoid subsidence. If placed laterally, the discs will probably angle toward the midline. Getting a paired posteriorly placed disc that is not parallel to allow the segment to move synchronously will be challenging, and a simple ball-and-socket design may not work in this situation. Figure 94–2 illustrates discs placed parallel and nonparallel in the disc space. Figure 94–3 illustrates the movement of the disc prosthesis developed by Disc Motion Technologies, when placed in a parallel and in a nonparallel manner, as seen in Figure 94–2. Figure 94–4 illustrates a posterior disc replacement system from Disc Motion Technologies, comprising a paired posteriorly inserted disc combined with a paired posterior dynamic stabilizer. The technique of insertion is similar to a bilateral TLIF. The posterior dynamic stabilizer can be used on its own as well, and the indications for this are discussed later. Matching the IAR of the motion segment during movement is important even when it is used on its own; otherwise, this will probably result in asynchronous movement in the disc above. The need to minimize wear is an important factor in material selection for the disc. Cobalt-chrome has a history of being used successfully in lumbar discs (e.g., Maverick disc [Medtronic Sofamor Danek, Memphis, TN]). It is not, however, MRI (magnetic RELATIVE MOTION AT L4-L5 UNDER 10.6 Nm (comparison of approaches) 7 6 5 4 3 2 1 0 Extension
Flexion Intact Parallel 20 deg
n n
F I G U R E 94–1. This model demonstrates the importance of matching the IAR (instantaneous axis of rotation) of the posterior disc and a posterior dynamic stabilizer.
F I G U R E 94–3. Graph illustrating the relative motion of the L4L5 motion segment before and after the insertion of a posterior lumbar disc and posterior dynamic stabilizer, with the discs inserted parallel and nonparallel. It shows that the Truedisc allows similar motion whether placed parallel or nonparallel.
738
P A R T
X I I
Controversies
X
X
X
X
n
F I G U R E 94–4. The Disc Motion Technologies Total Spinal Motion Segment (TSMS) replacement system comprising the paired Truedisc posterior lumbar disc and the paired Truedyne posterior dynamic stabilizer (PDS). The lateral connectors ensure a parallel placement of the Truedyne PDS devices and also house a ball-andsocket connector allowing for fine tuning of changes in the instantaneous axis of rotation during motion.
resonance imaging) compatible, which is a disadvantage. New materials which will have good wear characteristics and be MRI compatible are being tested (e.g., titanium-ceramics), but it will take time before they can be fully studied and introduced into clinical use. As far as the PDS device is concerned, a similar rationale applies. Cobalt-chrome will be preferable for its excellent wear characteristics if there are components that are subject to wear during motion (e.g., the Truedyne from DMT). OPERATIVE TECHNIQUE Posterior lumbar arthroplasty (PLA) is very similar to the bilateral TLIF approach, but there are some special considerations that need to be kept in mind. These, along with other general advice, are discussed in this section. The use of a good xenon 300-watt headlamp and loupes (2.5 magnification) adds to the safety and speed of this procedure and is routine in many practices. A bloodless field is desirable for this surgery. Careful patient positioning to avoid any pressure on the belly reduces venous pressure and bleeding. Hypotensive anesthesia is useful at reducing intraoperative venous bleeding. A skilled anesthetist can have a significant impact on blood loss. Understanding the anatomy of the epidural venous plexus is important so the veins can be cauterized before the bleeding starts. Bipolar cautery is used. There are usually large veins going around each pedicle and also coursing along the nerve roots and fanning across the disc space laterally. Gelfoam and Flowseal used intraoperatively help to stop any bleeding that cannot be controlled with a cautery. With these techniques, bleeding can be reduced to about 150 mL for a one-level procedure. The surgical approach is illustrated in Figure 94–5. A midline approach, sparing the interspinous ligament, is used. A 6-cm incision is usually enough for a one-level procedure. Exposure is only
The xs mark the pedicle screw insertion points. Do not expose TP’s. n F I G U R E 94–5. Features of the surgical technique of posterior lumbar arthroplasty. The xs mark the pedicle screw insertion points. Do not expose the transverse processes.
necessary to the edge of the facet joint. Exposing the transverse process is not needed. If the patent is obese, the pedicle screws are inserted through stab incisions lateral to the midline incision in order to get the proper medial angulation of the screws without having to enlarge the midline incision and overretract the soft tissues. Correct medial angulation is especially important at S1 where the L5 nerve root is at risk as it crosses over the ala of the sacrum. This does not require elaborate percutaneous instrumentation and can be done under direct vision. Correct pedicle screw placement is crucial to PLA. We advocate the use of an image intensifier to check that the screws are placed in the center of the pedicle. An off-center placement of the screw can result in early loosening or screw breakout. A visual and mechanical check of the medial pedicle wall after screw placement is important, as the pedicle probe is not foolproof. Some surgeons like to insert the screws after doing the decompression, so the pedicle can be visualized during insertion. A bilateral TLIF type approach allows access to the disc between the traversing and exiting nerve roots and avoids the need to retract the dura. The quickest way is to ostetomize the inferior facet first, the horizontal cut being at the level of the pars. The second cut is through the superior facet, just above the pedicle. With a Kerrison rongeur, bone is removed up to the pedicle above and below the disc, and the traversing and exiting nerve roots are deroofed. Care is taken to decompress the exiting nerve root above the disc. This also allows a manual inspection of the medial pedicle wall. Before cauterizing the epidural veins, check for conjoined nerve roots, which occur in 4% of patients. The disc is then incised. This should be done bilaterally before sequential spreaders are used. It is very important to cut the lateralmost part of the disc, as it can form a lateral tether preventing disc distraction and, worse, destruction of the end plate with the use of aggressive shavers. This is illustrated in Figure 94–6.
CHAPTER 94
X
X
Posterior Lumbar Arthroplasty
739
traditional PLIF approach, but the end plate strength is maximal laterally, rather than medially. Along with the disc, a posterior dynamic stabilizer (PDS) will be necessary to control motion of the discs and provide shear stability. The placement of this is important. The pedicle screws are placed in the usual manner. The PDS device needs to be lined up so its IAR will match that of the disc. CHALLENGES
Osteotomize lateral annulus to release lateral tether. n F I G U R E 94–6. When incising a disc, it is important to cut the lateralmost annulus to release a possible lateral tether.
End plate damage can lead to subsidence of the implant. The lateralmost disc is best cut using an osteotome, just above the pedicle below, taking care to protect the exiting nerve root. Shavers should not be used till the disc is opened to the required height, and even then they are used sparingly and carefully. A down-biting curette is also useful to remove the disc material. It is important to remove the lateralmost part of the annulus. One test of this is a clear view of the empty disc space, lateral to the traversing nerve root, without any retraction. A common problem is not enough room for the disc. To get around this use a disc trial. This creates the space the disc will occupy, and if it is not going in easily, one may need to remove more of the annulus from under the lateral edge of the dura. One should remember that the anterior annulus may have been torn during distraction, and so there is a risk to the anterior major vessels and also a chance that the disc can be inserted too far anteriorly and out into the retroperitoneal space. If a conjoined nerve root is found on one side, then the disc replacement is not possible, and one needs to convert this procedure to a fusion. The Disc Motion TrueDisc has a unique rail system, which makes the insertion of the spreaders and disc easy. The sizing of the disc is different from the way one sizes a cage. For the disc to move, we need to avoid overdistraction. If one is able to open the disc space to, say, 11 mm using spreaders, then a 10-mm disc prosthesis should be used, whereas an 11-mm fusion cage would probably be appropriate if one were doing a fusion. One must try to place the discs as parallel as possible and as laterally as possible. The discs must also be placed at the same depth from the posterior vertebral wall. Always make more room for the nerves by decompressing them, as they do swell after surgery due to handling, and relative stenosis may cause leg pain. Surgeons may prefer to take the midline out, only do a partial facetectomy, and place the discs more medially, as in the
Posterior lumbar arthroplasty is a new concept and there are a few disc systems in development. Clinical studies will begin in 2008. There are some challenges we can foresee and others that will emerge only after clinical use. Just as ALIF evolved into anterior lumbar arthroplasty, it is anticipated that the PLIF/TLIF procedure will evolve into posterior lumbar arthroplasty. The surgical approaches and techniques are familiar and well established for PLIF/TLIF procedures and the same will be applied to PLA. Matching the IAR of the disc and the posterior dynamic stabilizer will be a challenge for many reasons. Our understanding of the IAR in degenerative discs is incomplete, and there are variations between different degenerate segments. The IAR also changes in motion, and the system will need to adapt to that. Preventing subsidence through the end plate will be important. Strategies that may help here will include placing the discs laterally under the strongest part of the end plate, using the widest discs that can be safely inserted, and avoiding surgery in patients with osteoporosis. Disc retropulsion into the spinal canal is a potential problem. Features will need to be built into the disc to resist this. The PDS will help reduce this risk by controlling the movement of the disc and avoiding hyperflexion posteriorly. Putting the implants in with the disc slightly distracted to create appropriate tension will also help to avoid retropulsion. How much tension one needs to create is difficult to quantify at this stage. Too much distraction will result in a loss of movement, so the amount of distraction needs careful calibration, but there are no clear guidelines at this stage. The width of the discs will need to be balanced between the need to provide maximum end plate coverage and the ability to insert the disc past the neural structures without undue retraction. If paired discs are used, placing them parallel will be important. If placed laterally to take advantage of the strong end plate, they will invariably end up being angled medially. In this case, getting the discs to move, even when placed in a nonparallel manner, will require innovations in disc design. If the PDS system that is used in conjunction with the disc is anchored to the pedicle, then pedicle screw loosening will be an issue. This is because there will be torsion forces transmitted to the screw during motion. Mechanisms in the screw will need to be designed to counter this. How much movement in the motion segment should we aim for in PLA? Too little movement will cause additional stresses on the adjacent level. Too much movement will result in possibly more pain from the neural structures, loosening of the components, and wear issues. These questions need further study.
740
P A R T
X I I
Controversies
DISCUSSION
TABLE 94–3.
Posterior lumbar arthroplasty represents a natural evolution of lumbar arthroplasty, after anterior lumbar arthroplasty has become established as a clinical procedure. It aims to deal with the obvious clinical problems we have encountered with anterior arthroplasty. The posterior disc replacement, along with a dynamic stabilizer, aims to restore kinematics of the spine to near normal. There are many clinical questions being raised as total lumbar disc replacement continues to emerge in clinical practice. Many variables, such as effects on adjacent level degeneration, will take years to define. Finite element analysis allows for evaluation of motion and load changes which can be anticipated and help predict potential effects of clinical scenarios. The posterior lumbar discs need to function even when placed in a nonparallel manner. This simulates their insertion from a far lateral TLIF type approach, where the midline is spared and the facet removed to place the discs. This approach has additional advantages in that the neural structures need minimal retraction and there is a thorough decompression of the exiting nerve roots. Placing discs exactly parallel to each other is very difficult, as MRI scans of postoperative patients with implanted PLIF cages have shown, in clinical practice. Posterior lumbar arthroplasty is a true TDR and has several possible advantages to anterior lumbar arthroplasty. These are summarized in Table 94–2. The finite element analysis of the posterior disc and PDS system from Disc Motion Technologies (DMT, Inc., Boca Raton FL) shows that it is possible to design a system whereby the disc and PDS are matched as far as their center of rotation is concerned and that this results in near-normal motion at the implanted level, with no additional movement at the adjacent level. This was explained in detail by Goel and associates in their paper.7 It has also been corroborated by cadaver kinematic studies. Restoring angular motion with shear stability is one of the goals of PLA. Two separate meta-analyses of the literature8,9 suggest that PLIF gives better clinical results than ALIF. This may be because the PLIF deals with more pain generators. It is postulated these results may be similar for posterior lumbar arthroplasty, as compared to anterior arthroplasty, but clinical studies will be needed to confirm this. The term posterior lumbar arthroplasty could be extended to include the dynamic stabilization devices such as Isobar (Scient'x), TABLE 94–2. Arthroplasty
Comparison of Posterior and Anterior Lumbar
Posterior Lumbar Arthroplasty
Anterior Lumbar Arthroplasty
Deals with all three pain generators—disc, nerve, and facet joint Can be easily revised via an anterior lumbar interbody fusion Approach familiar to all surgeons Fewer contraindications than anterior arthroplasty Can be done even with facet degeneration and neural impingement
Deals only with disc Revision is difficult and hazardous Often needs a separate approach surgeon Applicable to only approximately 5% of patients Facet degeneration and neural impingement a contraindication; postoperative facet pain a possibility
1. 2. 3. 4. 5. 6.
Uses of Posterior Dynamic Stabilizers
For micromovement, which enhances a posterolateral fusion To stabilize, after decompression in spinal stenosis To protect a segment above a fusion In conjunction with a posterior disc replacement For the treatment of low back pain Stabilization of degenerative scoliosis.
Agile and Dynesys. The aim of these devices is to provide stability, and the slight movement they offer results in the device absorbing the loads and not breaking. This will avoid the need for posterolateral fusion. Other devices in this group (e.g., the Truedyne) allow the desired range of movement to be dialed into the device from 0 to 8 mm. These devices have multiple uses, as outlined in Table 94–3. These are the clinical scenarios in which surgeons are either using these devices already or contemplating their use. The clinical evidence for their use in these situations is being collected but has not yet been published. Each indication will require a different range of movement in the device, with minimal movement for the fusion indication (0–2 mm), 2 to 3 mm after decompression for spinal stenosis, 3 to 5 mm to protect a segment adjacent to a fusion, and 6 to 8 mm when used in conjunction with a posterior disc replacement. These figures are a clinical estimate at this stage, and further research is needed to quantify these more accurately. The aim should be not only to stabilize the segment without the need for a fusion but to allow enough movement in the device to avoid excessive adjacent segment movement and subsequent accelerated degeneration at that level. Allowing motion while preserving shear stability and matching the IAR of the motion segment is the challenge in designing these devices. The ideal surgical candidate for this procedure will be similar to that for a PLIF/TLIF procedure. To start with, however, it is recommended that the PLA be avoided in cases with a complete collapse of the disc with osteophytes. This is because once the disc height is restored in this group, movement in the segment will be limited. A potential problem is subsidence of the discs, as the end plate contact area is less than for anteriorly placed discs. This will be minimized with the lateral placement of the discs, where the end plate is strongest. The widest discs that can be accommodated will need to be placed in the discs. Other potential problems are similar to those seen with the PLIF/TLIF procedure, including neural irritation. CONCLUSION Posterior lumbar arthroplasty represents the next generation of lumbar arthroplasty devices, with several advantages over anteriorly placed discs. These advantages include a familiar approach, ease of revision, few contraindications, dealing with neural compression and facet degeneration in addition to disc degeneration, and applicability in a larger group of patients. It is important that the posterior disc be matched with a posterior dynamic stabilizer and that they both have a matched center of rotation.
CHAPTER 94
Posterior dynamic stabilizers are another group of posterior arthroplasty devices which can be used on their own or in conjunction with a posteriorly placed paired lumbar disc. They are used on their own in posterolateral fusion to stabilize a segment after a spinal fusion and to protect a segment above a fused level. Allowing the motion needed for the indication it is used for, preserving shear stability, and matching the IAR of the motion segment are desirable qualities in a PDS device.
REFERENCES 1. Fujiwara A, Tamai K, An HS, et al: The relationship between disc degeneration, facet joint osteoarthritis, and stability of the degenerative lumbar spine. J Spinal Disord 13(5):444–450, 2000. 2. Gries NC, Berlemann U, Moore RJ, Vernon-Roberts B: Early histologic changes in lower lumbar discs and facet joints and their correlation. Eur Spine J 9(1):23–29, 2000.
Posterior Lumbar Arthroplasty
741
3. Butler D, Trafimow JH, Andersson GB, et al: Discs degenerate before facets. Spine 15(2):111–113, 1990. 4. Tanaka N, An HS, Lim TH, et al: The relationship between disc degeneration and flexibility of the lumbar spine. Spine J 1(1):47–56, 2001. 5. Lotz JC, Ulrich JA: Innervation, inflammation, and hypermobility may characterize pathologic disc degeneration: Review of animal model data. J Bone Joint Surg Am 88(2):76–82, 2006 (review). 6. Rousseau MA, Bradford DS, Hadi TM, et al: The instant axis of rotation influences facet forces at L5/S1 during flexion/extension and lateral bending. Eur Spine J 15(3):299–307, 2006 (epub Sept 20, 2005). 7. Goel V, Kiapour A, Faizan A, et al: Finite element study of matched paired posterior disc implant and dynamic stabilizer (360 motion preservation system). SAS J 1:55–62, 2007. 8. Boos N, Webb JK: Pedicle screw fixation in spinal disorders: A European view. Eur Spine J 6(1):2–18, 1997 (review). 9. Turner JA, Herron L, Deyo RA: Meta-analysis of the results of lumbar spine fusion. Acta Orthop Scand Suppl 251:120–122, 1993.
CHAPTER
95
Cervical Arthroplasty with Myelopathy Jonathon R. Ball and Lali H.S. Sekhon
K E Y l l l
l
l
P O I N T S
The natural history of cervical spondylotic myelopathy is unclear. Surgery for cervical spondylotic myelopathy is of unproven benefit. Cervical arthroplasty is an alternative to fusion after anterior cervical decompression. Cervical arthroplasty may be beneficial in selected patients with cervical spondylotic myelopathy as long as careful patient selection is undertaken. Patient selection should be guided by an understanding of the pathophysiology of cervical spondylotic myelopathy and the benefits and limitations of cervical arthroplasty.
radiculopathy and myelopathy and highlighted the difference between acute disc protrusions and chronic degenerative changes in the cervical spine.2 The symptoms and signs exhibited by patients are variable and were grouped by Crandall and Batzdorf into five categories that covered the spectrum of spinal cord syndromes. It has been reported that gait abnormalities are among the earliest noticeable changes. More recent electrophysiologic studies have found that abnormalities of motor and somatosensory evoked potentials precede the development of clinically noticeable signs.3 Pathophysiology
Surgery for cervical myelopathy has been performed for over 50 years. The natural history of myelopathy is still unclear and the effects of our interventions sometimes unclear. Significant variations in surgical approaches to cervical disease, including cervical myelopathy, have been demonstrated in surveys of spine surgeons.1 The strategies to manage cervical myelopathy are manifold and include nonoperative management and surgical treatment using anterior and posterior approaches. Cervical arthroplasty introduces a new dimension into the management armamentarium, promising retained motion and decreased adjacent segment degeneration. The application of cervical arthroplasty to the treatment of cervical myelopathy is relatively new, and the long-term consequences are not entirely clear. This chapter will review the clinical and pathologic features of cervical myelopathy as well as more recent management strategies and finally the role of cervical arthroplasty in the surgical management of myelopathy. CERVICAL MYELOPATHY Presentation Cervical myelopathy refers to a symptom complex caused by compromise of the cervical spinal cord. Cervical spondylosis that leads to canal stenosis and cord compression is the most commonly encountered cause. The first clear delineation of this syndrome from other cervical diseases is attributed to Brain and associates in 1952 who stressed the distinction between the presentations of 742
The pathophysiology of cervical spondylotic myelopathy (CSM) remains incompletely understood. Current thinking is that static compressive elements cause dynamic effects with cervical motion. The forces developed in the cord substance together with potential vascular compromise combine to cause neural injury. Subcellular mechanisms of neural injury, similar to those elucidated in acute spinal cord injury, continue to be investigated. Presumably there is a point that is reached in terms of degree and severity of compression where cell death occurs, affecting both neurons and supporting elements. Consequent gliosis and scarring may contribute to a poor outcome after decompression. Cervical spondylosis is accompanied by structural changes that can lead to canal stenosis. A congenitally narrow cervical spinal canal is a predisposing factor for the development of CSM. The degree of stenosis has been quantified in a number of ways by different authors. The sagittal developmental diameter of the spinal canal is measured at the level of the midvertebral body to avoid inaccuracy by disease at the level of the disc space. In normal subjects, the diameter at levels from C3 and C7 is around 17 to 19 mm.4–6 Decreased developmental canal diameters are observed in patients with CSM. Some authors have proposed the use of ratios to evaluate spinal canal stenosis to account for variations that may occur with race, gender, or body size and radiographic distortion. The Torg/Pavlov ratio describes the relationship of the sagittal diameter of the spinal canal to the sagittal diameter of the vertebral body. Lower ratios have been found in patients with CSM.7 The degenerative process of cervical spondylosis produces anatomic changes that decrease the canal volume from all directions.
CHAPTER 95
Anterior compression from disc space disease rises as a result of disc herniation, osteophyte formation, and PLL (posterior longitudinal ligament) calcification while uncovertebral osteophytes can lead to anterolateral compromise. Synovial hypertrophy or osteophyte formation affecting the facet joints may cause lateral narrowing, and the ligamentum flavum can encroach on the canal posteriorly. Static compressive elements do not act alone in the pathophysiology of CSM. Dynamic changes that occur with cervical motion are believed to be an important contributor. In flexion, the cord is “draped” over anterior compressive elements and the cord is stretched. In extension, cord shortening leads to thickening of the spinal cord that combines with buckling of the ligamentum flavum to exacerbate posterior compression. Dynamic changes may exacerbate the vascular compromise that has been proposed to contribute to the development of CSM. Putative mechanisms include radicular feeder compromise with foraminal stenosis, anterior spinal artery compression in flexion, elongation of transversely directed terminal arteries, and compromise of venous drainage. Aside from a vascular cause, mounting evidence from experimental models suggests important biomechanical factors in CSM pathophysiology.8 Spondylotic changes deform and tether the cord. With movement, stretching of the cord causes increased intramedullary strain and shear forces that injure axons. Further subcellular mechanisms of neural injury have been proposed to include apoptosis, excitotoxicity, free radical formation, and cationic mediated injury.9,10 Natural History and Conservative Management The true natural history of untreated cervical myelopathy is not well established because all series are subject to bias from retrospective data collection, selection of patients for treatment, the use of nonoperative treatments, and the failure to use a validated clinical scale. These limitations mean that series purporting to describe the natural history may be able to describe the patterns of progression but will fail to estimate the true rates of progression in a patient population. The 1956 paper by Clarke and Robinson, often quoted as describing natural history, was based on the prodromal history of 120 patients before treatment and included a prospective cohort of 26 selected patients who received no treatment.11 They described three patterns of disease history—episodic deterioration, slow progressive deterioration, and prolonged stability/eventual deterioration—leading to the conclusion that the ultimate prognosis in most cases was poor. Lees and Turner in 1963 reported on 44 patients with myelopathy of whom 22 were followed for more than 10 years.12 The initial development of symptoms was often followed by a period of clinical stability or improvement. In those with longer follow-up, exacerbations of disease occurred with intervening periods of nonprogressive disability. A pattern of steady deterioration was rarely seen. The authors called for “reflection about the question of operation.” A 1966 paper by Roberts described spondylotic cervical myelopathy as a benign condition noting motor improvement in about a third of patients. However, a similar proportion deteriorated in the same time frame. Phillips rallied against this prevailing notion of cervical spondylotic myelopathy as a benign condition in his 1973 paper.13
Cervical Arthroplasty with Myelopathy
743
He criticized the methodology of earlier papers and, based on his good experience with anterior cervical decompression, recommended early surgery for the best results. Such advocacy of early surgery was often based on a theory that surgical delay may lead to permanent neurologic injury. Opponents claimed that surgical successes reflected patients who would have improved or stabilized without surgery if given enough time. Surgical Management Surgery has been proposed for CSM since its earliest definition. Generally speaking, surgery is utilized for those with severe or progressive symptoms.14 Numerous surgical techniques have been developed since, each purporting unique benefits. There is, however, little consensus between surgeons about the best approach. There is no class I evidence to support the superiority of surgical management.15 The only randomized controlled trial, reported by Kadanka and associates in 2000, randomized 48 patients with mild-tomoderate CSM (mJOA > 12) to conservative therapy or surgical management.16 The surgeries predominantly employed anterior approaches. At 2 years, no substantial difference in outcome was seen between groups.16 An analysis of an expanded series with longer follow-up attempted to identify predictors of good outcome through univariate analysis.17,18 Older age, wider canals, and normal central motor conduction time were predictive of response to conservative therapy. The degree of canal stenosis and worse preoperative function were more prevalent in surgical responders. On behalf of the Cervical Spine Research Society, Sampath published the results of a prospective, non-randomized, multicenter study of patients undergoing medical or surgical management for CSM.19 The treatment allocation was at the discretion of the participating surgeon. Despite having a poorer clinical status at baseline, surgical patients had improved outcome at follow-up. Surgical Approaches Surgical options in CSM involve both anterior and posterior approaches and have been developed to address both the static and dynamic causative components. Posterior Approaches
The initial description of spondylotic cervical myelopathy by Brain and associates2 included a description of a surgical series by Northfield. At that time, the operation performed was a posterior decompression by laminectomy with section of the dentate ligaments. This approach was based on the rationale that posterior decompression alone would not always allow sufficient relief of cord compression from anterior disease. Section of the dentate ligaments allowed further displacement of the cord away from compressive elements. Although dentate ligament section is no longer routine, it is now widely recognized that posterior decompression alone is not sufficient in all situations, particularly in the presence of significant anterior disease or kyphosis that drapes the cord over anterior compressive elements. Of a series of 100 patients with cervical spondylotic myelopathy reported by Ebersold in 1995, 51 underwent a laminectomy.20 Improvement was noted in about 70% of patients at early follow-up; however, 10%
744
P A R T
X I I
Controversies
of patients suffered a delayed deterioration. Several causes of deterioration after laminectomy have been proposed and include spinal instability, accelerated spondylotic changes, extradural scarring, and tethering. The incidence of postlaminectomy kyphosis was reported to be 21% in a study by Kaptain and associates and noted to be higher in patients with a preoperative loss of cervical lordosis.21 For this reason, it has been proposed that posterior decompression be accompanied by posterior fixation and fusion to prevent the development of deformity. The use of fusion techniques can also allow a wider decompression to be performed without chancing instability. This approach, involving wide laminectomy and lateral mass fixation, has been reported by Sekhon for patients with circumferential stenosis. In a report of this technique in 50 patients, he reported radiographic evidence of adequate decompression in all patients and a decrease in Nurick grade in the majority of patients. Progression of deformity was seen in only two patients.22 Another form of posterior decompression, laminoplasty, was pioneered by several Japanese groups. Various methods have been described but most involve detaching the laminae and repositioning them in an orientation that increases the diameter of the canal. These approaches purport to maintain spinal motion by preserving facet joints, prevent kyphosis through preservation of the posterior elements, and protect the dura from tethering to muscle. Kawaguchi and associates reported an impressive case series of lamininoplasty in 133 patients.23 Of the 84 patients who were alive and available for follow-up after a minimum of 10 years, 35 had their initial surgery for CSM. Improvement in clinical status, measured by the Japanese Orthopaedic Association (JOA) myelopathy score, was noted in 55.1% of subjects at last follow-up. The average preoperative score of 9.1 improved to 13.7 at 1 year and was maintained at 13.4 at last follow-up. A North American group has reported the results of cervical laminoplasty in 204 patients with CSM at an average follow-up period of 16 months.24 They reported a decrease in Nurick grade in 62% of patients. Radiographic progression of kyphosis was seen in two patients. Other variations on the posterior approach include a dorsolateral operation that involves the exposure and removal of one hemilamina and skip laminectomy where decompression is performed at discontinuous levels with preservation of muscle attachment at other levels.25,26 These procedures both aim to limit the disruption of spinal extensor musculature and decrease the risk of deformity. However, they are limited to CSM related to predominant posterior compression. Anterior Approaches
As noted earlier, it is widely accepted that the use of posterior decompressive techniques is not appropriate in all situations of CSM. This is particularly so when significant ventral disease exists or kyphotic deformity allows the spinal cord to be draped over anterior compressive elements. In fact, most surgeons embrace anterior techniques ahead of any posterior procedure, particularly for disease affecting fewer than three levels. Methods of anterior cervical spine surgery were initially described by Bailey and Badgley, Smith and Robinson, and Cloward.27–29 Historically, anterior approaches in the cervical spine have included anterior cervical
discectomy alone (ACD), anterior cervical discectomy with fusion (ACDF), and ACDF with anterior plating. These can be performed at single or multiple levels. For more extensive disease, including disease behind the vertebral bodies inaccessible through the disc space, cervical corpectomy and fusion can be performed. The substrate for fusion has included autograft, often from the iliac crest, human or animal allograft, or more recently bone graft substitutes and osteogenic factors. Intervertebral cages have been developed for graft containment and structural support, and the proliferation of different design in anterior cervical plates has necessitated the development of a classification scheme. Emery and associates reported the results of anterior cervical surgery in 108 patients with CSM at a minimum follow-up of 2 years.30 Operative treatment consisted of anterior discectomy, partial corpectomy, or subtotal corpectomy at one level or more, followed by placement of autogenous bone graft from the iliac crest or the fibula; 92% of subjects with motor weakness, 89% of subjects with sensory deficit, and 86% of those with gait abnormality showed some improvement postoperatively. One patient had marked neurologic deterioration after surgery and myelopathy recurred in five patients. CERVICAL ARTHROPLASTY Clear guidelines for the use of cervical arthroplasty surgery are lacking. The earliest clinical reports on cervical arthroplasty devices were dominated by patients with spondylotic myelopathy,31 but modern devices are recommended cautiously in this setting by manufacturers. Without consensus on best current management, the benefit of a new technology is difficult to assess and its introduction carries the risk of inappropriate use. However, a firm understanding of the pathophysiology of the disease, as already outlined, married with an appreciation of the principles of the technology, allows the formulation of a framework to guide the appropriate use of cervical arthroplasty in patients with spondylotic cervical myelopathy. Theoretical Considerations As with all surgical therapies for CSM, decompression of neural elements is the primary aim in cervical arthroplasty for myelopathy. Compressive elements can be anterior, posterior, or lateral and may occur in the context of a congenitally narrow canal. Significant compression by ventral disease accessible through the disc space approach is most suitable. The presence of ventral disease that extends behind the vertebral body and is inaccessible through the disc space may not be appropriate. Dorsal compressive elements cannot be removed through an anterior approach (Fig. 95–1). The constricting nature of a congenitally narrow canal is not addressed by anterior discectomy. The critical nature of adequate decompression must be emphasized. Given the role of dynamic effects in CSM, the presence of any residual compressive elements will be unacceptable. Such elements may be allowed in a fusion situation, when movement that causes dynamic compression is eliminated. If motion is preserved, such elements may impinge on the cord with movement and cause further neural injury. The unconstrained nature of many cervical disc prostheses limits their use in the presence of cervical
CHAPTER 95
Cervical Arthroplasty with Myelopathy
745
n
F I G U R E 95–1. Typical patient with cervical stenosis and potential outcomes from anterior fusion surgery and anterior cervical decompression and arthroplasty. In the fusion situation, over time, degenerative changes may develop at the adjacent level, whereas in the arthroplasty scenario possible posterior ligamentous thickening may worsen over time and cause recurrent stenosis. A, A hypothetical situation with predominantly ventral cord compression in a patient with some posterior ligamentous thickening as well. The two treatment algorithms are shown in (B) and (D) with either an arthroplasty or a fusion being performed. The hypothetical future situation is shown below this. In (C) the fusion is solid, no recurrent stenosis has occurred but there is some adjacent degeneration at the level above. In part (E) the arthroplasty is moving and unchanged. The adjacent level is intact. Note, however, that the posterior ligaments at the operated levels have thickened because of retained motion and recurrent stenosis is occurring. This scenario could also potentially cause foraminal stenosis with facet osteophytosis occurring over time.
A
B
D
C
E
deformity and instability, although correction of deformity may be an option with newer devices. Figure 95–2 illustrates a situation inappropriate for arthroplasty. A 44-year-old man with myelopathy had a congenitally narrow canal with significant dorsal and some ventral compression at multiple levels. Loss of lordosis is seen in the upper cervical spine. Arthroplasty would have addressed the ventral compression, but the combination of residual dorsal compression and a narrow canal may continue to cause compression, particularly in the presence of preserved motion. A posterior decompression and lateral mass fusion was performed from C3 to C6.
Figure 95–3 illustrates another situation in which arthroplasty may be unsuitable. This 76-year-old woman presented with myelopathy due to disc-osteophyte compression predominantly at C5C6 and C6-C7. This was associated with a kyphotic deformity that draped the spinal cord over the compressive elements. Adequate decompression may have been difficult through the disc space, and the unconstrained nature of many disc prostheses would not enable the restoration of cervical lordosis. Anterior decompression with multilevel corpectomy and local autografting into a PEEK (polyetheretherketone) cage was performed with anterior plating.
746
P A R T
n
X I I
Controversies
F I G U R E 95–2. This 44-year-old man presented with clinical cervical myelopathy, a congenitally narrow canal, and anterior and posterior compressive disease (A). In this scenario wide posterior laminectomy and fusion (B) lead to a successful decompression of the spinal cord (C).
n
F I G U R E 95–3. This 76-year-old woman presented with a progressive myelopathy, a focal radiologic kyphosis (A), and evidence of cord compression on magnetic resonance imaging (B). Three-level anterior corpectomy was effected with reduction of the kyphosis using a PEEK (polyetheretherketone) cage, local autograft, and plating (C). Subsequent posterior fixation was also performed. Again, this is not a case suitable to current arthroplasty technologies.
