AOSpine Manual: Principles And Techniques, Clinical Applications [1] 3131444819, 9783131444813

2,488 797 468MB

English Pages 0 [679] Year 2017

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

AOSpine Manual: Principles And Techniques, Clinical Applications [1]
 3131444819, 9783131444813

  • Commentary
  • 2 Volume Set
Citation preview

Volume 1 presents basic scientific and technical principles—it provides the reader with the scientific background to understand spine surgery and it teaches how to apply these surgical principles using the instrumentation necessary in a step-by-step manner with exceptional illustrations; some critical steps are explained using sequences from AOSpine teaching videos. PRINCIPLES AND TECHNIQUES (VOL 1) relates to the teaching of basic surgical knowledge and surgical techniques at AOSpine courses and acts as a foundation for the application of these principles in clinical practice.

www.aospine.org

Cover_PSpi_I_T03.indd 1

The Americas

Rest of World

(2-volume set)

(2-volume set)

ISBN 978-1-58890-557-4 (TPN)

ISBN 978-3-13-144481-3 (TPS)

Aebi | Arlet | Webb

Continuous systematic learning is essential for all spine surgeons endeavoring to improve their daily practice and enhance patient outcome. The mission of AOSpine is to share knowledge and expertise through accessible quality education.

AOSPINE MANUAL PRINCIPLES AND TECHNIQUES (VOL 1)

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES (VOL 1)

Max Aebi | Vincent Arlet | John K Webb

AOSPINE MANUAL PRINCIPLES AND TECHNIQUES (VOL 1) AOSPINE MANUAL CLINICAL APPLICATIONS (VOL 2)

Practical exercise video clips on DVD-ROM included

16.1.2007 17:06:53 Uhr

Volume 1 presents basic scientific and technical principles—it provides the reader with the scientific background to understand spine surgery and it teaches how to apply these surgical principles using the instrumentation necessary in a step-by-step manner with exceptional illustrations; some critical steps are explained using sequences from AOSpine teaching videos. PRINCIPLES AND TECHNIQUES (VOL 1) relates to the teaching of basic surgical knowledge and surgical techniques at AOSpine courses and acts as a foundation for the application of these principles in clinical practice.

www.aospine.org

Cover_PSpi_I_T03.indd 1

The Americas

Rest of World

(2-volume set)

(2-volume set)

ISBN 978-1-58890-557-4 (TPN)

ISBN 978-3-13-144481-3 (TPS)

Aebi | Arlet | Webb

Continuous systematic learning is essential for all spine surgeons endeavoring to improve their daily practice and enhance patient outcome. The mission of AOSpine is to share knowledge and expertise through accessible quality education.

AOSPINE MANUAL PRINCIPLES AND TECHNIQUES (VOL 1)

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES (VOL 1)

Max Aebi | Vincent Arlet | John K Webb

AOSPINE MANUAL PRINCIPLES AND TECHNIQUES (VOL 1)

Practical exercise video clips on DVD-ROM included

16.1.2007 17:06:53 Uhr

Max Aebi | Vincent Arlet | John K Webb

AOSPINE MANUAL PRINCIPLES AND TECHNIQUES (VOL 1)

Max Aebi | Vincent Arlet | John K Webb

AOSPINE MANUAL PRINCIPLES AND TECHNIQUES (VOL 1)

iv

Design and layout : nougat gmbh, CH-4056 Basel Illustrations: nougat gmbh, CH-4056 Basel Index: Jill Halliday BSc, Fellow of the Society of Indexers, GB-Diss, Norfolk IP21 4QT Video editing: fern media solutions gmbh, CH-8047 Zürich Production: AO Publishing, CH-8600 Dübendorf Library of Congress Cataloging-in-Publication Data is available from the publisher.

HAZARDS Great care has been taken to maintain the accuracy of the information contained in this publication. However, the publisher, and/or the distributor, and/or the editors, and/or the authors cannot be held responsible for errors or any consequences arising from the use of the information contained in this publication. Contributions published under the name of individual authors are statements and opinions solely of said authors and not of the publisher, and/or the distributor, and/or the AO Group. The products, procedures, and therapies described in this work are hazardous and are therefore only to be applied by certified and trained medical professionals in environments specially designed for such procedures. No suggested test or procedure should be carried out unless, in the user’s professional judgment, its risk is justified. Whoever applies products, procedures, and therapies shown or described in this work will do this at their own risk. Because of rapid advances in the medical sciences, AO recommends that independent verification of diagnosis, therapies, drugs, dosages, and operation methods should be made before any action is taken. Although all advertising material which may be inserted into the work is expected to conform to ethical (medical) standards, inclusion in this publication does not constitute a guarantee or endorsement by the publisher regarding quality or value of such product or of the claims made of it by its manufacturer. LEGAL RESTRICTIONS This work was produced by AO Publishing, Davos, Switzerland. All rights reserved by AOSpine International. This publication, including all parts thereof, is legally protected by copyright. Any use, exploitation or commercialization outside the narrow limits set forth by copyright legislation and the restrictions on use laid out below, without the publisher’s consent, is illegal and liable to prosecution. This applies in particular to photostat reproduction, copying, scanning or duplication of any kind, translation, preparation of microfilms, electronic data processing, and storage such as making this publication available on Intranet or Internet. Some of the products, names, instruments, treatments, logos, designs, etc, referred to in this publication are also protected by patents and trademarks or by other intellectual property protection laws (eg, “AO”, “ASIF”, “AO/ASIF”, “AOSpine“, “TRIANGLE/GLOBE Logo” are registered trademarks) even though specific reference to this fact is not always made in the text. Therefore, the appearance of a name, instrument, etc, without designation as proprietary is not to be construed as a representation by the publisher that it is in the public domain. Restrictions on use: The rightful owner of an authorized copy of this work may use it for educational and research purposes only. Single images or illustrations may be copied for research or educational purposes only. The images or illustrations may not be altered in any way and need to carry the following statement of origin “Copyright by AOSpine International, Switzerland”.

Copyright © 2007 by AOSpine International, Switzerland, Stettbachstrasse 10, CH-8600 Dübendorf Distribution by Georg Thieme Verlag, Rüdigerstrasse 14, DE-70469 Stuttgart and Thieme New York, 333 Seventh Avenue, New York, NY 10001, USA

Rest of World ISBN 978-3-13-144481-3 (TPS)

The Americas ISBN 978-1-58890-557-4 (TPN)

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

v

CONTRIBUTORS

EDITORS

AUTHORS

Max Aebi, MD, DHC, FRCSC Institute for Evaluative Research in Orthopaedic Surgery University of Bern Stauffacherstrasse 78 3014 Bern Switzerland

Mauro Alini, PhD AO Research Institute Clavadelerstrasse 8 7270 Davos Platz Switzerland

Christopher Cain, MD, FRACS, MBBS, FAOrthA Adelaide Spine Clinic 252 East Terrace Adelaide SA 5000 Australia

Juan Francisco Asenjo, MD Department of Anesthesia and McGill Pain Centre McGill University Health Centre 1650 Cedars Ave D10–152 Montreal QC H3G 1A4 Canada

Maximilian A Dambacher, MD University Hospital Balgrist Forchstrasse 340 8008 Zürich Switzerland

Vincent Arlet, MD Department Orthopaedic Surgery University of Virginia PO Box 800159 Charlottesville VA 22908-0159 USA John K Webb, FRCS, MBBS Spine Unit University Hospital Queen‘s Medical Centre Nottingham, NG7 2UH United Kingdom

Norbert Boos, MD, MBA Centre for Spinal Surgery University of Zürich University Hospital Balgrist Forchstrasse 340 8008 Zürich Switzerland

Jason C Datta, MD Sonoran Spine Center 2610 North 3rd Street Suite B Phoenix AZ 85004 USA

vi

Contributors

Stephen J Ferguson, PhD Institute for Surgical Technologies and Biomechanics University of Bern Stauffacherstrasse 78 3014 Bern Switzerland Brian JC Freeman, MD, FRCS (Tr&Orth) The Centre for Spinal Studies and Surgery Nottingham University Hospitals NHS Trust Queen’s Medical Centre Campus Nottingham, NG7 2UH United Kingdom Paul Heini, MD Department of Orthopaedic Surgery University of Bern Inselspital Murtenstrasse 3010 Bern Switzerland Travis Hunt, MD Central Kentucky Orthopaedics, PLC 1140 Lexington Road Georgetown KY 40324 USA

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Keita Ito, MD, ScD AO Research Institute Clavadelerstrasse 8 7270 Davos Platz Switzerland Michael E Janssen, DO Spine Education and Research Institute 9005 Grant Street, Suite #100 Denver CO 80229 USA Frank Kandziora, MD, PhD Spine Center Center for Musculoskeletal Surgery Charité University Hospital Augustenburgerplatz 1 13353 Berlin Germany

Dante G Marchesi, MD Orthopaedic Surgery and Traumatology Hirslanden Clinique Bois-Cerf Avenue d‘Ouchy 31 1006 Lausanne Switzerland H Michael Mayer, MD, PhD Spine Center Munich Orthozentrum München Harlachinger Strasse 51 81547 München Germany Jochen Meissner, MD Orthopaedic Hospital Speising Speisinger Strasse 109 1134 Wien Austria

Martin Krismer, MD Universitätsklinik für Orthopädie Medizinische Universität Innsbruck Anichstrasse 35 6020 Innsbruck Austria

Serge Nazarian, MD Service d‘Orthopédie et Chirurgie Vertébrale Hôpital de la Conception 147 Boulevard Baile 13005 Marseille France

Frank Langlotz, Dr Phil Dipl-Ing Institute for Surgical Technologies and Biomechanics University of Bern Stauffacherstrasse 78 3014 Bern Switzerland

Andreas G Nerlich, MD, PhD, MSc Klinikum München-Bogenhausen Englschalkingerstrasse 77 81925 München Germany

vii

Contributors

VIDEO EDITOR

Lutz-Peter Nolte, PhD Institute for Surgical Technologies and Biomechanics University of Bern Stauffacherstrasse 78 3014 Bern Switzerland Michael Ogon, MD Orthopaedic Hospital Speising Speisinger Strasse 109 1134 Wien Austria Jean Ouellet, MD, FRCSC McGill Scoliosis and Spine Centre McGill University Montreal Childrens Hospital 2300 Tupper Street, C-521 Montreal QC H3H 1P3 Canada Thomas Schlich, MD Department of Social Studies of Medicine McGill University 3647 Peel Street Montreal QC H3A 1X1 Canada

Christopher I Shaffrey, MD Department of Neurological Surgery University of Virginia PO Box 800212 Charlottesville VA 22908 USA Cyril Solari, MD Service d‘Orthopédie et Chirurgie Vertébrale Hôpital de la Conception 147 Boulevard Baille 13008 Marseille France Thomas Steffen, MD, PhD, MBA Orthopaedic Research Laboratory McGill University 687 Pine Avenue West, Room L4.69B Montreal QC H3A 1A1 Canada

John K O‘Dowd, FRCS Orth RealHealth Institute London, W6 9AR United Kingdom

viii

ix

FOREWORD/ACKNOWLEDGMENTS

This two-volume spine surgery textbook is the unique result of a tremendous effort made possible through the collaboration of more than 50 authors under the guidance of AO Publishing. Of course we would like to congratulate all the authors who contributed to this novel book. We know that for them each chapter represents hours and hours away from their busy clinical practices and/or families. We thank them for their commitment and steadfastness to deliver such excellent teaching material.