748
P A R T
X I I
Controversies
Figure 95–4 illustrates a situation in which arthroplasty was used. This 45-year-old woman presented with a clinical myelopathy due to anterior disc compression at C4-C5 and C5-C6. The canal was relatively capacious, and there was no significant dorsal compression. The degree of loss of lordosis was considered acceptable. Cervical disc arthroplasty was performed at two levels with good results. Clinical Studies It has been proposed that the ideal patient for insertion of an artificial disc prosthesis has a soft disc herniation causing neurologic symptoms or signs, motion at the involved segment with no evidence of instability or hypermobility, and an absence of osteoporosis or infection. Normal sagittal alignment with the absence of focal or global kyphosis is desirable. To date, the published clinical series of cervical disc arthroplasty has predominantly included patients with radiculopathy and myelopathy, secondary to acute disc herniations or degenerative spondylotic change. The initial series describing the Cummins/Bristol disc (produced in the hospital workshop at Frenchay Hospital, Bristol) included 16 patients
n
with myelopathy, 3 with radiculopathy, and 1 with severe neck pain.31 In reporting the initial use of the Bryan disc (Medtronic, Memphis, TN), Goffin described results in 53 patients with radiculopathy and 7 with myelopathy.32 Goffin and associates reported an expanded series of 146 patients including 43 two-level cases.33 They further reported 6-, 12-, and 24-month success rates of 90% (83/92), 86% (76/89), and 90% (44/49) for single-level surgery. Duggal and associates examined outcomes in 26 patients implanted with Bryan discs who had myelopathy or radiculopathy, most of which were favorable.34 Pimenta and associates reported their experience with the PCM (Cervitech, Inc., Rockaway, NJ) prosthesis in 40 patients with radiculopathy and 13 with myelopathy.35 This series yielded good or excellent outcomes greater than 90% at 3, 6, and 12 months. A study analyzing the use of ProDiscC (Synthes, West Chester PA), reported results in 16 patients with “symptomatic cervical spondylosis.”33 It included 4 patients with severe axial neck pain and 12 with established radiculopathies or myelopathies. Significant postoperative decreases in neck and arm pain intensity and frequency were noted. A similar decrease was recorded in the Oswestry Disability Index.
F I G U R E 95–4. This 45-year-old woman presented with bilateral arm weakness and a clinical picture of cervical spondylotic myelopathy. The preoperative magnetic resonance (MR) scan showed evidence of predominantly C4-C5 cord compression with a smaller disc at C5-C6 (A). The patient underwent uneventful two-level cervical arthroplasty at these levels with good decompression of the spinal cord on postoperative MR scanning (B).
CHAPTER 95
TABLE 95–1. Patient
Cervical Arthroplasty with Myelopathy
749
Results of Cervical Arthroplasty in 11 Patients with Spondylotic Myelopathy Showing Good Initial Outcomes
Preop Nurick
Postop Nurick
Levels
Follow-up (mos)
Preop Curvature
Postop Curvature
II I I III I I III III III I III
I I I I I II I I I I I
C5-C6 C5-C6 C6-C7 C6-C7 C5-C6, C6-C7 C4-C5, C3-C4 C4-C5 C4-C5, C5-C6,
31.9 29.3 24.6 18.0 17.2 13.7 15.9 11.8 11.7 10.4 18.0
Loss of lordosis Loss of lordosis Loss of lordosis Loss of lordosis Loss of lordosis Lordosis Kyphosis Lordosis Lordosis Lordosis Lordosis
Loss of lordosis Loss of lordosis Loss of lordosis Loss of lordosis Lordosis Lordosis Kyphosis Lordosis Loss of lordosis Kyphosis Lordosis
1 2 3 4 5 6 7 8 9 10 11
C6-C7 C6-C7 C5-C6 C6-C7
From Sekhon LH: Cervical arthroplasty in the management of spondylotic myelopathy: 18-month results. Neurosurg Focus 17(3):E8, 2004.
The use of Bryan disc arthroplasty exclusively in patients with cervical spondylotic myelopathy was reported in a series reported by Sekhon.36 Eleven patients with MRI-confirmed spinal cord compression or had a clinically confirmed myelopathy were included. A coexistent radiculopathy was present in eight subjects. Six patients had high signal in the spinal cord on T2-weighted imaging. Exclusion criteria included the presence of kyphotic deformity, severe multilevel spondylotic disc degeneration, spinal cord injury with possible instability, or pure radiculopathy secondary to posterolateral disc protrusion or foraminal stenosis. Average follow-up was 18.4 months, ranging from 10 to 32 months. Ten of the 11 (91%) patients had a good or excellent outcome and a statistically significant decrease in Nurick myelopathy scores. Further details of this cohort are presented in Table 95–1. Recomendations Cervical arthroplasty represents an exciting alternative to current strategies for the management of cervical myelopathy. Born out of trying to develop a solution to the shortcomings of cervical fusion surgery, cervical arthroplasty is itself presenting new challenges, some anticipated, others not, and others yet to be seen. Careful patient selection with, at this stage, avoidance of its application in situations of deformity, congenital stenosis, or multilevel disease is prudent, as is the long-term radiologic and clinical follow-up of these patients. Its ultimate role remains to be seen. REFERENCES 1. Pickett GE, Van Soelen J, Duggal N: Controversies in cervical discectomy and fusion: Practice patterns among Canadian surgeons. Can J Neurol Sci 31(4):478–483, 2004. 2. Brain WR, Northfield D, Wilkinson M: The neurological manifestations of cervical spondylosis. Brain 75(2):187–225, 1952. 3. Crandall PH, Batzdorf U: Cervical spondylotic myelopathy. J Neurosurg 25(1):57–66, 1966. 4. Payne EE, Spillane JD: The cervical spine: An anatomico-pathological study of 70 specimens (using a special technique) with particular reference to the problem of cervical spondylosis. Brain 80(4):571–596, 1957. 5. Adams CB, Logue V: Studies in cervical spondylotic myelopathy, II: The movement and contour of the spine in relation to the neural complications of cervical spondylosis. Brain 94(3):568–586, 1971. 6. Edwards WC, LaRocca H: The developmental segmental sagittal diameter of the cervical spinal canal in patients with cervical spondylosis. Spine 8(1):20–27, 1983.
7. Yue WM, Tan SB, Tan MH, et al: The Torg-Pavlov ratio in cervical spondylotic myelopathy: A comparative study between patients with cervical spondylotic myelopathy and a nonspondylotic, nonmyelopathic population. Spine 26(16):1760–1764, 2001. 8. Henderson FC, Geddes JF, Vaccaro AR, et al: Stretch-associated injury in cervical spondylotic myelopathy: New concept and review. Neurosurgery 56(5):1101–1113; discussion 1113, 2005. 9. Fehlings MG, Skaf G: A review of the pathophysiology of cervical spondylotic myelopathy with insights for potential novel mechanisms drawn from traumatic spinal cord injury. Spine 23(24):2730–2737, 1998. 10. Kim DH, Vaccaro AR, Henderson FC, Benzel EC: Molecular biology of cervical myelopathy and spinal cord injury: role of oligodendrocyte apoptosis. Spine J 3(6):510–519, 2003. 11. Clarke E, Robinson PK: Cervical myelopathy: A complication of cervical spondylosis. Brain 79(3):483–510, 1956. 12. Lees F, Turner JW: Natural history and prognosis of cervical spondylosis. Br Med J 5373:1607–1610, 1963. 13. Phillips DG: Surgical treatment of myelopathy with cervical spondylosis. J Neurol Neurosurg Psychiatry 36(5):879–884, 1973. 14. Edwards CC 2nd, Riew KD, Anderson PA, et al: Cervical myelopathy: Current diagnostic and treatment strategies. Spine J 3(1):68–81, 2003. 15. Fouyas IP, Statham PF, Sandercock PA: Cochrane review on the role of surgery in cervical spondylotic radiculomyelopathy. Spine 27 (7):736–747, 2002. 16. Kadanka Z, Bednarik J, Vohanka S, et al: Conservative treatment versus surgery in spondylotic cervical myelopathy: A prospective randomised study. Eur Spine J 9(6):538–544, 2000. 17. Kadanka Z, Mares M, Bednarik J, et al: Predictive factors for spondylotic cervical myelopathy treated conservatively or surgically. Eur J Neurol 12(1):55–63, 2005. 18. Kadanka Z, Mares M, Bednarik J, et al: Predictive factors for mild forms of spondylotic cervical myelopathy treated conservatively or surgically. Eur J Neurol 12(1):16–24, 2005. 19. Sampath P, Bendebba M, Davis JD, Ducker TB: Outcome of patients treated for cervical myelopathy: A prospective, multicenter study with independent clinical review. Spine 25(6):670–676, 2000. 20. Ebersold MJ, Pare MC, Quast LM: Surgical treatment for cervical spondylitic myelopathy. J Neurosurg 82(5):745–751, 1995. 21. Kaptain GJ, Simmons NE, Replogle RE, Pobereskin L: Incidence and outcome of kyphotic deformity following laminectomy for cervical spondylotic myelopathy. J Neurosurg 93(2):199–204, 2000. 22. Sekhon LH: Posterior cervical decompression and fusion for circumferential spondylotic cervical stenosis: Review of 50 consecutive cases. J Clin Neurosci 13(1):23–30, 2006. 23. Kawaguchi Y, Kanamori M, Ishihara H, et al: Minimum 10-year follow-up after en bloc cervical laminoplasty. Clin Orthop Relat Res 411:129–139, 2003. 24. Wang MY, Shah S, Green BA: Clinical outcomes following cervical laminoplasty for 204 patients with cervical spondylotic myelopathy. Surg Neurol 62(6):487–492; discussion 492–493, 2004.
750
P A R T
X I I
Controversies
25. Hidai Y, Ebara S, Kamimura M, et al: Treatment of cervical compressive myelopathy with a new dorsolateral decompressive procedure. J Neurosurg 90(2):178–185, 1999. 26. Shiraishi T: Skip laminectomy—A new treatment for cervical spondylotic myelopathy, preserving bilateral muscular attachments to the spinous processes: A preliminary report. Spine J 2(2):108–115, 2002. 27. Bailey RW, Badgley CE: Stabilization of the cervical spine by anterior fusion. J Bone Joint Surg Am 42-A:565–594, 1960. 28. Smith GW, Robinson RA: The treatment of certain cervical-spine disorders by anterior removal of the intervertebral disc and interbody fusion. J Bone Joint Surg Am 40-A(3):607–624, 1958. 29. Cloward RB: The anterior approach for removal of ruptured cervical disks. J Neurosurg 15(6):602–617, 1958. 30. Emery SE, Bohlman HH, Bolesta MJ, Jones PK: Anterior cervical decompression and arthrodesis for the treatment of cervical spondylotic myelopathy: Two- to seventeen-year follow-up. J Bone Joint Surg Am 80(7):941–951, 1998.
31. Cummins BH, Robertson JT, Gill SS: Surgical experience with an implanted artificial cervical joint. J Neurosurg 88(6):943–948, 1998. 32. Goffin J, Casey A, Kehr P, et al: Preliminary clinical experience with the Bryan Cervical Disc Prosthesis. Neurosurgery 51(3):840–845, 2002. 33. Goffin J, Van Calenbergh F, van Loon J, et al: Intermediate follow-up after treatment of degenerative disc disease with the Bryan Cervical Disc Prosthesis: Single-level and bi-level. Spine 28(24):2673–2678, 2003. 34. Duggal N, Pickett GE, Mitsis DK, Keller JL: Early clinical and biomechanical results following cervical arthroplasty. Neurosurg Focus 17(3):E9, 2004. 35. Bertagnoli R, Yue JJ, Pfeiffer F, et al: Early results after ProDisc-C cervical disc replacement. J Neurosurg Spine 2(4):403–410, 2005. 36. Sekhon LH: Cervical arthroplasty in the management of spondylotic myelopathy: 18-month results. Neurosurg Focus 17(3):E8, 2004.
CHAPTER
96
Cervical Arthroplasty Adjacent to Fusion, Multiple-Level Cases, and Hybrid Applications Paul C. McAfee, Matthew Scott-Young, and Rudolf Bertagnoli
K E Y l
l
l
l
l
P O I N T S
Cervical arthroplasty can be safely inserted in combination with or adjacent to anterior cervical discectomy and fusion (ACDF). Cervical arthroplasty can reduce the compensatory hypermobility found at cervical levels above ACDF. The outcomes of cervical arthroplasty adjacent to ACDF are comparable to virgin arthroplasty but clearly superior to multi-level cervical fusions. A hybrid procedure could be performed at two adjacent cervical levels if one level presented with translational instability. This level should be stabilized with fusion if there is 3.5 mm or greater translational instability. With progressively more and more vertebral levels of ACDF, the outcomes diminish and the complications increase—pseudarthrosis, instrumentation failure, reduced clinical success. This diminishing trend is not seen with progressive numbers of adjacent cervical arthroplasties.
The use of cervical arthroplasty devices in patients with symptomatic radiculopathy or myelopathy adjacent to a prior anterior cervical decompression and fusion (ACDF) is an attractive reconstructive option, obviating the need for a multilevel fusion surgery. This chapter reports a comparison of outcomes from patients with and without previous ACDF enrolled by five centers in a U.S. Food and Drug Administration (FDA) Investigative Device Exemption (IDE) clinical trial of the Porous-Coated Motion (PCM) artificial cervical disc (Cervitech, Inc., Rockaway, NJ) and PCM experience in Australia and Brazil. Retrospective reviews of ACDF have described the appearance of new degenerative disc disease (DDD) at levels adjacent to fused segments. Hilibrand1 has reported that adjacent motion segment disease occurred in 2.9% of patients annually. They found up to 25% of patients developed DDD at adjacent levels within 10 years of the initial surgery and that 7% to 15% of these patients required reoperation. In a recent study, Robertson and associates2 assessed the incidence of adjacent motion segment disease in patients treated with cervical fusion and arthroplasty. They found the incidence of new symptomatic DDD in the fusion group at 24 months was 13.9% (6.9% annually). In the arthroplasty group,
the incidence of adjacent motion segment disease was 1.3% (0.65% annually). The authors demonstrated that maintaining motion, rather than fusion, will prevent symptomatic adjacent disc disease and a reduction in the adjacent level radiologic indicators of disease at a 24-month postoperative interval. Gore3 studied lateral radiographic findings in asymptomatic persons with a 10-year follow-up to reveal that degenerative changes progress with age. He found 34% of asymptomatic patients with normal findings on the baseline cervical spine lateral x-rays developed degenerative disc disease 10 years later. He stated that 97% of patients with pre-existent disc degeneration showed progression over 10 years. Adjacent motion segment disease is defined as the development of a new radiculopathy or myelopathy or symptoms referable to a segment adjacent to a previously fused level in the cervical spine. It is thought to arise as a result of the biomechanical alterations and biologic changes that occur in the cervical spine subsequent to treatment. Schwab and associates4 found that motion compensation following arthrodesis was distributed through the unfused segments, with significant compensation at the segments adjacent to the fusion. They found significant differences occurring at the level above the fusion site for C3-C4 and C4-C5 in both flexion and extension. When the lower levels were fused, a significant amount of increased motion was observed at the levels immediately above and below the fusion. However, greater compensation occurred at the caudal segments than the cephalad segments for the lower level fusions (C5-C6 and C6-C7). This chapter focuses on the clinical incidence of adjacent motion segment degeneration and disease in the cervical spine adjacent to a cervical fusion. Hilibrand and associates1 published a report of 374 patients who had a total of 409 anterior cervical arthrodeses for the treatment of cervical spondylosis. They reported an average annual incidence of the development of adjacent level disease as 2.9%. Survivorship analysis predicted that 25% of the patients (95% confidence interval) who undergo ACDF would have new disease in an adjacent level within 10 years of the operation. They also found that the incidence of new disease at an 751
752
P A R T
X I I
Controversies
adjacent level was significantly lower following a multilevel arthrodesis, than it was following a single-level arthrodesis. The concept of the multifocal nature of cervical spondylosis was also introduced. Ishihara and associates5 followed 112 patients, clinically and radiologically, for 2 years. The incidence of adjacent motion segment disease/degeneration was 19% and they applied a KaplanMeyers Survivorship Analysis to follow the disease-free survivorship in the entire series of patients. They found an 84% incidence at 10 years. They also concluded that adjacent motion segment disease was more common in those who showed asymptomatic disc degeneration preoperatively. MATERIALS AND METHODS For the U.S. study on adjacent segment PCM, patients were enrolled into the initial non-randomized “training” and subsequent randomized investigational device arms of the multicenter clinical trial comparing PCM arthroplasty with ACDF. Inclusion and exclusion criteria specified patients between the ages of 18 and 65 with symptomatic cervical radiculopathy or myelopathy at one level, unresponsive to at least 6 weeks of nonsurgical therapy, or experiencing progressive neurologic symptoms. Of 103 patients enrolled in the IDE study at these sites, 19 patients had previous ACDF at a single adjacent level, and 84 patients had the PCM as a primary intervention. The Neck Disability Index (NDI), neck and arm Visual Analog Scale (VAS) scores, and recorded complications and adverse events for both groups were compared. There was a minimum of 12 months follow-up, the average follow-up being 26 months (range 46–12 months). Fifty-one patients (31 female and 20 male) were included in the international study group. Entry into the study group required symptomatic evidence of adjacent motion segment disease that required surgical intervention. The primary diagnosis of radiculopathy occurred in 36 patients and myelopathy in 15 patients; 36 patients had a prior one-level fusion, 9 had a prior two-level fusion, and 6 had a prior three-level fusion. Surgeons from Brazil, Australia, and the United States participated in this study. Each surgeon was asked to participate in the treatment of adjacent motion segment degeneration with total disc replacement (TDR), provided the indications were appropriate. FDA regulations restricted patient selection and the treatment options in the United States. The surgeons in Australia and Brazil were able to treat extensive and complex disease and perform one or more TDRs at adjacent levels, as indicated. Thirty-six one-level PCMs were performed adjacent to fusions (20 from the United States, 11 from Brazil, and 5 from Australia). Nine two-level PCMs were performed (7 in Brazil and 2 in Australia). Six three-level PCMs were performed (4 in Brazil and 2 in Australia). The outcome measures that were utilized included the NDI, VAS, SF-36, and neurologic status. Scores were recorded preoperatively and at 3 months, 6 months, 1 year, and 2 years after surgery where applicable. Cervical lordosis was measured from the posterior surface of the body of C2 in a line parallel to the posterior surface of the body of C7. The postoperative lordosis was measured at the postoperative intervals and the change in lordosis was determined. Radiographic outcomes included flexion and extension and lateral bending range of motion. Disc space height restoration was
i) ACDF uninstrumented ii) ACDF/internal fixation Plate Cage PMMA Cage/plate iii) Congenital fusion iv) Arthroplasty with HO v) Posterior cervical fusion with instrumentation
12 17 4 3 1 1 1 2
n F I G U R E 96–1. Types of prior fusion mechanisms, listing prior instrumented as well as uninstrumented anterior cervical discectomy and fusion (ACDF) procedures.
measured, the presence of heterotopic ossification was noted, and the translational and rotational stability was assessed. The prior fusion mechanism is illustrated in Figure 96–1; 60% of patients had an ACDF with a form of internal fixation, 30% had an uninstrumented fusion, and 10% had either a posterior spinal cervical fusion, congenital fusion, or a spontaneous fusion around a previous TDR. Figure 96–2 shows the number of levels fused, compared to the number of levels replaced. As expected, the majority of the fusions involved C4 to C7. Sixty-two levels had undergone arthrodesis at some stage in the past. Fifty-nine prostheses were inserted in the distribution shown in Figure 96–2 and represent treatment of symptomatic discs adjacent to a fusion. RESULTS U.S. Prospective Randomized Study Among the 19 patients in the U.S. study with previous “adjacent” fusion, there were 6 males and 13 females with a mean age of 49.1 years. There were 50 males and 34 females with a mean age of 47.0 years comprising the “primary” patients. There were no significant differences in age, height, weight, or body mass index (BMI). Average time of surgery was 97 minutes in the primary group, and 91 minutes in the adjacent to fusion group (P > 0.1). Estimated blood loss was 79 mL and 76 mL in the primary and adjacent to fusion groups, respectively (P > 0.1). Follow-up at 6 months was reached by 68 primary patients and 15 adjacent patients, and at 12 months by 41 primary patients and 11 adjacent patients. Clinical outcomes were similar between groups at all time points. Revision for subsidence, misalignment, or device migration occurred in 2 of 84 patients in the primary group, and in 2 of 19 patients in the adjacent to fusion group. One device migration not requiring revision occurred in the adjacent to fusion group. Other reported complications including dysphagia and postoperative neck or arm symptoms were comparable in both groups (Table 96–1). International Study For the international study, the average length of the procedure was 91 minutes. The estimated blood loss was 56 mL on average. The average length of stay was 1.65 days. The mean preoperative cervical lordosis was 2.65 degrees ( 32 to 25 degrees). The mean postoperative lordosis was 12.3 degrees ( 17 to 30 degrees).
CHAPTER 96 n
F I G U R E 96–2. A comparison of the number of levels fused with the number of vertebral levels replaced with a Porous-Coated Motion device in the international study.
Levels fused
C2-C3 C3-C4 C4-C5 C5-C6 C6-C7 C7-T1
Levels replaced
2 7
1
12
10
26
12
14
15
1
19 2
Time
Baseline 6 weeks 12 weeks 26 weeks 52 weeks
Outcomes after PCM Arthroplasty*
NDI (Primary/ Adjacent)
Neck VAS (Primary/ Adjacent)
Arm VAS (Primary/ Adjacent)
27/27 14/14 12/11 11/11 10/11
67/73 25/31 24/30 24/33 23/24
69/78 21/28 23/21 23/26 26/25
*United States study only. NDI, Neck Disability Index; PCM, Porous-Coated Motion; VAS, Visual Analog Scale.
C2-C3 C3-C4 C4-C5 C5-C6 C6-C7 C7-T1
mean improvement in NDI in the United States was 81%. The mean improvement in Brazil was 45%. The range of flexion and extension motion at the level of prosthesis was a mean of 8.5 degrees (4 to 20 degrees). There was no evidence of translational or rotational instability. All patients were neurologically intact at the conclusion of the study. The disc heights were restored to the heights of the implant (6.5 mm, 7.2 mm, and 8.5 mm). There was no loss of disc height or subsidence noted. The mean SF-36 physical component summary (PCS) increased to 70% from 38%, representing an 86% improvement. The mean SF36 mental component summary (MCS) increased to 66% from 43%, representing a 34% improvement. In relation to the significant improvements that were seen in both the physical and mental scores in the SF-36 data, significant improvements occurred within the first 3 months. Between 3 and 6 months, there was no statistical improvement, and between 6 and 12 months, there was a moderate COMBINED DATA 90 VAS NDI
80 70 Average score
The mean improvement was 9.4 degrees of cervical lordosis ( 15 to 23 degrees). The clinical follow-up period varied from 12 to 46 months, the average follow-up time being 26 months. There was one revision case that occurred in Australia where the patient had a two-level PCM inserted above a previously uninstrumented C5-C6 fusion. Six weeks postoperatively, the patient was involved in a rear-end motor vehicle accident, sustaining a whiplash injury. Radiographs were taken that showed an anterior lip fracture of the superior surface of C5 with no migration of the inferior end plate of the prosthesis anteriorly. The patient was advised to have a reduction of the fracture to prevent any late mechanical instability. In the operating theater, the end plate was found to be securely fixed to the device. This required removal of the device in order to reduce the fracture and then a new prosthesis was inserted with an inferior flange and two screws. The patient has had an excellent result. In this international multicenter study, a mean improvement in the VAS of 60.9% was recorded (Fig. 96–3). In terms of the specific regional results, the VAS improvement in Australia was 79.6% (Fig. 96–4). The mean improvement in VAS in the United States was 80%. The mean improvement in VAS in Brazil was 51%. In this study, the average mean improvement in the NDI was 54% (Fig. 96–3). In regards to regional variations, the mean improvement in NDI in Australia was 53% (Fig. 96–5). The
TABLE 96–1.
753
Cervical Arthroplasty Adjacent to Fusion, Multiple-Level Cases, and Hybrid Applications
60 50 40 30 20 10 0 Pre-op
3 mo
6 mo
12 mo
Time period The combined Neck Disability Index (NDI) and Visual Analog Scale (VAS) clinical outcomes in the international study.
n
F I G U R E 96–3.
P A R T
754
X I I
Controversies
VAS AUSTRALIA 90 80 Average score
70 60 50 40 30 20 10 0 Pre-op
3 mo
6 mo
12 mo
Time period n F I G U R E 96–4. Visual Analog Scale (VAS) scores for the patients in Australia. Notice that the speed of improvement is not quite as rapid as for virgin-level cervical arthroplasties, even from the same surgeon’s experience.
NDI AUSTRALIA 60
Average score
50 40 30 20 10 0 Pre-op
3 mo
6 mo
12 mo
Time period The Neck Disability Index (NDI) results of the patients with adjacent level fusions in Australia.
n
F I G U R E 96–5.
improvement, once again. The data, combined with the NDI data and VAS data, highlight the concept that the patient's pain is initially reduced fairly dramatically but that it takes some time to adjust their functional impairments. There is often a lag between the two. It also highlights the importance of a functional rehabilitation program following surgery. Long-term follow-up is also essential to determine whether the results deteriorate with time. DISCUSSION Cervical arthroplasty adjacent to a previous cervical fusion, as an alternative to multilevel fusion, is an attractive clinical option.6–8 Although the level adjacent to a prior cervical fusion is subject to altered biomechanical forces, the PCM arthroplasty was well tolerated at these levels. The initial clinical results of PCM arthroplasty adjacent to a prior fusion were excellent and comparable to the outcomes after primary PCM surgery. Wigfield9 reported the results of a 2-year pilot study where the new Frenchay Artificial Cervical Joint was inserted in 15 patients. This was a clinically prospective cohort study. All patients had a
previous adjacent level surgical fusion or congenital spinal fusion or radiologic evidence of adjacent level degenerative disc disease. Improvements in the VAS scores showed a 45% improvement, NDI a 31% improvement, and SF-36 PCS quotient 14% improvement. The conclusions from the study were that the prosthesis maintains physiologic motion, that the device maintained its position, and the improvements in the functional scores were acceptable. Yue and associates10 recently reported the treatment of adjacent motion segment degeneration with the ProDisc-C (Synthes, West Chester, PA) TDR in a prospective study with a minimum of 2 years follow-up and which included 15 patients (Fig. 96–6). They concluded that the results were inferior to historical control subjects. The study showed that the application of a ProDisc-C device for the treatment of adjacent motion segment degeneration only achieved 80% of what a ProDisc-C device achieved in the management of a one-level cervical DDD. The treatment of adjacent motion segment disease above a fusion has a variety of solutions. A cervical TDR serves to replace a symptomatic degenerative disc, restore the functional biomechanical properties of the motion segment, and protect the neurologic structures. Treatment of adjacent motion segment degeneration with cervical arthroplasty is a very challenging environment, for many reasons (Fig. 96–7). First, these patients are physically and emotionally distraught as they are faced with further surgery that they in all probability did not expect. Second, the patient, by definition, has had a history of prior surgery. As a general rule, all patients with prior spine operations have reduced clinical outcomes compared to patients with virgin procedures. This is why every U.S. prospective randomized clinical arthroplasty trial with the exception of the PCM excludes patients with even one prior cervical arthrodesis. The prior surgery leads to scarring, both externally and internally, which then increases the complication rate. Third, these patients often have internal fixation devices, such as plates, that will need to be removed before the insertion of a prosthesis. This may require a larger dissection than an approach to one level. Often the previously fused level has been inappropriately fixed in a kyphotic position (Fig. 96–8) and the end plates can be damaged by malpositioned screws. The mechanism of the prior fusion methods in this study is described in Figure 96–1. The variety of instrumentation used increases the complexity of the surgery. As can be seen from the preoperative lordotic angles, considerable kyphosis made these cases extremely challenging, and despite the complicated presenting disease, PCM prostheses successfully restored some element of the cervical lordosis. It also restored stability to the cervical segments, based on the flexion and extension and rotation shown radiologically. It also preserved 8.9 degrees of flexion and extension mobility. Despite the complexity of the revision cases, the PCM device and its modularity appeared to achieve the goals of the hypothesis of this study. There was also a lower incidence of dysphagia and soft tissue retraction in the cervical arthroplasty cases compared to the ACDF with instrumentation procedures (Fig. 96–9). The PCM international study is the largest study, to date, investigating prospectively the value of arthroplasty in adjacent motion segment disease. The overall results are encouraging
CHAPTER 96
Cervical Arthroplasty Adjacent to Fusion, Multiple-Level Cases, and Hybrid Applications
755
19.7 degrees pre-op motion Adjacent level is challenging 1. Excessive motion 2. Previous surgery 3. Prior instrumentation 4. Disc height collapse 5. Localized kyphosis
A 10.0 degrees post-op motion Prosthesis has to ratchet down or restrict the mobility
B n
F I G U R E 96–6. This patient had excessive motion at C4-C5 measuring 19.7 degrees on flexionextension. A, Hypermobility is seen adjacent to a prior two-level arthrodesis, instrumented at C5-C6 and uninstrumented at C6-C7. The adjacent level problems include excessive compensatory motion, previous surgery, prior instrumentation, disc height collapse, and localized kyphosis. B, The purpose of a cervical arthroplasty is not necessarily to increase motion but rather to make the motion more physiologic within the neutral zone. The C4-C5 level was successfully replaced with a ProDisc-C cervical arthroplasty. Now the postoperative flexion-extension motion is more physiologic at 10.1 degrees, and the patient is clinically much improved.
and discussion is essential in relation to the regional variations in the results. In terms of the American results, these show significant improvements in VAS and NDI and are as good as the historical control results for one-level cervical TDR with the PCM device. It should be noted that, from an FDA perspective, the IDE criteria for the American patients had to be met in that they needed to present with radiculopathy and corresponding neurologic deficit, confirmed by an MRI compressive lesion. The Brazilian and Australian cases were not constrained by the IDE criteria and, as a result, more complex disease was included at these two sites.
Based on the biomechanical data, the clinical studies indicating adjacent motion segment degeneration/disease, and the various solutions to the problem that the surgeons had at their disposal, a study was performed to assess the safety and efficacy of the PCM prosthesis in the treatment of adjacent motion segment disease. In this study of 51 cervical arthroplasty procedures, there was one revision, following a trauma, resulting in one device-related complication, and no approach-related complications. No adjacent motion segment disease had developed adjacent to cervical TDR. There were no cases of heterotopic ossification, and there was no evidence of any neurologic deterioration. Considering the degree
756
P A R T
X I I
Controversies
5 months post-op Head-on MVA collision Fx C5 vertebral body
Erect extension
A
n F I G U R E 96–7. This 57-year-old man from Australia had a prior adjacent level fusion at C5-C6. A, Five months after a two-level Porous-Coated Motion (PCM) device arthroplasty at C3-C4 and C4-C5, he was involved in a head-on motor vehicular accident and fractured the cephalad C5 vertebral end plate. B, The loose prosthesis and C5 fracture were successfully treated by inserting a flanged version of the PCM prosthesis with fixation with two screws. One year later he had preserved cervical motion and was asymptomatic.
CHAPTER 96
Cervical Arthroplasty Adjacent to Fusion, Multiple-Level Cases, and Hybrid Applications
757
Case 33–28 months F/U
Mean for 51 cases = 2.65 to 12.3 degrees lordosis n
F I G U R E 96–8. This 39-year-old woman had prior lordotic “cages” at C3-C4 and C6-C7 and presented
with severe myelopathy and cervical kyphosis. Two-level cervical decompressison and Porous-Coated Motion (PCM) arthroplasties at C4-C5 and C5-C6 successfully corrected 30 degrees of kyphosis and restored 27 degrees of flexion-extension motion. Overall, the 51 patients in the international study improved from a mean of 2.65 degrees of kyphosis to 12.3 degrees of kyphosis. This patient is clinically neurologically intact at 28 months’ follow-up.
of difficulty of this group of patients, the clinical success rate, the complication rate, and the revision rate, the author's consensus is that the PCM is a device that is safe and effective for use in patients with adjacent motion segment degeneration in the cervical spine. The hypothesis was verified showing a significant improvement in functional scores and a reduction in pain scores. In terms of the
n
American results, the results are as good as historical control results when using the PCM prostheses. Yue and associates reported that treating adjacent motion segment degeneration with ProDisc-C prostheses, the results were only 80% of historical controls.10 This highlights the versatility of the PCM prosthesis in catering for the variety of anatomic and pathologic configurations
F I G U R E 96–9. Lower incidence of dysphagia with total disc replacement versus anterior cervical discectomy and fusion (ACDF). A, This 45-year-old man presented with dysphagia for the last 9 months following ACDF with a poorly placed anterior plate. He could only swallow by leaning forward. His Gastrografin swallow on the left demonstrates the location of esophageal compression from the upper margin of the cervical plate.