We are very grateful to Zoe Koh and her team for the outstanding artwork. A special thanks goes to Sandro Isler whose patience and meticulousness has delivered a didactic design and layout. This book is unique—the excellent quality illustrations allow the reader to better visualize, understand, and prepare for a successful surgery. Last but not least our thanks go to the people who have been involved in the proofreading of the text in English, the video editing, the indexing, and the fi nal production/printing of this book.

Furthermore, we would like to express our gratitude to those in the AO Publishing office in Dübendorf for their outstanding support, work, time, and commitment to this book. Doris Straub Piccirillo has been dedicated to this book with all her heart and soul from the very onset of the project to its fi nal completion. Without her determination and patience with the different authors (who regularly turned their chapters in late), this book would have never been possible. Max Aebi

Vincent Arlet

John K Webb

x

TABLE OF CONTENTS 1

PREFACE—AO EDUCATION AND TEACHING CONCEPT

2

INTRODUCTION—AO PRINCIPLES APPLIED TO THE SPINE

3

HISTORY OF SPINE SURGERY WITHIN AO

4

BIOMECHANICS OF THE SPINE THOMAS STEFFEN …………………………………………………………… Introduction ……………………………………………………………………………………………………… General biomechanics of the spinal motion segment and the spinal organ THOMAS STEFFEN ………… Biomechanics of spinal stabilization STEPHEN J FERGUSON, THOMAS STEFFEN ……………………………………

4.1 4.2

5 5.1 5.2 5.3 5.4

6 6.1 6.2 6.3 6.4 6.5 6.6 6.7

……………

1

MA X AEBI ……………………………………

7

MA X AEBI, VINCENT ARLET, JOHN K WEBB

THOMAS SCHLICH, MA X AEBI

…………………………………… 15

29 31 33 53

BIOLOGY OF THE SPINE NORBERT BOOS ……………………………………………………………………… 71 Introduction ……………………………………………………………………………………………………… 73 Biology of the motion segment NORBERT BOOS ……………………………………………………………… 77 Aging and pathological degeneration MAURO ALINI, KEITA ITO, ANDREAS G NERLICH, NORBERT BOOS ……………… 87 Biology of the osteoporotic spine MA XIMILIAN A DAMBACHER, NORBERT BOOS ………………………………… 103 Biology of fusion with bone and bone substitutes DANTE G MARCHESI …………………………………… 117

SURGICAL ANATOMY OF THE SPINE SERGE NAZARIAN …………………………………………………… Introduction …………………………………………………………………………………………………… Upper cervical spine SERGE NAZARIAN, CYRIL SOLARI …………………………………………………………… Lower cervical spine SERGE NAZARIAN, CYRIL SOLARI …………………………………………………………… Cervicothoracic junction SERGE NAZARIAN, CYRIL SOLARI ……………………………………………………… Thoracic spine SERGE NAZARIAN, CYRIL SOLARI …………………………………………………………………… Thoracolumbar junction SERGE NAZARIAN, CYRIL SOLARI ……………………………………………………… Lumbar spine and lumbosacral junction SERGE NAZARIAN, CYRIL SOLARI …………………………………… Sacrum SERGE NAZARIAN, CYRIL SOLARI ……………………………………………………………………………

129 131 135 147 161 173 193 205 235

xi

Table of contents

7 7.1 7.2 7.2.1 7.2.2 7.2.3 7.2.4 7.2.5 7.2.6 7.3 7.3.1 7.3.2

SPINAL INSTRUMENTATION VINCENT ARLET ……………………………………………………………… 241 Introduction …………………………………………………………………………………………………… 243 Modularity of spinal instruments (systems) MA X AEBI …………………………………………………… 247 Cervical spine Modularity and evolution of instrumentation for the cervical spine MICHAEL E JANSSEN, JASON C DATTA, VINCENT ARLET ……………………………………………………………………… Upper cervical spine VINCENT ARLET, JASON C DATTA …………………………………………………………… Middle and lower cervical spine JASON C DATTA, MICHAEL E JANSSEN ………………………………………… Craniocervical junction MICHAEL E JANSSEN …………………………………………………………………… Cervicothoracic junction BRIAN JC FREEMAN, FRANK KANDZIORA ………………………………………………… Cervical laminoplasty CHRISTOPHER I SHAFFREY …………………………………………………………………

253 265 289 305 315 329

7.3.3 7.3.4 7.3.5

Thoracolumbar and sacropelvic spine Modularity of the universal spine system CHRISTOPHER CAIN ……………………………………………… Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative) DANTE G MARCHESI …………………………… Universal spinal instrumentation for deformity CHRISTOPHER CAIN ……………………………………… Instrumentation for the degenerative thoracolumbar spine DANTE G MARCHESI ……………………… Fixation of the sacrum and pelvis TRAVIS HUNT, VINCENT ARLET ………………………………………………

357 391 423 457

7.4 7.4.1 7.4.2 7.4.3 7.4.4 7.4.5

Special techniques and instrumentation Concept of MISS/LISS MA X AEBI ……………………………………………………………………………… Bone harvesting tool MA X AEBI ……………………………………………………………………………… Vertebroplasty PAUL HEINI ……………………………………………………………………………………… Fixation technology in osteoporosis PAUL HEINI …………………………………………………………… Spondylolysis, spondylolisthesis—reduction and stabilization VINCENT ARLET, MARTIN KRISMER …………

467 485 491 507 519

7.5 7.5.1 7.5.2

Motion-preserving technology Arthroplasty in cervical spine surgery Arthroplasty in lumbar spine surgery

…………………………………… 543 …………………………………………………… 555

JOCHEN MEISSNER, MICHAEL OGON H MICHAEL MAYER

337

xii

Table of contents

8

COMPUTER-ASSISTED SURGERY

9

ANESTHESIA FOR SPINE SURGERY

FRANK LANGLOTZ, LUTZ-PETER NOLTE

JUAN FRANCISCO ASENJO

……………………………………… 571

……………………………………………… 589

GLOSSARY ……………………………………………………………………………………………………………… INDEX

617

…………………………………………………………………………………………………………………… 642

xiii

To access additional material available for this e-book, please go to MediaCenter.thieme.com. You will be asked to login to this site using your thieme.com account credentials. If you don’t have a thieme.com account, please register first. Click on “All titles” at the top of the page to find the additional material for your e-book and enter the following access code to gain access: 86JP-T2F7-C7GW-USJ5

PREFACE—AO EDUCATION AND TEACHING CONCEPT

1

3

Max Aebi, Vincent Arlet, John K Webb

1

PREFACE—AO EDUCATION AND TEACHING CONCEPT

Surgical education is the main purpose of the AO Organization and for this reason great efforts have been made to translate the knowledge gained as well as the skills to a new generation of surgeons, who want to expand their expertise in spine surgery. The experience and knowledge acquired by teaching at AOSpine courses soon resulted in growing popularity within the AO community and beyond. Some of the orthopedic surgeons have successfully used the AO Principles for fracture management (chapters 2 Introduction—AO principles applied to the spine, 3 History of spine surgery within AO) to treat primarily spinal fractures and dislocations. Those concepts helped to significantly reduce the postoperative immobilization time with all its consequences and undoubtedly contributed to the development of an efficient treatment of spinal injuries, which is similar to what had previously occurred in AO for the treatment of injuries of the lower and upper extremities. The experiences gained with the treatment of spinal fractures were then applied to surgical concepts of nontraumatic spinal disorders by these pioneering surgeons.

The first AO spine manual (AO ASIF Principles in Spine Surgery), which was published in 1998 [1], was the first attempt to structure spine surgery into a body of knowledge that standardized surgical techniques and indications with step-by-step guidelines/procedures. However, in the early nineties Aebi and Webb began, together with several faculty members, to systematically restructure the spine courses and to abandon the classic “frontal” lecture and teaching style: From that time on every teaching session had to be based on one or more concrete case(s). The method of applying the surgical concepts and techniques was then interactively elaborated between the faculty and participants on the basis of those concrete cases. The lecturer became a tutor and tried to involve everybody in the teaching and learning process. The role of the lecturer was to bring the discussion of clinical cases into the context of established knowledge by

4

alluding to current peer-reviewed literature and guidelines. So, current knowledge became the scaffold onto which the actual concrete case was built by “thinking” and formulating the case through. The concept of the interactive course was born and it became more sophisticated with the help of a lot of enthusiastic teachers and learners, and today it is the rule in all the AOSpine courses. However, an interactive case discussion and the development of an exemplary teaching case are only possible when the participants have a basic level of knowledge, which has to be acquired by systematic learning. This basic knowledge of the scientific backgrounds of what spine surgeons do in their clinical practice and the technical applications are the tools and ingredients for good practice. With these realities in mind, in 2001 the editors started to design a completely new AOSpine Manual, in which the experience gained over years in teaching and learning would transpire in a textbook. Considering what had been done up until then, this undertaking meant a complete paradigm shift. Surgical book chapters have been designed over the years according to a very classic and repetitive format made of an intro duction, epidemiology, pathophysiology, clinical diagnosis, radiographic evaluation, surgical indications, surgical techniques, and one or two case examples to illustrate the chapters. Modern electronic working tools with the “copy and paste” mentality have translated into a uniform reproduction of these book chapters, where each new author

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

would copy the previously published chapter, make a few modifications and publish a copycat of the previous chapter. Therefore, each chapter would resemble the chapters of the most eminent leader in their field, and “politically correct” chapters avoiding controversies, discussion, and debate became the norm in most spinal textbooks because publishing a new spinal textbook with an attractive title was more important than the very content or teaching value of it. This is the type of textbook the editors obviously did not wish to write. Therefore, this now completely restructured and newly written AOSpine Manual is divided into two major volumes: • Volume 1—Principles and Techniques—features the basic scientific and technical principles, including all the handling of the implants and instruments emerging from these principles. • Volume 2—Clinical Applications—features the case discussions with a fi nal chapter on how to measure outcome in clinical practice. Volume 1 is a comprehensive manual intended to aid in the understanding of the scientific backgrounds of spine surgery and the systematic teaching of surgical techniques and applications as they are communicated today in the AOSpine courses. This volume is the most comprehensive book on surgical procedures, which deals with almost all the established surgical techniques and principles applied today in spine surgery. The technical handling is abundantly illustrated in order to facilitate a step-by-step understanding of the surgical procedures. This volume represents a significant expansion and improvement of its predecessor, the AO ASIF Principles in Spine Surgery.