758
P A R T
X I I
Controversies
n
F I G U R E 96–9. Cont’d Lower incidence of dysphagia with total disc replacement versus anterior cervical discectomy and fusion (ACDF). B, The patient presented with C6-C7 HNP, and at surgery the esophageal compression was visualized and required a difficult dissection off the hardware. C, After removal of the cervical plate and successful C6-C7 cervical discectomy and Porous-Coated Motion (PCM) arthroplasty, not only was the patient’s right arm radiculopathy relieved but his dysphagia completely resolved. The PCM device has no anterior profile compared to anterior cervical plates and screws. The severity of dysphagia has been found to correlate with the thickness of the anterior cervical plate instrumentation.
that may be present in an adjacent motion segment that is degenerate and painful. In terms of the Australian results in comparison to historical control findings within one author's (MSY's) practice, the VAS scores and NDI scores at 12 months of follow-up were exactly the same. However, it appears that the patients treated for adjacent motion segment degeneration took longer to recover following
the surgery than the historical control subjects. In terms of the Brazilian data, this also illustrates the fact that multilevel total disc replacements often do better than single-level disc replacements. In regard to the 51 patients within this international study of PCM applied to adjacent levels, the complexity of the preoperative surgery and the complexities of the preoperative diagnosis tend to lower the results below the historical control results.
CHAPTER 96
Cervical Arthroplasty Adjacent to Fusion, Multiple-Level Cases, and Hybrid Applications
CONCLUSIONS It is clear that cervical adjacent motion segment degeneration represents a hostile environment for the revision surgeon, first, because of prior surgery and, second, because of the altered mechanics. This is a multicenter prospective descriptive study that showed a mean improvement of 60.9% in VAS and a mean improvement of 54% in NDI. The safety and efficacy of the device in this biomechanically challenging situation have been proved. Nevertheless, long-term follow-up of these devices is essential. REFERENCES 1. Hilibrand AS, Carlson GD, Palumbo MA, et al: Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg Am 81(14):519–528, 1999. 2. Robertson JT, Papadopoulos SM, Traynelis VC: Assessment of adjacent segment disease in patients treated with cervical fusion or arthroplasty: A prospective 2-year study. J Neurosurg Spine 3:417–423, 2005. 3. Gore DR: Roentgenographic findings in the cervical spine in asymptomatic persons: A ten-year follow-up. Spine 26(22):2463–2466, 2001.
759
4. Schwab JS, DiAngelo DJ, Foley KT: Motion compensation associated with single-level cervical fusion: Where does the lost motion go? Spine 31(21):2439–2448, 2006. 5. Ishihara H, Kanamori M, Kawaguchi Y, et al: Adjacent segment disease after anterior cervical interbody fusion. Spine J 4(6):624–628, 2004. 6. McAfee PC, Cunningham BW, Hayes V, et al: Biomechanical analysis of rotational motions after disc arthroplasty. implications for patients with adult deformities. Spine 31(19):S152–S160, 2006. 7. McAfee PC, Geisler FH, Saiedy SS, et al: Revisibility of the Charite Artificial Disc: Analysis of 688 patients enrolled in the US IDE Study of the Charite Artificial Disc. Spine 31(11):1217–1226, 2006. 8. McAfee PC, Geisler FH, Scott-Young M (eds): Roundtables in Spine Surgery: Complications and Revision Strategies in Lumbar Spine Arthroplasty. St. Louis, MO, Quality Medical Publishing, 2005. 9. Wigfield CC, Gill SS, Nelson RJ, et al: The new Frenchay Artificial Cervical Joint: Results from a two-year pilot study. Spine 15:27 (22):2446–2452, 2002. 10. Yue J, Bertagnoli R, Fenk-Mayer A, et al: The treatment of symptomatic adjacent segment degeneration after cervical fusion with total disc arthroplasty utilising the ProDisc-C Prosthesis: A prospective study with 2 year minimum follow-up. Spine J 6:1S–161S, 2006. Proceedings of the NASS 21st Annual Meeting, Seattle, WA, Sept. 28, 2006.
CHAPTER
97
The Future of Motion Preservation Stephen H. Hochschuler and Donna D. Ohnmeiss
K E Y l
l
l
l
l
P O I N T S
We are already seeing new design concepts for total disc replacements, and nucleus replacements are being evaluated. One of the next phases of motion preservation will be combining existing technologies such as anterior and posterior dynamic stabilization. Treatment will be improved through increased understanding of pain origins, disc biochemistry, and mechanical function of the spine. While spine technology grows, its development and acceptance will have to be founded in critical cost-effectiveness studies, and the key may be identifying specific subgroups of patients who are the best candidates for specific treatments rather than large-scale use of any emerging technology. Many metal implants and surgeries may eventually become obsolete and be replaced by biologic and pharmacologic therapies.
OVERVIEW Achieving the goal of providing the best care for patients with back and neck pain requires advancements in many areas. These advancements include a better understanding of the pain generators, more reliable diagnostic evaluations (preferably noninvasive), comprehensive understanding of the physiologic effects of various treatment alternatives, and, of course, the treatments themselves. As demonstrated in the various chapters of this book, there are a myriad of emerging treatment options for patients with low back or neck pain. Total disc replacement (TDR) is becoming a commonly used alternative to fusion in patients with symptomatic disc degeneration. There are already new concepts for disc designs being evaluated. Completely new treatment alternatives are also under investigation. There are and will likely continue to be, however, many challenges in the ongoing development and acceptance of these new treatments. In this chapter, we will present ongoing developments in spine technology, including those in the near future as well as those that may emerge years in the future. In addition, challenges to the development and use of these technologies will be discussed. NEW KNOWLEDGE One of the venues by which care may be improved is through advances in our understanding of the spine and pain. This area 760
includes understanding the tissues involved, the chemistry and biomechanics of pain mechanisms, the physiologic effect of various treatments, and the influence of behavioral and psychological factors on pain. Through increased knowledge in each of these areas, more comprehensive and accurate diagnostic evaluations and treatment plans can be derived. The expanding number of treatment options will drive the need for better diagnostics to maximally match the treatment to the individual patient's problem. Fully accomplishing the goal of providing the best possible treatment for each patient will require many years of careful data documentation to assess and reassess indications and outcomes for subgroups of patients undergoing various motion-preserving treatments. Without a doubt, as knowledge of disc tissue increases so will the treatment options. Not all disc problems are created equal. Much too often in the literature and clinical reports, too many terms are used interchangeably when in actuality there are distinct differences in the various disc diseases. It is likely that true disc degeneration is a normal part of aging. However, disc herniation, disruption, and other problems, which are more typically associated with clinical pain, should not be considered a part of the natural aging process.
PAIN MECHANISMS Traditionally, the exact mechanisms of back pain have been poorly understood and often without definitive diagnostic evaluation. One of the reasons for the spine patient being so difficult to treat is that there are many possible origins for pain in the same body region. For example, pain in the buttock region may be related to disc, facets, sacroiliac joints, hips, nerve roots, pyriformis muscle, or any combination of these structures. Noninvasive clinical evaluations with high specificity and sensitivity have yet to be developed to identify pain origin in many patients. Most typically, combinations of clinical signs and symptoms, imaging, and injections are used to confirm some pain origins and rule out others. However, it is likely that in many patients some problems go undiagnosed or the diagnosis is delayed until after treatment for one problem and symptoms persist and further diagnostics are pursued. On the other hand, it is not practical to expose the
CHAPTER 97
majority of patients to extensive and expensive injections to fully evaluate every possible source of symptoms. More research is needed in the area of creating inexpensive diagnostic evaluations with high rates of sensitivity and specificity for the various possible origins of back pain. These advances will enhance the matching of treatment with problems, reduce the potential for treating a structural problem not related to symptoms, accelerate the patient's treatment plan, decrease costs, and potentially reduce the chances of patients developing chronic back pain behaviors. Ongoing research is leading to enhanced understanding of spinerelated pain. Discussed here are some of the newer studies that are enhancing our understanding of the tissue and biochemistry of disc-related pain. Not only are some of these studies providing information on discal tissue and pain generators but these results may also have important implications in the development of motion-preserving technologies. This may be particularly true for therapies targeting tissue regeneration, in that the cellular and structural changes that take place within the disc may limit or at least affect the performance of these emerging treatments. Some of the biologic therapies have been investigated in ideal laboratory conditions or in animals with normal discs, but not in models with the type of ruptures and degenerative changes that are being identified in anatomic and biochemical studies of painful discs encountered clinically. Several inflammatory mediators have been identified in relation to back and radicular pain. Among them are substance P, prostaglandin E2, nitric oxide, TNF-a, and interleukins 1, 6, and 8. The exact role of these agents in producing back and leg pain is the topic of much investigation, but a more specific understanding of these inflammatory mediators and their role in different pain situations is developing. For example, Burke and associates found that disc tissue removed from patients with discogenic pain had greater expression of prostaglandin E2, and interleukins 1, 6, and 8.1 In addition to enhanced understanding of back pain–related inflammatory mediators, a greater knowledge concerning specific tissue-related changes that occur in an injured disc is being developed. Coppes and associates reported that in discs that were positive on discography, compared to normal discs, there was more extensive innervation in the disc.2 In particular, these nerve fibers were substance P immunoreactive. In a similar study of discographically positive discs, Freemont and associates found nerve fibers passing into the inner annulus and into the nucleus of some specimens. These fibers expressed substance P. No nerve fibers were found in the control discs.3 Further investigation in this area involved analyzing disc tissue removed during surgery from patients with positive discograms, aging patients undergoing surgery for stenosis, and control tissue from cadavers.4 The authors found that a zone of vascularized granulation tissue had formed along the painful disc tears passing from the nucleus to the outer annulus. Nerve growth was noted along the granulation tissue. Substance P, neurofilament, and vasoactive-intestinal peptide immunoreactive nerve fibers in the painful discs were more extensive than in the control discs. The same investigators differentiated asymptomatic degenerated discs, painful discs, and normal discs by comparing the expressions of basic fibroblast growth and its receptor and transforming growth factor-b1 and its receptor using immunohistochemistry methods.5 The primary differences were
The Future of Motion Preservation
761
that along the granulation tissue of the painful discs, there were strong expressions of basic fibroblast growth factor, transforming growth factor-b1, and their receptors, as well as proliferating cell nuclear antigen. These were only weakly noted in the nonpainful discs and in tissue from areas without granulation tissue in the painful disc. There was no expression of these agents in the control discs. Also, differentiating the painful discs was the large number of macrophages and mast cells. Along with these investigations of disc degeneration, results from an animal model found an association between disc degeneration and calcification of the end plate.6 Such calcification may interfere with the diffusion of nutrients to the disc and be related to degeneration or reduced healing potential after an intradiscal injury. All of these studies provide an increased understanding of painful disc degeneration. Their findings may provide insight as to challenges that are likely to be encountered when trying to develop tissue engineering based therapies. PAIN IMAGING One emerging area of research that has the potential to greatly impact all types of pain-related treatment, including those in the spine, is pain imaging of the brain. This application was developed in the 1990s and was based on PET (positron emission tomography) and SPECT (single photon emission computed tomography) imaging. Now it is possible to use functional MRI (magnetic resonance imaging) to perform imaging of the brain to investigate patterns associated with changes in pain. This technology has application in pharmacologic studies.7 It is anticipated that fMRI will become a valuable tool in evaluating responses to various spinal treatment interventions, both nonoperative and operative. This evaluation should compliment emerging back pain therapies such as inhibitors for TNF-a and other inflammatory mediators that have been implicated in chronic back pain. BIOMECHANICS Traditionally in the spine, a biomechanical relationship has been the primary explanation sought for many of the types of back pain. However, there remains a chasm between biomechanical concepts and information that can be applied during the treatment of an individual patient. The concept of the neutral zone has been widely accepted. However, in terms of treating an individual, no clinically executable definition has been developed. In relation to total disc replacement, there has been discussion of the center of rotation of various implants as well as the merits of constrained, semi-constrained, and unconstrained devices. However, there has been no compelling evidence provided concerning how these concepts influence patient outcomes. A stronger relationship between biomechanical knowledge, theory, and the clinical situation needs to be developed. New implants such as the Theken disc open the door for exploration into the loads on the disc in everyday use. This device is an artificial disc with electronics that allows monitoring of the loads and movement of the implant. The data are read with an external unit. There will likely be important information gained on the influence of implant position within the disc space and
762
P A R T
X I I
Controversies
the load-bearing capability demanded from it. Implants such as this, incorporating microelectronics, will allow the collection of “real life” data on how spines function. This type of information may have a significant impact in improving computerized models used for spine research as well as changing the biomechanical testing protocols for spinal implants due to the more direct information gained on the demands on the disc in everyday life. This device has been implanted in a few patients outside the United States and data are currently being collected on the functioning of the implant. CURRENT TECHNOLOGIES Total Disc Replacement Total disc replacement may no longer qualify as a “new technology.” It was first used clinically in the 1980s. There are long-term data available from Europe supporting that these implants are safe and effective following more than 10 years of use.8 All the current designs to date depend on the basic concept of motion through the articulation of curved surfaces. However, the next generation is being introduced that is made of materials with properties that more closely mimic the properties of the natural disc. This concept was described years ago by Lee and associates9 and is currently coming to fruition with clinical trials being initiated. Rather than allowing motion through sliding surfaces, these new devices allow motion through the compression of the materials making up the disc. Nucleus Replacement Nucleus replacements can be divided into three basic types: injectable, soft implants, and hard implants. The very first nuclear implant was the stainless steel ball described by Fernstrom. These were implanted into the disc space after discectomy, and they reportedly produced results superior to discectomy alone, although in some patients, they eventually subsided into the vertebral bodies. The first modern-day nucleus replacement to undergo widespread use, and now clinical trials, is based on a hydrogel as first described as a discal implant by Ray and called the PDN (prosthetic disc nucleus; Raymedica, Inc., Minneapolis, MN).10,11 Initially there were problems with displacement, but changes in the implant design should address this problem. Just as the CHARITÉ total disc replacement (DePuy Spine, Raynham, MA) was based on studies of the motion of the natural disc, the PDN was based on the hydrostatic properties of the disc nucleus. Several nucleus replacements exist that are based on hydrogel technologies. These implants should absorb water after being implanted into the nuclear cavity and subsequently mimic the natural disc nucleus. In addition to hydrogels and other “elastic” materials that allow motion through compression of the implanted device, one mechanical nucleus replacement has been introduced. The NeuDisc (Replication Medical, Cranbury, NJ) is similar in concept to the current total disc replacements by allowing motion through sliding articulation between concave and convex surfaces. One concern with nuclear implants is that most require the implant to pass through the disc annulus. There is concern that this may possibly initiate degeneration of the annulus. However,
if the artificial nucleus can reduce the load on the annulus, this may not be of clinical relevance. This potential problem may be avoided altogether with the tran-S1 technique of approaching the disc space through the sacrum and thus not violating any of the annular layers while providing access to the nucleus. Posterior Dynamic Stabilization In addition to the disc replacements, there have been numerous dynamic posterior stabilization devices developed in recent years. Many of these have incorporated adaptations from existing pedicle screw technology with flexible rods and screws. The primary mechanical goal of these implants is to allow controlled motion. Pedicle anchoring has also been used for new devices such as the TFAS (Archus Orthopedics, Redmond, WA), TOPS (Impliant Spine, Princeton, NJ), and Anatomic Facet Replacement System (Facet Solutions, Logan, UT). One additional posterior dynamic device is the M-Brace (Advanced Spinal Technologies, Boca Raton, FL), which is designed to be implanted using a minimally invasive technique. These new technologies are currently being evaluated in clinical trials. The future will undoubtedly bring more variations of pedicle screw dynamic devices such as that being developed by Innovative Spinal Technologies (IST, Mansfield, MA). Interspinous Spacers Another type of dynamic posterior instrumentation is the interspinous spacer. These devices are designed to be minimally invasive and provide distraction. The first of these devices to receive FDA approval for sale in the United States was the X-STOP (Kyphon, Inc., Sunnyvale, CA). This device was investigated for the treatment of spinal stenosis and favorable results were achieved.12 It provides symptomatic relief by distracting the foramen and decreasing the potential for nerve root compression. Currently, several other types of spinous process spacers are undergoing formal clinical evaluation, including the Wallis (Abbot Spine, Austin, TX), Diam (Medtronic Sofamor Danek, Memphis, TN), and coflex (Paradigm Spine LLC, New York, NY). Some of these devices will be investigated to determine their effectiveness in the treatment of symptomatic disc degeneration by unloading the disc. The concept of this type of technology, decreased cost (compared to open decompression or fusion), minimally invasive surgical approaches, and possible applications in patients who are not good candidates for the alternative treatment (such as multilevel decompression or fusion in elderly patients for treating stenosis) may set a new standard for emerging implant technologies. Spinous Process Plate In the arena on minimally invasive implants, an interspinous plate has been described.13,14 The goal of the device is to provide the same degree of stability to the lumbar spine as pedicle screws and rod systems, but to achieve this using a minimally invasive surgical technique. The plate appears to be simpler to implant than percutaneous pedicle screws. A biomechanical study found that the plate provides stability very similar to bilateral pedicle screws and rods.14 Results of a clinical study found that the plate could be implanted in significantly less time and with less blood loss compared to pedicle screw placement through an open approach
CHAPTER 97
or a technique using a tubular retractor.13 Additionally, the implant did not result in an increased rate of pseudoarthrosis of the anterior interbody fusion graft. Percutaneous Osteoporotic Fracture Treatment One of the advances in recent years has been the methods for treating osteoporotic spinal fractures. However, there is potential for improvement in these treatments that will likely be addressed in the near future. Two issues are the density of the PMMA (polymethyl/methacrylate) that is currently used for the treatment and the possibility for the material to leak through fractures in the vertebrae. A strong, less dense injectate is desirable due to the potential for fracture of the adjacent vertebrae when the dense PMMA is injected. It may be possible to use materials similar to existing bone graft supplements to inject into the fractured body. This should help to reduce the problem of adjacent vertebral body fracture. Also emerging in the treatment arena for spinal fractures are technologies to re-establish the vertebral body height such as the systems from Crosstrees Medical (Boulder, CO) and SpineWave Inc. (Shelton, CT). Combined Motion-Preserving Technologies Just as there became a role for combined anterior/posterior fusion, there will likely be a role of combined anterior/posterior dynamic stabilization. A biomechanical study investigating such a combination has already been undertaken.15 A CHARITÉ device combined with the TOPS device posteriorly was compared to the intact spine. This combined construct provided good results. No clinical application of combined motion retaining technologies, however, has been found. Development and evaluation of combined disc and posterior arthroplasty systems are under way such as the Total Spinal Motion Segment system (Disc Motion Technologies, Boca Raton, FL), the flexible spine segment replacement system being developed by Flexuspine, Inc. (Tyler, TX). EMERGING NEW THERAPIES One of the most exciting areas for the treatment of back pain is the possibility of treating back pain using pharmacologic or biologic agents. As discussed earlier in this chapter, a better understanding of the disc and its role in pain is continually developing. From this knowledge, biologic and pharmacologic therapies may be possible. This area holds the promise of directly addressing chemically mediated pain as well as regenerating deteriorating disc tissue. This technology well may make many of the metallic implants obsolete. In the pharmacologic arena, therapies targeting pain caused by inflammatory mediators may be used. Some investigation in this area has already been undertaken for disc herniation. Clinical investigation using a TNF-a inhibitor originally yielded favorable results; however, results of a randomized trial found it provided no benefit over placebo.16,17 A study with a small series has been published on the perispinal injection of etanercept, a TNF-a inhibitor, with very favorable results.18 However, it should be noted that the authors included favorable response as an inclusion criteria for the study reported and therefore the number of
The Future of Motion Preservation
763
patients in whom the injection was beneficial versus not beneficial cannot be determined. Painful disc ruptures are associated with neovascularization in a zone of granulation tissue that forms along annular tears.4 Based on this finding, combined with the reported neurotropic effect of methylene blue, investigators conducted a prospective clinical trial to determine if it could be used to treat painful disc disruption. A series of 24 patients with chronic back pain but no radicular symptoms, and for whom interbody fusion was considered to be indicated based on positive discography, underwent methylene blue intradiscal injection. The authors reported significant improvement after the injections that was sustained during a minimum 12-month follow-up. There were no complications in the series. Further investigation in a large series of patients is needed to verify these encouraging findings. The intradiscal injection of steroids has also been investigated.20 The results of the randomized trial found that this treatment yielded no significant benefit. Other approaches have been described for the injection therapy of disc-related pain. Some of the effects of disc injury and degeneration are dehydration of the disc and reduced proteoglycans. To treat these effects, the injection of glucosamine and chondroitin sulfate combined with hypertonic dextrose and dimethyl sulfoxide (DMSO) has been investigated.21 This analysis was undertaken in a series of 30 patients who had a positive discogram. At a mean follow-up of 12 months, the overall group improved significantly based on pain and function questionnaires. However, the authors noted that 43% of patients had little or no improvement and the remaining 57% of patients had very good outcomes. Factors related to the poor outcomes were failed surgery, stenosis, and long-term disability. The authors also noted that patients had moderate to severe pain 48 to 72 hours following the injections. Biologics and Tissue Engineering A host of studies has discussed various potential ways to restore disc tissue. Some of these methods are based on the concept of preventing, or at least slowing the rate of, disc degeneration. Other possible treatments have the goal of either replacing or regenerating disc tissue. The following is an overview of some of the most recent reports related to the biologic therapy of the intervertebral disc. BMP has been shown to increase bony fusion rates in interbody fusion procedures. There is also a potential for BMP to result in enhanced disc tissue growth. In a recent study, apoptosis was induced by TNF-a or hydrogen peroxide in disc cells removed during surgery.22 The authors found among the cells that were pretreated with BMP, apoptosis was retarded, which may help guard against degeneration. There is interest in determining if osteogenic protein-1 (OP-1) can be used to regenerate disc tissue. In a bovine study, it was found that OP-1 significantly increased the proteoglycan content of cells from the disc nucleus, inner annulus, and outer annulus.23 Another strategy emerging for discal repair is providing a scaffold to guide and encourage restorative cell growth. This type of tissue restorative therapy was recently described in a study using a rabbit model.24 The authors found that using atelocollagen scaffolds, disc nucleus cells were responsive to growth factors.
764
P A R T
X I I
Controversies
This type of work lays the groundwork for further development of tissue engineering using scaffolds. Wilke and associates performed a biomechanical study to determine the feasibility of using a scaffold consisting of a collagen matrix seeded with cells to be placed into the nuclear cavity.25 They found that the implant restored disc space height, although there was a significant problem with device extrusion. The authors indicated that overcoming extrusion of implanted devices is one of the primary biomechanical challenges that must be overcome in order for the scaffold-based therapies to be optimized. They suggested that a means by which to seal the annulus may help reduce the problem of extrusion. One strategy for addressing painful disc degeneration following disc injury is to treat patients very early after rupture of the disc annulus. To investigate one such possible treatment, disc injury was induced in an ovine model.26 PMMA microspheres and collagen were then injected along the tear. Sacrifice of the animals revealed that the microspheres maintained their position and were encapsulated. The tear was healed with the growth of the new collagen and without severe inflammatory response. There was no evidence of neovascularization along the tear. While the treated discs went on to heal, the control disc showed signs of progressive degeneration. The authors concluded that the injection slowed or prevented the degeneration that typically occurs after disc injury and suggested that further investigation for early disc therapy was warranted. Based on the potential of stem cells to grow various tissue types, there has been an interest in spine applications. In a small clinical trial of 10 patients with discogenic pain, Haufe and Mork injected hematopoietic stem cells obtained from the patient's pelvis into the disc.27 The authors reported that this treatment was ineffective in reducing pain. Another therapy that has been evaluated in a clinical series is disc chondrocyte transplantation.28 Following success in animal studies, this treatment was undertaken in humans. Patients were randomized to discectomy alone versus discectomy combined with autologous disc chondrocyte transplantation into the disc. The transplantation required an injection performed after the chondrocytes were expended for 12 weeks. The patients were followed for 2 years. The authors reported that the pain relief was greater in the chondrocyte treated group compared to the discectomy only group. Also, disc hydration was maintained in the chondrocyte group, unlike the control group in which it decreased over time. The authors concluded that the treatment was feasible and prevented ongoing degeneration following discectomy. Another possible intradiscal therapy is the injection of growth factor into the disc. This has been studied using a rabbit model with favorable results reported.29 The authors found that an injection of OP-1 restored the biomechanical properties of the intervertebral disc. A very good update and review of potential molecular therapies for disc degeneration was recently published by Yoon and Patel.30 The authors described four categories of molecules currently being investigated for possible therapeutic application in intradiscal therapy. These included anticatabolic, mitogens, morphogen, and intracellular regulators. As the authors noted, in vitro data are available for these molecules, although few have been evaluated in animal models evaluating their potential as a treatment for disc problems.
As seen in other chapters in this book, there are promising new technologies for regeneration of disc tissue. Exactly which of these biologics will emerge as the best for painful disc disruption will require much investigation. It may be that some of the biologics will be combined with mechanical annular repair devices or even the interspinous devices. Such combined therapies would allow unloading of the disc, creating an environment that may maximize the chances for tissue regeneration. Unloading the Disc One well-accepted contributor to disc injury and degeneration has been the mechanical load put upon the disc. Although the disc is dependent upon load to stimulate the diffusion of nutrients into the disc, there is a point at which the disc may become overloaded, resulting in injury. It has long been known that during periods of load bearing the disc loses hydration, while during periods of rest, or unloading, the disc rehydrates and its height increases. The effect of loading and unloading the disc has been studied in animal models using external fixation to compress and unload the disc.31 The investigators found that distraction of the disc space resulted in rehydration, stimulated extracellular matrix gene expression, and increased in the number of protein-expressing cells. These data support that there may be a role for mechanical disc unloading in conjunction with cell-based intervertebral disc therapies to help create an environment favorable for stimulating cell growth. As previously discussed, some of the interspinous spacers may play a role in reducing the load on the intervertebral disc. Data concerning the unloading of the disc with dynamic stabilization devices are starting to emerge. Putzier and associates reported that Dynesys (Zimmer Spine, Inc., Warsaw, IN) had a protective effect on the disc following nucleotomy.32 It cannot be clearly determined if the benefit was related to disc unloading, controlled motion, or a combination of the two. CHALLENGES Small-scale early investigations of various potential therapeutics, as discussed earlier, are starting to provide insight into possible new treatments for back pain. However, the results need to be interpreted with care. For several of these studies, the patients undergoing treatment are very select and the results may not be achieved with less homogeneous patients. With medication as well as BMP applications for spinal fusions, the dosing may have a significant impact on the effectiveness of the treatments investigated. Much more research will be needed to determine which emerging therapies are beneficial and to refine the dosing. Like all technologies, biologics face many challenges. While the results from many studies conducted in laboratories using only cell cultures in extremely controlled conditions yield promising results, which one will emerge as effective in adult patients? What are the long-term safety concerns of some of the proposed gene-based therapies? There may also be a lot of work required to determine the window of opportunity with respect to timing the intervention in relation to pain onset. As described by Peng and associates,4,5,33,34 as the disc is injured and the degenerative process progresses, there are changes in the composition of the disc tissue. These changes
CHAPTER 97
include neovascularization and the formation of granulation tissue along the tears in the annulus. The long-term influence of these structural changes within the disc on efforts to regenerate disc tissue or for nuclear repair remains to be seen. Also, as attempts are made to encourage the disc to regenerate tissue, a question arises as to the quality of the tissue that will be regenerated. As described by Roberts and associates,35 some disc cells demonstrate senescence. This occurred much more frequently in herniated discs than in normal discs or discs related to discogenic pain. While this finding provides new insight into cell biology, as the authors discussed, it also has very important implications in tissue engineering and intradiscal gene therapy in that if such cells are used, the desired regeneration may not occur. This consequence is particularly important considering that the highest rate of senescence was found in herniated discs. It has been suggested that disc tissue removed surgically to relieve back pain could be used for tissue engineering; however, these findings suggest that such tissue may be less than optimal for therapeutic interventions. Getting cells to grow in ideal laboratory conditions is a first step, but the real question remains: Which will grow within the human body to an extent that will provide favorable clinical results? There are also many environmental factors to consider in attempting to regenerate disc tissue in back pain patients and relieve pain: Considering that the end plates are the conduit for nutrition to the disc, what is the role of end plate assessment, such as Modic changes, in determining if the environment will support new cell growth? If the end plates became too dense, either leading to degeneration in the first place or as a result of degeneration, can enough nutrition be transferred to the disc to sustain regenerative cell growth? What is the ideal biomechanical environment for new cells to grow within the disc? Too much load may compromise cell life, but some load is needed to stimulate tissue growth. Can the load within the disc be manipulated by devices such as intraspinous spacers or other implants to create the ideal biomechanical environment for new cell growth? In the future, is there a potential role for microelectromechanical systems (MEMS) to assess the load in the disc space to provide information concerning if the ideal environment has been achieved and maintained? If therapy involves regeneration of the existing disc tissue, is there a potential problem in that the factors that originally led to deterioration in the first place will result in the failure of the regenerated tissue as well? Will changes that take place following disc injury, such as neovascularization and the formation of granulation tissue, lead to failed attempts to regenerate healthy tissue and diminish the patient's pain? Or once these changes in the disc have occurred, is it too late to use biologics to treat disc-related symptoms? CURRENTLY EMERGING HIGH-TECH TREATMENTS Over the past several years, image-guided systems and robotics have been adapted to spine applications. A few articles have been published on the development of image-guided and robotic
The Future of Motion Preservation
765
systems.36–40 This technology was primarily introduced in routine spine surgery for pedicle screw placement, but potential applications for disc arthroplasty are being investigated. The imageguided systems have not yet gained widespread acceptance. One concern is the accuracy of the registration systems. Even slight misalignment has the potential to create significant malpositioning of the instrumentation. As with any electronic technology, each advance in development provides significant improvement over its predecessor. A remaining question is addressing the potential variation introduced by the patient positioning required for surgery on the accuracy of the registration when based on images with the patient lying supine in a scanner or when the relationship of the vertebral bodies has been altered during the surgical procedures, as occurs with distraction of the disc space. Perhaps the greatest challenge for these devices will be addressing the costeffectiveness of their utilization. If a specific image scanning protocol has to be employed to generate the images to be used with the registry system, this also increases the overall cost of these systems. The data needed will be related to demonstrating if the benefits gained by more accurate device placement potentially achieved by this technology can overcome its costs. Electronics and Other Downstream Technologies We are already seeing the incorporation of electronics into spinal devices such as the Theken disc. This type of microtechnology may be incorporated into other implants to provide valuable information related to the load demands and function of various implants. The next phase of electronics in spinal implants may be those that actually respond to changes in their environment or play a role in stimulating tissue or dispensing pharmacologic agents. These microelectronics and possibly nanotechnology may be programmed in such a way as to sense changes in their environment and respond to such. For example, a device may be programmed to sense an acute increase in neurotoxic agents being released from the disc and respond by releasing a predetermined dosage of anti-inflammatory medication. Other devices may also be programmed to respond to acute changes in load, that is, to trigger the implant to provide more resistance during activities that place greater loads on the spinal column and then become less rigid during rest times. Perhaps one of the applications of nanotechnology in the spine is for an extremely small device to be programmed to detect neurotoxic agents and destroy them. Microelectromechanical systems (MEMS) are extremely small “smart” devices. That is, they can be programmed to respond to changes in their environment, usually triggered by chemical or mechanical alterations. The devices include a circuit board and may incorporate motors, sensors, gears, pumps, and other items needed to perform the desired task. Several applications of MEMS technology in the spine have been described. These included measuring load or pressure within a bone graft or fusion cage.41 This information may provide insight into the biomechanics of bone healing and modeling in a fusion mass. Devices utilizing MEMS technology may also act on the information they collect. For example, if the device senses a pattern or trigger for pseudoarthrosis in a fusion mass, it may be programmed to
766
P A R T
X I I
Controversies
release BMP at a certain point to maximize the chances of achieving a bony union without any other surgical intervention. Aebersold described the design of a MEMS device to be used with posterior internal fixation combined with posterior fusion.42 The device incorporates a telemetry unit that records stresses on the fusion rods. The theory is that reduced stress on the rod is indicative of fusion mass incorporation. Such a device could provide a noninvasive means to assess fusion that is more reliable than currently used radiographs. Perhaps one of the most important potential applications of MEMS and nanotechnology is in the area of spinal cord injury. The knowledge base is increasing dramatically in the area of understanding the function and growth patterns of the spinal cord. It may be possible to use MEMS to stimulate and guide the growth of neural tissue in such a way as to repair or replace damaged tissue, thus restoring function to a person with a devastating spinal cord injury. Some of the potential challenges in the process of incorporating MEMS technology in medical applications have been discussed.41 The human body is a harsh environment for many materials, including those used in the design of MEMS devices. The humidity is high and the body often mounts a response, including inflammation and scarring, that may compromise the performance of these implants.
Many are much better educated than patients in the past. They now have access to the same abstracts, meeting proceedings, and often literature that previously had primarily been limited to care providers. The Internet also provides many sources of information on various web pages that provide good quality information on anatomy, various diagnoses, and treatment options. But for all the benefits, the Internet also provides the opportunity for misinformation and biased communication. There is no regulation for quality control on the web. This situation can have detrimental effects for patients seeking information upon which to make their decisions about treatment options. In addition, there is the potential for sales of unfounded therapies touted by “experts” and numerous testimonials. One responsibility care providers should take on is to help patients locate unbiased web pages. This assistance should help patients become better educated on their spine-related problem and the care options available to them. Computers have also made it possible to provide patients much better education in the office than what was previously available. In our clinic, for example, we have created several interactive Power Point educational programs with voice overlay. Such materials require time to produce but provide patients with visual as well as audio explanations that are easy to understand. The materials can also be produced in handouts or CDs for patients to take home to review the content or to share with their family members.
Patient Selection Although there is great excitement over developing new technologies, old facts must be considered. As always, patient selection is vital for good clinical outcome for any spine surgery, including motion-preserving technologies. As our diagnostic evaluations improve and more knowledge is gained concerning the role of chemical irritation of neural tissue, the role of facet and sacroiliac joints, as well as the disc itself, clinical outcomes will likely improve as well. One well-documented component of clinical outcome in spinal surgery is the strong role of psychological factors.43 Preoperative screening has been shown to be helpful in differentiating patients who do well following surgery from those who do not.44 Even with TDR, it has recently been found that the single factor most strongly related to the extremes of total disc replacement was the length of time off work prior to surgery. Factors that were not related were device placement, age, gender, level operated, body mass index, previous discectomy, as well as preoperative pain and function scores. More extensive understanding and evaluation of psychosocial factors should improve the chances of the full potential of new treatment technologies to be reached. Some of the less invasive biologic and pharmacologic interventions may have the added benefit of being administered early after pain onset. This timing may help to improve results by providing treatment before the patient lapses into a pattern of chronic pain behavior. As Technology Changes, so Do Patients Not only is technology changing the treatments available to back pain patients, it is changing our patients themselves. The ready access to the Internet has created a new generation of patients.
Limiting Factors Although the future of spine technology is exciting, there are also factors potentially limiting its fruition. Perhaps the greatest factors are cost, safety, and proven effectiveness. It is clear that new “addon technologies” such as image guidance combined with existing pedicle screw or TDR add cost to the procedures. The outstanding question remains: Is the new technology cost-effective? It is possible that such technology is cost-effective in subgroups of patients but not if used for all cases. The key will be to identify the subgroups in which it is cost-effective and avoid its routine use unless proven to be cost-effective for all cases. We can learn much from the history of other spine technologies. In the report by Bono and Lee, despite the enthusiasm over pedicle screws, there are not much data to support that they resulted in a significantly improved clinical outcome.46 Although the implants inevitably add to the cost of the surgery, the technology might not be to blame. As discussed here, responsible use and employment of well-defined selection criteria will be key for the benefit of the new treatments to be realized. In the past, FDA approval was typically enough for insurance approval for a procedure. Presently, however, the standards seem to have been raised with unclear guidance for the type and amount of data needed to gain widespread acceptance from insurers for reimbursement of new technologies. To address this problem as well as concerns over the rising cost of health care in general, there will be a new responsibility for manufacturers to investigate the costeffectiveness of new technologies. Although it is expeditious for manufacturers to get 510(k) approval from the U.S. Food and Drug Administration (FDA) to market their new implants, this leaves them without adequate data to demonstrate the effectiveness
CHAPTER 97
of their new products. The use of registries for new product use after FDA approval will likely continue to increase. Another important arena will be the precise matching of technology to patients. We have seen numerous times the very successful results of FDA-regulated or other initial studies of devices, only to be followed several years later by reports of failures or results no better than those reported prior to the new technology. This result is likely based on not following the stringent inclusion/exclusion criteria used in the early studies or the users of the technology not being as well trained as needed to be as successful as the early users. The keys to success for any of the emerging technologies will be physician training and employment of appropriate selection criteria. There are already growing concerns over the rapidly rising cost of health care. No palatable end to this trend is foreseeable. Considering the high cost of back pain, there is a tremendous need to control the cost of spine care. This appears to almost be in conflict with the inevitable costs associated with developing new technology. In spinal care, we are already seeing the emergence of quality assurance programs such as that being established by the NCQA (National Committee for Quality Assurance). This same group has already launched similar programs for cardiology and diabetes. Spine care providers are also becoming familiar with the concepts and terminology related to Pay-for-Performance. These new initiatives may help to drive what is needed to fully evaluate new technologies, that is, comprehensive data collection. This process will also be facilitated by the increasing use of electronic medical records and other computerized data capture systems. Regardless of how good a treatment is, a single one will not emerge as best due to the various origins of back pain. The real key to optimizing treatment will be the matching of the best treatment for a particular well-defined spinal problem. This goal is also the key to maximizing cost-effectiveness, which may be one of the greatest challenges facing the ongoing incorporation of new technologies into everyday spine practice. However, achieving the optimal matching of treatment to pain generators will require developing not only new treatments but also better diagnostics to more accurately identify the source(s) of each patient's symptoms. Responsible Use Although these are exciting times in spine surgery, technology must be embraced responsibly by all involved. The primary responsibility is with physicians, but manufacturers also play a significant role in responsibly introducing technology for clinical use. Physician users of technologies must be trained appropriately in technique as well as patient selection. With new initiatives being introduced, such as quality assurance programs for spine care providers, patient case registries for all providers, electronic patient data capture, and electronic medical records, there will be greater opportunity for care providers to monitor and report results, both successful and not successful, to other care providers. Through the use of registries and comprehensive data capture, it is likely that greater insight will be gained into indications and contraindications, complications, and the results of off-label use.
The Future of Motion Preservation
767
We have seen the high rates of fusion reported for BMP which has been approved by the FDA to be marketed for use in tapered interbody fusion cages. However, as a result of its success, many have been tempted to use it off-label for other applications. In the cervical spine, its use has been associated with several reports of increased incidence of difficulties swallowing in up to 42% of patients and consequently other complications.47,48 Naraghi and associates reported that the incidence of complications was significantly less when the dosage was reduced.47 They also suggested that fibrin glue used anteriorly and posteriorly may reduce the incidence of complications. As these physicians did, all physicians using devices off-label need to document and share their experiences, both good and bad, to improve patient care and to reduce the risk to patients when using materials for which the application has not been previously investigated. Physicians using new technologies also have a responsibility to report less than desirable results and problems encountered with its use. Only through this system of open reporting can problems be identified and addressed and, most importantly, patient safety optimized. Using the BMP and fusion experience again, there are now reports of problems in the lumbar spine that were not identified in the early studies using this graft material.49 Bone resorption defects in the vertebral bodies next to the intervertebral space where BMP was used were noted in 69% of the study group. Just as this problem, as well as those discussed concerning the use of BMP in the cervical spine, is emerging as use of this new technology expands, there is no doubt that as we move into an age of new implant materials and pharmacologic and biologic spinal therapies, they too will create unanticipated problems or events that did not occur in the clinical trials. Only through vigilant data collection by users and open and honest reporting can such problems be identified and addressed. As has been seen in recent years, there will be more effort to address concerns related to finances and the relationship between manufacturers, investors, and physicians. Cessation of these relationships will lead to slower development and acceptance of technologies. However, with the increasing number of reported ethics violations, the concern is certainly legitimate and needs to be addressed. The Internet has brought with it great capabilities for information exchange and education. However, it has also brought with it the possibility for patients to get erroneous information or develop unrealistic expectations based on the enthusiasm for new technologies. We have already seen this with patients interested in TDR. In their desperation to find relief, patients do not understand the finer details of selection criteria for new technology. Instead, they just want relief and have seen something about a new breakthrough in spine care. Only by making quality, unbiased educational material available to patients can we help them become better consumers. For years, web sites such as spine-health.com and spineuniverse.com have provided quality information to the public. Professional societies such as the North American Spine Society (NASS), American Academy of Orthopaedic Surgeons (AAOS), and Spine Arthroplasty Society (SAS) are increasingly making materials available to the public. CLOSING This is a time of great changes in spine surgery. Technology is growing rapidly with many promising treatments on the horizon
768
P A R T
X I I
Controversies
in an environment of changing health care ethics and reimbursement requirements. Patients are becoming more educated concerning their treatment options than ever before. The new technologies hold the promise of further developing minimally invasive surgery, motion-retaining devices, the repair or regeneration of degenerated tissue, and microelectronics that can monitor and respond to changing conditions in the spinal microenvironment. As with all exciting prospects, not all of the individual devices and treatments will evolve into successful therapies for clinical use. Some will fail biomechanical tests, and for others the promising results seen in ideal laboratory conditions will not be realized in humans. Also, enthusiasm for new treatments must be tempered with responsibility. The bar will be raised for the acceptance of new technologies. Some manufacturers will have to expand their clinical trials to not only prove the new technology not inferior to currently accepted care but also to prove its superiority. Also, cost-effectiveness must be addressed. This may mean that the spine market will be less attractive to investors, as the cost-effectiveness may require a lesser return on investment than what may be earned in other investment arenas. Physicians using the new technologies must also take on the responsibility of monitoring their results, particularly when using new treatments offlabel, and reporting any complications or undesirable events related to the use of such technologies. Issues related to the financing of new technology development and relationships between industry and physicians will have to be addressed to ensure as much impartiality as possible, while allowing fair sharing of profits from intellectual properties. Although there are biologic and social challenges to overcome, the future of spine medicine is bright with many emerging options that should offer reduced treatment-related morbidity, yield good results, and offer earlier effective interventions for patients.
REFERENCES 1. Burke JG, Watson RW, McCormack D, et al: Intervertebral discs which cause low back pain secrete high levels of proinflammatory mediators. J Bone Joint Surg Br 84:196–201, 2002. 2. Coppes MH, Marani E, Thomeer RT, et al: Innervation of “painful” lumbar discs. Spine 22:2342–2349, 1997. 3. Freemont AJ, Peacock TE, Goupille P, et al: Nerve ingrowth into diseased intervertebral disc in chronic back pain. Lancet 350:178– 181, 1997. 4. Peng B, Wu W, Hou S, et al: The pathogenesis of discogenic low back pain. J Bone Joint Surg Br 87:62–67, 2005. 5. Peng B, Hao J, Hou S, et al: Possible pathogenesis of painful intervertebral disc degeneration. Spine 31:560–566, 2006. 6. Peng B, Hou S, Shi Q, et al: The relationship between cartilage endplate calcification and disc degeneration: An experimental study. Chin Med J (Engl) 114:308–312, 2001. 7. Borras MC, Becerra L, Ploghaus A, et al: fMRI measurement of CNS responses to naloxone infusion and subsequent mild noxious thermal stimuli in healthy volunteers. J Neurophysiol 91:2723– 2733, 2004. 8. Lemaire JP, Carrier H, Sariali el H, et al: Clinical and radiological outcomes with the CHARITÉ artificial disc: A 10-year minimum follow-up. J Spinal Disord Tech 18:353–359, 2005. 9. Lee CK, Langrana NA, Parsons JR, et al: Development of a prosthetic intervertebral disc. Spine 16:S253–S255, 1991.
10. Klara PM, Ray CD: Artificial nucleus replacement: Clinical experience. Spine 27:1374–1377, 2002. 11. Ray CD: The PDN prosthetic disc-nucleus device. Eur Spine J 11 (2):S137–142, 2002. 12. Zucherman JF, Hsu KY, Hartjen CA, et al: A multicenter, prospective, randomized trial evaluating the X STOP interspinous process decompression system for the treatment of neurogenic intermittent claudication: Two-year follow-up results. Spine 30:1351–1358, 2005. 13. Wang JC, Haid RW Jr, Miller JS, et al: Comparison of CD HORIZON SPIRE spinous process plate stabilization and pedicle screw fixation after anterior lumbar interbody fusion. Invited submission from the Joint Section Meeting on Disorders of the Spine and Peripheral Nerves, March 2005. J Neurosurg Spine 4:132–136, 2006. 14. Wang JC, Spenciner D, Robinson JC: SPIRE spinous process stabilization plate: Biomechanical evaluation of a novel technology. Invited submission from the Joint Section Meeting on Disorders of the Spine and Peripheral Nerves, March 2005. J Neurosurg Spine 4:160–164, 2006. 15. Cunningham BC, Hu N, Beatson H, et al: Biomechanical evaluation of a posterior dynamic stabilization system combined with total disc replacement. International Meeting on Advance Spine Technologies. Athens, Greece, 2006. 16. Korhonen T, Karppinen J, Malmivaara A, et al: Efficacy of infliximab for disc herniation-induced sciatica: One-year follow-up. Spine 29:2115–2119, 2004. 17. Korhonen T, Karppinen J, Paimela L, et al: The treatment of disc herniation-induced sciatica with infliximab: Results of a randomized, controlled, 3-month follow-up study. Spine 30:2724–2728, 2005. 18. Tobinick EL, Britschgi-Davoodifar S: Perispinal TNF-alpha inhibition for discogenic pain. Swiss Med Wkly 133:170–177, 2003. 19. Peng B, Zhang Y, Hou S, et al: Intradiscal methylene blue injection for the treatment of chronic discogenic low back pain. Eur Spine J 16:33–38, 2007. 20. Simmons JW, McMillin JN, Emery SF, et al: Intradiscal steroids: A prospective double-blind clinical trial. Spine 17:S172–S175, 1992. 21. Klein RG, Eek BC, O'Neill CW, et al: Biochemical injection treatment for discogenic low back pain: A pilot study. Spine J 3:220– 226, 2003. 22. Wei A, Chung S, Brisby H, et al: Bone morphogenic protein rescues human intervertebral disc cells in vitro from apoptosis. Spine Arthroplasty Society. Montreal, Canada, 2006. 23. Zhang Y, An HS, Song S, et al: Growth factor osteogenic protein-1: Differing effects on cells from three distinct zones in the bovine intervertebral disc. Am J Phys Med Rehabil 83:515–521, 2004. 24. Lee K-I, Moon E-S, Kim H, et al: Tissue engineering of the intervertebral disc with cultured nucleus pulposus cells using atelocollagen scaffold and growth factors. Spine Arthroplasty Society. Montreal, Canada, 2006. 25. Wilke HJ, Heuer F, Neidlinger-Wilke C, et al: Is a collagen scaffold for a tissue engineered nucleus replacement capable of restoring disc height and stability in an animal model? Eur Spine J 15(15):433– 438, 2006. 26. Taylor W, Ozgur B: Biologic collagen agent repairs mid annular defects and prohibits early stage disc degeneration in the ovine model. Spine Arthroplasty Society. Montreal, Canada, 2006. 27. Haufe SM, Mork AR: Intradiscal injection of hematopoietic stem cells in an attempt to rejuvenate the intervertebral discs. Stem Cells Dev 15:136–137, 2006. 28. Meisel HJ, Siodla V, Ganey T, et al: Clinical experience in cell-based therapeutics: Disc chondrocyte transplantation: A treatment for degenerated or damaged intervertebral disc. Biomol Eng 24:5–21, 2007. 29. Miyamoto K, Masuda K, Kim J, et al: Intradiscal injections of osteogenic protein-1 restore the viscoelastic properties of degenerated intervertebral discs. North American Spine Society. Seattle, WA, 2006.
CHAPTER 97
30. Yoon ST, Patel NM: Molecular therapy of the intervertebral disc. Eur Spine J 15(15):379–388, 2006. 31. Guehring T, Omlor GW, Lorenz H, et al: Disc distraction shows evidence of regenerative potential in degenerated intervertebral discs as evaluated by protein expression, magnetic resonance imaging, and messenger ribonucleic acid expression analysis. Spine 31:1658– 1665, 2006. 32. Putzier M, Schneider SV, Funk JF, et al: The surgical treatment of the lumbar disc prolapse: Nucleotomy with additional transpedicular dynamic stabilization versus nucleotomy alone. Spine 30:E109–114, 2005. 33. Peng B, Hou S, Wu W, et al: The pathogenesis and clinical significance of a high-intensity zone (HIZ) of lumbar intervertebral disc on MR imaging in the patient with discogenic low back pain. Eur Spine J 15:583–587, 2006. 34. Peng B, Wu W, Li Z, et al: Chemical radiculitis. Pain 127:11–16, 2007. 35. Roberts S, Evans EH, Kletsas D, et al: Senescence in human intervertebral discs. Eur Spine J 15(15):312–316, 2006. 36. Holly LT, Bloch O, Johnson JP: Evaluation of registration techniques for spinal image guidance. J Neurosurg Spine 4:323–328, 2006. 37. Lieberman IH, Togawa D, Kayanja MM, et al: Bone-mounted miniature robotic guidance for pedicle screw and translaminar facet screw placement, Part I: Technical development and a test case result. Neurosurgery 59:641–650, 2006. 38. Richards PJ, Kurta IC, Jasani V, et al: Assessment of CAOS as a training model in spinal surgery: A randomised study. Eur Spine J 16:239–244, 2007. 39. Thomale UW, Kneissler M, Hein A, et al: A spine frame for intraoperative fixation to increase accuracy in spinal navigation and robotics. Comput Aided Surg 10:151–155, 2005.
The Future of Motion Preservation
769
40. Wolf A, Shoham M, Michael S, et al: Feasibility study of a mini, bone-attached, robotic system for spinal operations: Analysis and experiments. Spine 29:220–228, 2004. 41. Benzel E, Ferrara L, Roy S, et al: Micromachines in spine surgery. Spine 29:601–606, 2004. 42. Aebersold JMW: Design and development of a MEMS-based capacitive bending strain sensor and a biocompatible housing for a telemetric strain monitoring system. Department of Mechanical Engineering. Louisville, KY, University of Louisville, 2005. 43. Block AR, Gatchel RJ, Deardoff WW, et al: The Psychology of Spine Surgery. Washington, DC, American Psychological Association, 2003. 44. Block AR, Ohnmeiss DD, Guyer RD, et al: The use of presurgical psychological screening to predict the outcome of spine surgery. Spine J 1:274–282, 2001. 45. Siddiqui S, Guyer RD, Zigler JE, et al: Factors related to the 20 best and 20 worst 24-month outcomes of total disc replacement in prospective FDA-regulated trials. North American Spine Society. Seattle, WA, 2006. 46. Bono CM, Lee CK: Critical analysis of trends in fusion for degenerative disc disease over the past 20 years: Influence of technique on fusion rate and clinical outcome. Spine 29:455–463, 2004. 47. Naraghi F, Wolfer L, Lewis M, et al: BMP-2 dose correlates with increased postoperative edema and swallowing complications after cervical fusion. North American Spine Society. Seattle, WA, 2006. 48. Shields LB, Raque GH, Glassman SD, et al: Adverse effects associated with high-dose recombinant human bone morphogenetic protein-2 use in anterior cervical spine fusion. Spine 31:542–547, 2006. 49. McClellan JW, Mulconrey DS, Forbes RJ, et al: Vertebral bone resorption after transforaminal lumbar interbody fusion with bone morphogenetic protein (rhBMP-2). J Spinal Disord Tech 19:483–486, 2006.
Index
Note: Page numbers followed by f indicate figures; t indicate tables; b indicate boxes.
A AAV (adeno-associated virus), as vector for gene therapy, 674–675, 675f Abrasion, 53, 60f, 61 Abscess, psoas, after anterior exposure of lumbar spine, 153 AccuFlex system, 31 ACD (anterior cervical discectomy), for cervical myelopathy, 744 ACDF. (See Anterior cervical discectomy and fusion (ACDF). AcroFlex device, 15, 15f, 26 baboon model of, 68, 70f, 70t, 71f, 72f Active subsystem, of spinal stability, 476 Activ-L Artificial Disc, 346–352 clinical data on, 350, 351f design of, 346–351 chisel protection sleeve in, 346–347, 349f end plate fixation options in, 347f midline marker in, 346–347, 349f parallel distraction, trial implant, and size verification instrument in, 346–347, 348f polyethylene inlay in, 347f revision instruments in, 347, 350f size(s) in, 346, 348f spiked vs. keeled versions in, 347f translation and rotation equals mobilization in, 347f two-level, 351f disc space preparation for, 351 discussion of, 351–352 indications/contraindications for, 346 operative technique for, 351 positioning for, 351 scientific testing of, 347–351 dynamic fatigue, 348–349 particle characterization in, 349–350 static, 347–348 wear simulator, 349–350 ADCT. See Autologous disc chondrocyte transplant (ADCT). Adeno-associated virus (AAV), as vector for gene therapy, 674–675, 675f Adhesion, 53, 60, 60f Adjacent segment cervical arthroplasty, 751–759 discussion of, 754–758, 755f, 756f, 757f materials and methods for, 752, 752f, 753f results of, 752–754 in international study, 752–754, 753f, 754f in U.S. prospective randomized study, 752, 753t Adjacent segment degeneration and disease cervical arthroplasty for, 751–759 discussion of, 754–758, 755f, 756f, 757f
materials and methods for, 752, 752f, 753f results of, 752–754 in international study, 752–754, 753f, 754f in U.S. prospective randomized study, 752, 753t Dynesys Spinal System for, 465 epidemiology of, 751 IsoBar TTL Dynamic Rod Stabilization for, 483, 484, 488–489 pathophysiology of, 751 after spinal fusion, 118–125 cervical, 120–121, 121t defined, 118 fusion biomechanics and cervical, 119–120 lumbar, 120 lumbar, 121–123, 122t animal models of, 641t, 644 fusion biomechanics and, 120 spinal biomechanics and, 118–119, 119f, 120f Wallis Stabilization System for, 529–530, 530f, 532 AF. See Annulus fibrosis (AF). AFRS. See Anatomic Facet Replacement System (AFRS). Alby, Albert, 484 ALIF (anterior lumbar interbody fusion), mini-, positioning for, 155–156, 156f ALL (anterior longitudinal ligament) anatomy of, 37–39, 38f All-in-1 guide, for Maverick Total Disc Replacement, 356, 357f, 360 Alloy(s) cobalt-chromium, 58 material properties of, 55–59, 56t, 57t new, 59 stainless steel, 57 new, 59 tantalum, 59 titanium, 58–59 new, 59 zirconium, 59 ALPA. See Anterolateral transpsoatic approach (ALPA). Alumina ceramics, material properties of, 57 A-MAV version, of Maverick Total Disc Replacement, 354, 354f American Society for Testing and Materials (ASTM) guidelines, for biomechanical testing, 46–50 Anas acuta, as animal model for disc degeneration, 640, 641t Anatomic Facet Replacement System (AFRS), 577–580
advantages/disadvantages of, 580 clinical presentation and evaluation for, 578–579, 579f complications of, 579 description of, 578, 578f discussion of, 580 indications/contraindications for, 577–578 operative technique for, 579 and posterior lumbar arthroplasty, 735 postoperative care for, 579 rationale for, 577 scientific testing of, 578, 578f, 579f Anesthetic discogram, functional, 106–108 complications of, 108 results of, 108 technique of, 106–108, 107f, 108f Angle of influence (AOI), 37f measurement of, 695–696, 698f Angular motion, with posterior lumbar arthroplasty, 737 Animal model(s), 63–73 anatomic and kinematic considerations with of cervical spine, 64, 65f of lumbar spine, 65, 67f of Aquarelle hydrogel disc nucleus, 427–429 of disc degeneration, 637–648 with adjacent level lumbar fusion, 641t, 644 with annulotomy, 641t, 644–645 axial loading in, 641t, 643 categories of, 640–646, 641t chemically or genetically induced, 641t, 645–646 due to chemonucleolysis, 641t, 645 Chinese hamster as, 640, 641t discussion of, 646 dog as, 640–643, 641t due to end plate injury, 641t, 645 experimentally induced, 641t, 643–646 due to fibronectin fragment injection, 641t, 645 intervertebral disc anatomy and, 639 knockout mice as, 641t, 645–646 due to muscle stimulation, 641t, 645 naturally occurring, 640–643, 641t pintail mouse as, 640, 641t with postural change, 641t, 643 primates as, 641t, 643 rabbit as, 640 requirements for and selection of, 639–640 with resection of spinal process or facet joint, 641t, 643–644 sand rat as, 640, 641t due to smoking, 641t, 645 surgically or physically induced, 641t, 643–645 tail suspension, 641t, 643 torsional injury, 641t, 643
771
772
Index
Animal model(s) (Continued) experimental endpoints with, 64 importance of, 63–64 of lumbar disc arthroplasty, 68, 70f, 70t, 71f, 72f of lumbar dynamic posterior fixation, 68–72, 72f, 73f of lumbar nucleoplasty, 65–68, 69f, 70f of NUBAC disc arthroplasty device, 447–448, 448f of particulate wear debris, 72–73, 73f, 74f of PCM arthroplasty, 64–65, 66f, 67f regulations on, 63 Ankylosing spondylitis, as contraindication for cervical disc arthroplasty, 186 Annular defect, location of, 630 Annular repair Barricaid device for, 629–636 BioDisc Nucleus Pulposus Replacement for, 431–434 Inclose Surgical Mesh System for, 621–628 lumbar discectomy vs., 629–630 Annulotomy animal models of disc degeneration due to, 641t, 644–645 for lateral lumbar total disc replacement, 369–370, 370f Annulus fibrosis (AF) anatomy and physiology of, 5, 6, 21, 639 in functional spinal unit, 37f, 41 pathologic changes in, 6–7 poor healing of, 630 Anterior cervical approach, 39, 39f, 178–181 complications of, 287–294, 288t alignment- and stability-related, 288t, 294 approach- and decompression-related, 288t, 289 fixation technique–related, 288t, 290–293, 292f, 293f, 294f, 295f implantation technique–related, 288t, 289–290, 291f movement-related, 288t, 294 revision for (See Cervical revision) Anterior cervical discectomy (ACD) for cervical myelopathy, 744 for CerviCore Intervertebral Disc prosthesis, 241–242, 241f Anterior cervical discectomy and fusion (ACDF) adjacent segment degeneration after, 120–121, 121t Bryan artificial disc vs., 193 cervical arthroplasty adjacent to, 751–759 discussion of, 754–758, 755f, 756f, 757f materials and methods for, 752, 752f, 753f results of, 752–754 in international study, 752–754, 753f, 754f in U.S. prospective randomized study, 752, 753t for cervical myelopathy, 744 complications of, 287–294, 288t alignment- and stability-related, 288t, 294 approach- and decompression-related, 288t, 289 fixation technique–related, 288t, 290–293, 292f, 293f, 294f, 295f implantation technique–related, 288t, 289–290, 291f movement-related, 288t, 294 revision for (See Cervical revision) disadvantages of, 185 Anterior cervical spine, anatomy of, 39–40, 39f Anterior exposure, of lumbar spine, 139–147 advantages of, 146–147 comment on, 146–147 complications of, 147, 148–154, 149t miscellaneous, 149t, 153 neurologic, 149t, 151–153, 152f
rate of, 149t urologic, 149t, 150, 151f vascular, 148–150, 149f, 149t, 150f entry into retroperitoneal space in, 141–142, 143f, 144f, 145f mobilization of blood vessels in, 142–146, 145f, 146f other technical considerations in, 146 patient preparation for, 141 skin incision for, 141, 142f Anterior longitudinal ligament (ALL) anatomy of, 37–39, 38f Anterior lumbar arthroplasty, posterior vs., 740, 740t Anterior lumbar interbody fusion (ALIF), mini-, positioning for, 155–156, 156f Anterior lumbar spine anatomy of, 41–43, 42f, 43f retroperitoneal approach to, 139–147 advantages of, 146–147 anatomy of, 41, 42f comment on, 146–147 complications of, 147, 148–154, 149t miscellaneous, 149t, 153 neurologic, 149t, 151–153, 152f rate of, 149t urologic, 149t, 150, 151f vascular, 148–150, 149f, 149t, 150f entry into retroperitoneal space in, 141–142, 143f, 144f, 145f mobilization of blood vessels in, 142–146, 145f, 146f other technical considerations in, 146 patient preparation for, 141 skin incision for, 141, 142f Anterior pseudomeningocele, after anterior exposure of lumbar spine, 153 Anterior retroperitoneal approach (ARPA), for HydraFlex device, 409 Anterolateral transpsoatic approach (ALPA), to lumbar spine, 155–158 advantages of, 158 annulus incision and annulotomy in, 156, 158f complications of, 158 incision and disc exposure in, 155–156, 157f, 158f indications for, 155 limitation of, 155 nucleus excision and wound closure in, 157, 158f positioning for, 155–157, 156f, 157f AOI (angle of influence), 37f measurement of, 695–696, 698f AOI (arc of influence), 37f measurement of, 695–696, 698f Aortic arch, in cervical revision, 278–280, 280f Aquarelle hydrogel disc nucleus, 423–430 advantages/disadvantages of, 429b clinical presentation and evaluation for, 429 complications of, 429 description of, 423–425, 424f design of, 424–425 development of, 423 discussion of, 429–430 function of, 423–424, 424f indications/contraindications for, 424 operative techniques for, 429 postoperative care for, 429 scientific testing of, 425–429 animal studies in, 427–429 for biocompatibility, 425 for biomechanics, 426–427 for fatigue, 427, 427f for swelling pressure, 425–426, 426f Arc of influence (AOI), 37f measurement of, 695–696, 698f