1

5

Preface—AO education and teaching concept

Volume 2 is a newly designed book based on the interactive courses. This means that every surgical application is based on a concrete case. In other words, volume 2 is a compilation of typical clinical cases out of the day-to-day spinal practice. The whole spinal pathology is presented in the different chapters, and in each chapter representative cases have been chosen to develop the rationale for treatment, the indications, the contraindications, the argumentation for a technique or against one, and the outcome. More than a textbook of recipes for spine surgeries, this allows for an understanding of the concepts, approaches, and controversies that can exist in treating patients with spinal ailments. In order to achieve such a task, each author was asked to write according to the aforementioned format. This extraordinary undertaking was made possible thanks to the contribution of more than 50 leaders in their field who gathered from their clinical practice, case examples that would bring up the best teaching and learning points. It is obvious that this concept of writing necessitated a lot of editing to achieve a uniform structure that would communicate all this complex information. It is not without saying that such a revolution in textbook writing was naturally associated with a painful adaptation and learning curve for each of the authors, who more often than not were asked to rewrite their initially submitted chapters according to the desired format. Nevertheless, this book has become a huge resource including many typical cases we encounter in our daily work, as well as the algorithms used to help treat these cases. This movement away from the classic textbook configuration aims to involve the reader intellectually into the treatment of each a specific

case. This is truly a new concept that does not just compile and present established knowledge, it invites the reader to get involved, not as a passive spectator, but as a potentially active treating surgeon. It is only logical that at the end of such a book methodologies on how to measure clinical outcomes [2] and on how to follow up clinical cases are demonstrated. In the chapter about the spine registry the AOSpine Manual maintains the tradition of AO that was built on four pillars, one of which includes the documentation of the outcome of cases. The need for modern documentation systems which allow comparison of cases will become increasingly needed to justify certain clinical procedures. The editors and authors hope that with this book new ground will be broken and a new era of educational communication will open up. The editors would like to thank all the authors and contributors who accepted the challenge of writing a clinical application textbook. Their contributions will no doubt greatly enhance the education and teaching of a new generation of spine surgeons.

BIBLIOGRAPHY Aebi M, Thalgott JS, Webb JK (1998) AO ASIF Principles in Spine Surgery. Berlin, Heidelberg, New York: Springer-Verlag. 2. Chapman JR, Hanson BP, Dettori JR, et al (2007) Spine Outcomes Measures and Instruments . Stuttgart New York: Thieme Verlag. 1.

INTRODUCTION—AO PRINCIPLES APPLIED TO THE SPINE

2

8

2

INTRODUCTION—AO PRINCIPLES APPLIED TO THE SPINE

1

AO organization ………………………………………………………………………………………………

2

Development ………………………………………………………………………………………………… 11

3

Education ……………………………………………………………………………………………………… 11

4

Research ……………………………………………………………………………………………………… 12

5

Bibliography …………………………………………………………………………………………………… 12

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

9

9

Max Aebi

2

1

INTRODUCTION—AO PRINCIPLES APPLIED TO THE SPINE

AO ORGANIZATION

The AO organization has attained its reputation as a world leader in musculoskeletal trauma through its innovative and systematic approach to operative fracture treatment which started in the mid 1950s. Over the years, the AO organization has become a foundation with a mission based on four fundamental pillars: development, research, education, and documentation. These four elements are all geared toward a fracture treatment that is based on stable internal fi xation to allow faster bone healing and early motion of the injured limb. Thus, early mobilization, active rehabilitation, and early reintegration of the patient into their original social and working environment are made possible. AO has allocated enormous resources to making fracture treatment better, more standardized, reproducible, and rational; these improvements are based on scientific research throughout the whole continuum from the bench (basic science), to the bed side (clinical application), and beyond (outcome).

These principles were transferred to treating primarily spinal fractures within this environment and context, while the enormous potential of stable internal fi xation for the whole spectrum of spinal disorders had not yet been recognized. Compared to long-bone fractures, spinal fractures are almost always “articular” injuries affl icting the motion segment (ie, the functional unit, FU, of the spine), involving the disc, facet joints, ligamentous structures, and at least one of the adjacent vertebrae. Even today, these injuries cannot be “reconstructed” in an anatomical fashion like a long-bone fracture, but they need to be treated with a fusion, a measure that nature would fi nally have done over a prolonged period of time in many cases by itself. In most of the cases, the spinal fracture treatment ends with an arthrodesis of two or more vertebral bodies, in which all the principles of modern fracture treatment propagated by AO can be applied: optimal mechanical conditions for fitting two or more parts together, anatomical alignment, stable internal fixation, compression between the parts

10

fitted together, and early mobilization and muscular exercise to avoid the fracture “disease”. In the very few cases of spinal trauma where pure fracture principles can be applied, fusion is obviously avoided (odontoid fracture, traumatic spondylolysis of C2 and lumbar vertebrae). The early spine surgeons within AO developed surgical concepts and implants, which allowed a comprehensive treatment of spinal fractures: anterior and posterior plating of the cervical spine, pedicular fi xation, and anterior plating as well as rod systems for thoracolumbar fractures. All of these are concepts, which hold true today, although many implants have been adapted under market pressures into more sophisticated or “modern” implants. In the late 1960s and 1970s the AO surgeons used the Harrington instrumentation, which was originally developed for deformity surgery. They quickly understood the significance of the pedicle as an anatomical entity for anchoring instrumentation to the spine as introduced by Judet and Roy-Camille. Magerl, and later Dick, however, made it possible to shorten the extent of the fi xation on the spine by linking the pedicle screws in an angle-stable fashion to a vertical bar (rod); fi rst with the external fi xator, and later on with the internal fi xator. For the fi rst time this technology allowed the surgeon to limit the fi xation to one or two segments, thereby preserving the mobility of healthy motion segments. The intrinsic stability of these new systems allowed for early mobilization of the patient without a cast or corset, which are detrimental to the stabilizing muscles of the spine and highly uncomfortable for the patient. While AO temporarily borrowed the Harrington instrumentation to deal with fractures, the AO surgeons started to use the original trauma implants and concepts in nontraumatic conditions, foremost for degenerative diseases, pathological fractures (tumors, infections), and finally in

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

deformities. The leading principles were: reconstruction of the anatomy, stable fi xation respecting fundamental mechanical principles of the spine, short fi xation to maintain a maximum number of mobile motion segments, and early mobilization. In 1981 a patient with an anterior lumbar interbody fusion usually had to stay in bed and in hospital for 4–6 weeks. Today, the patient can leave the hospital 3–4 days after surgery wearing a soft brace for comfort if needed. In 2003, a spine section was formally created within the AO Foundation with a great amount of autonomy in day-to-day business and ultimately its own legal structure (2006). This organization has fi nally implemented all four pillars of the AO Foundation: • Development of implants based on sound scientific principles. • Education. • Research network with clear research priorities and strategies. • Documentation systems with specific registries.

2

Introduction—AO principles applied to the spine

2

DEVELOPMENT

Having been influenced and stimulated by the early success in spinal trauma surgery, in 1987 Aebi and Webb had a vision to design and develop a philosophy of spine surgery. This philosphy included a spinal system, which not only addresses trauma, but the whole spinal pathology from the occiput to the sacrum, and from the front as well as from the back. These spine surgeons utilized what they had learned from the AO founders: to develop precise, simple instrumentation with as few instruments and implants as possible, but of the highest quality, which allowed for a reproducible surgical treatment of the spine. This led to an enormous expansion of the possibilities and indications in spine surgery. Such a large development, at least initially, represented a challenge for the teaching and education of a small spine group at the time. More and more surgeons adhered to the AOSpine cause, forming a group of surgeons who paralleled their trauma colleagues in developing their own less invasive spine surgery concept. This included cages, bone substitutes, and sophisticated retractor systems in combination with optical systems (technology integration in the true sense of word). At this time, the group did not jump on the bandwagon of endoscopic spine surgery. 10 years later, when endoscopic spine surgery was recognized as a fad, nonAO surgeons and producers switched to the AO concepts by imitating, for instance, the retractor system. This is just one example of how AO’s fi rst spine surgeons, in the tradition of the AO Foundation, acted as the leaders and initiators of new trends in spine surgery.

11

3

EDUCATION

Although the AO’s spine surgeons adapted very quickly to the concept of trauma courses, very soon they developed new educational pathways by introducing interactive case-based courses and live surgery courses. This leadership, and the success of such educational formats, were soon to be adopted in other subspeciality courses. It also became necessary to compile a manual of AO principles in spine surgery in order to bring all the knowledge of broad variety and high versatility in spine surgery together. This handbook quickly appeared in four languages (English, Spanish, Mandarin, and Japanese) [1]. Due to the very rapid accumulation of new technology and new knowledge the “AOSpine Manual” was to be followed quickly by this more comprehensive book, which you now hold in your hands. One of the major tasks in education today is to offer teaching and learning modules that benefit not only orthopedic surgeons, but neurosurgeons as well. Spine surgery has become a subspecialty, which can be reached through different curricula: orthopedic surgery and neurological surgery.

12

4

RESEARCH

Whereas AO institutions and AO-linked clinical departments have contributed significantly to the assessment of new implants—biomechanically and clinically—the AOSpine group has decided to use its resources in a very focused manner and to explore as a fi rst priority the aging and degenerated disc with the possibility of replacing it with biological and/or mechanical devices, thus, maintaining the function of a motion segment. Today’s market is overflowing with new mechanical prostheses with very poorly established background data. To assure a sustained development and growth it is paramount that sound scientific knowledge is established before adding other implants. The AOSpine group has decided to put most of the available resources for research in centers that have already created specific knowledge and expertise, and which can be linked as a spinal reference network for research. The second priority is the aging spine, specifically the bone (osteoporosis) and ways in which it can be repaired. The challenges of the future are manifold: • To maintain the integrity of a medical and scientific organization that is driven by real innovative thinking and conceptual approaches. • To develop a strong presence in the spine world, by maintaining the values and valid principles of the AO Foundation. • To open the AO’s community to an extended spine world by building and maintaining an integrated basic and clinical science and research network. This should cover the real needs of an aging society in spinal care for which new products have to be developed for the benefit of the patients.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

• To incorporate nonfusion technology into the armamentarium from within the minefield of today‘s possible technologies and the evidence-based needs of the patients. • To support these concepts by sound clinical outcome research and participation in national and international registers, where a benchmarking of activity and an assessment of quality standards is possible. • To further develop the high level of educational activities, which are based on a 50-year effort of excellence by the AO community. There are tremendous tasks waiting ahead for the AOSpine group in order to match the success story of the original AO organization.

5

BIBLIOGRAPHY

1. Aebi M, Thalgott JS, Webb JK (1998) AO ASIF Principles in Spine Surgery. Berlin, Heidelberg, New York: Springer-Verlag

13

Max Aebi

HISTORY OF SPINE SURGERY WITHIN AO

3 2

16

3

HISTORY OF SPINE SURGERY WITHIN AO

1

Introduction …………………………………………………………………………………………………… 17

2

Spine surgery ………………………………………………………………………………………………… 17

3

AO ……………………………………………………………………………………………………………… 20

4

Spine surgery within AO …………………………………………………………………………………… 21

5

Bibliography

………………………………………………………………………………………………… 26

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

17

Thomas Schlich, Max Aebi

3

1

HISTORY OF SPINE SURGERY WITHIN AO

INTRODUCTION

This introductory chapter is a survey of the development of spine surgery within AO. It starts with a more general history of spine surgery, putting special emphasis on fractures and their internal fi xation. More detailed accounts, as well as numerous references to original publications, can be found in the works quoted in the literature list at the end of the chapter. The second part of the chapter provides a short introduction into AO’s history [1]. Finally, in the third part of this chapter, both topics converge in the history of the technical and institutional development of spine surgery within AO.

2

SPINE SURGERY

Traditionally, injuries and deformities of the vertebral column were treated by nonoperative management. The medical literature on the topic starts with the Hippocratic Corpus, which mentions traction devices, and continues to refer to these treatments throughout later periods until the emergence of operative treatment methods in the nineteenth century. Early in the nineteenth century posterior decompression by laminectomy was frequently discussed, though only very rarely performed because of pain, inevitable infection, and poor results. Pain and infection became manageable when anesthesia and aseptic operation methods were gradually introduced later in the century. Simultaneously, diagnostic localization of spinal pathology was much improved, fi rst by new physical examination methods, and then by the introduction of radiographic investigation [2].

18

The fi rst attempts at stabilizing the spine with metalwork date back to the late nineteenth century. In 1887 William F Wilkins (1848–1945) operated on a 6-day-old child with a fracture dislocation of T12 on L1 and fi xed the two vertebrae together with silver wire, which he passed around the pedicles of T12 and L1. In 1891 Berthold E Hadra (1842–1903) in Austin, Texas, described a case in which he stabilized a fracture dislocation of the cervical spine through wiring of the spinous processes. Other operations followed so that by the end of the century spine surgery was considered an option in the management of spinal disorders and injuries [2, 3]. In 1909 Fritz Lange (1864–1952) of Munich was the fi rst to use rods to stabilize the spine. He reported on the use of steel rods fi xed to the spinous processes with silk and then later with silver wire. The history of spinal fusion began in New York with Fred H Albee (1876–1945) and Russel A Hibbs (1869– 1932) who, independently of each other, published the results of their respective techniques in 1911. Subsequently, various techniques for achieving spinal fusion were suggested and tried out. But despite sporadic reports about similar operations, the following decades were a time of caution in the field [4]. In the 1950s, efforts were being made to improve the results of lumbar fusions by supplementing them with internal fi xation. In 1944 Donald King (1903–1987), of San Francisco, described the technique of screw fusion of the facets, and in 1959 Harold Boucher of Vancouver reported an improvement of King’s technique with the fi rst description of the use of facet-pedicle screws [2, 4]. In England in the 1950s, early attempts were being made to use metalwork for the correction of scoliosis by Allan (1955) and Roaf (1966). In 1960 Paul Harrington (1911–1980) in Houston, Texas, introduced his hook and rod fi xation system,

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

fi rst used for deformities, later also applied to fractures. The combination of effective instrumentation and bony fusion dramatically reduced the incidence of pseudarthroses and greatly improved the results, making this the standard method for scoliosis correction and fusion throughout the world for more than a quarter of a century. This instrumentation also made it possible to surgically reduce and hold spinal fractures [2, 5, 6]. However, the burgeoning of spine surgery in the second half of the twentieth century also depended on advances in other medical specialties, in particular in the field of imaging. At the time of the Second World War, myelography was developed, and in the 1960s and 1970s CT scanning and magnetic resonance imaging revolutionized the field. At the same time important developments in the field of anesthesia and intensive care meant that surgeons could safely extend the scale of their operative procedures [2, 4]. One disadvantage of the Harrington system was that the rod was too straight and that fi xation depended on the security of the hooks at each end of the rod. A number of surgeons recognized that, the spine being a segmental structure, effective internal fi xation should also be segmental. The development of segmental systems not only provided much more secure fi xation and much better correction of spinal deformity, it also allowed the surgeon to reproduce the normal sagittal curves of the vertebral column [2]. The earliest forms of segmental fi xation for spinal deformity with the help of wiring were developed in the 1950s in Spain and Portugal and perfected by Eduardo Luque in Mexico. Spinal fi xation was also being approached from a different angle in order to facilitate a fi xation of the front of the spine.

3

19

History of spine surgery within AO

In the 1960s Dwyer in Australia developed an anterior segmental spinal fi xation system which involved the correction of scoliosis using vertebral body screws and a flexible cable. The Dwyer system was then modified and improved through the use of a threaded rod by Zielke in Germany [2]. Harrington’s and Zielke’s techniques were the most important treatment methods for spinal fractures right into the 1980s [7]. A decisive improvement with regard to stability was reached with the use of a combination of plate osteosynthesis and pedicle screws. In 1963 Raymond Roy-Camille in Paris started to fi x vertebral fractures by using posterior plates with pedicle screws [8, 9]. It was not until 1970 that Roy-Camille published his method. From the mid-1980s, pedicle screws gained widespread use and numerous analogous systems have since been developed. In order to try to address the mechanics of injuries at the thoracolumbar junction, specific instrumentation systems have been developed [2]. In 1977 Magerl introduced a new biomechanical principle when he presented a fi xateur externe, in which pedicle screws were fi xed outside the body with a special rod system connected to the screws providing angular stability [10]. Walter Dick in Basel took up the idea and modified it by making the screws shorter and placing the rods inside the body, beside the spine, resulting in a fi xateur interne as also proposed by Kluger in 1980 [11, 12]. The fi rst operation with the fi xateur interne took place in 1982 in Basel. This innovation enhanced the practicability of the system. Both these fi xators had an advantage; they enabled the fi xation of a shorter portion of the spine than other techniques leaving more flexibility for the patient. The internal fi xator inspired the creation of many similar devices worldwide. Magerl as well as Dick did their work within the context of AO, the surgeons’ association that had revolutionized fracture care by making internal fi xation a routine surgical procedure.

Segmental instrumentation for use in scoliosis treatment was further developed by Yves Cotrel and Jean Dubousset (1985) in Paris. They developed a double-rod system using fi rst hooks and later a combination of hooks and pedicle screws. Since then, many different spinal fi xation systems have been put on the market. What they all have in common is that they are segmental and fi xed to the spine by wires, hooks, screws, or a combination of all three [2, 4]. Stabilization of the cervical spine required special techniques. At the uppermost part of the neck, the direct stabilization of the dens axis by a screw was fi rst ventured by Friedrich Magerl in February 1978 in a tumor patient, a technique which was also independently described in Japan by Nakanishi in 1978. The standardized technique was then developed and published by Jörg Böhler [13] and later modified with the use of cannulated screws over K-wires by the Bernese group [14]. In order to stabilize an anterior interbody fusion with a bone graft in the cervical spine, Jörg Böhler had started to use “AO plates” in 1964. In 1970 Orozco and Llovet [15] introduced the use of an H-plate for this indication, a technique also described by Sénégas [16]. After production of these types of plates applying AO techniques had begun, the procedure became standard treatment. The original design was later modified by Erwin Morscher, who used titanium locking screws as applied in maxillofascial surgery. These screws achieve enough purchase in the cancellous bone of the vertebra so that it is no longer necessary to include the opposite cortical bone close to the spinal cord as was the case in previous systems. This system became known under the name of the cervical spine locking plate (CLSP) and was a great success [9, 17]. In 1979 Magerl also developed a hook plate system for the cervical spine based on the tension band principle, of which results were published later in 1991 [18]. The unique way in which screws are placed

20

3

into the lateral mass in this context remained under the name, Magerl technique of “articular mass screw fi xation”, as opposed to the Roy-Camille technique, even after the hook-plate system was replaced by other devices. With the development of cervical fi xation systems for anterior and posterior procedures, open surgical stabilization of these injuries has become more common than external fi xation with devices such as the halo vest [2].

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

AO

Techniques of internal fi xation of broken bones already existed in the fi rst half of the twentieth century. However, apart from intramedullary nailing of the femur and nailing of the hip, these techniques never became accepted elements in the surgical repertoire. The main problems were the lack of quality control and standardization with regard to the material used, and, even more importantly, the lack of control over its application. Thus, as soon as surgeons other than those who had invented the new techniques started to use them, results deteriorated and complications abounded. As a result, osteosynthesis was widely regarded as being unsuitable for widespread use. This was the situation when in 1958 thirteen young Swiss surgeons under the leadership of Maurice E Müller, Martin Allgöwer, Robert Schneider, and Hans Willenegger joined together in a surgical association they called AO for “Arbeitsgemeinschaft für Osteosynthesefragen” (Association for the Study of Internal Fixation and Fractures). Their aim was to promote improvement in fracture care by introducing internal fi xation under the tightly controlled conditions of a close-knit group of surgeons. Working with the engineers Robert Mathys and later Fritz Straumann, Maurice E Müller began designing and developing standardized osteosynthesis equipment. In 1960 the leading AO surgeons established Synthes AG Chur, Switzerland. The company was to hold and exploit patents and trademarks of the AO equipment. For this purpose Maurice E Müller handed over all his patents without compensation. This principle became standard policy and all future patents on special instruments and implants developed by AO surgeons have automatically been transferred to Synthes AG Chur free of charge. Synthes AG Chur licensed out these patents to two, later three, exclusive producers under the trademark Synthes. The producers agreed to pay a certain percentage of royalties on all sold items which were to be used to support the AO’s research and educational

3

21

History of spine surgery within AO

4

activities. In the 1970s, the two AO producers Mathys and Straumann (whose company later became Stratec) were joined by a third producer, Synthes USA, led by Hansjörg Wyss. While the producers’ royalties remained the fi nancial basis of the AO’s scientific activities, the group’s policy was determined by the surgeons in the AO, who created a Technical Commission (TK) for surveillance of research and product policy. Only devices approved by the TK were allowed to be included in the equipment. In an attempt to guide the use of these instruments and implants even beyond the small circle of AO surgeons, in the early 1960s AO started to offer instructional courses, fi rst in the Swiss resort of Davos, and later in other countries as well. AO surgeons were extremely successful in convincing colleagues all over the world of the usefulness of their system. By the end of the 1970s AO had outgrown its original structures. In order to adapt to its new size and international activity, the group was reorganized. As its new institutional basis the AO Foundation was established in 1984. The Foundation fully owns Synthes AG Chur. Since 1992 it has been based in a purpose-built headquarters in Davos, which also houses the AO’s various functional divisions: research, development, documentation, and education [1].