ARPA (anterior retroperitoneal approach), for HydraFlex device, 409 Arterial thrombosis, due to anterior exposure of lumbar spine, 149–150 Arthrodesis arthroplasty vs., 11–12 Cosmic device for extension of existing, 491, 494f Cosmic device in combination with, 491, 493f Articulating plate devices, 13–14, 14f Artificial cervical disc. See Cervical disc arthroplasty. Artificial lumbar disc. See Lumbar disc arthroplasty. Ascending lumbar vein, in anterior exposure of lumbar spine, 143–146, 146f, 149, 150f ASTM (American Society for Testing and Materials) guidelines, for biomechanical testing, 46–50 Asymptomatic subjects, abnormal MRI findings in, 97–98, 98f, 98t Autologous disc chondrocyte transplant (ADCT), 680–683, 764 case study of, 680, 682f clinical presentation and evaluation of, 681–682, 681f, 682f complications of, 682 discussion of, 682–683 indications/contraindications for, 680 operative techniques for, 680–681 postoperative care for, 681 rationale for, 680, 681f Axial compression bending, of Orthobiom Spinal System dynamic, 713 static, 713 Axial loading animal models of disc degeneration due to, 641t, 643 of Orthobiom Spinal System, 713 Axial low back pain, posterior lumbar arthroplasty for, 736 Axial neck pain, after cervical laminoplasty, 600 Axial rotation, constraints to, 695, 697t AXIENT Dynamic Fixation System, 31–32, 32f AXIENT TOTAL Dynamic Fixation System, 31–32
B Baboon model of Aquarelle hydrogel disc nucleus, 427–429 of disc degeneration, 641t, 643 of lumbar disc arthroplasty, 68, 70f, 70t, 71f, 72f of lumbar dynamic posterior fixation, 68–72, 72f, 73f of lumbar nucleoplasty, 65–68, 69f, 70f of lumbar spine, 65, 67f of NUBAC disc arthroplasty device, 447–448, 448f Back pain biomechanics of, 761–762 discogenic, 98 concordant vs. nonconcordant, 104t definite, 103 indeterminate, 104 mechanisms of, 761 probable, 104 unequivocal, 98 Wallis Stabilization System for, 526–527, 528f, 529f facet joint–mediated, 98 with flexion vs. extension, 97–98 growth factor therapy and, 659 inflammatory mediators of, 761 low axial, posterior lumbar arthroplasty for, 736 costs of, 132, 135
Index Back pain, low (Continued) PercuDyn device for, 511 societal costs of, 662 after lumbar total disc arthroplasty, 389–394 due to disease progression, 392–393 early-onset, 393 due to failure of prosthesis, 390–392 due to failure to achieve surgical goals, 390 due to incorrect procedure, 389–390, 391f late-onset, 394 due to poor patient selection, 389 subacute, 393–394 due to technique-related problems, 390, 392f, 393f Baguera prosthesis, 24, 24f Ball bearing devices, 12, 12f Ball-and-socket devices, 13, 13f Barricaid device, 629–636 clinical data on, 634–635 for disc height maintenance, 634–635, 635f from initial postmarketing study, 634, 634f for symptomatic recurrent herniation, 634, 634f, 635 description of, 630–631, 631f, 632f design of, 630 discussion of, 635 indications/contraindications for, 635 operative technique for, 631 preclinical testing of, 631–634 benchtop, 633 on cadavers, 632, 632f, 633 for expulsion from disc space, 631–633 for fatigue, 633 for intradiscal pressure changes, 633 for vertebral body movement, 633 rationale for, 629–630 Beta-titanium (beta-Ti) alloys, 59 Biocompatibility, 54 of Aquarelle hydrogel disc nucleus, 425, 429 of DIAM system, 519, 521t of NUBAC disc arthroplasty device, 444 of TranS1 Percutaneous Nucleus Replacement, 437 BioDisc Nucleus Pulposus Replacement, 431–434 advantages/disadvantages of, 434b clinical presentation and evaluation for, 433, 434f complications of, 433 description of, 431–432, 432f discussion of, 433 indications/contraindications for, 431 operative techniques for, 433, 433f postoperative care for, 433 scientific testing/clinical outcomes of, 432–433, 433f BioDur 109, 59 Biodurability, of NUBAC disc arthroplasty device, 444 BioFlex System, 31 Biologic devices, 18–19 Biologic therapy, for disc degeneration, 763–764 Biomechanical model, 685–694 basic biomechanics of, 688 basic structure of, 688–690, 689f complexity of, 687 discussion of, 693–694 electromyography in, 688–689, 689f, 693f enhancements and improvements in, 691–693 finite element analysis in, 691 flexible discs in, 693f importation of subject-specific images into, 688, 693, 693f MSC.ADAMS software in, 691, 692f, 692t output measures of, 692t performance and validation of, 690, 690t, 691f realism of, 687 repeatability and variability of, 690–691 sensitivity to nonphysical stress of, 691
skeletal representation of, 692f uses of, 687 "virtual surgery" with, 691–693, 693f Biomechanical testing, 45–51 of Aquarelle hydrogel disc nucleus, 426–427 ASTM guidelines for, 46–50 challenges of, 46 clinical relevance of, 46 of coflex implant, 537 configurations for, 48–49, 49f and continuous spectrum of stability, 48, 48f experimental designs for, 48, 48f history of, 45–46 of IsoBar TTL Dynamic Rod Stabilization, 484–485, 485f loading and displacement protocols in, 46, 47, 47f, 48 measurement parameters for, 50 of NUBAC disc arthroplasty device, 445–446 range of motion in, 46, 47f, 48, 50 for rotational stability, 697–701, 701f, 703f for spinal instability, 46 standardized, 46–50, 47f of TranS1 Percutaneous Nucleus Replacement, 437 Biomechanics basic, 688 fusion, and adjacent level degeneration cervical, 119–120 lumbar, 120 and future of motion preservation, 761–762 spinal, and adjacent level degeneration, 118–119, 119f, 120f Body, in tribologic system, 59, 60f Bone mineral density (BMD), lumbar fusion and, 618 Bone morphogenetic protein (BMP), for disc regeneration, 763, 767 Bone morphogenetic protein-2 (BMP-2), for disc regeneration, 650, 675–676, 676f Bone morphogenetic protein-7 (BMP-7), for disc regeneration, 676, 676f, 677f after chemonucleolysis, 655–657, 656f in vivo studies of, 651–655, 653f, 654f, 655f Bone scintigraphy, for lumbar nonfusion surgery, 77 Bonit coating, of Cosmic screw, 491–492 Bovine serum albumin (BSA), in BioDisc Nucleus Pulposus Replacement, 431–432 Boyd, Larry, 354 Bryan and Kunzler screw-in prosthesis, 16, 16f Bryan Cervical Disc, 26, 193–198 with cervical myelopathy, 748, 749 complication(s) of, 193–194 heterotopic ossification as, 299f description of, 193, 194f discussion of, 197–198 indications/contraindications for, 193 operative technique for, 195–196 confirmation of placement in, 196, 197f disc insertion in, 194f, 196, 196f, 197f end plate preparation in, 195–196, 196f patient positioning for, 186–188, 195 persistent pain after, 299f, 301f postoperative care for, 197 preoperative assessment for, 195 scientific testing/clinical outcomes of, 193–195, 195f BSA (bovine serum albumin), in BioDisc Nucleus Pulposus Replacement, 431–432 Bullivant design, 15 Buttner-Janz, Karin, 14, 309
C C-ABC (chondroitinase ABC), for chemonucleolysis, 655–656, 656f CAdisc-C prosthesis, 27
773
CAdisc-L prosthesis, 27, 27f Canine models, for disc degeneration, 640–643, 641t, 645 Caprine model of cervical spine, 64, 65f of PCM arthroplasty, 64–65, 66f, 67f, 203–204, 207f Carotid sheath, in anterior cervical approach, 39, 39f Carotid tubercle, in cervical revision, 278 Caspar distractor, with CerviCore Intervertebral Disc prosthesis, 241, 241f Cauthen screw-in dowel design, 16, 16f Cauthen slit-and-rib design, 13–14, 14f CD HORIZON AGILE Dynamic Stabilization Device, 31, 32f CD HORIZON LEGACY PEEK Rod, 31 Cell transplantation therapy, 662–672 concerns in interpreting animal models of, 670 with mesenchymal stem cells in vivo experiment with, 667–670, 668f, 669f sources of cells for, 664–665, 664f with nucleus pulposus cells in vitro study of, 665, 665f, 666f in vivo experiment with, 665–667, 667f sources of cells for, 663, 664 obstacles to clinical application of, 670 potential cell sources for, 663–665, 664f rationale for, 662 safety issues with, 670 therapeutic scheme for, 662–663, 663f Center of rotation (COR) and coupled motion, 119, 120f of Maverick Total Disc Replacement, 355, 355f, 356 in measurement of rotational stability, 695–696, 698f patient position and patient effort and, 90t, 92t reference data for, 90–93, 93f Centurion disc, 338 Ceramics, material properties of, 57 CerPass device, 254–257 advantages/disadvantages of, 257b clinical presentation and evaluation for, 256 complications of, 257 description of, 255, 255f discussion of, 257 indications/contraindications for, 255 operative technique for, 256–257 postoperative care for, 257 rationale and design criteria for, 254–255 scientific testing/clinical outcomes of, 255–256, 256f Cervical adjacent segment degeneration, after spinal fusion, 120–121, 121t Cervical approach(es) anterior, 39, 39f, 178–181 posterior, 182 Cervical arthroplasty. See Cervical total disc arthroplasty. Cervical artificial disc replacement. See Cervical total disc arthroplasty. Cervical coupled motion, 118, 119f Cervical disc arthroplasty. See Cervical total disc arthroplasty. Cervical disc herniation, Prestige Cervical Disc for, 199 Cervical disc replacement. See Cervical total disc arthroplasty. Cervical facets, 566 Cervical fusion, adjacent segment degeneration after, 120–121, 121t cervical arthroplasty for, 751–759 discussion of, 754–758, 755f, 756f, 757f materials and methods for, 752, 752f, 753f results of, 752–754, 753f, 753t, 754f fusion biomechanics and, 119–120
774
Index
Cervical kyphosis after cervical laminoplasty, 600 and cervical revision, 278–280, 281f, 293f, 294 as contraindication to cervical nonfusion surgery, 83f persistent pain due to, 297–298, 298f Cervical laminectomy cervical laminoplasty vs., 600–603 for cervical myelopathy, 743–744 Cervical laminoplasty, 593–603 biomechanical studies of, 598 case report on, 601–603, 602f cervical disc replacement with, 601 for cervical myelopathy, 744 clinical outcomes of, 599 development of, 595–596 discussion of, 600–603 French door (double-door, midline opening) case study of, 601–603, 602f Kurokowa modification of, 597 outcome of, 599 technique of, 597, 598f Tomita modification of, 597, 598f, 599f Hirabayashi open-door outcome of, 599 technique of, 597, 597f, 598f vs. laminectomy and fusion, 600 vs. multilevel anterior corpectomy and discectomy, 600 postoperative complications of, 600 range of motion in, 600–601 vs. in cervical arthroplasty, 601 rationale for, 595 techniques of, 596–598 Z-plasty, 597 Cervical lordosis, loss of, after cervical laminoplasty, 600 Cervical myelopathy Bryan artificial disc for, 193 cervical disc arthroplasty with, 185–186, 742–750 clinical studies of, 748–749, 749t indications for, 81, 82 recommendations on, 749 theoretical considerations for, 744–748, 745f, 746f, 747f, 748f cervical laminoplasty for, 595 natural history and conservative management of, 743 NeoDisc device for, 228f, 229f pathophysiology of, 742–743 presentation of, 742 Prestige Cervical Disc for, 199 surgical management of anterior approaches for, 744 background of, 742, 743 posterior approaches for, 743–744 Cervical nonfusion surgery contraindications for, 82–83, 83f, 83t, 85f indications for, 82, 82t patient evaluation for, 80–84 diagnostic studies in, 82 history in, 80–81, 81f physical evaluation in, 81–82, 82t Cervical radiculopathy Bryan artificial disc for, 193, 194f cervical disc arthroplasty for, 185–186 NeoDisc device for, 228f Prestige Cervical Disc for, 199 Cervical revision, 277–296 carotid tubercle in, 278 with cervical kyphosis, 278–280, 281f common mistakes with, 282 complications of, 284, 284f, 285f dislocation of temporomandibular joint in, 282, 283f
esophagus in, 280–282, 282f fascial planes in, 278 Horner's syndrome due to, 282, 283f hypoglossal nerve in, 280, 281f incision for, 277, 279f indications for, 287–288, 288t alignment- and stability-related, 288t, 294 approach- and decompression-related, 288t, 289 fixation technique–related, 288t, 290–293, 292f, 293f, 294f, 295f implantation technique–related, 288t, 289–290, 291f movement-related, 288t, 294 marginal mandibular branch of facial nerve in, 278, 278f, 280f motion preservation with, 282, 284f operative technique for, 277–283 positioning for, 277, 278f postoperative management of, 284 radiographic imaging during, 277 range of motion with, 290, 292f recurrent laryngeal nerves in, 278 structures of concern with, 278f superior laryngeal nerve in, 282 trial range of motion with, 283 vascular pitfalls of, 278–280, 280f vertebral arteries in, 282, 283f Cervical spine anatomy of, 178 animal models of, 64, 65f constraints to axial rotation in, 695, 697t extensive immobility of, as contraindication to cervical nonfusion surgery, 83f Cervical spondylotic myelopathy (CSM) cervical disc arthroplasty with, 185–186, 742–750 clinical studies of, 748–749, 749t recommendations on, 749 theoretical considerations for, 744–748, 745f, 746f, 747f, 748f cervical laminoplasty for, 595 natural history and conservative management of, 743 pathophysiology of, 742–743 presentation of, 742 surgical management of, 743 anterior approaches for, 744 posterior approaches for, 743–744 Cervical total disc arthroplasty adjacent segment degeneration after, 121 adjacent to fusion, 751–759 discussion of, 754–758, 755f, 756f, 757f materials and methods for, 752, 752f, 753f results of, 752–754 in international study, 752–754, 753f, 754f in U.S. prospective randomized study, 752, 753t anterior approach for, 178–181 Bryan artificial disc for, 193–198 CerPass device for, 254–257 with cervical laminoplasty, 601 case study on, 601–603, 602f with cervical myelopathy, 185–186, 742–750 clinical studies of, 748–749, 749t recommendations on, 749 theoretical considerations for, 744–748, 745f, 746f, 747f, 748f CerviCore Intervertebral Disc prosthesis for, 238–246 closure for, 180, 181f complications of, 287–288, 288t, 289–294 alignment- and stability-related, 288t, 294 approach- and decompression-related, 288t, 289 fixation technique–related, 288t, 290–293, 292f, 293f, 294f, 295f
implantation technique–related, 288t, 289–290, 291f movement-related, 288t, 294 revision for (See Cervical revision) contraindications for, 186, 186t decompression in, 179, 189, 190f, 191f disc placement in, 179–180, 180f two-level, 179–180, 181f disc space preparation in, 179, 179f, 188–189, 188f DISCOVER Artificial Cervical Disc for, 267–271 distraction pin placement in, 178–179, 179f exposure in, 188 goals of, 185 goat model of, 64–65, 66f, 67f incision for, 178–179 indications for, 185–186, 186t, 287–288 keel cutting in, 179–180, 180f key points for, 179t Kineflex|C Cervical Disc for, 258–266 M6 Artificial Cervical Disc for, 272–276 midline verification in, 179–180, 180f, 188, 188f Mobi-C for, 231–237 NeoDisc device for, 221–230 outcome study of, 288–289 discussion of, 294–296 materials and methods for, 288 patient selection for, 288, 289f results of, 289–294 patient positioning for, 178–179, 179f, 186–188, 187f persistent pain after, 297–302 due to device failure, 298–301, 301f due to facet disease, 298 due to heterotopic ossification, 298f, 299f, 300 due to inadequate decompression, 297 due to kyphotic deformity, 297–298, 298f Porous Coated Motion device for, 202–213 posterior approach for, 182 Prestige Cervical Disc for, 199–201 ProDisc-C for, 214–220 radiographic work-up for, 185–186 range of motion in, 601 revision of (See Cervical revision) Secure-C device for, 247–253 sizing in, 179–180, 180f, 181f, 189 unconstrained (See Porous Coated Motion [PCM] arthroplasty) Cervical vertebrae, anatomy of, 40–41, 40f, 41f Cervical vertebral bodies anatomy of, 40, 40f height and width of, 41 CerviCore Intervertebral Disc prosthesis, 238–246 complications of, 245 contraindications for, 238 description of, 24–25, 238–240, 239f discussion of, 245–246 indications for, 238 operative technique for, 240–243 device insertion in, 243, 243f disc space distraction and discectomy in, 241–242, 241f fin track preparation in, 242–243, 243f identification and marking of midline in, 240–241, 240f, 241f initial endplate preparation in, 242, 242f insertion of reference pins in, 241, 241f positioning and surgical approach/exposure in, 240 sizing in, 242, 242f postoperative care and rehabilitation with, 243–245, 244f removal of, 245 scientific testing/clinical outcomes of, 240
Index CHARITÉ Artificial Disc, 309–317 baboon model of, 68, 70f, 70t, 71f, 72f clinical trials of, 126–131 discussion of, 128–130 introduction to, 126–127 materials and methods for, 127 results of, 127–128, 129f, 617, 618f statistical methods for, 127, 128f complication(s) of, 313–316 due to device migration, 313 subsidence into vertebral column as, 315 due to wear debris, 313 description of, 14, 14f, 23, 23f, 310–311, 310f discussion of, 316 effect on facet joints of, 315–316 in hybrid surgery, 316 indications/contraindications for, 309–310, 316 lateral revision of due to malpositioning, 384–385, 385f due to pars fracture, 383–384, 384f, 385f long-range performance of, 315 operative technique for, 313, 314f postoperative care for, 313 with previous abdominal surgery, 310 with previous spinal surgery, 310 range of motion with, 311, 312, 312t rationale and development of, 309 revision of, 380 rotational testing of, 700, 701f scientific testing/clinical outcomes of, 311–312 titanium calcium phosphate–coated (See Porous Coated Motion [PCM] arthroplasty) CHARITÉ Evolution, 23 Chassaignac's tubercle, in cervical revision, 278 Chemonucleolysis animal models of disc degeneration due to, 641t, 645 regeneration of degenerated discs after, 655–657, 656f Chinese hamster model, for disc degeneration, 640, 641t Chondrocyte transplant, 680–683, 764 case study of, 680, 682f clinical presentation and evaluation of, 681–682, 681f, 682f complications of, 682 discussion of, 682–683 indications/contraindications for, 680 operative techniques for, 680–681 postoperative care for, 681 rationale for, 680, 681f Chondroitin sulfate (CS) and hydrogen disc nucleus, 423 intradiscal injection of, 763 Chondroitinase ABC (C-ABC), for chemonucleolysis, 655–656, 656f Cigarette smoking, animal models of disc degeneration due to, 641t, 645 Circumferential lumbar spinal stenosis, 720, 721f Claudicatio spinalis, Cosmic device for, 491, 492f Claudication, neurogenic Coflex device for, 534, 535 X-STOP for, 541 Cleveland, David, 12, 452 Clinical trials, of lumbar artificial disc, 126–131 discussion of, 128–130 introduction to, 126–127 materials and methods for, 127 results of, 127–128, 129f, 617, 618f statistical methods for, 127, 128f Cobalt-chromium (CoCr) alloys, material properties of, 58 Cobb method, of intervertebral motion measurement, 87 Coculture system, for nucleus pulposus and mesenchymal stem cells, 665, 665f, 666f
coflex implant, 534–540 clinical presentation and evaluation for, 538–539, 538t contraindications for, 535–536 description of, 33, 33f, 534–535, 535f indications for, 535 NeuDisc plus, 610, 610f operative technique for, 539–540, 540f with ProDisc-L prosthesis, 606, 607f rationale for, 534 scientific testing/clinical outcomes of, 536–538 biomechanical wing test in, 537 compression testing in, 536 finite element analysis and mechanical evaluation of physiologic loading in, 536–537 in vivo deflection and loading analysis in, 538 in vivo loading environment in, 537–538 mechanical testing in, 536, 539f for range of motion, 537 torsional analysis (wing bending) in, 536 wear characteristics in, 537 COL2a1 knockout mice, as animal model for disc degeneration, 641t Combined motion-preserving technologies, future of, 763 Common iliac arteries, 41, 42f in anterior exposure of lumbar spine, 142–143, 145f, 149, 149f Common iliac veins, in anterior exposure of lumbar spine, 143, 145f, 149, 149f Complex mechanical devices, 17, 17f Compression testing, of coflex implant, 536 Compression trabeculae, 4–5, 5f Computed tomography (CT) scans, for lumbar nonfusion surgery, 77 Computer-assisted techniques, for intervertebral motion measurement, 87–88, 88f Connectors, for posterior dynamic stabilization, 30–32, 30t with combination metal and plastic rods, 31–32 in AXIENT Dynamic Fixation System, 31–32, 32f in CD HORIZON AGILE Dynamic Stabilization Device, 31, 32f in Isobar TTL Dynamic Rod, 31, 31f in NFlex system, 31, 31f in Stabilimax NZ system, 32 with metal rods, 31 with plastic rods, 30–31, 31f Constrained cervical prosthesis, 267–268 Constrained lumbar prosthesis. See FlexiCore Intervertebral Disc. Contralateral herniation, and annular repair, 630 COR. See Center of rotation (COR). Corrosion, 54–55 factors controlling, 54 kinetic barriers to, 54–55 resistance to, 53 thermodynamics of, 54 Cortoss, in osteoporotic patients, 721–722, 723f Cosmic system, 490–499 advantages/disadvantages of, 496 complications with, 495–496 contraindications for, 491 description of, 491–493 angle of screw fixation in, 493, 496f flexion and extension in, 492–493, 496f rod in, 495, 498f screw coating in, 491–492 screws in, 30, 30f, 491, 495f discussion of, 496–498 indication(s) for, 491 in combination with arthrodesis, 491, 493f degenerative disc disease as, 491, 493f
775
as extension of existing arthrodesis, 491, 494f recurrent disc herniation as, 491 symptomatic lumbar stenosis as, 491, 492f operative technique for, 495, 497f, 498f postoperative care with, 495 rationale for, 490–491 scientific testing/clinical outcomes of, 493–495 Cost(s) of low back pain, 132, 135 of motion preservation surgery, 134–135 Cost effectiveness of motion preservation technology, 132–138, 766–767 discussion of, 136–137 future trends in, 135–136 issues in measuring, 132–134 ways to improve, 134–135 of X-STOP, 134, 547 Counterbody, in tribologic system, 59, 60f Coupled motion center of rotation and, 119, 120f cervical, 118, 119f lumbar, 118, 119f Coupling, of motion preservation device, 7, 7f Crane principle, and DIAM system, 520 Creep, of polymers, 55 Crevice corrosion, 53 Cribriform cartilage plate, 4–5, 4f, 5f Cricetulus griseus, as animal model for disc degeneration, 640, 641t Critical load, of spine, 4 Cross-sectional area, of muscle, 689 CS (chondroitin sulfate) and hydrogen disc nucleus, 423 intradiscal injection of, 763 CSM. See Cervical spondylotic myelopathy (CSM). CT (computed tomography) scans, for lumbar nonfusion surgery, 77 Cummins, Brian, 199 Cyst, facet (synovial), herniated nucleus pulposus and contralateral, 615
D Dampener, in IsoBar TTL Dynamic Rod Stabilization, 484, 485–486 Damping, 22 DASCOR device, 395–406 anesthesia for, 402 baboon model of, 65–68, 69f, 70f clinical presentation and evaluation for, 402 complications of, 405–406 description of, 29, 30f, 398, 398f, 399f discussion of, 406 indications/contraindications for, 397–398 operative techniques for, 402–405, 404f, 405f positioning for, 402 postoperative care for, 405 scientific testing/clinical outcomes of, 398–402 animal model and biomaterials safety testing in, 400–401 of axial compressive fatigue strength, 399, 400f of biological response to wear debris, 401 biomechanical testing in, 400, 400f carcinogenicity testing in, 401 of durability and particle analysis, 399 of functional biodurability, 401–402 mechanical bench testing in, 399 David parallel distractor, in lumbar disc arthroplasty, 307, 308f Decompression, inadequate, in cervical arthroplasty, 297 Decompressive laminectomy, X-STOP vs., 545, 545t Deep venous thrombosis, due to anterior exposure of lumbar spine, 150
776
Index
Deflection analysis, of coflex implant, 538 Degenerative disc disease (DDD) adjacent segment (See Adjacent segment degeneration and disease) animal models for, 637–648 with adjacent level lumbar fusion, 641t, 644 with annulotomy, 641t, 644–645 axial loading in, 641t, 643 categories of, 640–646, 641t chemically or genetically induced, 641t, 645–646 due to chemonucleolysis, 641t, 645 Chinese hamster as, 640, 641t discussion of, 646 dog as, 640–643, 641t due to end plate injury, 641t, 645 experimentally induced, 641t, 643–646 due to fibronectin fragment injection, 641t, 645 intervertebral disc anatomy and, 639 knockout mice as, 641t, 645–646 due to muscle stimulation, 641t, 645 naturally occurring, 640–643, 641t pintail mouse as, 640, 641t with postural change, 641t, 643 primates as, 641t, 643 rabbit as, 640 requirements for and selection of, 639–640 with resection of spinal process or facet joint, 641t, 643–644 sand rat as, 640, 641t due to smoking, 641t, 645 surgically or physically induced, 641t, 643–645 tail suspension, 641t, 643 torsional injury, 641t, 643 Aquarelle hydrogel disc nucleus for, 465–467 autologous disc chondrocyte transplant for, 680–683, 764 case study of, 680, 682f clinical presentation and evaluation of, 681–682, 681f, 682f complications of, 682 discussion of, 682–683 indications/contraindications for, 680 operative techniques for, 680–681 postoperative care for, 681 rationale for, 680, 681f biologic changes during, 662 biologics and tissue engineering for, 763–764 cell transplantation therapy for, 662–672 concerns in interpreting animal models of, 670 with mesenchymal stem cells in vivo experiment with, 667–670, 668f, 669f sources of cells for, 664–665, 664f with nucleus pulposus cells in vitro study of, 665, 665f, 666f in vivo experiment with, 665–667, 667f sources of cells for, 663, 664 obstacles to clinical application of, 670 potential cell sources for, 663–665, 664f rationale for, 662 safety issues with, 670 therapeutic scheme for, 662–663, 663f Cosmic device for, 491, 493f Dascor device for, 397 Dynesys Spinal System for, 465–467, 468 gene therapy for, 673–679 bone morphogenetic protein-2 in, 675–676, 676f bone morphogenetic protein-7 in, 676, 676f, 677f delivery systems for, 674–675 future considerations for, 678 goal of, 673, 675 interleukin-1 in, 678 of intervertebral disc, 675–678
Lim mineralization protein-1 in, 674, 676–677, 677f matrix metalloproteinases in, 674, 677–678, 678f schematic representation of, 675f Sox 9 in, 677 target genes for, 673–674 transforming growth factor-b1 in, 675, 675f, 676f growth factor injection therapy for, 649–661 with bone morphogenetic protein-2, 650 with growth and differentiation factor-5 in adolescent rabbits, 656–657, 657f, 658f in vitro evidence of, 650, 651f in mature rabbits, 657, 658f vitro evidence for, 650, 650f, 651f, 652f vivo studies of, 650–651 and pain, 659 with platelet-rich plasma, 650, 652f possible limitations of, 657–659 with recombinant human osteogenic protein-1 after chemonucleolysis, 655–657, 656f in vitro evidence of, 650, 650f in vivo studies of, 651–655, 653f, 654f, 655f hybrid constructs for, 605 HydraFlex nucleus replacement system for, 407 IsoBar TTL Dynamic Rod Stabilization for, 483–484 Kineflex device for, 338 lateral lumbar total disc replacement for, 368 Maverick total disc replacement for, 353 multilevel, 725, 728 proximate, 728, 733f NUBAC disc arthroplasty for, 442 NuCore injectable nucleus for, 418 PercuDyn device for, 511 ProDisc-L device for, 318 proposed mechanism of, 673, 674f Satellite Spinal System for, 453 TranS1 percutaneous nucleus replacement for, 435 unloading of disc for, 764 Degenerative spondylolisthesis, X-STOP with, 544–545 Device for Intervertebral Assisted Motion (DIAM), 517–522 advantages/disadvantages of, 522 biocompatibility of, 519, 521t clinical presentation and evaluation for, 520–521 complications of, 521 Crane principle and, 520 description of, 33, 519, 520f discussion of, 522 dynamic interspinous process technology and, 519–520 future development with, 522 indications/contraindications for, 519 instrumentation for, 519, 520f operative technique for, 521, 522f postoperative treatment of, 522 results of, 521 scientific testing of, 519 DEXA (dual x-ray absorptiometry) scanning, for lumbar nonfusion surgery, 77 DHI. See Disc height index (DHI). Diagnostic injections, 98–102 facet blocks, 101–102, 101f, 102f selective nerve root block, 99–101, 99f Diffuse idiopathic skeletal hyperostosis (DISH), as contraindication for cervical disc arthroplasty, 186 Dimethylsulfoxide (DMSO), intradiscal injection of, 763 Disc arthroplasty vs. arthrodesis, 11–12 biomechanical challenges of, 12
history of, 12 rationale for, 12 Disc degeneration. See Degenerative disc disease (DDD). Disc height, with NUBAC disc arthroplasty device, 445–446, 446f Disc height index (DHI) with gene therapy, 676, 676f growth factor injection therapy and growth and differentiation factor-5 for, 656–657, 657f, 658f osteogenic protein-1 for, 651–655, 653f after chemonucleolysis, 655–657, 656f with mesenchymal cell transplantation, 667–668, 669f Disc height maintenance, with Barricaid device, 634–635, 635f Disc height reduction, after discectomy, 418 Disc height requirement, for NUBAC disc arthroplasty device, 442 Disc herniation BioDisc nucleus pulposus replacement for, 431 Dynesys Spinal System for, 465, 466f, 468 recurrent with Barricaid device, 634, 634f, 635 Cosmic device for, 491 Satellite Spinal System for, 453 Disc Motion Technologies (DMT), 737, 738f, 739, 740 Disc prolapse, large, recurrent, or at transition levels, Wallis Stabilization System for, 528–529, 529f, 530f Disc regeneration. See Intervertebral disc (IVD) regeneration. Disc replacement devices, total. See Total disc replacement [TDR] devices. Disc restoration, with Wallis Stabilization System, 532 Disc ruptures, neovascularization with, 763 Disc stabilization arthroplasty, Satellite Spinal System for, 452–462 advantages/disadvantages of, 460–461 adverse effects of, 453, 454t biomechanics of, 454, 455f contraindications for, 453, 453tf description of, 454 development of, 452–453 indications for, 453 operative technique for, 456–460 closure in, 460 distraction and disc removal in, 456, 458f exposure in, 456, 458f final preparation in, 457–460, 459f implantation in, 460, 460f patient positioning for, 456, 458f trial in, 456–457, 458f, 459f patient selection for, 453–454 postoperative care for, 460 scientific testing/clinical outcomes of, 454–460, 456t, 457f Disc unloading, for disc injury or degeneration, 764 Disc water content, with mesenchymal cell transplantation, 668, 669f Discectomy anterior cervical for cervical myelopathy, 744 for CerviCore Intervertebral Disc prosthesis, 241–242, 241f and fusion (See Anterior cervical discectomy and fusion [ACDF]) disc height reduction after, 418 lumbar vs. annular repair, 629–630 for FlexiCore Intervertebral Disc, 331 in lumbar total disc arthroplasty, 305–306, 306f
Index Discectomy (Continued) micro-, 162–165, 164f, 165f multilevel anterior corpectomy and, cervical laminoplasty vs., 600 problems with, 7 video-assisted, 162–165, 164f, 165f Discocerv prosthesis, 24–25, 25f Discogenic pain, 98 concordant vs. nonconcordant, 104t definite, 103 indeterminate, 104 mechanisms of, 761 probable, 104 unequivocal, 98 Wallis Stabilization System for, 526–527, 528f, 529f Discogram, functional anesthetic, 106–108 complications of, 108 results of, 108 technique of, 106–108, 107f, 108f Discography, 102–105 cervical spine, 103 criteria for positive result in, 103–104, 103t, 104t diagnostic validity of, 104–105 historical background of, 102–103 indications for, 103 for lumbar nonfusion surgery, 77–78 lumbar spine, 103 DISCOVER Artificial Cervical Disc, 267–271 advantages/disadvantages of, 271b clinical presentation and evaluation for, 269–270 coating of, 268–269, 269f complications of, 271 description of, 25, 25f, 267–269, 268f design considerations in, 267–268 discussion of, 271 indications/contraindications for, 267 instruments with, 269 operative technique for, 270–271, 270f postoperative care for, 271 scientific testing/clinical outcomes of, 269 two-level, 181f wear testing of, 269 DISH (diffuse idiopathic skeletal hyperostosis), as contraindication for cervical disc arthroplasty, 186 Distraction decompression, Wallis Stabilization System for, 530–531 DMSO (dimethylsulfoxide), intradiscal injection of, 763 DMT (Disc Motion Technologies), 737, 738f, 739, 740 Dorsal approach, to total disc replacement, 27 Double-door laminoplasty case study of, 601–603, 602f Kurokowa modification of, 597 outcome of, 599 technique of, 597, 598f Tomita modification of, 597, 598f, 599f Downey prosthesis, 15 DSS. See Dynamic Stabilization System (DSS). Dual x-ray absorptiometry (DEXA) scanning, for lumbar nonfusion surgery, 77 Dubois, Gilles, 465 Ductility, 54, 54f Dynamic Artificial Disc, 23–24, 24f Dynamic axial compression bending, of Orthobiom Spinal System, 713 Dynamic compression testing, of Coflex implant, 536 Dynamic fatigue testing, of Kineflex Disc, 339–340 Dynamic interspinous process technology, and DIAM system, 519–520 Dynamic lumbar posterior stabilization. See Lumbar posterior dynamic stabilization.