SPINE SURGERY WITHIN AO

Initially, AO’s activities did not include internal fi xation of the spine. This negligence reflects a general trend. Operative treatment of spinal fractures using internal fi xation was developed considerably later than similar techniques in the upper and lower extremities. Part of this delay had to do with the relative complexity of the anatomy and function of the spine requiring a certain level of specialization. The often complicated and multifactorial patterns of injury in vertebrae took more time to understand than those of the arms and legs. Also, operative approaches to the spine are much more difficult to achieve than for other body structures and were thus standardized relatively late. Technical complexity was also the reason why principles for fi xation of the vertebrae and special implant systems took a relatively long time to be developed [7]. Another issue that needs further examination, however, is the significance of cultural, economic, and social factors contributing to the rise of spine surgery. Such factors might also help to explain differences in the geographic spread of highly sophisticated medical techniques in bone surgery [19]. In the early 1960s in Central Europe, the vertebral column had not yet become the subject of a surgical subspecialty. After Maurice E Müller had been appointed director of the newly established Clinic of Orthopedic Surgery in St Gallen in 1960, spine surgery including surgery of the injured spine was part of the clinic’s operative repertoire. Bernhard G Weber was especially interested in the field and started to develop techniques such as spondylodesis with the help of internal fi xation, fi rst using wire cerclage, and later using plates. In 1968 the Austrian surgeon Friedrich Magerl came to St Gallen. He had been an AO adherent since his fi rst encounter with the AO technique during his training period in Graz, Austria, where he had already worked on spine surgery. Shortly after his move to St Gallen, surgeons there started to use long-bone plates,

22

applying the AO techniques for stabilization of vertebral fractures. In this context Magerl developed his special techniques and invented the fi xateur externe for the spine, mentioned above, a principle which was used for the fi rst time on June 10, 1977 on a motorbike driver with an unstable fracture of the third lumbar vertebra. Together with the engineer Fridolin Schläpfer, Magerl developed the external fi xator into an element of “AO equipment”. In October 1980 he also had created the fi rst prototype of an internal version of the device, which was further developed later by Dick in Basel under the influence of Kluger from Germany as mentioned above. At that time, besides St Gallen, Basel was the most important center in Switzerland of surgery in spinal trauma. Being the location of the Swiss Paraplegic Center, many patients in need of spine surgery came to be concentrated in Basel. When Max Aebi created the fi rst specialized academic spine unit at the Department of Orthopaedic Surgery of the University of Bern in 1984, where spinal research, development, and clinical application became focused, this center triggered the further development of spine surgery in Switzerland, and with the cooperation of John Webb from Nottingham (UK), on an international scale as well. In the AO’s documents, vertebral fractures start to appear as a subject of discussion during the 1970s [1]. In 1980 a number of interested surgeons formed a group within the AO and suggested the addition of a special unit for spine surgery to the annual AO course in Davos. The following year the fi rst spine symposium took place as a subunit of the course [20]. It was organized by Magerl and Morscher in 1981, and in January 1983 the fi rst independent course was organized in Davos. The list of participants includes numerous directors of trauma units, many of them AO surgeons without spine surgery background.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Initially, AO surgeons and producers were rather reluctant to spend much time or money on specialized instruction in spine surgery. Given the low number of specialists existing at the time, they did not expect spinal instruments and implants to be in great demand. These sorts of instruments, they thought, would never become mass articles; to have them in the program was more a question of prestige than of commercial consideration [1]. The traditional AO members did not really recognize the future potential of spine surgery for the AO. In Urs Heim’s book “The AO Phenomenon” of 2001, spine surgery is not even mentioned [21]. In fact, AO had underestimated the dynamics in this field. More and more AO symposia, workshops, and courses, however, were organized. By the mid-1980s the subject had given rise to a new dynamic subspecialty within surgery. Its members were now able to operate on a variety of conditions that had not been amenable to effective treatment before. It is an indication of AO’s increase in interest that Magerl was awarded the AO prize for his work on external and internal fi xation of spinal fractures in 1984 [1]. In the early 1980s, the informal spine group within AO organized itself more formally. Within the AO organization itself, a specialty TK led by Erwin Morscher, was established. Cooperation with American spine surgeons was initiated, most notably among them Rae Jacobs (1937–1988) of Kansas University, who together with Robert Mathys had developed a spinal hook-rod system, basically a modified Harrington system, in the late 1970s. After Magerl’s external fi xator, Jacobs’ rod was one of the fi rst AO developments in spine surgery. Jacobs was also the founding chairman of AO’s “North American Spine Study Group”.

3

23

History of spine surgery within AO

In the 1980s AO surgeons specializing in the spine continued to develop techniques and devices. In 1986 Max Aebi and Röbi Mathys Jr invented a modified version of the internal fi xator clamp. Today, this clamp is still part of the trauma module of the USS. A significant achievement was also the creation of an internationally acknowledged classification system for spinal injuries, which was fi nally published in 1994 [22]. A major impulse for the development of the spinal armamentarium was started after the modification of the internal fi xator. At the behest of Martin Allgöwer, parallel to the AO’s Spine TK, a special task force for development of spine surgery was set up in November 1986 because the regular Spine TK was too slow in keeping pace with the market developments in spinal surgery. It was initially comprised of Max Aebi and John Webb as surgeons, and Robert Frigg as engineer. Its members met every 8 weeks, starting in January 1987, to develop the universal spine system (USS), which became the flagship of spinal instrumentation in the early 1990s. The USS as a segmental instrumentation and based on pedicle fi xation, either by screws or special screw-hooks, is comprised of three surgical modules that can be used to address different pathological groups: treatment of deformity, degenerative disorders, and spinal trauma. Stratec supported the group and contributed the engineer Fridolin Schläpfer, however, AO did not invest primarily in further development of spine surgery. Simultaneously, a number of compatible devices to supplement the USS were developed or initiated by the Spine Task Force, such as a system allowing for anterior fi xation (Ventrofi x) or another one for inclusion of the sacrum. In the fi rst half of the 1990s the anterior intervertebral cage concepts were developed and with them the less invasive access technology for spine surgery (John Thalgott, Max Aebi, and Thomas Steffen). It was the AO surgeons specializing in spine surgery who encouraged

the producers to build their specialist units for the development of new devices for spine surgery. In 1991 Synthes Spine was officially launched as a company in the USA to serve the North American market. The initial product line consisted of items developed and manufactured by the sister companies in Europe and imported by Synthes Spine for sale in America. Inspired by the AO Manual of Osteosynthesis the AO Principles in Spine Surgery was published in 1998 [23]. For the fi rst time a monography was published, where the experience of the specialized courses in spine surgery and the concept and “philosophy” of spinal instrumentation was condensed into one book. Before, spinal principles and techniques were almost not represented in the AO Manual of Internal Fixation—in the second edition of the 409-page book, only a 2-page section about spinal fracture treatment was included, whereas in the third edition Aebi and Webb [24] were asked to author an entire chapter, which was comprised of 55 pages. The 243-page book, AO Principles in Spine Surgery, however, was published in 4 languages (English, Spanish, Mandarin, and Japanese). Within AO, a new generation of spine surgeons came into the leadership when on January 1st, 1991 John K Webb took over as chairman of the Spine Surgery Specialty TK (AO Annual Report 1991, 13). 9 years later, after an intense preparation initiated by Max Aebi and John K Webb, a more radical restructuring of the institutionalization of spine surgery within the AO took place. In December 2000 the AO Specialty Board for Spine Surgery was inaugurated. It was the fi rst specialty group within the new concept of AO to allow for more autonomy of clearly defi ned specialty groups within the AO Foundation [25]. Designed to give the specialty groups more autonomy, it was to provide an environment which would cater to the specific needs in education, research, and

24

development of the field. Under the chairmanship of John K Webb the board consisted of six specialist spine surgeons responsible for education, research, and development policy. The AO Spine Expert Group started off being organized quite autonomously, while reporting to the AOVA. Instead of a permanent TK, however, flexible task forces were responsible for new developments. At the same time, it was supported by an Education Committee and a Research and Strategy Committee. Day-to-day matters became the responsibility of the Executive Committee and the Executive Directors (AO Foundation Annual Report 2001). The rationale for this increase in autonomy resulted from ever more visible differences in approach and clientele between AO as such and those surgeons committed to spine surgery. On a global level, most of spine surgery is being done by specialized orthopedic and neurological surgeons, not by trauma surgeons, who are AO’s natural clientele. Therefore, AO does not have the same prestige among spine surgeons as among trauma surgeons. Also, trauma accounts for only around 16% of spine surgery in general, the rest being degenerative disorders, tumors, etc. This means that the issues and priorities that occupy the community of spine surgeons differ very much from those of the majority of AO surgeons, who come from fracture care and traumatology. In order to set their own priorities the spine surgeons within AO felt they needed to control more of the steadily growing portion of royalties the producers paid for implants sold in their specific field. This was also plausible on the basis of the unprecedented increase in sales of spinal implants and instruments, meaning that more and more of AO’s total royalty income originated from that segment of the market, a development which in turn was based on the scientific and educational efforts made by the AO’s spine surgeons. The establishment of a relatively autonomous