Dynamic Neutralization System for the Spine. See Dynesys Spinal System. Dynamic stabilization defined, 484, 513 market segmentation for, 501, 501f Dynamic Stabilization System (DSS), 472–475 description of, 472, 473f discussion of, 474–475 operative technique for, 473–474, 475f rationale for, 472 scientific testing/clinical outcomes of, 472–473, 474f Dynardi prosthesis, 23–24, 24f Dynesys Spinal System, 463–471 advantages/disadvantages of, 471 baboon model of, 68–72, 72f, 73f clinical experience with, 468 complications of, 471 description of, 30–31, 31f, 467, 467f development of, 465 discussion of, 471 indications/contraindications for, 465–467, 466f operative technique for, 468–470, 469f in posterior lumbar arthroplasty, 740 postoperative care for, 471 with ProDisc-L prosthesis, 605, 606f subsequent implantation of, 609, 609f for three-level procedure, 607, 608f scientific testing of, 467–468 Dysphagia, with CerviCore Intervertebral Disc prosthesis, 245 Dysphonia, with CerviCore Intervertebral Disc prosthesis, 245
E Economic impact of low back pain, 132 of motion preservation technology, 132–138 discussion of, 136–137 future trends in, 135–136 issues in measuring, 132–134 ways to improve, 134–135 eDisc. See Theken eDisc. Elastic limit, 53f Elasticity defined, 54 modulus of, 54 Elastomer devices, 18, 18f Elastomeric cervical artificial disc. See NeoDisc device. Elastomeric core, of Theken eDisc, 364, 364f Elastomeric lumbar artificial disc. See Theken eDisc. Elderly considerations for disc arthroplasty in, 720, 721f, 722f interspinous spacer devices in, 721 with osteoporosis, 721–722, 723f posterior dynamic lumbar stabilization in, 720–721 Electromyography (EMG), in biomechanical model, 688–689, 689f, 693f Electronics in spinal devices, 765–766 of Theken eDisc, 365–366, 366f End plate injury, animal models of disc degeneration due to, 641t, 645 End plate loading, with posterior lumbar arthroplasty, 737 Endoscopic posterolateral (transforaminal) approach, to lumbar spine. See Lumbar endoscopic posterolateral [transforaminal] approach. Endoscopic spine surgery system, 168, 169f Erickson and Griffith device, 15, 15f
777
Esophagus in anterior cervical approach, 39, 39f in cervical revision, 280–282, 282f Etanercept, for back pain, 763 Expulsion test, of NUBAC disc arthroplasty device, 446 Extension, pain with, 97–98 External forces, 688 Extreme lateral interbody fusion (XLIF) approach for lumbar revision, 383–388 discussion of, 387 illustrative cases on, 383–386 due to malpositioned keeled implant, 386, 386f, 387f, 388f nonkeeled implant, 384–385, 385f due to pars fracture, 383–384, 384f, 385f rationale for, 383 for lumbar total disc replacement, 368–372 advantage(s) of, 370 ease of insertion and device placement as, 370, 371f end plate support as, 202, 372f lower risk of approach-related complications and morbidity as, 370 safer reversibility as, 370 stability as, 370 disadvantages of, 370 indications for, 368, 369f results of, 368 surgical technique for, 368–370 closure in, 370 disc space preparation in, 369–370, 370f insertion of device in, 370, 371f positioning and surgical exposure in, 368–369, 369f, 370f
F Facet arthropathy, persistent pain after lumbar total disc arthroplasty due to, 390 Facet arthroplasty, for posterior dynamic stabilization, 32–33, 32t Facet arthrosis, as contraindication for cervical disc arthroplasty, 186 Facet blocks, 101–102, 101f false-negative, 102 false-positive, 102 technique for, 101–102, 102f Facet capsular ligaments (FCLs), 38f Facet complex, in functional spinal unit, 36, 37, 37f Facet cyst, herniated nucleus pulposus and contralateral, 615 Facet degeneration, grading system for, 565–566, 567f Facet disease, persistent pain after cervical arthroplasty due to, 298 Facet joint(s) anatomy and biomechanics of, 21–22, 565, 566f damage to, 565–566 effect of CHARITÉ Artificial Disc on, 315–316 morphology of, 566, 568f Facet joint capsules, innervation of, 565 Facet joint osteoarthritis, FENIX Facet Resurfacing Implant for, 585–586 Facet joint resection, animal models of disc degeneration due to, 641t, 643–644 Facet joint–mediated pain, 98 Facet syndrome, 565–566 Facet/end plate ratio (FER), 696, 699f Facial nerve, marginal mandibular branch of, in cervical revision, 278, 280f FAD. See Functional anesthetic discogram (FAD). Failed back surgical syndrome (FBSS), 377, 378 Failure strength, 53f
778
Index
Far lateral approach, to minimally invasive lumbar surgery, 161–162, 162f, 163f Fatigue, 53 surface, 60f, 61 Fatigue strength, 53 Fatigue testing of Aquarelle hydrogel disc nucleus, 427, 427f of Barricaid device, 633 of NUBAC disc arthroplasty device, 447 of TranS1 Percutaneous Nucleus Replacement, 437 FBSS (failed back surgical syndrome), 377, 378 FCLs (facet capsular ligaments), 38f FDA (Food and Drug Administration), regulation of medical implants by, 47–48 FEA. See Finite element analysis (FEA). FEM (finite element method), in biomechanical model, 691–693 Femoral nerve injury, in anterior exposure of lumbar spine, 151, 152f, 153 FENIX Facet Resurfacing Implant, 585–592 advantages/disadvantages of, 591 complications of, 591 contraindications for, 586 description of, 585, 586, 586f discussion of, 591f, 592 indications for, 585–586 operative technique for, 588–590, 590f, 591f postoperative care for, 590 rationale for, 585, 586f scientific testing of, 586–588, 587f, 589f, 589t FENIX Inferior Facet Resurfacing component, 586, 586f, 590 FENIX Locking Screw, 586, 586f FENIX Superior Facet Resurfacing component, 586, 586f, 590 FER (facet/end plate ratio), 696, 699f Fernstrom Ball, 12, 12f, 443 Fernstrom, Ulf, 8, 12, 453 Fibronectin fragment injection, animal models of disc degeneration due to, 641t, 645 Financial impact of low back pain, 132 of motion preservation technology, 132–138 discussion of, 136–137 future trends in, 135–136 issues in measuring, 132–134 ways to improve, 134–135 Finite element analysis (FEA) in biomechanical model, 691 of coflex implant, 536–537 of IsoBar TTL Dynamic Rod Stabilization, 485–486, 486f, 486t, 487f of TOPS device, 553, 555f Finite element method (FEM), in biomechanical model, 691–693 Fixed dome devices, 13, 13f Flexibility testing, of FENIX Facet Resurfacing Implant, 586–588, 587f, 589f, 589t FlexiCore Intervertebral Disc, 330–337 clinical data on, 335, 335f complications of, 336 contraindications for, 331 description of, 24–25, 330, 331f discussion of, 336 indications for, 330 operative technique for, 331–334 closure in, 334 discectomy and end plate preparation in, 331 distraction (disc height restoration) in, 331–332, 332f, 333f insertion of disc in, 332–333, 333f, 334f repositioning (or extraction) in, 334, 334f, 335f surgical approach in, 331 verification of size and placement in, 334, 334f
postoperative care for, 334 preoperative planning for, 331 Flexion, pain with, 97–98 Flexion-extension of FENIX Facet Resurfacing Implant, 586, 587f of Orthobiom Spinal System dynamic, 714 static, 714 Flexion-extension radiographs for lumbar nonfusion surgery, 76–77 with TOPS device, 559–560, 564f Fluid-filled bag devices, 17–18 fMRI (functional magnetic resonance imaging), and future of motion preservation, 761 Food and Drug Administration (FDA) Investigational Device Exemption (IDE) studies by (See Investigational Device Exemption [IDE] study[ies]) regulation of medical implants by, 47–48 Foraminal herniated nucleus pulposus, dynamic pedicle-screw stabilization with nucleus replacement for, 614 Forces external, 688 internal, 688 Forward bending, in biomechanical model, 690t Fracture(s) osteoporotic percutaneous treatment of, 763 polymethyl/methacrylate (PMMA) for, 763 pars, after lumbar total disc replacement, lateral revision of, 383–384, 384f, 385f Fracture point, 53f Fracture toughness, 53, 53f Freedom Lumbar Disc, 26, 26f French door laminoplasty case study of, 601–603, 602f Kurokowa modification of, 597 outcome of, 599 technique of, 597, 598f Tomita modification of, 597, 598f, 599f Frenchay Artificial Cervical Joint, for adjacent segment disease, 754 Functional anesthetic discogram (FAD), 106–108 complications of, 108 results of, 108 technique of, 106–108, 107f, 108f Functional magnetic resonance imaging (fMRI), and future of motion preservation, 761 Functional spinal unit (FSU), 21 anatomy of, 36–39, 37f, 38f biomechanical testing of, 45–51 ASTM guidelines for, 46–50 challenges of, 46 clinical relevance of, 46 configurations for, 48–49, 49f and continuous spectrum of stability, 48, 48f experimental designs for, 48, 48f history of, 45–46 loading and displacement protocols in, 46, 47, 47f, 48 measurement parameters for, 50 range of motion in, 46, 47f, 48, 50 for spinal instability, 46 standardized, 46–50, 47f motion preservation devices and, 22 roles of, 565 Fusion adjacent segment degeneration after, 118–125 defined, 118 fusion biomechanics and cervical, 119–120 lumbar, 120 lumbar, 121–123, 122t
animal models of, 641t, 644 fusion biomechanics and, 120 spinal biomechanics and, 118–119, 119f, 120f anterior cervical discectomy and (See Anterior cervical discectomy and fusion [ACDF]) anterior lumbar interbody, positioning for, 155–156, 156f arthroplasty vs., 11–12 cervical disc arthroplasty adjacent to, 751–759 discussion of, 754–758, 755f, 756f, 757f materials and methods for, 752, 752f, 753f results of, 752–754 in international study, 752–754, 753f, 754f in U.S. prospective randomized study, 752, 753t cost effectiveness of, 133–134 endpoints and success criteria for, 94 extreme lateral interbody (See Extreme lateral interbody fusion [XLIF] approach) herniated nucleus pulposus adjacent to, 615 lumbar adjacent segment degeneration after, 121–123, 122t animal models of, 641t, 644 fusion biomechanics and, 120 and bone mineral density, 618 same segment degeneration after, 123 and simultaneous total disc replacement, 617–620 biomechanics of, 617–618 clinical experience with, 618–619, 618f rationale for, 617 motion-preserving device with, 605 problems with, 7 Fusion biomechanics, and adjacent level degeneration cervical, 119–120 lumbar, 120
G GDF-5. See Growth and differentiation factor-5 (GDF-5). GDF-8 knockout mice, as animal model for disc degeneration, 641t Gene therapy, 673–679 bone morphogenetic protein-2 in, 675–676, 676f bone morphogenetic protein-7 in, 676, 676f, 677f delivery systems for, 674–675 future considerations for, 678 goal of, 673, 675 interleukin-1 in, 678 of intervertebral disc, 675–678 Lim mineralization protein-1 in, 676–677, 677f matrix metalloproteinases in, 677–678, 678f schematic representation of, 675f Sox 9 in, 677 target genes for, 673–674 tissue inhibitors of matrix metalloproteinase in, 674 transforming growth factor-b1 in, 675, 675f, 676f Genetic knockout mice, as animal models for disc degeneration, 641t, 645–646 Genitofemoral nerve injury, in anterior exposure of lumbar spine, 151–153, 152f Genitourinary complications, of anterior exposure of lumbar spine, 149t, 150, 151f Glucosamine, intradiscal injection of, 763 Glutaraldehyde, in BioDisc Nucleus Pulposus Replacement, 431–432 Goat model of cervical spine, 64, 65f of PCM arthroplasty, 64–65, 66f, 67f, 203–204, 207f Goffin, Jan, 292f Goniometer, in biomechanical model, 689
Index Gore-Tex (polytetrafluoroethylene) in Barricaid device, 631 material properties of, 55 Gottingen pig model, for disc degeneration, 641t Graf, Henry, 484 Graf Ligament System, 15, 16f, 30, 484 Graham ball-and-socket device, 13, 13f Great vessels, of abdomen and pelvis, 41, 42f Growth and differentiation factor-5 (GDF-5), for intervertebral disc regeneration in adolescent rabbits, 656–657, 657f, 658f in vitro evidence of, 650, 651f in mature rabbits, 657, 658f Growth factor(s), in maintenance of matrix homeostasis, 649 Growth factor injection therapy, 649–661, 764 with bone morphogenetic protein-2, 650 with growth and differentiation factor-5 in adolescent rabbits, 656–657, 657f, 658f in vitro evidence of, 650, 651f in mature rabbits, 657, 658f vitro evidence of, 650, 650f, 651f, 652f vivo studies of, 650–651 and pain, 659 with platelet-rich plasma, 650, 652f possible limitations of, 657–659 with recombinant human osteogenic protein-1 after chemonucleolysis, 655–657, 656f in vitro evidence of, 650, 650f in vivo studies of, 651–655, 653f, 654f, 655f
H HA (hydroxyapatite) coating, of Maverick Total Disc Replacement, 355, 356f HA (hydroxyapatite) spacers, in French door laminoplasty, 597, 598f, 599f HAM (helical axis of motion), 92 Hamster models, for disc degeneration, 640, 641t Hard plate/hard core devices, 14–15, 14f, 15f Hard plate/soft core devices, 15, 15f, 16f Harmon, Paul, 7, 12, 452, 453 Health insurance coverage, for motion preservation technology, 133, 135–136 Helical axis of motion (HAM), 92 Hematomas, rectus sheath, after anterior exposure of lumbar spine, 153 Herniated nucleus pulposus (HNP) adjacent to fusion, 615 at apex of scoliotic curve, 614–615 and contralateral facet (synovial) cyst, 615 dynamic pedicle-screw stabilization with nucleus replacement for, 614–615 Dynesys Spinal System for, 468 foraminal, necessitating vigorous unilateral medial facetectomy, 614 large central, 615 recurrent, 615 with spondylolysis without listhesis, 614 Heterotopic ossification (HO), after cervical arthroplasty, 298f, 299f, 300, 301 Hirabayashi open-door laminoplasty outcome of, 599 technique of, 597, 597f, 598f History, in patient evaluation for cervical nonfusion surgery, 80–81, 81f for lumbar nonfusion surgery, 75–76 HNP. See Herniated nucleus pulposus (HNP). HO (heterotopic ossification), after cervical arthroplasty, 298f, 299f, 300, 301 Horner's syndrome, in cervical revision, 282, 283f Hybrid nonfusion technique(s), 604–611 case studies on, 605 cervical laminoplasty as, 593–603 CHARITÉ Artificial Disc in, 316
classification of, 604–605, 605t contraindications for, 605–611 discussion of, 611 dynamic pedicle-screw stabilization with nucleus replacement as, 612–616 Dynesys system after ProDisc-L prosthesis in, 609, 609f Dynesys system plus ProDisc-L prosthesis in, 605, 606f for three-level procedure, 607, 608f indications for, 605 for interspinous implants with nucleus replacement, 610 multilevel, 604–605, 605t multistage, 604, 605t NeuDisc plus coflex system in, 610, 610f for posterior dynamic stabilization and lumbar total disc replacement, 605 ProDisc-L prosthesis plus coflex system in, 606, 607f simultaneous lumbar fusion and total disc replacement as, 617–620 biomechanics of, 617–618 clinical experience with, 618–619, 618f rationale for, 617 single-level, 604, 605t single-stage, 604, 605t HydraFlex device, 407–410 advantages/disadvantages of, 409b clinical presentation and evaluation for, 408–409 complications of, 409 description of, 407, 408, 408f discussion of, 409–410 indications/contraindications for, 407 operative technique for, 409 postoperative care for, 409 scientific testing/clinical outcomes of, 408 Hydrogel, defined, 423 Hydrogel nucleus replacement Aquarelle, 423–430 advantages/disadvantages of, 429b clinical presentation and evaluation for, 429 complications of, 429 description of, 423–425, 424f design of, 424–425 development of, 423 discussion of, 429–430 function of, 423–424, 424f indications/contraindications for, 424 operative techniques for, 429 postoperative care for, 429 scientific testing of, 425–429 animal studies in, 427–429 for biocompatibility, 425 for biomechanics, 426–427 for fatigue, 427, 427f for swelling pressure, 425–426, 426f BioDisc, 431–434 advantages/disadvantages of, 434b clinical presentation and evaluation for, 433, 434f complications of, 433 description of, 431–432, 432f discussion of, 433 indications/contraindications for, 431 operative techniques for, 433, 433f postoperative care for, 433 scientific testing/clinical outcomes of, 432–433, 433f HydraFlex, 407–410 advantages/disadvantages of, 409b clinical presentation and evaluation for, 408–409 complications of, 409 description of, 407, 408, 408f
779
discussion of, 409–410 indications/contraindications for, 407 operative technique for, 409 postoperative care for, 409 scientific testing/clinical outcomes of, 408 NeuDisc, 411–416 advantages/disadvantages of, 415 description of, 412, 412f, 413f design of, 411–412 discussion of, 415 operative technique for, 414–415, 415f scientific testing of, 412–414 for biocompatibility, 413 for clinical outcomes, 413–414, 414f for endurance, 412–413 for expulsion, 413 Hydroxyapatite (HA) coating, of Maverick Total Disc Replacement, 355, 356f Hydroxyapatite (HA) spacers, in French door laminoplasty, 597, 598f, 599f Hyperostosis, diffuse idiopathic skeletal, as contraindication for cervical disc arthroplasty, 186 Hypogastric nerve plexus, 41 Hypoglossal nerve, in cervical revision, 280, 281f
I IAR. See Instantaneous axis of rotation (IAR). Ibo ball-and-socket prosthesis, 13, 13f IDE study. See Investigational Device Exemption (IDE) study. IL-1 (interleukin-1), in gene therapy, 678 Ileus, after anterior exposure of lumbar spine, 153 Iliac arteries, 41, 42f in anterior exposure of lumbar spine, 142–143, 145f, 149, 149f Iliac veins, in anterior exposure of lumbar spine, 143, 145f, 149, 149f Iliocaval junction, 41 Iliohypogastric nerve injury, in anterior exposure of lumbar spine, 151, 152, 152f Ilioinguinal nerve injury, in anterior exposure of lumbar spine, 151, 152, 152f Iliolumbar vein, 41 in anterior exposure of lumbar spine, 143–146, 146f, 149, 150f Images, subject-specific, in biomechanical model, 688, 693, 693f Impact strength, 53 In situ polymerization, 18 Inclose Surgical Mesh System, 621–628 advantages/disadvantages of, 627–628 clinical presentation and evaluation for, 625, 625f complications of, 626–627 description of, 623–624, 624f discussion of, 627–628 indications/contraindications for, 623 operative techniques for, 625–626, 626f postoperative care for, 626 rationale for, 623 scientific testing/clinical outcomes of, 624, 624f Infection, lumbar revision due to, 378 Inferior facet burr, for FENIX Facet Resurfacing Implant, 588–589 Inflammatory mediators, of back and radicular pain, 761 In vivo loading, of Total Facet Arthroplasty System, 567 Injection diagnostics, for lumbar nonfusion surgery, 78 Innovative Spinal Technologies (IST) Dynamic Stabilization device, 500–504 advantages/disadvantages of, 504 allowed motions of, 502, 503f
780
Index
Innovative Spinal Technologies (IST) Dynamic Stabilization device (Continued) description of, 501–502 discussion of, 504 DS FM version of, 502, 503f indications/contraindications for, 500 vs. interspinous spacers, 501 operative technique for, 504 Paramount MM version of, 502, 503f pedicle screw kinematics of, 501, 502f pedicle-to-pedicle measurements through range of motion for, 501, 502f rationale for, 500 scientific testing/clinical outcomes of, 502–504 In-Space Interspinous Distraction Device, 33, 34f Instantaneous axis of rotation (IAR), 92 with IsoBar TTL Dynamic Rod Stabilization, 486, 488–489 with posterior lumbar arthroplasty, 737, 737f, 739 Insurance reimbursement, for motion preservation devices, 766–767 Interfacial medium, in tribologic system, 60, 60f Interleukin-1 (IL-1), in gene therapy, 678 Internal forces, 688 Interpositional arthroplasty, Zyre Facet Replacement Device for, 581–584 complications of, 584 description of, 582, 582f indications/contraindications for, 581–582 operative techniques for, 583–584 rationale for, 581 scientific testing/clinical outcomes of, 582, 583f, 584f Interspinous distraction devices, mechanism of action of, 519–520 Interspinous implant(s), 33, 33t with function by itself, 33, 33f with nucleus replacement, 610, 610f without own function, 33, 33f, 34f Interspinous ligaments (ISLs), 38f Interspinous spacer devices cost effectiveness of, 134 in elderly, 721 future of, 762 indications/contraindications for, 78 Interspinous "U", 534 Interspinous-based lumbar posterior dynamic stabilization coflex implant for, 534–540 Device for Intervertebral Assisted Motion (DIAM) for, 517–522 Wallis Stabilization System for, 523–533 X-STOP for, 541–548 Intertransverse approach, to minimally invasive lumbar surgery far lateral, 161–162, 162f, 163f paramedian, 160–161, 161f Intervertebral disc(s) (IVDs) anatomy and physiology of, 5–7, 11, 21–22, 639 in functional spinal unit, 36–37, 37f maintenance of matrix homeostasis in, 649 noncritical vertical loading of, 5–6, 6f pathologic changes in, 6–7, 6f, 11 roles of, 11–12 as source of pain, 98 Intervertebral disc (IVD) degeneration. See Degenerative disc disease (DDD). Intervertebral disc prostheses. See Total disc replacement (TDR) devices. Intervertebral disc (IVD) regeneration challenges of, 764–765 growth factor injection therapy for, 649–661 with bone morphogenetic protein-2, 650 with growth and differentiation factor-5 in adolescent rabbits, 656–657, 657f, 658f
in vitro evidence of, 650, 651f in mature rabbits, 657, 658f in vitro evidence for, 650, 650f, 651f, 652f in vivo studies of, 650–651 and pain, 659 with platelet-rich plasma, 650, 652f possible limitations of, 657–659 with recombinant human osteogenic protein-1 after chemonucleolysis, 655–657, 656f in vitro evidence of, 650, 650f in vivo studies of, 651–655, 653f, 654f, 655f Intervertebral motion, preoperative excessive, 93–94 Intervertebral motion measurements accuracy and reproducibility of with computed-assisted techniques, 87–88, 88f with manual methods, 87 requirements for, 85–87 balance of motion in, 86 component motion in, 86 endpoints and success criteria for, 93–95 patient positioning and patient effort for, 88–90, 91f and patient selection criteria, 86–87 quality control for, 88–90, 90t, 92t quality of motion parameters for, 86 quantity of motion parameters for, 86 reference data for, 90–93, 90t, 92t, 93f, 93t INTRA device, 522 Intra-articular block, 101 Intradiscal pressure(s), and annular repair, 630 Intradiscal pressure changes, with Barricaid device, 633 Intrinsic Therapeutics Barricaid device. See Barricaid device. Invasive diagnostic tool(s), 97–117 diagnostic injections as, 98–102 facet block, 101–102, 101f, 102f selective nerve root block, 99–101, 99f discography as, 102–105, 103t, 104t functional anesthetic discogram as, 106–108, 107f, 108f mechanism of, 98–102 rationale for, 97–98, 98f, 98t temporary external transpedicular fixation as, 105–106, 105f Investigational Device Exemption (IDE) study(ies) of CHARITÉ Artificial Disc, 311–312 of Maverick Total Disc Replacement, 357–358, 358t of Porous Coated Motion (PCM) arthroplasty, 202 of ProDisc-C artificial cervical disc, 216, 216t, 217f, 218f of ProDisc-L artificial lumbar disc, 320, 322t of Secure-C device, 247, 248t, 249–251 ISLs (interspinous ligaments), 38f IsoBar TTL Dynamic Rod Stabilization, 483–489 clinical outcomes with, 486–487 complications of, 488 dampener in, 484, 485–486 description of, 31, 31f, 484, 484f design of, 484, 485f discussion of, 488–489 indications/contraindications for, 484 operative technique for, 487–488 in posterior lumbar arthroplasty, 740 postoperative care for, 488 rationale for, 483–484 scientific testing of, 484–487 axial compression in, 485–486 axial displacement and stiffness in, 486, 486f biomechanical, 484–485, 485f centers of rotation in, 486f finite element analysis in, 485–486, 486f, 486t, 487f flexion-extension radiographs in, 487f, 488f
instantaneous axis of rotation in, 486, 488–489 load deformation in, 484–485, 485f stress values in, 486, 488 Young's modulus and Poisson's ratio in, 485–486, 486t IST. See Innovative Spinal Technologies (IST). IVDs. See Intervertebral disc(s) (IVDs).
J Japanese Orthopaedic Association (JOA) score for cervical laminoplasty, 599 for Wallis Stabilization System, 526f Joubert, M. J., 8
K Keeled prosthesis, progression of scoliosis with, 696f Keratin sulfate (KS), and hydrogen disc nucleus, 423 Kineflex Disc, for lumbar total disc arthroplasty, 338–345 advantages/disadvantages of, 344b biomechanics of, 340 clinical testing of, 340 clinical presentation and evaluation in, 340–341, 341t, 342 recovery and patient satisfaction in, 341, 342f complications of, 344 description of, 23, 24f, 338–339, 339f discussion of, 344–345 indications/contraindications for, 338 insertion of, 342–344, 342f operative techniques for, 341–344, 342f postoperative care for, 344 preclinical testing of, 339–340 for dynamic and shear fatigue, 339–340 for static compression and static shear, 339 for wear, 340 rationale for, 338 Kineflex|C Cervical Disc, 258–266 advantages/disadvantages of, 265b clinical presentation and evaluation for, 261 complications of, 264–265 current treatment modalities vs., 258–259 description of, 23, 24f, 259, 259f, 260f imaging of, 263, 263f, 264f, 265f indications/contraindications for, 258 operative technique for, 261–262 anesthesia in, 262 distraction in, 262, 262f end plate sizing in, 262, 262f final placement in, 263, 263f initial insertion in, 262, 263f midline slot cut in, 262, 263f midline verification in, 262, 262f positioning in, 262 placement of, 180f postoperative care for, 264 scientific testing/clinical outcomes of, 260–261 Kinematic testing of NUBAC disc arthroplasty device, 444–445 of Total Facet Arthroplasty System, 567, 570f Klippel-Feil syndrome, cervical revision for, 290, 291f, 293f Knockout mice, as animal models for disc degeneration, 641t, 645–646 Kostuik device, 16–17, 17f Kotani prosthesis, 18, 18f KS (keratin sulfate), and hydrogen disc nucleus, 423 Kurokowa modification, of French door laminoplasty, 597 Kyphosis, cervical after cervical laminoplasty, 600 and cervical revision, 278–280, 281f, 293f, 294
Index Kyphosis, cervical (Continued) fixed, as contraindication to cervical nonfusion surgery, 83f persistent pain due to, 297–298, 298f
L Laryngeal nerve(s) recurrent, 39–40 in cervical revision, 278 superior, 39 Lateral approach, to total disc replacement, 27 Lateral bending in biomechanical model, 690t of FENIX Facet Resurfacing Implant, 586, 587f of Orthobiom Spinal System, 713–714 Lateral lumbar total disc replacement, 368–372 advantage(s) of, 370 ease of insertion and device placement as, 370, 371f end plate support as, 202, 372f lower risk of approach-related complications and morbidity as, 370 safer reversibility as, 370 stability as, 370 disadvantages of, 370 indications for, 368, 369f results of, 368 for revision, 383–388 discussion of, 387 illustrative cases on, 383–386 due to malpositioned keeled implant, 386, 386f, 387f, 388f nonkeeled implant, 384–385, 385f due to pars fracture, 383–384, 384f, 385f rationale for, 383 surgical technique for, 368–370 closure in, 370 disc space preparation in, 369–370, 370f insertion of device in, 370, 371f positioning and surgical exposure in, 368–369, 369f, 370f Lateral shear, in biomechanical model, 691f LBP. See Low back pain (LBP). Leg pain, history of, 76 Ligaments, of spine, 37–38, 38f Ligamentum flavum (LF), 38–39, 38f Lim mineralization protein-1 (LMP-1), in gene therapy, 676–677, 677f LMM (lumbar motion monitor), in biomechanical model, 689 Load displacement, 46, 47, 47f, 48 Load sharing, of Total Facet Arthroplasty System, 568 Loading analysis, of coflex implant, 538 Loading environment, of coflex implant, 537–538 Low back pain (LBP) axial, posterior lumbar arthroplasty for, 736 costs of, 132, 135 PercuDyn device for, 511 societal costs of, 662 LSO (lumbar-sacral orthosis), with Satellite Spinal System, 460 LSS. See Lumbar spinal stenosis (LSS). Lumbar adjacent segment degeneration, 121–123, 122t animal models of, 641t, 644 fusion biomechanics and, 120 Lumbar arthrodesis, adjacent segment degeneration after, 121–123, 122t Lumbar arthroplasty. See Lumbar total disc arthroplasty. Lumbar artificial disc. See Lumbar total disc arthroplasty. Lumbar coupled motion, 118, 119f
Lumbar disc disease, multilevel, 725, 728 proximate, 728, 733f Lumbar discectomy vs. annular repair, 629, 629–630 for FlexiCore Intervertebral Disc, 331 in lumbar total disc arthroplasty, 305–306, 306f Lumbar dynamic posterior fixation, baboon model of, 68–72, 72f, 73f Lumbar endoscopic posterolateral (transforaminal) approach, 167–178 anesthesia for, 169 clinical presentation and evaluation for, 168 complications of, 174–176 description of device for, 168, 169f discussion of, 176 future considerations with, 176–177 indications/contraindications for, 168 instrument placement in, 173–174, 173f needle placement in, 171–172 performing discectomy in, 174 position for, 169, 170f postoperative care with, 174 procedure for, 169–171, 171f, 172f rationale for, 167–168, 168f scientific testing/clinical outcomes of, 168 Lumbar facet joints anatomy and biomechanics of, 566 referred pain from, 101, 101f Lumbar facet replacement Anatomic Facet Replacement System (AFRS) for, 577–580 FENIX Facet Resurfacing Implant for, 585–592 Total Facet Arthroplasty System (TFAS) for, 565–576 Total Posterior Arthroplasty System (TOPS) for, 549–564 Zyre Facet Replacement Device for, 581–584 Lumbar facet syndrome, 101, 565–566 Lumbar fusion adjacent segment degeneration after, 121–123, 122t animal models of, 641t, 644 fusion biomechanics and, 120 and bone mineral density, 618 and simultaneous total disc replacement, 617–620 biomechanics of, 617–618 clinical experience with, 618–619, 618f rationale for, 617 Lumbar motion monitor (LMM), in biomechanical model, 689 Lumbar nonfusion surgery, patient evaluation for, 75–79 bone scintigraphy in, 77 CT scans in, 77 discography in, 77–78 flexion-extension radiographs in, 76–77 history in, 75–76 injection diagnostics in, 78 MRI in, 77 physical examination in, 76, 76f plain radiographs in, 76, 77f Lumbar nucleoplasty, baboon model of, 65–68, 69f, 70f Lumbar nucleus replacement. See Nucleus replacement. Lumbar partial disc replacement. See Nucleus replacement. Lumbar posterior dynamic stabilization coflex implant for, 534–540 Cosmic system for, 490–499 Device for Intervertebral Assisted Motion (DIAM) for, 517–522 Dynamic Stabilization System for, 472–475 Dynesys Spinal System for, 463–471 in elderly, 720–721
781
IsoBar TTL Dynamic Rod Stabilization, 483–489 IST Dynamic Stabilization device for, 500–504 NFlex system for, 505–510 with nucleus replacement, 612–616 anatomic, biomechanical, and pathologic considerations for, 612–613 contraindications for, 615 discussion of, 615–616 indications for, 614–615 primary, 614–615 secondary, 615 options for, 615, 616t rationale for, 613 review of literature on, 613–614 options for, 616t PercuDyn device for, 511–516 Stabilimax NZ system for, 476–482 Wallis Stabilization System for, 523–533 X-STOP for, 541–548 Lumbar revision, 377–382 advantages/disadvantages of, 381b anesthesia for, 380 anterior, 373–376 advantages of, 373–376, 373 approach considerations for, 375–376, 380 at different level than index level, 376 L4-L5, 376, 380 L5-S1, 375–376, 375f, 380 risks of, 373 of CHARITÉ Artificial Disc, 380 clinical presentation and evaluation for, 378–379 complications of, 381 contraindications for, 378 discussion of, 381 distraction system in, 380 extreme lateral interbody fusion (XLIF) approach for, 383–388 discussion of, 387 illustrative cases on, 383–386 due to malpositioned keeled implant, 386, 386f, 387f, 388f nonkeeled implant, 384–385, 385f due to pars fracture, 383–384, 384f, 385f rationale for, 383 indications for, 374, 377–378, 379f intraoperative photograph of key structures in, 374f of keeled devices, 380 left retroperitoneal, 380 operative technique for, 379–381 for device not removed, 381 for device requiring removal, 380 overview of, 377 patient preparation for, 374–375, 374f, 375f postoperative care after, 381 prevention of, 374 rate of, 378 right retroperitoneal, 380 time, procedure, complication (TPC) system for, 377 timing of, 373 transperitoneal, 375 Lumbar same segment degeneration, after spinal fusion, 123 Lumbar spinal stenosis (LSS) circumferential, 720, 721f coflex implant for, 534, 535 Cosmic device for, 491, 492f Dynesys Spinal System for, 465, 466f, 468 TOPS device for, 3, 551–552 Total Facet Arthroplasty System (TFAS) for, 565 Wallis Stabilization System for, 530–531 X-STOP for, 541
782
Index
Lumbar spine animal models of, 65, 67f anterior anatomy of, 41–43, 42f, 43f anterior exposure of, 139–147 advantages of, 146–147 comment on, 146–147 complications of, 147–154, 149t miscellaneous, 149t, 153 neurologic, 149t, 151–153, 152f rate of, 149t urologic, 149t, 150, 151f vascular, 148–150, 149f, 149t, 150f entry into retroperitoneal space in, 141–142, 143f, 144f, 145f mobilization of blood vessels in, 142–146, 145f, 146f other technical considerations in, 146 patient preparation for, 141 skin incision for, 141, 142f anterolateral transpsoatic approach to, 155–158 advantages of, 158 annulus incision and annulotomy in, 156, 158f complications of, 158 incision and disc exposure in, 155–156, 157f, 158f indications for, 155 limitation of, 155 nucleus excision and wound closure in, 157, 158f positioning for, 155–157, 156f, 157f constraints to axial rotation in, 695, 697t minimally invasive posterior approaches to, 159–166 far lateral, 161–162, 162f, 163f midline, 159–160, 160f muscle-splitting, 162–165, 164f, 165f vs. open surgery, 159 paramedian, 160–161, 161f paraspinal, 160–165 via tubular retractor system, 162–165, 164f, 165f rotational stability of, 695 Lumbar spondylosis Activ-L artificial disc for, 346 history of, 75–76 Lumbar total disc arthroplasty Activ-L Artificial Disc for, 346–352 adjacent to scoliotic deformity, 705–710 classification of, 705–706, 706t clinical background and methods of, 705–706, 706t discussion of, 706–709 results of, 706 Oswestry Disability Index in, 706, 706f for type 2a patient, 706, 708f, 709f, 710f for type Ia patient, 706, 707f for type Ib patient, 706, 707f, 708f Visual Analog Scale in, 706, 706f baboon model of, 68, 70f, 70t, 71f, 72f CHARITÉ Artificial Disc for, 309–317 clinical trials of, 126–131 discussion of, 128–130 introduction to, 126–127 materials and methods for, 127 results of, 127–128, 129f, 617, 618f statistical methods for, 127, 128f disc space distraction in, 306–307, 306f, 307f, 308f disc space exposure for, 305, 306f discectomy in, 305–306, 306f in elderly, 720, 721f, 722f end plate preparation in, 307 FlexiCore Intervertebral Disc for, 330–337 goals of, 75 hybrid constructs for, 605 indications/contraindications for, 78, 617