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Spine Group within the AO was thus the consequence of the dynamic development of the last two decades during which spine surgery became increasingly important not only in AO, but rather in musculoskeletal surgery in general. On behalf of AOSpine the AO Foundation hired a professional business manager (Michael Piccirillo), who has been instrumental in introducing business expertise, transparent processes, and marketing management to AOSpine in order to enable the development of an exciting platform and value proposition that appeals to spine specialists across the globe. This development reached its culmination in the formation of the AOSpine Specialty Group in the summer of 2003 during the Trustees Meeting in Crete, after Max Aebi took over the 3-year Chairmanship of the Spine Specialty Board in January 2003. From then on the AOSpine surgeons had a template of governance contract with the AO Foundation, where the role of the AOSpine Specialty Board and Group within AO was defi ned. Now an unprecedented growth in the activities of the AOSpine Group was to take place and a membership scheme established. With a fast growing number of new members, the group began several projects, which are presently underway and are not yet finished. There was a branding process implemented with the development of the logo, AOSpine, which is now consistently applied. Furthermore, AOSpine started a determined regionalization process with four major world regions (AOSpine North America, AOSpine Latin America, AOSpine Europe, and AOSpine Asia Pacific). Within the regions, country chapters or subregional sections were created. The authority and responsibility for education, documentation, clinical research, and development was delegated to the region and the AOSpine International Board concentrated on developing research strategies, general educational strategies, dealing with the TK

3

25

History of spine surgery within AO

system, scientific marketing, and coordinating leadership. Innovative initiatives emerged from this process like the socalled Spine Research Network with priority research concentrating on the degeneration and regeneration of the motion segment [26]. Several such initiatives have also been later adapted to the overall AO. By January 2006, after Michael E Janssen from Denver, USA, took over as Chairman, the board was expanded by two further members, one responsible for research, and one for scientific marketing. The leadership structure was further refi ned and established, and the new Chairman was fi nally able to negotiate in spring 2006 a separate legal entity for AOSpine, a goal which has been long worked for. The AOSpine has matured over all these years and has now become a global spine organization with ambitious goals for the future (www.aospine.org). The development and innovation priorities have changed. AOSpine, whose ideas have been developed and translated together with the industrial partners into viable products is no longer a “brotherhood” [1] of pioneers, but a modern global organization of many different surgeons and scientists, who all share their knowledge and ideas. It is probably not to be expected that ground breaking innovations will be made, since AOSpine has—like the whole spine community—for the last few years focused on nonfusion technology such as disc replacement. This technology has not been developed within AOSpine, but has been acquired by Synthes Inc. and brought from the industrial side into the AO (although a development group within the AOSpine was in the process of developing a novel concept of metal-on-metal intervertebral disc replacement).

The future development in spine surgery will be focused on further refi nement of existing devices, pushing the technology to the edge, and the next step of fundamental changes in spine surgery will most probably come from the biosciences.

26

5

1.

2.

BIBLIOGRAPHY

Schlich T (2002) Surgery, Science and Industry. A Revolution in Fracture Care, 1950s–1990s. Houndmills, Basingstoke and New York: Palgrave Macmillan. Dove J (2002) Evolution of spinal surgery.

Klenerman L (ed), The evolution of orthopaedic surgery. London: Royal Society of Medicine 3.

Press, 159–166. Hadra BE (1975) Wiring of the Vertebrae as a

Means of Immobilization in Fracture and Potts’ Disease, Series “The Classic”, Clinical

4.

5.

6.

7.

Orthopaedics and Related Research 112, 4–8. Originally published in 1891. Wiltse LL (1987) History of Lumbar Spine Surgery. Lumbar Spine Surgery. Techniques and Complications . St Louis: The C.V. Mosby Company, 5–23. Kahn A (1986) Current Concepts of Internal Fixation. Dunsker SB, Schmidek HH, Frymoyer J, Kahn A (eds). The Unstable Spine. Orlando: Grune&Stratton, 45–83. Wiltse LL (1991) The History of Spinal Disorders. Frymoyer JW (ed), The Adult Spine: Principles and Practice. New York: Raven Press, 3–41. Kinzl L, Arand M, Hartwig E (1997) Wirbelsäulenverletzungen. Oestern HJ, Probst J

(eds), Unfallchirurgie in Deutschland, Bilanz und Perspektiven . Berlin: Springer Verlag, 511–522. Roy-Camille R, Saillant G, Berteaux D, et al (1976) Osteosynthesis of thoracolumbar spine fractures with metal plates screwed through the vertebral pedicles. Reconstr Surg Traumatol ; 15:2–16. 9. Povacz F (2000) Geschichte der Unfallchirurgie. Berlin: Springer Verlag. 10. Magerl F (1985) Der Wirbel-Fixateur externe. 8.

Weber BG, Magerl F (eds) Fixateur Externe. AO-Gewindespindel-Fixateur. Berlin: Springer Verlag. 11. Dick W, Kluger P, Magerl F, et al (1987) A new device for internal fi xation of thoracolumbar and lumbar spine fractures: the “fi xateur interne”. Paraplegia ; 23(4):225–232.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

12. Dick W (1987) Innere Fixation von Brust- und Lendenwirbelfrakturen . Bern: Hans Huber Verlag. 13. Böhler J (1982) Anterior stabilization for acute fractures and non-unions of the dens. J Bone Joint Surg Am ; 64(1):18–27. 14. Etter C, Coscia M, Jaberg H, et al (1991) Direct anterior fi xation of dens fractures with a cannulated screw system. Spine ; 16(3Suppl): S25–32. 15. Orozco Delclos R, Llovet TJ (1970) Osteosynthesis en las fracturas de raquis cervical: nota de technica. Rev Orthop Traumatol ; 14: 285–288. 16. Sénégas J, Gauzère JM (1977) Traitement des lésions par voie antérieure. Rev Chir Orthop Reparatrice Appar Mot ; 63(5):466–469. 17. Morscher E, Sutter F, Jenny H, et al (1986) [Anterior plating of the cervical spine with the hollow screw-plate system of titanium.] Chirurg ; 57(11):702–707. 18. Jeanneret B, Magerl F, Ward EH, et al (1991) Posterior stabilization of the cervical spine with hook plates. Spine ; 16(3 Suppl):S56–63. 19. Schlich T (2002) Degrees of Control: The Spread of Operative Fracture Treatment with Metal Implants. A Comparative Perspective on Switzerland, East Germany and the USA, 1950s-1990s. Stanton J (ed) Innovations in

Health and Medicine. Diffusion and Resistance in the Twentieth Century. London and New York: Routledge, 106–125. 20. Schneider R (1983) 25 Jahre AO-Schweiz,

Arbeitsgemeinschaft für Osteosynthesefragen 1958–1983. Davos: Arbeitsgemeinschaft für Osteosynthesefragen. 21. Heim UF (2001) The AO Phenomenon . Bern, Göttingen, Toronto, Seattle: Hans Huber. 22. Magerl F, Aebi M, Gertzbein SD, et al (1994) A comprehensive classifi cation of thoracic and lumbar injuries. Eur Spine J ; 3(4), 184–201. 23. Aebi M, Thalgott JS, Webb JK (1998) AO ASIF Principles in Spine Surgery. Berlin Heidelberg New York: Springer-Verlag.

24. Aebi M, Webb JK (1991) The Spine.

Müller ME, Allgöwer M, Willenegger H, Schneider R (eds), Manual of Internal Fixation . 3rd ed. Berlin Heidelberg: Springer-Verlag, 627–682. 25. Schatzker J (2000) Presidential address . Vancouver: AO Trustees Meeting. 26. Spine Research Network: Global collaboration in studying the spine. AO Foundation Global Research Report 2005 :26–27.

27

Thomas Steffen

BIOMECHANICS OF THE SPINE

4

31

Thomas Steffen

4

BIOMECHANICS OF THE SPINE

INTRODUCTION

This chapter provides basic anatomical and biomechanical knowledge of the spine. Anatomy uses descriptive language to teach physical structure and biomechanics applies engineering methods to the analysis of human motion. This chapter will also explore what biomechanics can teach us about spinal function. Figuring out the relationship between structure and function is the key to grasping how the healthy spine functions properly. Knowledge of this relationship is also important when trying to rationalize why a specific type of instrumentation used to alter the spine’s mechanical behavior is believed to ultimately help the patient. Understanding the anatomy and biomechanics of the spine helps physicians determine the likely source of a patient‘s spinal complaint. The diagnostic workup of a spine patient is often challenging—due mostly to the complexity of the spinal

anatomy, the multisourced generation of pain, and the sheer number of joints involved. The physician needs to have deep and thorough knowledge to be able to narrow a patient’s complaint down to a specific problem, which is often mechanical in nature. It is like everything else in life: We will see only what we are looking for, we will only fi nd what we already know, and we will only have a grasp of what we understand. It is the author’s goal to provide the reader with a certain level of biomechanical knowledge so that they can become experts in diagnosing spinal problems. Using this knowledge and combining it with a history and physical examination can guide clinicians in determining the likely cause of a patient‘s spinal complaints and the ways in which it can be treated.

32

4.1

GENERAL BIOMECHANICS OF THE SPINAL MOTION SEGMENT AND THE SPINAL ORGAN

1

Introduction …………………………………………………………………………………………………… 33

2 2.1 2.2 2.3 2.4 2.5 2.6

Motion segment ……………………………………………………………………………………………… Vertebra ……………………………………………………………………………………………………… Facet joints …………………………………………………………………………………………………… Intervertebral disc …………………………………………………………………………………………… Muscles ………………………………………………………………………………………………………… Ligaments ……………………………………………………………………………………………………… Spinal cord, nerve roots ……………………………………………………………………………………

3

Spinal motion ………………………………………………………………………………………………… 45

4

Bibliography …………………………………………………………………………………………………… 51

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

34 34 36 37 43 44 44

33

Thomas Steffen

4

BIOMECHANICS OF THE SPINE

4.1

GENERAL BIOMECHANICS OF THE SPINAL MOTION SEGMENT AND THE SPINAL ORGAN

1

INTRODUCTION

The spine is essentially a curved stack of 33 vertebrae, which can be divided based on structural differences into five distinct regions: cervical (7 vertebrae), thoracic (12 vertebrae), lumbar (5 vertebrae), sacral (5 fused vertebrae), and coccygeal (4 fused vertebrae). An obvious difference between the distinct regions is the curvature in the sagittal plane. The thoracic and sacral regions of the spine feature a kyphotic curvature. Kyphotic curvatures are considered primary because they already exist at birth. Later, to allow the growing child an upright posture, a secondary lordotic curvature develops in the cervical and lumbar regions. Individual vertebrae, including all interposed structures, show reasonable similarities along the entire spine. This is mostly true, except for the upper cervical spine which has an adapted anatomy to allow for larger head movements, and for

the sacral and coccygeal regions in which mobility was largely lost due to aberrant discs. The remaining spine, reaching from C3 to S1, is often referred to and looked upon, for diagnostic and therapy-related decisions, as being composed of individual motion segments. The subsequent sections describe the individual structural components of such motion segments, their interplay with neighboring components, und how this ultimately enables the spine to fulfi ll a complex function. The primary purpose of the entire spine is to provide axial support for the head and trunk, while allowing for bending and twisting movements, and to protect neural structures encased in a bony canal running from the head to the sacrum. Not unique to the spine, but nevertheless significant, is its contribution to blood cell formation through a noteworthy total mass of bone marrow.