initial preoperative preparation for, 305, 306f Kineflex Disc for, 338–345 lateral approach to, 368–372 lumbar fusion and simultaneous, 617–620 biomechanics of, 617–618 clinical experience with, 618–619, 618f rationale for, 617 Maverick Total Disc Replacement for, 353–362 midline marking for, 305–306, 306f Mobidisc prosthesis for, 326–329 multilevel, 725–734 case studies on for five levels, 728, 732f for four levels, 728, 731f for three levels, 728, 730f for two levels, 728, 729f clinical outcomes for, 727 complications of, 727, 727f contraindications for, 726 description of device for, 726, 726f discussion of, 728–733, 733f indications for, 726 materials and methods for, 727 operative techniques for, 726 patient selection for, 725 postoperative care for, 726–727 proximate, 728, 733f patient positioning for, 305, 306f persistent pain after, 389–394 due to disease progression, 392–393 early-onset, 393 due to failure of prosthesis, 390–392 due to failure to achieve surgical goals, 390 due to incorrect procedure, 389–390, 391f late-onset, 394 due to poor patient selection, 389 subacute, 393–394 due to technique-related problems, 390, 392f, 393f posterior, 735–741 anatomic considerations for, 735–736 anterior vs., 740, 740t biomechanical consideration(s) for, 736–738 angular motion as, 737 extent of motion as, 737 loading on end plate as, 737 matching instantaneous axis of rotation as, 737, 737f, 739 minimization of wear as, 737–738 parallel vs. nonparallel placement as, 737, 737f challenges of, 739 device for, 737, 738f discussion of, 740, 740t goals of, 736, 736t, 740 history of, 735 operative technique for, 738–739, 738f, 739f pre-operative considerations for, 305–308 ProDisc-L artificial lumbar disc for, 318–325 remobilization in, 306–307, 307f revision of (See Lumbar revision) rotational instability with, 695–705, 695, 697f scientific testing of, 695–700 anatomic measurements in, 695–696, 700 biomechanical testing in, 697–701, 701f, 703f discussion of, 701–704 results of, 700–701 scoliosis as contraindication to, 695 Theken eDisc for, 363–367 Lumbar vertebrae, anatomy of, 43–44, 43f Lumbar vertebral bodies anatomy of, 43, 43f height and width of, 43–44 Lumbar-sacral orthosis (LSO), with Satellite Spinal System, 460
M M6 Artificial Cervical Disc, 272–276 animal implantation studies of, 273 biocompatibility of, 273 biomechanical evaluation of, 273, 274f clinical experience with, 274–275, 274f, 275f description of, 26, 26f, 272, 273f discussion of, 275 operative technique for, 273–274, 274f scientific testing of, 272–273 Magnetic resonance imaging (MRI) abnormal findings in asymptomatic subjects on, 97–98, 98f, 98t in biomechanical model, 689 functional, and future of motion preservation, 761 in patient evaluation for cervical nonfusion surgery, 82 for lumbar nonfusion surgery, 77 Main device, 17, 17f Malpositioned lumbar implant, lateral revision due to keeled, 386, 386f, 387f, 388f nonkeeled, 384–385, 385f Marginal mandibular branch, of facial nerve, in cervical revision, 278, 280f Marketing, of spine implants, 133 Marnay, Thierry, 14, 318 Material property(ies), 52–55 biocompatibility as, 54 of ceramics, 57 corrosion resistance as, 53 ductility as, 54, 54f of metals, 57–59 modulus of elasticity as, 54 of polymers, 55–57 strength as, 53, 53f toughness as, 53, 53f wear resistance as, 53–54 Matrix homeostasis, growth factors and other biologic molecules in, 649 Matrix metalloproteinases (MMPs), in gene therapy, 674, 677–678, 678f Maverick Total Disc Replacement, 353–362 advantages/disadvantages of, 361 All-in-1 guide for, 356, 357f, 360 A-MAV (direct anterior approach) version of, 354, 354f bone-prosthesis contact surface of, 355, 356f center of rotation of, 355, 355f, 356 clinical presentation and evaluation for, 357–358, 358t complications of, 361 description of, 24–25, 354–356, 354f discussion of, 361 end plates of, 355–356, 355f, 356f hydroxyapatite coating of, 355, 356f indications/contraindications for, 346, 354f lordosis angle of, 359–360 malpositioned, lateral revision of, 386, 386f, 387f, 388f metal-on-metal articulation of, 354–357, 355f O-MAV (oblique approach) version of, 354, 355f operative technique for, 358–360, 359f, 360f postoperative care for, 360–361 scientific testing/clinical outcomes for, 356–357 sizes of, 355–356, 356f, 359 MaXcess retractor, for lateral lumbar total disc replacement, 368–369, 369f, 370f MBB (medial branch block), 101–102, 102f M-Brace, 762 McKenzie, Alvin, 8, 452, 453 Mean center of rotation (MCR), with Mobi-C, 234, 236, 236f
Index Mechanical pain, lumbar revision due to, 378–379 Mechanical testing, of Coflex implant, 536, 539f MED (microendoscopic discectomy) system, 162–165, 164f, 165f Medial branch block (MBB), 101–102, 102f MEMS (microelectromechanical systems), 765–766 Merino sheep model, for disc degeneration, 641t Mesenchymal stem cell (MSC) transplantation in vivo experiment with, 667–670, 668f, 669f sources of cells for, 664–665, 664f Metal(s) corrosion of, 54–55 factors controlling, 54 kinetic barriers to, 54–55 resistance to, 53 thermodynamics of, 54 material properties of, 55–59 Metal ball arthroplasty, 7–9, 8f, 9f Metal-on-metal (MOM) articulation, of Maverick Total Disc Replacement, 354–357, 355f Methylene blue, for disc ruptures, 763 METRx system, 163–164 Michelin fractal airless metal tire, 4–5, 5f Microdiscectomy, in midline approach to minimally invasive lumbar surgery, 159–160, 160f Microelectromechanical systems (MEMS), 765–766 Microelectronics, of Theken eDisc, 365–366, 366f Microendoscopic discectomy (MED) system, 162–165, 164f, 165f Midline approach, to minimally invasive lumbar surgery, 159–160, 160f Midline opening laminoplasty case study of, 601–603, 602f Kurokowa modification of, 597 outcome of, 599 technique of, 597, 598f Tomita modification of, 597, 598f, 599f Mini-anterior lumbar interbody fusion (Mini-ALIF), positioning for, 155–156, 156f Minimally invasive posterior approaches, to lumbar spine, 159–166 far lateral, 161–162, 162f, 163f midline, 159–160, 160f muscle-splitting, 162–165, 164f, 165f vs. open surgery, 159 paramedian, 160–161, 161f paraspinal, 160–165 via tubular retractor system, 162–165, 164f, 165f Minimally invasive procedures, cost effectiveness of, 136 Minipig models, for Orthobiom Spinal System, 714, 715f, 718f MLDD (multilevel lumbar disc disease), 725, 728 proximate, 728, 733f MMPs (matrix metalloproteinases), in gene therapy, 674, 677–678, 678f Mobi-C Cervical Disc Prosthesis, 231–237 complications of, 234–236 degrees of freedom of, 231, 233f description of, 231, 232f, 233f discussion of, 236–237 mean center of rotation with, 234, 236, 236f operative technique for, 234 postoperative care for, 234 radiologic evaluation of, 233, 234, 235f, 236f scientific testing/clinical outcomes of, 231–232, 233f, 234f, 234 Mobidisc prosthesis, 326–329 advantages/disadvantages of, 328 clinical presentation and evaluation of, 327, 328t complications of, 328 description of, 24, 326–327, 326f design of, 326, 326f discussion of, 328 operative technique for, 327–328
postoperative care for, 328 scientific testing/clinical outcomes of, 327 Mobile inlay in Activ-L Artificial Disc, 347f in Mobidisc prosthesis, 327f Modulus of elasticity, 54 MOI (multiplicities of infectivity), in gene therapy, 675 Molecular therapies, for disc degeneration, 764 MOM (metal-on-metal) articulation, of Maverick Total Disc Replacement, 354–357, 355f Motion analysis. See Intervertebral motion measurements. Motion preservation biomechanical challenges of, 12 changing patient pool in, 766 future of, 760–769 biomechanics and, 761–762 challenges for, 764–765 current technologies and, 762–763 currently emerging high-tech treatments and, 765–767 emerging new therapies and, 763–764 new knowledge and, 760 overview of, 760 pain imaging and, 761 pain mechanisms and, 760–761 history and evolution of, 11–20 limiting factors for, 766–767 patient selection in, 766 responsible use of, 767 Motion preservation device(s) articulating plate, 13–14, 14f ball bearing, 12, 12f ball-and-socket, 13, 13f biologic, 18–19 biomechanical testing of (See Biomechanical testing) biomechanics of, 7, 7f classification of, 12–19, 21–35 complex mechanical/vertebral body replacement, 17, 17f core questions on, 11 fixed dome, 13, 13f fluid-filled bag, 17–18 and functional spinal unit, 22 with fusion, 605 hard plate/hard core, 14–15, 14f, 15f hard plate/soft core, 15, 15f, 16f historical background of, 7–8, 8f in situ polymerization, 18 intervertebral metal ball arthroplasty as, 8–9, 8f, 9f lessons learned on, 19 for nucleus replacement (See Nucleus replacement) overview of, 22–33 for posterior dynamic stabilization (See Posterior dynamic stabilization [PDS]) prerequisites for, 7 regulation of, 47–48 requirements for, 7–9, 7f, 23 role of surgeons in design of, 132–133 screw-in dowel, 16, 16f simple elastomer/polymer, 18, 18f spiral nucleoplasty, 18, 18f spring and piston, 16–17, 17f for total disc replacement (See Total disc replacement [TDR] devices) Motion preservation surgery basis for, 1–10 goals of, 75 patient selection for, 7 Motion preservation technology(ies) combined, future of, 763 cost effectiveness of discussion of, 136–137 future trends in, 135–136
783
issues in measuring, 132–134 ways to improve, 134–135 socioeconomic impact of, 132–138 Motion segment analysis, of TOPS device, 553, 556f Motion testing, of FENIX Facet Resurfacing Implant, 586–588, 587f, 589f, 589t Motor nerve root palsy, after cervical laminoplasty, 600 Mouse models, for disc degeneration, 641t due to axial loading, 641t, 643 knockout, 641t, 645–646 naturally occurring, 640, 641t due to postural change, 641t, 643 MRI. See Magnetic resonance imaging (MRI). MSC (mesenchymal stem cell) transplantation in vivo experiment with, 667–670, 668f, 669f sources of cells for, 664–665, 664f MSC.ADAMS software, in biomechanical model, 691, 692f, 692t Multidirectional flexibility testing, of NUBAC disc arthroplasty device, 445, 445f, 448, 448f Multilevel anterior corpectomy and discectomy, cervical laminoplasty vs., 600 Multilevel hybrid constructs, 604–605, 605t Dynesys system plus ProDisc-L prosthesis in, 607, 608f Multilevel lumbar disc arthroplasty, 725–734 case studies on for five levels, 728, 732f for four levels, 728, 731f for three levels, 728, 730f for two levels, 728, 729f clinical outcomes for, 727 complications of, 727, 727f contraindications for, 726 description of device for, 726, 726f discussion of, 728–733, 733f indications for, 726 materials and methods for, 727 operative techniques for, 726 patient selection for, 725 postoperative care for, 726–727 proximate, 728, 733f Multilevel lumbar disc disease (MLDD), 725, 728 proximate, 728, 733f Multiplicities of infectivity (MOI), in gene therapy, 675 Muscle cross-sectional area, 689 Muscle gain, in biomechanical model, 689, 690 Muscle recruitment patterns, 688 Muscle stimulation, animal models of disc degeneration due to, 641t, 645 Muscle-splitting paraspinal approach, to minimally invasive lumbar spinal surgery, 162–165, 164f, 165f Myelopathy, cervical Bryan artificial disc for, 193 cervical disc arthroplasty with, 185–186, 742–750 clinical studies of, 748–749, 749t indications for, 81, 82 recommendations on, 749 theoretical considerations for, 744–748, 745f, 746f, 747f, 748f natural history and conservative management of, 743 NeoDisc device for, 228f, 229f pathophysiology of, 742–743 presentation of, 742 Prestige Cervical Disc for, 199 surgical management of anterior approaches for, 744 background of, 742–744 posterior approaches for, 743–744 Myopathy, cervical, cervical laminoplasty for, 595
784
Index
N Neck Disability Index (NDI) for Bryan Cervical Disc, 195f for cervical arthroplasty adjacent to fusion, 753, 753f, 753t, 754f for M6 Artificial Cervical Disc, 274–275, 274f for Mobi-C, 233–234, 234f, 235f for NeoDisc device, 227 for Porous Coated Motion arthroplasty, 204 for Secure-C device, 249–251, 251f Neck fascia, in cervical revision, 278 Neck pain, 80, 81 NeoDisc device, 29, 221–230 advantages/disadvantages of, 230b animal data on, 225–227, 226f clinical presentation and evaluation for, 227, 228f, 229f complications of, 230 description of, 222–223, 223f discussion of, 230 indications/contraindications for, 222 operative technique for, 227–230 postoperative care for, 230 rationale and design criteria for, 221–222 scientific testing/clinical outcomes of, 223–225 for axial compression, 224 benchmark mechanical evaluation in, 223–224 biomechanical, 225 for pushout (expulsion), 225 for shear compression, 224–225 for wear, 223–224 wear debris from, 226–227 Neovascularization, with disc ruptures, 763 Nerve entrapment, due to lumbar total disc arthroplasty, 390, 392f NeuDisc hydrogel implant, 411–416 advantages/disadvantages of, 415 with coflex system, 610, 610f description of, 412, 412f, 413f design of, 411–412 discussion of, 415 operative technique for, 414–415, 415f scientific testing of, 412–414 for biocompatibility, 413 for clinical outcomes, 413–414, 414f for endurance, 412–413 for expulsion, 413 Neural subsystem, of spinal stability, 476 Neurogenic claudication, Coflex device for, 534, 535 Neurogenic intermittent claudication (NIC), X-STOP for, 541 Neurogenic leg pain, after lumbar total disc arthroplasty, 393 Neurologic complications of anterior exposure of lumbar spine, 149t, 151–153, 152f of CerviCore Intervertebral Disc prosthesis, 245 Neuropathic pain, lumbar revision due to, 378–379 Neurotoxicity rabbit model, of particulate wear debris, 72–73, 73f, 74f NeuroVision guidance, for lateral lumbar total disc replacement, 368–369, 369f Neutral zone (NZ) in intervertebral motion measurement, 89 Stabilimax NZ system and, 476, 477, 477f TOPS device and, 552, 552f and Wallis Stabilization System, 523–524, 524f New Zealand white rabbit model, for disc degeneration, 641t, 643 NFlex system, 505–510 advantages/disadvantages of, 508 clinical applications of, 507–508, 509f description of, 31, 31f, 505–506, 506f
pedicle screws in, 505, 506f rod in, 505, 506f sleeve in, 505, 506, 506f discussion of, 508–509 rationale for, 505 scientific testing of, 506–507 cadaver studies in, 507, 508t static compression bending in, 506–507, 507t static tension and compression in, 507, 508f, 508t static torsion in, 507, 507t NIC (neurogenic intermittent claudication), X-STOP for, 541 Nitinol, in Barricaid device, 631 Non-constrained lumbar disc prosthesis. See Mobidisc prosthesis. Noninferiority study design, for lumbar arthroplasty, 127, 128 Nonparametric test, for lumbar arthroplasty, 127, 128 Nonviral vectors, for gene therapy, 674 NP. See Nucleus pulposus (NP). NR. See Nucleus replacement (NR). NUBAC disc arthroplasty device, 442–451 advantages/disadvantages of, 450–451 clinical presentation and evaluation of, 448, 449f complications of, 450 description of, 443–444, 444f, 445f design rationale for, 442 discussion of, 451 indications/contraindications for, 442–443 operative techniques for, 448–450 postoperative care for, 450 scientific testing of, 444–448 animal study in, 447–448, 448f for biocompatibility and biodurability, 444 for disc height and biomechanics, 445–446, 446f for expulsion, 446 for kinematics, 444–445 for multidirectional flexibility, 445, 445f, 448, 448f for static and fatigue strength, 447 for wear, 446–447 Nucleoplasty lumbar, baboon model of, 65–68, 69f, 70f spiral, 18, 18f Nucleus pulposus (NP) anatomy and physiology of, 5–6, 6f, 639 in functional spinal unit, 36, 37f herniated adjacent to fusion, 615 at apex of scoliotic curve, 614–615 and contralateral facet (synovial) cyst, 615 dynamic pedicle-screw stabilization with nucleus replacement for, 614–615 Dynesys Spinal System for, 468 foraminal, necessitating vigorous unilateral medial facetectomy, 614 large central, 615 recurrent, 615 with spondylolysis without listhesis, 614 pathologic changes in, 6–7 Nucleus pulposus (NP) cells, cell transplantation therapy with in vitro study of, 665, 665f, 666f in vivo experiment with, 665–667, 667f sources of cells for, 663, 664 Nucleus pulposus (NP) implant, baboon model of, 65–68, 69f, 70f Nucleus replacement (NR) Aquarelle hydrogel disc nucleus for, 423–430 BioDisc Nucleus Pulposus Replacement for, 431–434 classification of devices for, 27–30, 29t, 397, 398f
DASCOR device for, 395–406 dynamic pedicle-screw stabilization with, 612–616 future of, 762 interspinous implants with, 610, 610f NeuDisc hydrogel implant for, 411–416 NUBAC disc arthroplasty device for, 442–451 NuCore Injectable Nucleus for, 417–422 options for, 616t PDN-SOLO and HydraFlex devices for, 407–410 primary function of, 613 Satellite Spinal System for, 452–462 TranS1 percutaneous, 435–441 NuCore Injectable Nucleus, 417–422 description of, 29, 29f, 418, 419t development of, 417 discussion of, 421 early clinical results with, 419–420, 420f, 421f indications/contraindications for, 418 operative technique for, 420 preclinical testing of, 419–420, 419f rationale for, 417 NZ. See Neutral zone (NZ).
O ODI. See Oswestry Disability Index (ODI). O-MAV version, of Maverick Total Disc Replacement, 354, 355f One-component prostheses, for total disc replacement, 26–27 AcroFlex, 26 Bryan, 26 CAdisc-L and CAdisc-C, 27, 27f eDisc, 27, 27f Freedom Lumbar Disc, 26, 26f M6 , 26, 26f Physio-L, 26, 26f OP-1. See Osteogenic protein-1 (OP-1). OPDG (Oswestry Low Back Pain Disability Questionnaire), for autologous disc chondrocyte transplant, 681, 681f Open-door laminoplasty outcome of, 599 technique of, 597, 597f, 598f OPLL. See Ossification of the posterior longitudinal ligament (OPLL). Orthobiom Spinal System, 711–719 complications of, 716–717 contraindications for, 712 description of, 711–712, 712f functional animal study of, 714–715, 715f indications for, 712 operative technique for, 715–716 outcomes of, 717–719, 718f, 718t with mobile fixation, 717–718, 718f, 718t with rigid fixation, 717, 718f, 718t postoperative care for, 716 rationale for, 711 scientific testing of, 712–714 component testing in, 712–713 construct testing in, 713–714 Orthopaedic Feature Stabilization, 88 Ossification, heterotopic, after cervical arthroplasty, 298f, 299f, 300, 301 Ossification of the posterior longitudinal ligament (OPLL) cervical laminoplasty for, 595, 599 as contraindication for cervical disc arthroplasty, 186 as contraindication to cervical nonfusion surgery, 85f Osteoarthritis, FENIX Facet Resurfacing Implant for, 585–586
Index Osteogenic protein-1 (OP-1) in gene therapy, 656, 676f, 677f for intervertebral disc regeneration, 763 after chemonucleolysis, 655–657, 656f in vivo studies of, 651–655, 653f, 654f, 655f Osteoporosis as contraindication for cervical disc arthroplasty, 186 spinal arthroplasty with, 721–722, 723f Osteoporotic fracture(s), percutaneous treatment of, future of, 763 Oswestry Disability Index (ODI) for Activ-L Artificial Disc, 350 for CHARITÉ Artificial Disc, 311, 358, 358t for Cosmic system, 494 for DASCOR device, 402, 403f for Dynamic Stabilization System, 473 for FlexiCore Intervertebral Disc, 335, 335f for IsoBar TTL Dynamic Rod Stabilization, 487 for Kineflex Disc, 340, 341, 341f, 345 for lumbar disc arthroplasty, 127–130, 128f, 129f adjacent to scoliotic deformity, 706, 706f multilevel, 727 for Maverick Total Disc Replacement, 357, 358, 358t for Mobidisc prosthesis, 327, 328t for NUBAC disc arthroplasty device, 448, 449f for NuCore Injectable Nucleus, 420 for PercuDyn device, 512, 512f, 513t for ProDisc-C artificial cervical disc, 216, 217f for ProDisc-L artificial lumbar disc, 320 for simultaneous lumbar fusion and total disc replacement, 618–619 for TOPS device, 559, 563f for TranS1 Percutaneous Nucleus Replacement, 435, 437 for Wallis Stabilization System, 525–526, 526f, 527f, 531–532 for X-STOP, 544 Oswestry Low Back Pain Disability Questionnaire (OPDG), for autologous disc chondrocyte transplant, 681, 681f Ovine model, NeoDisc device in, 225–226, 226f Oxidation, and corrosion, 54 Oxide film, as kinetic barrier to corrosion, 54–55
P Pain discogenic, 98 concordant vs. nonconcordant, 104t definite, 103 indeterminate, 104 mechanisms of, 761 probable, 104 unequivocal, 98 Wallis Stabilization System for, 526–527, 528f, 529f facet joint–mediated, 98 with flexion vs. extension, 97–98 growth factor therapy and, 659 inflammatory mediators of, 761 after lumbar total disc arthroplasty, 389–394 due to disease progression, 392–393 early-onset, 393 due to failure of prosthesis, 390–392 due to failure to achieve surgical goals, 390 due to incorrect procedure, 389–390, 391f late-onset, 394 due to poor patient selection, 389 subacute, 393–394 due to technique-related problems, 390, 392f, 393f Pain imaging, and future of motion preservation, 761
Pain mechanisms, and future of motion preservation, 760–761 Papio cynocephalus anubis, as animal model for disc degeneration, 641t, 643 Paramedian approach, to minimally invasive lumbar surgery, 160–161, 161f Paraspinal approaches, to minimally invasive lumbar surgery, 160–165 far lateral, 161–162, 162f, 163f muscle-splitting, 162–165, 164f, 165f paramedian, 160–161, 161f via tubular retractor system, 162–165, 164f, 165f Pars defect, persistent pain after lumbar total disc arthroplasty due to, 391f Pars fracture, after lumbar total disc replacement, lateral revision of, 383–384, 384f, 385f Particulate wear debris, neurotoxicity rabbit model of, 72–73, 73f, 74f Passive subsystem, of spinal stability, 476 Patient effort, in intervertebral motion measurement, 88–90, 91f Patient evaluation for cervical nonfusion surgery, 80–84 diagnostic studies in, 82 history in, 80–81, 81f physical evaluation in, 81–82, 82t for lumbar nonfusion surgery, 75–79 bone scintigraphy in, 77 CT scans in, 77 discography in, 77–78 flexion-extension radiographs in, 76–77 history in, 75–76 injection diagnostics in, 78 MRI in, 77 physical examination in, 76, 76f plain radiographs in, 76, 77f Patient positioning, for intervertebral motion measurement, 88–90, 91f Patient selection, criteria for, 135, 136–137 Patil device, 16, 17f PCM arthroplasty. See Porous Coated Motion (PCM) arthroplasty. PCS (physical component scores), for Wallis Stabilization System, 531 PDN-SOLO device. See Prosthetic Disc Nucleus (PDN)-SOLO device. PDR (Percutaneous Disc Reconstruction) device, 30 Pedal pulses, with anterior exposure of lumbar spine, 148 Pedicle screw(s) in Orthobiom Spinal System, 715–716 in posterior lumbar arthroplasty, 738, 738f in TOPS device, 552–553, 554f load on, 553–554, 557f operative technique for, 553, 555 Pedicle screw–based lumbar posterior dynamic stabilization Cosmic system for, 490–499 Dynamic Stabilization System for, 472–475 Dynesys Spinal System for, 463–471 IsoBar TTL Dynamic Rod Stabilization, 483–489 IST Dynamic Stabilization device for, 500–504 NFlex system for, 505–510 with nucleus replacement, 612–616 anatomic, biomechanical, and pathologic considerations for, 612–613 contraindications for, 615 discussion of, 615–616 indications for, 614–615 primary, 614–615 secondary, 615 options for, 615, 616t rationale for, 613 review of literature on, 613–614
785
PercuDyn device for, 511–516 Stabilimax NZ system for, 476–482 PEEK-on-PEEK self-mating material couple, in NUBAC disc arthroplasty device, 446–447 PEEK-OPTIMA. See Polyetheretherketone (PEEK)-OPTIMA. PercuDyn device, 511–516 advantages/disadvantages of, 513 clinical presentation for, 512–513, 514f clinical studies on, 512, 512t ODI outcomes in, 512, 512f, 513t radiologic evaluation in, 512, 514f VAS outcomes in, 512, 513f, 513t complications of, 513–514 contraindications for, 511 description of, 511, 512f discussion of, 514–515 indications for, 511 operative technique for, 513 postoperative care for, 513 rationale for, 511 scientific testing of, 512 Percutaneous Disc Reconstruction (PDR) device, 30 Percutaneous Nucleus Replacement (PNR), 435–441 advantages/disadvantages of, 440 clinical outcomes with, 437, 438f complications of, 440 description of, 30, 435–436, 436f discussion of, 440–441 indications/contraindications for, 435 patient preparation and operative technique for, 437–440 postoperative care for, 440 rationale for, 435 scientific testing of, 437 Percutaneous osteoporotic fracture treatment, future of, 763 Personalized hybrid EMG-assisted/finite element biomechanical model, 685–694 basic biomechanics of, 688 basic structure of, 688–690, 689f complexity of, 687 discussion of, 693–694 electromyography in, 688–689, 689f, 693f enhancements and improvements in, 691–693 finite element analysis in, 691 flexible discs in, 693f importation of subject-specific images into, 688, 693, 693f MSC.ADAMS software in, 691, 692f, 692t output measures of, 692t performance and validation of, 690, 690t, 691f realism of, 687 repeatability and variability of, 690–691 sensitivity to nonphysical stress of, 691 skeletal representation of, 692f uses of, 687 "virtual surgery" with, 691–693, 693f PG. See Proteoglycan(s) (PG). Pharmacologic agents, for back pain, 763–764 Physical component scores (PCS), for Wallis Stabilization System, 531 Physical examination for cervical nonfusion surgery, 81–82, 82t for lumbar nonfusion surgery, 76, 76f Physio-L prosthesis, 26, 26f Physiologic loading, of coflex implant, 536–537 Pig models for disc degeneration, 641t, 644–645 for Orthobiom Spinal System, 714, 715f, 718f Pintail mouse model, for disc degeneration, 640, 641t Pisharodi expandable device, 16–17, 17f Pitting corrosion, 53
786
Index
PLA. See Posterior lumbar arthroplasty (PLA). Plain radiographs for cervical nonfusion surgery, 82 for lumbar nonfusion surgery, 76, 77f Platelet-rich plasma (PRP), for intervertebral disc regeneration, 650, 652f PLIF (posterior lumbar interbody fusion), 736, 739 PLL. See Posterior longitudinal ligament (PLL). PMMA. See Polymethyl/methacrylate (PMMA). PNR (Percutaneous Nucleus Replacement), 435–441 Poisson's ratio, for IsoBar TTL Dynamic Rod Stabilization, 485–486, 486t Polyetheretherketone (PEEK)-OPTIMA in NUBAC disc arthroplasty device, 443–444 biocompatibility and biodurability of, 444 in Wallis Stabilization System, 523, 524f Polyethylene, ultra-high-molecular-weight, material properties of, 56 Polymer(s), material properties of, 55–56 Polymer devices, 18, 18f Polymethyl/methacrylate (PMMA) for disc injury, 764 material properties of, 55 for osteoporotic fractures, 763 Polytetrafluoroethylene (PTFE, Teflon, Gore-Tex) in Barricaid device, 631 material properties of, 55 Polyvinyl alcohol (PVA), in Aquarelle hydrogen disc nucleus, 425 animal study of, 427–429 biocompatibility of, 425, 429 swelling pressure testing of, 425–426, 426f Porous Coated Motion (PCM) arthroplasty, 202–213 for adjacent segment disease, 751–759 discussion of, 754–758, 755f, 756f, 757f materials and methods for, 752, 752f, 753f results of, 752–754 in international study, 752–754, 753f, 754f in U.S. prospective randomized study, 752, 753t advantages/disadvantages of, 212–213, 212b augmented, 202–203 bearing surface of, 203 with cervical laminoplasty, 601–603, 602f with cervical myelopathy, 748 clinical presentation and evaluation for, 204, 205t complications of, 211–213, 212f, 287–294, 288t alignment- and stability-related, 288t, 294 approach- and decompression-related, 288t, 289 fixation technique–related, 288t, 290–293, 292f, 293f, 294f, 295f implantation technique–related, 288t, 289–290, 291f movement-related, 288t, 294 constrained version of (PCM-C), 203f, 205t, 206f for corpectomies (PCM Corpomotion), 203f description of, 202–204 discussion of, 213 with excentric flange (PCM-EF), 203f failure rate for, 211–213, 212f "flanged" version of, 288, 289f goat model of, 64–65, 66f, 67f, 203–204, 207f indications/contraindications for, 204, 205t, 288 Investigational Device Exemption study of, 202 in level 5-6, 213f lordotic (PCM-L), 203f with modular flange, 203f modularity of, 202, 203f, 289f operative techniques for, 204–210, 209f postoperative care after, 211 "press-fit" version of, 288
scientific testing/clinical outcomes for, 204, 208f with titanium fusion cage (PCM-Ti), 203f trial range of motion check for, 294f Posterior approaches, to minimally invasive lumbar surgery, 159–166 far lateral, 161–162, 162f, 163f midline, 159–160, 160f muscle-splitting, 162–165, 164f, 165f vs. open surgery, 159 paramedian, 160–161, 161f paraspinal, 160–165 via tubular retractor system, 162–165, 164f, 165f Posterior cervical approach, 182 Posterior dynamic stabilization (PDS). See Lumbar posterior dynamic stabilization. Posterior ligamentous disruption, as contraindication to cervical nonfusion surgery, 83f Posterior longitudinal ligament (PLL) in cervical revision, 288, 291f in functional spinal unit, 37–39, 38f in Klippel-Feil syndrome, 291f in lumbar disc arthroplasty, 307, 307f ossification of cervical laminoplasty for, 595, 599 as contraindication for cervical disc arthroplasty, 85f, 186 Posterior lumbar arthroplasty (PLA), 735–741 anatomic considerations for, 735–736 anterior vs., 740, 740t biomechanical consideration(s) for, 736–738 angular motion as, 737 extent of motion as, 737 loading on end plate as, 737 matching instantaneous axis of rotation as, 737, 737f, 739 minimization of wear as, 737–738 parallel vs. nonparallel placement as, 737, 737f challenges of, 739 device for, 737, 738f discussion of, 740, 740t goals of, 736, 736t, 740 history of, 735 operative technique for, 738–739, 738f, 739f Posterior lumbar interbody fusion (PLIF), 736, 739 Posterior tangent method, of intervertebral motion measurement, 87 Posterolateral endoscopic approach, to lumbar spine. See Lumbar endoscopic posterolateral [transforaminal] approach. Postural change, animal models of disc degeneration due to, 641t, 643 Presacral percutaneous nucleus replacement. See Percutaneous Nucleus Replacement (PNR). Prestige Cervical Disc, 24–25, 199–201 advantages/disadvantages of, 201 clinical presentation and evaluation for, 200 complications of, 201 description of, 199, 200f development of, 199 discussion of, 201 indications/contraindications for, 199 operative techniques for, 200 postoperative care with, 200–201 scientific testing/clinical outcomes for, 199–200 Prestige LP, 199–200, 200f Prestige ST, 199–200, 200f Prevertebral fascia, in anterior cervical approach, 39, 39f Primate models, for disc degeneration, 641t, 643 ProDisc design, 14, 15f ProDisc-C artificial cervical disc, 214–220 for adjacent segment disease, 754, 755f
advantages/disadvantages of, 220b with cervical myelopathy, 748 complications of, 219, 219t description of, 24, 25f, 214–215, 215f, 216f disc heights of, 214–215 discussion of, 220, 220f historical background of, 215 indications/contraindications for, 214, 215t kinematics of, 214, 216f for multilevel disc replacement, 220 operative technique for, 216–219, 218f, 219f postoperative care for, 219 scientific testing/clinical outcomes of, 215–216, 216t, 217f, 218f ProDisc-L artificial lumbar disc, 318–325 adjacent to scoliotic deformity, 705–706 advantages/disadvantages of, 325 with coflex system, 606, 607f complications of, 320–325, 324t description of, 24, 25f, 318–319, 319f discussion of, 325 with Dynesys system, 605, 606f subsequent implantation of, 609, 609f for three-level procedure, 607, 608f historical background of, 318 indications/contraindications for, 318, 319t multilevel, 320, 321f, 726, 726f operative technique for, 320, 322f postoperative care for, 320 scientific testing/clinical outcomes of, 320, 321f in IDE trial, 320, 322t Proportional limit, 53f Prosthetic Disc Nucleus (PDN)-SOLO device, 407–410 advantages/disadvantages of, 409b clinical presentation and evaluation for, 408–409 complications of, 409 description of, 29, 407, 408, 408f discussion of, 409–410 indications/contraindications for, 407 operative technique for, 409 postoperative care for, 409 scientific testing/clinical outcomes of, 408 Proteoglycan(s) (PG), and Aquarelle hydrogel disc nucleus, 423 Proteoglycan (PG) synthesis bone morphogenetic protein-2 and, 650 with gene therapy, 675–676, 676f growth and differentiation factor-5 and, 650, 651f in nucleus pulposus cell activation, 650, 666f platelet-rich plasma and, 650, 652f recombinant human osteogenic protein-1 and, 650, 650f, 655 Provocative discography. See Discography. Proximate multilevel lumbar disc disease, 728, 733f PRP (platelet-rich plasma), for intervertebral disc regeneration, 650, 652f Psammomys obesus, as animal model for disc degeneration, 640, 641t Pseudoarthrosis, functional definition of, 85–86 Pseudomeningocele, after anterior exposure of lumbar spine, 153 Psoas abscess, after anterior exposure of lumbar spine, 153 Psychological problems, and persistent pain after lumbar total disc arthroplasty, 389 Psychosocial stress, biomechanical model sensitivity to, 691 PTFE (polytetrafluoroethylene) in Barricaid device, 631 material properties of, 55 PVA. See Polyvinyl alcohol (PVA).