34

4

Biomechanics of the spine

2

MOTION SEGMENT

2.1

VERTEBRA

Vertebral anatomy—regional differences

A vertebra is composed of the vertebral body and posterior elements, which include the paired pedicles, superior and inferior articular processes with interposed intraarticular mass, the lamina, transverse processes, and the singular spinous process. Vertebrae are primarily composed of cancellous bone, an anisotropic viscoelastic material. Fortunately, for noninjurious values of strain and over a wide range of strain rates, cancellous bone behaves elastically. The vertebral body has an approximated cylindrical shape. A thin shell of increased density trabecular bone surrounds a core of cancellous bone. Posterior elements are made from true cortical and cancellous bone. Regional differences between vertebrae are obvious: • Typical cervical vertebrae have a small broad body, a large triangular canal, laterally directed pedicles lying just anterior to the transverse foramen, and medially directed laminae ending in a bifid spinous process. The seventh cervical vertebra (C7), with its long spinous process palpable through the skin, is known as the vertebra prominens. The lowest portion of the nuchal ligament attaches to this spinous process. • Thoracic vertebrae have a triangular cross section at the cranial levels, but they gradually become more circular at the caudal levels. Most vertebrae feature paired superior and inferior demifacets for the rib head articulation. The canal is small relative to the body and is circular in outline, but there are no transverse foramina in the thoracic transverse processes. The typical spinous

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

processes are long, straight and narrow. They overlap like roof tiles. Transverse processes are equally prominent and articulate with the tubercle of the ribs. The superior articular processes are vertical, flat, and face backward and laterally. • Lumbar vertebrae have a large size vertebral body. The vertebral foramen is triangular, and the spinous process is hatchet-shaped and blunt. The superior articular surfaces are vertical, curved, and face backward and medially inward. The posterior rim features a mammillary process. • The sacrum is typically formed from five fused sacral vertebrae and is triangular in shape. It has a superior, posterior, anterior, and lateral surface. The anterior surface is curved. The superior surface is formed by the superior end plate of the fi rst sacral vertebra. Vertebral bodies are much wider transversely than anteroposteriorly. The S1 superior articular processes are concave and directed posteromedially, to be congruent with the L5 lower articular processes. • The lower end of the sacrum is often fused with the coccyx, the four lowest vertebrae which are small and rudimentary. The coccygeal vertebrae, of course, form the tail in animals, with the aberrant human version of it gently reminding us of our origins. The coccyx can be a common source of postpartum pain, but also as a result of falling on the buttocks.

How is axial load being carried?

Between 70% and 90% of static axial load is carried by the cancellous vertebral body. The role of the shell and core in providing mechanical strength varies with age. They carry, modulated by sagittal posture variations, the remaining axial load. Processes serve as lever arms to provide mechanical advantage for muscles inserting along their surfaces.

4.1

35

General biomechanics of the spinal motion segment and the spinal organ

Vertebrae are loaded in series. Caudal vertebrae must support a greater share of the body weight and this accounts for an increasing cross-sectional area of the vertebral bodies. In healthy adults the bone density remains reasonably constant throughout the entire spine. Ultimate failure of the individual vertebral bodies similarly increases from cranial to caudal vertebrae (Table 4.1-1).

Spinal region

C3–7

T1– 6

T 7/8

T9 –12

L1–5

Streng th ( N )

1,600

2,000

2,300

3,600

5,600

Table 4.1-1 Approximate ultimate compressive strength values of different vertebral bodies for a nonosteoporotic adult male.

Fig 4.1-1 Compressive strength (yield stress, left graph) and stiffness (Young modulus, right graph) of vertebral cancellous bone as a function of bone density (2nd order polynomial function curve fit). The data were recorded from 12.5 mm diameter cancellous bone plugs compressed at a 0.1% strain/sec rate. Fractional bone volume was measured from bone samples of the same vertebral body (mirrored site) using the Archimedes principle.

36

4

Biomechanics of the spine

Normal vertebral cancellous bone has a bone density of about 15%. Axial compressive material properties for the cancellous bone of a normal vertebral body (ie, 15–18% fractional bone volume) are estimated around 5 MPa yield stress and around 300 MPa elastic modulus ( Fig 4.1-1). The strength and elastic modulus (Young modulus) of cancellous bone are roughly dependent on its density to the second power. That is, a 25% decrease in density may result in a 50% decrease in strength. The dense shell takes a greater share of the load as the cancellous core density is gradually lost due to osteoporosis. In severe osteoporotic conditions, bone density can be drastically reduced, maybe as low as one third of its original density. Not surprisingly, the overall load carrying capacity of osteoporotic vertebrae can then be reduced by almost a magnitude. From initially being fairly uniform, with progressive loss of bone density, regional differences in trabecular bone strength become more obvious. Trabecular bone strength decreases from the anterior to the posterior and from the medial to the lateral. Thanks to the bony end plate, the axial load is more uniformly distributed across the cancellous bone cross section. The strongest part of the bony end plate is the peripheral epiphyseal ring, which makes this region best suited to resist localized axial loads. Intervertebral spacers (cages) best resist subsidence when they are seated on this epiphyseal ring [1].

2.2

FACET JOINTS

Facet orientation and joint loads

The paired facet joints form, along with the intervertebral disc, the intervertebral joint. Whereas the disc is a fibrocartilaginous syndesmosis, the facet joints are diarthrotic joints with sliding cartilaginous surfaces lubricated with synovial fluid. Facet joints channel and limit the range of motion in anteroposterior shear and axial rotation directions ( Fig 4.1-2 ), which is an important and differentiating aspect of spinal regions. • Cervical spine: facet orientation is roof-tile shaped, coupling lateral bending and axial rotation motions in an opposite direction (bending the head to the left results automatically in an axial rotation to the right). • Thoracic spine: coronal plane orientation with a slight inward tilt in the transversal plane, permitting easy axial rotation movements with the center of rotation projected into the vertebral body. • Lumbar spine: sagittal plane orientation of facet joint surfaces, effectively blocking axial rotation movements. Facet joints carry, in an upright standing posture, between 10% and 20% of the axial body load. In hyperextension, the joint load increases up to 30%. In a flexed posture, the facet joints carry up to 50% of the anterior shear load (compressive loads transmitted by surface contact and tensile loads resisted by joint capsule). Facet joint capsules are highly innervated and have been shown to be a source of low back pain. Facet joint tropism

Facet joint tropism is defi ned as asymmetry in the facet joint angles, with one joint having a more coronal orientation than the other. It has been reported [2] that the incidence of tropism

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

4.1

General biomechanics of the spinal motion segment and the spinal organ

in patients with degenerative disc disease is higher than in the normal population. When tropism is present, the segment tends to rotate toward the more oblique facet when axial loads are applied. Rotation resulting from the joint asymmetry can place additional torsional stress on the anulus fibrosus, thus, possibly contributing to intervertebral disc injury.

2.3

INTERVERTEBRAL DISC

Disc anatomy

The nucleus pulposus is located approximately in the disc’s center. Through a transitional zone the disc’s appearance gradually changes toward the periphery, with concentric annular fiber layers making up its outer border. The nucleus, particularly with age, is difficult to delimit. It covers an estimated 30%–50% of the cross-sectional area of the total disc. The healthy nucleus contains almost exclusively type II collagen fibers in an aqueous gel rich in proteoglycans. The latter attracts water, leading to natural swelling of the nucleus. Water content in the normal nucleus decreases from about 90% of its total volume during the fi rst year of life to around 70% at old age. Water is gradually replaced with a nondirectional fibrous matrix, associated with an overall loss in tissue elasticity and an increase in stiffness. The end plates are composed of a dense layer of trabecular bone, further covered with a layer of hyaline cartilage ( Fig 4.1-3 ). Vascular channels within the vertebral bodies have been observed to run directly along the end plates, representing the predominant nutrient source for the adult disc cells. Some blood vessels approach the annulus at the periphery but do not penetrate the disc. A healthy intervertebral disc is the human body’s largest avascular structure. The cartilaginous end plates Fig 4.1-2 Ranges of motion (ROM) for each spinal motion segment of the entire spine. Values are given separately for flexion/extension, lateral bending, and axial rotation. Significant differences are apparent for the cervical, thoracic, and lumbar spine. These differences are largely the result of distinct regional facet joint orientations.

37

38

4

Biomechanics of the spine

undergo progressive calcification with age, which impedes nutrition and contributes to a progressive degeneration of the disc throughout adulthood.

regions to less than twelve in the posterior regions. The individual fi ber layers are in the order of 50–300 µm thick, with the outer layers generally being smaller.

The anulus fibrosus is composed of concentric layers of collagen fiber bundles. The fiber orientations alternate from layer to layer, with the fibers generally oriented at an angle of approximately 30° with respect to the horizontal plane and in any two adjacent layers at 120° with respect to each other. The fiber orientation from outer to the inner annulus gradually becomes more horizontal. The anterior anulus fibrosus band is thickest, the posterolateral and, most significantly, the posterior anulus fi brosus band is thinnest. The number of distinct fiber layers varies from over twenty in the anterior

The fibers are almost exclusively of type I collagen for the outer annular portions, but they gradually change to a 40% type I and 60% type II fiber ratio for the inner portions. With degeneration, type I collagen fibers are replaced with type II fibers.