Index
Q Quality assurance, 767 Quantitative motion analysis. See Intervertebral motion measurements. Quebec Back Pain Disability Scale (QBPD), for autologous disc chondrocyte transplant, 681–682, 682f
R Rabbit model(s) of cell transplantation therapy with mesenchymal stem cells, 667–670, 668f, 669f with nucleus pulposus cells, 665–667, 667f for disc degeneration, 641t due to adjacent level lumbar fusion, 641t, 644 due to annulotomy, 641t, 644 due to axial loading, 641t, 643 due to fibronectin fragment injection, 641t, 645 due to muscle stimulation, 641t, 645 naturally occurring, 640, 641t due to resection of spinal process or facet joint, 641t, 644 due to smoking, 641t, 645 due to torsional injury, 641t, 643 of growth factor injection therapy after chemonucleolysis, 655–657, 656f with growth and differentiation factor5, 656–657, 657f, 658f with osteogenic protein-1, 651–655, 653f, 654f, 655f NeoDisc device in, 226–227 Orthobiom Spinal System in, 714–715 of particulate wear debris, 72–73, 73f, 74f Radicular pain, inflammatory mediators of, 761 Radiculopathy, 80, 82 after lumbar total disc arthroplasty, 393 Randomized clinical trials, of lumbar artificial disc, 126–131 discussion of, 128–130 introduction to, 126–127 materials and methods for, 127 results of, 127–128, 129f, 617, 618f statistical methods for, 127, 128f Range of motion (ROM), 21–22 and annular repair, 630 in biomechanical testing, 46, 47f, 48, 50 in cervical arthroplasty, 601 in cervical laminoplasty, 600–601 in cervical revision, 290, 292f of coflex implant, 537, 538 and neutral zone, 476, 477 of NUBAC disc arthroplasty device, 445, 445f, 446, 448, 448f of TOPS device, 553, 556f of Total Facet Arthroplasty System, 568–569, 570f Rat models, for disc degeneration, 641t due to axial loading, 641t, 643 due to chemonucleolysis, 645 naturally occurring, 640, 641t due to postural change, 641t, 643 due to resection of spinal process or facet joint, 643 due to smoking, 641t, 645 RBSS (refractory back surgical syndrome), 377–378 Recentering, with Kineflex Disc, 338–339 Recombinant human osteogenic protein-1 (rhOP-1), for intervertebral disc regeneration, 651–655, 653f, 654f, 655f after chemonucleolysis, 655–657, 656f Rectus sheath hematomas, after anterior exposure of lumbar spine, 153
Recurrent disc herniation with Barricaid device, 634, 634f, 635 Cosmic device for, 491 Recurrent laryngeal nerve (RLN), 39–40 in cervical revision, 278 Reference data, for intervertebral motion, 90–93, 90t, 92t, 93f, 93t Referred pain, from lumbar facet joints, 101, 101f Refractory back surgical syndrome (RBSS), 377–378 Region of interest (ROI) methods, of intervertebral motion measurement, 87 Reitz, H., 8 Retrograde ejaculation, due to anterior exposure of lumbar spine, 150, 151f Retroperitoneal approach, to anterior lumbar spine, 139–147 advantages of, 146–147 anatomy of, 41, 42f comment on, 146–147, 149t complications of, 147, 148–154 miscellaneous, 149t, 153 neurologic, 149t, 151–153, 152f rate of, 149t urologic, 149t, 150, 151f vascular, 148–150, 149f, 149t, 150f entry into retroperitoneal space in, 141–142, 143f, 144f, 145f mobilization of blood vessels in, 142–146, 145f, 146f other technical considerations in, 146 patient preparation for, 141 skin incision for, 141, 142f Revision. See Cervical revision; Lumbar revision. rhOP-1 (recombinant human osteogenic protein-1), for intervertebral disc regeneration, 651–655, 653f, 654f, 655f after chemonucleolysis, 655–657, 656f RLN (recurrent laryngeal nerve), 39–40 in cervical revision, 278 Robinson, Robbie, 277, 279f ROI (region of interest) methods, of intervertebral motion measurement, 87 ROM. See Range of motion (ROM). Rotation, of motion preservation device, 7 Rotational stability, with lumbar arthroplasty, 695–705, 695, 697f scientific testing of, 695–700 anatomic measurements in, 695–696, 700 biomechanical testing in, 697–701, 701f, 703f discussion of, 701–704 results of, 700–701 Rotational testing with lumbar arthroplasty, 695–700 anatomic measurements in, 695–696, 700 biomechanical testing in, 697–701, 701f, 703f discussion of, 701–704 results of, 700–701 of single-level and double-level CHARITÉ Artificial Disc, 700, 701f of soft tissue preparation for disc replacement, 697–700 uncovertebral joint analysis in, 700–701, 703f
S Sabitzer and Fuss device, 13, 13f Safety analysis, of TOPS device, 560 Sagittal balance in cervical revision, 294 loss of, 297–298 with X-STOP, 545 Salib and Pettine articulating plate design, 13, 14f
787
Salvage procedures, dynamic pedicle-screw stabilization with nucleus replacement for, 615 Samani, Jacques, 534 Same segment degeneration, lumbar, after spinal fusion, 123 Sand rat model, for disc degeneration, 640, 641t Satellite Spinal System, 452–462 advantages/disadvantages of, 460–461 adverse effects of, 453, biomechanics of, 454, 455f contraindications for, 453, 453t description of, 454 development of, 452–453 indications for, 453 operative technique for, 456–460 closure in, 460 distraction and disc removal in, 456, 458f exposure in, 456, 458f final preparation in, 457–460, 459f implantation in, 460, 460f preoperative planning and patient positioning for, 456, 458f trial in, 456–457, 458f, 459f patient selection for, 453–454 postoperative care for, 460 scientific testing/clinical outcomes of, 454–460, 456t, 457f Schellnack Buttner-Janz (SB) device, 14, 14f Schellnack, Kurt, 309 Sciatica, Wallis Stabilization System for, 528–529, 529f, 530f Scoliosis classification of, 705 as contraindication to lumbar disc arthroplasty, 695 defined, 711 iatrogenic, 695, 697f with keeled prosthesis, 696f Orthobiom Spinal System for, 711–719 complications of, 716–717 contraindications for, 712 description of, 711–712, 712f functional animal study of, 714–715, 715f indications for, 712 operative technique for, 715–716 outcomes of, 717–719, 718f, 718t with mobile fixation, 717–718, 718f, 718t with rigid fixation, 717, 718f, 718t postoperative care for, 716 rationale for, 711 scientific testing of, 712–714 component testing in, 712–713 construct testing in, 713–714 Scoliotic curve, herniated nucleus pulposus at apex of, dynamic pedicle-screw stabilization with nucleus replacement for, 614–615 Scoliotic deformity, lumbar disc replacement adjacent to, 705–710 classification of, 705–706, 706t clinical background and methods of, 705–706, 706t discussion of, 706–709 results of, 706 Oswestry Disability Index in, 706, 706f for type 2a patient, 706, 708f, 709f, 710f for type Ia patient, 706, 707f for type Ib patient, 706, 707f, 708f Visual Analog Scale in, 706, 706f Screw(s) in Cosmic system, 30, 30f, 491, 495f pedicle (See Pedicle screw[s]) for posterior dynamic stabilization, 30, 30f Screw coating, in Cosmic system, 491–492 Screw fixation angle, in Cosmic system, 493, 496f Screw-in dowel devices, 16, 16f
788
Index
Secondary gain, and persistent pain after lumbar total disc arthroplasty, 389 Secure-C Cervical Artificial Disc, 24, 247–253 advantages/disadvantages of, 253b clinical presentation and evaluation for, 249–251, 250t, 251t, 252f Neck Disability Index in, 249–251, 251f Visual Analog Scale in, 249–251, 251f complications of, 251–253 configurations of, 248 description of, 247–248, 248f discussion of, 253 indications/contraindications for, 247, 248t operative technique for, 249, 250f postoperative care for, 249 range of motion of, 247–248, 251 Selective nerve root block (SNRB), 99–101 complications of, 100–101 diagnostic value of, 100 indications for, 99 injection location with, 99–100 technique of, 99, 99f therapeutic value of, 100 vs. transforaminal block, 99, 99f Self-retaining retractor (Syn-Frame), in anterior exposure of lumbar spine, 142, 145f Semi-constrained cervical arthroplasty, 267–268 See also ProDisc-L artificial lumbar disc. Semi-constrained lumbar arthroplasty. See Activ-L Artificial Disc; ProDisc-L artificial lumbar disc. Semilunar line of Douglas, in anterior exposure of lumbar spine, 141–142, 143f Sensors, of Theken eDisc, 365–366, 366f SF-36. See Short form 36 (SF-36). Shear fatigue testing, of Kineflex Disc, 339–340 Shear loading, in biomechanical model, 691f Sheep model for disc degeneration, 641t NeoDisc device in, 225–226, 226f Shock absorption, 22 Short form 36 (SF-36) for cervical arthroplasty adjacent to fusion, 753–754 for IsoBar TTL Dynamic Rod Stabilization, 487 for NuCore Injectable Nucleus, 420f for Wallis Stabilization System, 525–526, 527f, 528f Simple elastomer/polymer devices, 18, 18f Single-level hybrid constructs, 604, 605t Slit-and-rib design, 13–14, 14f SLN (superior laryngeal nerve), 39 Smith-Robinson approach, revision of. See Cervical revision. Smoking, animal models of disc degeneration due to, 641t, 645 SNRB. See Selective nerve root block (SNRB). Socioeconomic impact of low back pain, 132 of motion preservation technology, 132–138 discussion of, 136–137 future trends in, 135–136 issues in measuring, 132–134 ways to improve, 134–135 Soft stabilization, defined, 484 Sox 9, in gene therapy, 677 Spherical partial disc replacement, 452–462 advantages/disadvantages of, 460–461 adverse effects of, 453, 454t biomechanics of, 454, 455f contraindications for, 453, 453t description of, 454 development of, 452–453 indications for, 453 operative technique for, 456–460 closure in, 460
distraction and disc removal in, 456, 458f exposure in, 456, 458f final preparation in, 457–460, 459f implantation in, 460, 460f patient positioning for, 456, 458f trial in, 456–457, 458f, 459f patient selection for, 453–454 postoperative care for, 460 scientific testing/clinical outcomes of, 454–460, 453t, 457f Spinal biomechanics, and adjacent level degeneration, 118–119, 119f, 120f Spinal column, 4–7 intervertebral disc in, 5–7, 6f vertebral body in, 4–5, 4f, 5f Spinal decompression, problems with, 7 Spinal fusion. See Fusion. Spinal process resection, animal models of disc degeneration due to, 641t, 643–644 Spinal stability, subsystems of, 476 Spinal stabilization, problems with, 7 Spinal stenosis, lumbar circumferential, 720, 721f coflex implant for, 534, 535 Cosmic device for, 491, 492f Dynesys Spinal System for, 465, 466f, 468 TOPS device for, 3, 551–552 Total Facet Arthroplasty System (TFAS) for, 565 Wallis Stabilization System for, 530–531 X-STOP for, 541 Spinal unit loads, in biomechanical model, 689 Spine critical load of, 4 dimensions of, 4 physiologic curve of, 22 Spine arthroplasty, defined, 21 Spine arthroplasty devices. See Motion preservation device(s). Spine loading, in biomechanical model, 690, 690t, 691f Spine testers, configurations for, 48–49, 49f Spinous process plate, future of, 762–763 Spiral nucleoplasty, 18, 18f Spondylolisthesis degenerative, X-STOP with, 544–545 Dynesys Spinal System for, 465, 468 persistent pain after lumbar total disc arthroplasty due to, 390 Spondylolysis herniated nucleus pulposus with, dynamic pediclescrew stabilization with nucleus replacement for, 614 persistent pain after lumbar total disc arthroplasty due to, 390, 391f Spondylosis, lumbar Activ-L artificial disc for, 346 history of, 75–76 Spondylotic cervical myelopathy cervical disc arthroplasty with, 185–186, 742–750 clinical studies of, 748–749, 749t recommendations on, 749 theoretical considerations for, 744–748, 745f, 746f, 747f, 748f cervical laminoplasty for, 595 natural history and conservative management of, 743 pathophysiology of, 742–743 presentation of, 742 surgical management of, 743 anterior approaches for, 744 posterior approaches for, 743–744 Sprague-Dawley rat models, for disc degeneration, 641t, 643 Spring and piston devices, 16–17, 17f
Stabilimax NZ system, 476–482 clinical data on, 480, 481f description of, 32, 477, 477f dual-spring mechanism in, 479f dynamic connectors in, 477, 478f end connector in, 478f pedicle screws in, 477, 478f, 479f development and biomechanical testing of, 476–477 discussion of, 480 indications/contraindications for, 477 operative procedure for, 477–480 device placement in, 477–480, 479f, 480f exposure in, 477 positioning in, 477 rationale for, 476, 477f, 478f Stability, of motion preservation device, 7 Stainless steel alloys material properties of, 57–58 new, 59 Static axial compression bending, of Orthobiom Spinal System, 713 Static compression testing of coflex implant, 536 of Kineflex Disc, 339 Static shear testing, of Kineflex Disc, 339 Static strength test, of NUBAC disc arthroplasty device, 447 Static torsion of coflex implant, 536 of Orthobiom Spinal System, 714 Static torsion bending, of Orthobiom Spinal System, 712–713 Statistical outcome interpretation, of randomized clinical trials, 126–131 Steel ball arthroplasty, 7–9, 8f, 9f Stem cell(s), for disc injury, 764 Stem cell transplantation, mesenchymal, in vivo experiment with, 667–670, 668f, 669f Sternocleidomastoid muscle, in anterior cervical approach, 39, 39f Steroids, intradiscal injection of, 763 Strength, 53, 53f Strength testing, of Total Facet Arthroplasty System, 567–568, 569f Stress analysis, of TOPS device, 553, 555f Stress-strain curve, 53, 53f Stubstad implant, 18, 18f Student's t-test, for lumbar arthroplasty, 127 Subject-specific images, in biomechanical model, 688, 693, 693f Subsidence, 721–722, 723f with multilevel lumbar disc arthroplasty, 727, 727f Superior facet burr, for FENIX Facet Resurfacing Implant, 589 Superior hypogastric plexus injury, in anterior exposure of lumbar spine, 150, 151, 151f, 152f Superior laryngeal nerve (SLN), 39 Superior Spacer, 33, 34f Supraspinous ligaments, 38f Surface fatigue, 60f, 61 Sus scrofa, as animal model for disc degeneration, 641t, 644–645 Swelling pressure, with Aquarelle hydrogel disc nucleus defined, 423 and equilibrium water content, 423–424, 424f scientific testing of, 425–426, 426f Sympathetic chain injury, in anterior exposure of lumbar spine, 151, 151f, 152f, 153 Syn-Frame, in anterior exposure of lumbar spine, 142, 145f Synovial cyst, herniated nucleus pulposus and contralateral, 615
Index
T Tail suspension, animal models of disc degeneration due to, 641t, 643 Tantalum (Ta) alloys, material properties of, 59 TDA (total disc arthroplasty) cervical (See Cervical disc arthroplasty) cost effectiveness of, 134 lumbar (See Lumbar total disc arthroplasty) TDR. See Total disc replacement (TDR). Teflon (polytetrafluoroethylene) in Barricaid device, 631 material properties of, 55–56 Temporary external transpedicular fixation (TETF), 105–106 complications of, 106 indications for, 105 results of, 105–106 technique of, 105, 105f Temporomandibular joint, in cervical revision, 282, 283f Tensile strength, 53, 53f Tension trabeculae, 4–5, 5f TETF. See Temporary external transpedicular fixation (TETF). TFAS. See Total Facet Arthroplasty System (TFAS). TGF-b1 (transforming growth factor-b1), in gene therapy, 675, 675f, 676f Theken eDisc, 363–367 advantages/disadvantages of, 367b description of, 27, 27f, 364, 364f discussion of, 367 elastomeric core of, 364, 364f event-triggered recording of loads for, 366–367, 366f first-generation artificial discs vs., 363, 364f hydrolytic stability of, 364–365, 365f indications/contraindications for, 363 load sharing by, 364, 365f microelectronics of, 365–366, 366f operative techniques for, 367 postoperative care for, 367 retained tensile strength of, 364–365, 365f scientific testing/clinical outcomes of, 364–367 stiffness comparison of, 364, 365t Third bodies, in tribologic system, 60 Three-component prostheses, for total disc replacement, 23–24 Baguera, 24, 24f CHARITÉ, 23, 23f Dynardi, 23–24, 24f Kineflex and Kineflex|C, 23, 24f Mobidisc, 24 Secure C, 24 Three-Dimensional Fabricube (3DF) Device, baboon model of, 68, 70f, 70t, 71f Ti (titanium) alloys, material properties of, 58–59 TiCaP (titanium calcium phosphate)-coated CHARITÉ device. See Porous Coated Motion (PCM) arthroplasty. TiCaP (titanium calcium phosphate) porous ingrowth, 202 Time, procedure, complication (TPC) system, for lumbar revision, 377 Tissue engineering, for disc degeneration, 763–764 Tissue inhibitors of matrix metalloproteinase (TIMPs), in gene therapy, 674, 677–678 Tissue regeneration, 764–765 Titanium (Ti) alloys material properties of, 58–59 new, 59 Titanium calcium phosphate (TiCaP)-coated CHARITÉ device. See Porous Coated Motion (PCM) arthroplasty.
Titanium calcium phosphate (TiCaP) porous ingrowth, 202 TLIF (total lumbar interbody fusion), 736, 738, 739 TNF-a (tumor necrosis factor-a) inhibitor, for back pain, 763 Tomita modification, of French door laminoplasty, 597, 598f, 599f TOPS. See Total Posterior Arthroplasty System (TOPS). Torsion fatigue testing, of coflex implant, 536 Torsion testing of coflex implant, 536 of FENIX Facet Resurfacing Implant, 586–588, 587f Torsional injury, animal models of disc degeneration due to, 641t, 643 Total disc arthroplasty (TDA) cervical (See Cervical disc arthroplasty) cost effectiveness of, 134 lumbar (See Lumbar total disc arthroplasty) Total disc replacement (TDR) cervical (See Cervical total disc arthroplasty) cost effectiveness of, 134 future of, 762 lumbar (See Lumbar total disc arthroplasty) Total disc replacement (TDR) devices, 23–27, 23t advantages of, 27 with dorsal approach, 27 fixation of, 27 one-component, 26–27 AcroFlex, 26 Bryan, 26 CAdisc-L and CAdisc-C, 27, 27f eDisc, 27, 27f Freedom Lumbar Disc, 26, 26f M6 , 26, 26f Physio-L, 26, 26f three-component, 23–24 Baguera, 24, 24f CHARITÉ, 23, 23f Dynardi, 23–24, 24f Kineflex and Kineflex|C, 23, 24f Mobidisc, 24 Secure C, 24 two-component, 24–26 CerviCore, 24–25 Discocerv, 24–25, 25f DISCOVER, 25, 25f FlexiCore, 24–25 Maverick, 24–25 Prestige, 24–25 ProDisc-L and ProDisc-C, 24, 25f with ventrolateral or lateral approach, 27 Total Facet Arthroplasty System (TFAS), 565–576 biomechanics of, 565–566, 566f, 567f, 568f clinical presentation for, 569 complications of, 575 description of, 32, 566–569, 568f, 569f indications/contraindications for, 566 operative techniques for, 569–572 alignment verification and locking of cephalad arms in, 575, 575f, 576f caudal cup selection in, 571, 572f caudal implant assembly in, 572–575 caudal stem selection in, 570, 571f cement mixing and preparation in, 573, 574f cementing of caudal stem/cup assembly in, 573–574 cephalad arm selection in, 571–572, 573f cross-arm assembly in, 574, 574f cross-arm assembly insertion and cementing of cephalad arms in, 574–575, 575f exposure in, 570 facet removal in, 570
789
final preparation of pedicle channels in, 572 housing implant selection in, 572, 573f pedicle preparation in, 570 positioning in, 569–570 and posterior lumbar arthroplasty, 735 postoperative care for, 575 quality of motion with, 568, 570f rationale for, 565 scientific testing of, 566–569 for in vivo loading, 567 kinematic, 567, 570f for load sharing, 568 for range of motion, 568–569, 570f for strength, 567–568, 569f for wear, 568 Total lumbar interbody fusion (TLIF), 736, 738, 739 Total Posterior Arthroplasty System (TOPS), 549–564 clinical outcomes of, 557–560 mean Zurich Claudication Questionnaire score in, 559, 563f on Oswestry Disability Index, 559, 563f patient follow-up distribution in, 559, 563t patients and methods in, 557–559, 563t radiographic findings in, 559–560, 564f results in, 559–560 safety analysis in, 560 Visual Analog Scale in, 559, 563f description of, 32, 552–553, 553f, 554f design parameters of, 552–553, 553f, 554f discussion of, 561–563 operative technique for, 554–557 claw-armed holder in, 556–557, 561f degree of resection in, 555, 558f fluoroscopic confirmation of positioning in, 556–557, 562f implant trial in, 555, 558f incision in, 555 locking nut in, 556–557, 562f patient preparation in, 554–555 pendulum guide for angle of pedicle cannulation in, 555, 559f slotted extension sleeves in, 556, 560f targeting jig in, 556, 561f and posterior lumbar arthroplasty, 735 postoperative care of, 557 rationale for, 551–552, 552f scientific testing/clinical outcomes of, 553–554 biomechanical in vitro motion segment analysis in, 553, 556f finite element analysis in, 553, 555f load on pedicle screws in, 553–554, 557f Total Spinal Motion Segment (TSMS), 737, 738f Toughness, 53, 53f TPC (time, procedure, complication) system, for lumbar revision, 377 Trabeculae, 4–5, 4f compression, 4–5, 5f tension, 4–5, 5f Trabecular densification, with NUBAC disc arthroplasty device, 448 Trachea, in anterior cervical approach, 39, 39f TranS1 Percutaneous Nucleus Replacement. See Percutaneous Nucleus Replacement (PNR). Transforaminal block, 99, 99f Transforaminal endoscopic approach, to lumbar spine. See Lumbar endoscopic posterolateral [transforaminal] approach. Transformation matrix, 88 Transforming growth factor-b1 (TGF-b1), in gene therapy, 675, 675f, 676f Transition syndrome, persistent pain after lumbar total disc arthroplasty due to, 392–393
790
Index
Translation, of motion preservation device, 7 Transpedicular fixation, temporary external, 105–106 complications of, 106 indications for, 105 results of, 105–106 technique of, 105, 105f Transpsoas approaches, to lumbar disc spaces, 41–43, 43f Trans-sacral percutaneous nucleus replacement. See Percutaneous Nucleus Replacement (PNR). Tribochemical reactions, 60f, 61 Tribologic stresses, 59, 60f Tribologic system, 60, 60f TrueDisc posterior lumbar disc, 737, 738f, 739 Truedyne posterior dynamic stabilizer, 737, 738f Trunk loading, in biomechanical model, 690, 690t, 691f Trunk moments, in biomechanical model, 689, 690 TSMS (Total Spinal Motion Segment), 737, 738f Tubular retractor system, for minimally invasive lumbar surgery, 162–165, 164f, 165f Tumor necrosis factor-a (TNF-a) inhibitor, for back pain, 763 Twisting velocity, in biomechanical model, 690, 690t, 691f Two-component prostheses, for total disc replacement, 24–26 CerviCore, 24–25 Discocerv, 24–25, 25f Discover, 25, 25f FlexiCore, 24–25 Maverick, 24–25 Prestige, 24–25 ProDisc-L and ProDisc-C, 24, 25f
U Ultra-high-molecular-weight polyethylene (UHMWPE), material properties of, 56–57 Uncinate processes, 40–41 Uncinate resection, in anterior cervical approach, 189, 191f Unconstrained cervical disc prosthesis, 267–268 Porous Coated Motion (PCM) arthroplasty as, 202–213 Unconstrained lumbar disc prosthesis, Mobidisc, 326–329 Unconstrained lumbar disc prosthesis Kineflex, 338–345 Uncovertebral joint(s), 40–41, 40f Uncovertebral joint analysis, 700–701, 703f Upright gait, adaptation to, 22 Ureteral stents, in anterior exposure of lumbar spine, 150 Urologic complications, of anterior exposure of lumbar spine, 149t, 150, 151f
V VAS. See Visual Analog Scale (VAS). Vascular complications, of anterior exposure of lumbar spine, 148–150, 149f, 149t, 150f Vector-mediated delivery, of gene therapy, 674 Venous anomalies, in anterior exposure of lumbar spine, 143, 146f Venous lacerations, due to anterior exposure of lumbar spine, 150 Ventrolateral approach, to total disc replacement, 27 Vertebral arteries, 41, 41f in cervical revision, 282, 283f Vertebral body(ies) anatomy of, 4–5, 4f, 5f
cervical anatomy of, 40, 40f height and width of, 41 compression and rebound of, 4–5, 5f lumbar anatomy of, 43, 43f height and width of, 43–44 pathologic changes in, 6–7, 6f Vertebral body movement, with Barricaid device, 633 Vertebral body replacement, 17, 17f Vertebral end plate(s), 36–37 classification of, 76, 77f Viart and Marin device, 15, 16f Video-assisted discectomy, 162–165, 164f, 165f Villette, Louis, 318 Viral vectors, for gene therapy, 674 "Virtual surgery," with biomechanical model, 691–693, 693f Viscoelasticity, of polymers, 55–56 Visual Analog Scale (VAS) for Activ-L Artificial Disc, 350 for autologous disc chondrocyte transplant, 681–682 for Bryan Cervical Disc, 194–195, 195f for cervical arthroplasty adjacent to fusion, 753, 753f, 753t, 754f for CHARITÉ Artificial Disc, 311 for Cosmic system, 494 for DASCOR device, 402, 403f for DIAM system, 521 for Dynamic Stabilization System, 473 for FlexiCore Intervertebral Disc, 335, 335f for lumbar disc arthroplasty, 127–130, 128f, 129f adjacent to scoliotic deformity, 706, 706f multilevel, 727 for Mobi-C, 233–234, 233f, 235f for Mobidisc prosthesis, 327, 328t for NeoDisc device, 227 for NUBAC disc arthroplasty device, 448, 449f for NuCore Injectable Nucleus, 420, 420f for PercuDyn device, 512, 513f, 513t for Porous Coated Motion arthroplasty, 204 for ProDisc-C artificial cervical disc, 216, 217f for ProDisc-L artificial lumbar disc, 320 for Secure-C device, 249–251, 251f for simultaneous lumbar fusion and total disc replacement, 618–619, 618f for TOPS device, 559, 563f for TranS1 Percutaneous Nucleus Replacement, 435, 437 for Wallis Stabilization System, 525–526, 525f, 526f, 531 Vitallium spheres, 452
indication(s) for, 526–531 disc support and discogenic back pain as, 526–527, 528f, 529f distraction decompression as, 530–531 large or recurrent disc prolapse or prolapse at transitional levels as, 528–529, 529f, 530f support and protection of levels adjacent to fusion as, 529–530, 530f potential disc restoration with, 532 rationale for, 523 safety of, 531 scientific testing of, 523–525, 524f, 525f Watkin's technique, for minimally invasive lumbar surgery, 161–162, 162f, 163f Wear defined, 59 type of, 60 Wear analysis, 59–62 abrasion in, 60f, 61 adhesion in, 60, 60f surface fatigue in, 60f, 61 system approach to, 59–60, 60f tribochemical reactions in, 60f, 61 wear appearances in, 60 wear countermeasures in, 61 wear mechanisms in, 60, 60f wear modes in, 60 wear testing for, 61–62, 61f Wear appearances, 60 Wear characteristics, of coflex implant, 537 Wear countermeasures, 61 Wear damage, 60 Wear debris, 59 with NUBAC disc arthroplasty device, 447 particulate, neurotoxicity rabbit model of, 72–73, 73f, 74f Wear mechanisms, 60, 60f Wear modes, 60 Wear pattern, 60 Wear rate, of NUBAC disc arthroplasty device, 447 Wear resistance, 53–54 Wear testing, 61–62, 61f of Kineflex Disc, 340 of NUBAC disc arthroplasty device, 446–447 of Total Facet Arthroplasty System, 568 Wilcoxon Rank Sum Test, for lumbar arthroplasty, 127 Wiltse technique, for minimally invasive lumbar surgery, 160–161, 161f Wing bending, of coflex implant, 536 Wistar rat model, for disc degeneration, 641t, 643 Wolff's law, 4–5 Wound infection, after anterior exposure of lumbar spine, 153
W
X
Wallis Stabilization System, 523–533 adjacent level preservation with, 529–530, 530f, 532 clinical benefit of, 531–532 clinical outcomes with, 525–526 analgesic use for, 525–526, 527f Japanese Orthopaedic Association score for, 526f Oswestry Disability Index for, 525–526, 526f, 527f, 531–532 physical component scores for, 531 SF-36 for, 525–526, 527f, 528f Visual Analog Scale for, 525–526, 525f, 526f, 531 description of, 33, 33f, 523, 524f development of, 523
X-STOP, 541–548 advantages/disadvantages of, 547 background of, 541 clinical outcomes of, 543–544, 544t vs. decompressive laminectomy, 545, 545t with degenerative spondylolisthesis, 544–545 European experience of, 544 German registry of, 544, 544t sagittal balance in, 545 complications of, 547 cost effectiveness of, 134, 547 description of, 33, 542–543, 542f, 543f discussion of, 547 indications/contraindications for, 542 operative technique for, 545–547, 546f postoperative care for, 547
Index X-STOP (Continued) rationale for, 541–542 scientific testing of, 543 XLIF approach. See Extreme lateral interbody fusion (XLIF) approach.
Y Yeung Endoscopic Spine Surgery (YESS) technique, 167–178 anesthesia for, 169 clinical presentation and evaluation for, 168 complications of, 174–176 description of device for, 168, 169f discussion of, 176 future considerations with, 176–177
indications/contraindications for, 168 instrument placement in, 173–174, 173f needle placement in, 171–172 for NeuDisc device, 415 performing discectomy in, 174 position for, 169, 170f postoperative care with, 174 procedure for, 169–171, 171f, 172f rationale for, 167–168, 168f scientific testing/clinical outcomes of, 168 Yield point, 53f Yield strength, 53f Young's modulus, for IsoBar TTL Dynamic Rod Stabilization, 485–486, 486t Yuan articulating plate design, 13, 14f
791
Z Zirconia ceramics, material properties of, 57 Zirconium (Zr) alloys, material properties of, 59 Z-plasty, 597 Zurich Claudication Questionnaire (ZCQ) score for TOPS device, 559, 563f for X-STOP, 544, 544t, 545, 545t Zyre Facet Replacement Device, 581–584 complications of, 584 description of, 582, 582f indications/contraindications for, 581–582 operative techniques for, 583–584 rationale for, 581 scientific testing/clinical outcomes of, 582, 583f, 584f