AF

ER

NP

CEP BEP

VB

Fig 4.1-3 Vertebral body (VB) with bony end plate (BEP) and cartilaginous end plate (CEP). The intervertebral disc is comprised of the layers of the anulus fibrosus (AF) and the central nucleus pulposus (NP). Sharpey fibers insert from the anulus fibrosus directly into the epiphyseal ring (ER).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Many fiber layers are discontinuous. This fact may be responsible for interlamellar stress peaks predisposing the annulus to fail with the formation of circumferential or radial tears. SEM (scanning electron microscopy) imaging has shown the fibers in the inner third of the annulus to interconnect loosely with the cartilaginous end plate; the fibers in the outer portion are fi rmly bonded to the epiphyseal ring of the bony vertebral end plate. Thus, the inner annulus is most prone to initial mechanical failure. Seemingly, intervertebral disc morphology predisposes injury through sites of high stress. Disc cells are not as readily serviced with nutrients as other tissues in the body, a critical factor that only gets worse with age. Degeneration and/or injury decrease the functional ability of the disc to distribute axial forces through hydrostatic pressure. Finally, degenerative changes are accompanied by an ingrowth of nerve fibers into the outer annular regions, sensing and transmitting pain. Anulus fibrosus is made to resist hoop stresses

The intervertebral disc is an inhomogeneous, anisotropic, porous, and nonlinearly viscoelastic structure. Mechanical characterization of discs can either be performed on isolated anulus fi brosus or nucleus pulposus material, or on intact whole discs. Since the annulus is physiologically loaded

4.1

39

General biomechanics of the spinal motion segment and the spinal organ

in tension (at least for the nondegenerated disc), its tensile properties are best documented. Test data on whole discs reflect the predominant compression loading to which they are subjected in vivo and their exhibited viscoelastic behavior.

to nondegenerated ones. Tensile failure strength of the annulus shows a similar trend, with the maximum values of healthy discs being around 5–10 MPa for loads applied in a circumferential direction.

Quasistatic tensile modulus and failure strength have been determined for small rectangular anulus fi brosus specimens. The modulus of the outer annulus is greater in the circumferential direction compared to the vertical direction (approximately 3–4 MPa compared to 0.5 MPa for circumferential and vertical directions, respectively). The anterior annulus consistently reveals a larger tensile modulus than the posterior annulus, regardless of depth and loading direction. This implies a potential weakness of the posterolateral annulus, making it more prone to bulging or protrusion of disc material. Degenerated discs have lower moduli compared

In a healthy disc, with the annular fibers cyclically loaded in circumferential directions, the endurance limit for such hoop stresses is around 1.5 MPa, but it can drop quickly for degenerated discs. The annulus is weakest in a radial direction, with tensile strength values consistently below 0.5 MPa. The annulus is poorly designed to resist tensile radial forces which tend to separate the laminar layers. When the annular band is being compressed on the side of bending (eg, anteriorly during flexion), inner fiber layers are bulging inward and outer layers are bulging outward, in effect separating annular fiber layers ( Fig 4.1-4 ). Fig 4.1-4 Transversal disc maps [3] showing internal disc strain during compression in extended posture, measured for the posterocentral (PC), posterolateral (PL), centrolateral (CL), anterolateral (AL), and anterocentral (AC) annular regions. All data are from normal discs ( Thompson grade I or II). The left side displays strain (expressed in %, with standard deviations in brackets) recorded in a circumferential direction, the right side in a radial direction. Red areas were identifi ed to demonstrate signifi cant strain in a tensile mode, blue areas in a compressive mode. Whereas the circumferential tensile strain in the annular regions is expected, the tensile radial strains in the PL and PC regions are not. Tensile radial strains were even amplifi ed in degenerated discs. They might be responsible for the progressive annular fi ber delamination seen in degenerated discs.

40

4

Biomechanics of the spine

Nucleus pulposus pressure varies with external loads

Due to the high proteogycan content of the disc matrix, the nucleus pulposus (NP) has a base swelling pressure of about 0.1–0.3 MPa [4]. When a normal disc is cut through the center, the NP is immediately protruding from the cutting surface. When the disc is put in a saline solution the NP matrix will continue to swell.

Besides this base pressure, the NP pressure measured at its center is greatly modulated by external trunk loads and paraspinal muscle tension balancing those external loads. Nachemson [5] and later Wilke [6] both measured intradiscal pressure in healthy subjects using an invasive method ( Fig 4.1-5 ). Moderate activities such as walking or stair climbing, compared to upright standing, may already double the NP pressure. Carrying a 20 kg load, depending on the technique used, can increase the NP pressure by a factor of four.

Fig 4.1-5 Intradiscal pressure measurements using a percutaneous needle placed in the center of the disc [5, 6]. Pressures are normalized to the average pressure recorded in an upright standing position. The pressure may rise four to five times for moderate physical activity such as the lifting of a 20 kg weight. Knee and lower back positions play a major role.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

4.1

General biomechanics of the spinal motion segment and the spinal organ

McNally and Adams [7] introduced a new technique called “stress profi lometry” that allowed them to quantify pressure within the nucleus and annulus under axial load application. A strain-gauged sensor mounted to a needle was passed incrementally through the disc along a straight path (midsagittal or mid-coronal plane). Measurements were obtained also for flexion, extension, and side bending. The authors found abnormal stress concentrations in the posterior annulus [8], which suggests a predisposition for the prolapse of an intervertebral disc (IVD) at this location. This stress peak in the posterior annulus was found to increase even further after minor damage to the trabecular arcades was induced by supramaximal compression to the vertebra. The author [9] measured intradiscal pressures using three needles with a total of nine strain-gauged pressure sensors mounted on them. The needles were placed in the anterior, as well as in the left and right posterolateral disc regions, with the sensors positioned in the intermediate disc zone between nucleus and annulus. “Pressure maps” were recorded for axial loads applied in the flexed or extended postures, combined with axial rotation ( Fig 4.1-6 ). The largest pressure increase was found in the posterolateral regions during flexion, predominantly on the side of axial rotation (eg, left posterolateral region for left axial rotation). It seems that the disc’s posterolateral regions are subjected to even higher stress peaks when axial rotation is added to the loading modality. Volume shifts in the nucleus pulposus matrix

The water content of the matrix is variable and represents an equilibrium between two opposing pressures: mechanical, which dehydrates the gel-like matrix; and swelling of the hydrophilic proteoglycans, which causes the matrix to absorb fluid. Changes in the load applied to the motion segment will disturb the equilibrium and subsequently cause a net outward fluid

flow, until a new balance is reached. Net fluid flow can occur within the disc (eg, from the anterior to the posterior areas) or from the disc to the outside. Fluid exchange between the disc and the surrounding tissue occurs through both the periannular route [10] and through the end-plate route [11, 12]. In this scope, diurnal changes in the intervertebral disc height can easily be explained. At night, in a supine position, reduced axial load acting on the disc allows relatively unopposed swelling of the disc‘s matrix. During the daytime, in a predominantly upright position, the absorbed fluid is again expelled from the disc. Apart from circadian rhythms modulating the IVD volume, postural load changes can be responsible for an internal fluid shift [13]. Because the IVD‘s permeability is very low [14], a considerable amount of fluid shift can only be achieved by large changes in postural loads acting over long periods of time. Maintaining a specific posture over an extended period (as might be the case in a workplace) will produce fluid shifts within the IVD. These shifts in turn can influence the spine‘s mechanical behavior, and, more importantly, increase the disc‘s vulnerability to localized mechanical overload [15]. Disc mechanical failure

The intradiscal pressure in a healthy individual is proportional to the compressive load applied to the motion segment. The maximum pressure is about 1.5 times the applied force, divided by the disc’s transverse cross-sectional area. Because of this proportionality, the disc pressure (in a healthy disc) can be used to estimate compressive loading of the spine. Pressure causes the end plate to bulge about 0.5 mm toward the vertebral body [16]. Excessive disc load, seen mostly during flexion and while carrying loads, may fracture the end plate’s central region.

41

42

4

Biomechanics of the spine

A degenerated or aged disc undergoes gradual changes. Common structural changes include loss of nucleus pulposus volume, disc height loss, radial fi ssures, circumferential clefts and rim tears in the annulus, inward bulging of the inner annulus, increased radial (outward) bulging of the annulus, reduced disc height, and possible end-plate defects with disc material herniated into the adjacent vertebral body. A healthy disc contains a relatively soft, hydrated hydraulic nucleus pulposus, which distributes load and stress evenly between

vertebrae. The degenerated disc with a diminished hydrostatic region exhibits high stress concentrations in the posterior annular regions [3]. Disc degeneration also affects other elements of the motion segment. Through disc height loss and apparent changes in axial load transmission, the facet joints become incongruent and may become mechanically overloaded. They gradually develop arthritic changes as is the case with all diarthrotic joints. Also, loss of disc height will initially result in ligament Axial rotation

anterior

inner middle outer

ipsilateral a Fig 4.1-6a–b Intradiscal pressure increase measured in a multisegmental cadaveric lumbar spinal test setup. Pressures were recorded at 10 Nm with a left-sided axial rotation load, recorded either in a neutral, fl exed, or extended posture [9]. Three needles were placed

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

contralateral

b in the anterior and both posterolateral regions. For each needle three sensors were spread out across the disc’s transitional zone (b). The highest pressure increase was seen in the posterolateral region, facing the side of axial rotation (ie, ipsilateral).

4.1

General biomechanics of the spinal motion segment and the spinal organ

laxity, hypermobility, and a loss in segmental stiffness. Only much later, with almost complete collapse of the disc space and the formation of bridging osteophytes, the segmental mobility will be reduced and the stiffness increased. Ultimately, the segment will spontaneously fuse. As disc structure deteriorates so does disc function. Annular fibers progressively fail through fissuring. The degrading nucleus pulposus no longer transfers pressure across the disc’s center. Consequently, the annulus which is usually loaded in tension due to hoop stress, no longer fails mechanically in tension, because it becomes primarily loaded in compression. In healthy discs, the underlying end plate will fail due to compressive loading before the disc does, but this is no longer the case in the degenerated disc. The remaining nuclear material is pressed against the annulus, this causes bulging and ultimately a disc prolapse (herniation). Disc prolapses occur mostly in the cervical and lower lumbar spine. Bulging or herniated discs may compress nerve roots or the thecal sac. Gross structural disruption certainly appears to represent mechanical failure, but tissue composition is usually altered and it is not clear whether the disc structure is weakened by biochemical changes or whether those changes represent a response to mechanical failure. Further cadaveric experiments have attempted to link mechanical factors to disc degeneration by showing that disc prolapse and radial fissures can be simulated in apparently normal discs if the loading is sufficiently severe [17, 18]. Animal experiments show that biological degeneration always occurs after minimal structural damage was induced by means of a scalpel blade stab incision into the annulus [19, 20]. There is considerable interest in identifying biochemical and metabolic abnormal ities in degenerated disc tissues, but these abnormalities may be the consequences of disc failure rather than the cause.

2.4

MUSCLES

Along with ligaments, muscles initiate and guide spinal movements. Larger muscles are important in balancing external forces. Small segmental muscles, too small in their cross section to produce relevant force, are densely packed with muscle spindles and therefore believed to be important for proprioception of the spine. The spinal musculature may be divided, based on location, into six major groups: • Posterior spinal muscles (erector spinae, multifidus, lumbar part of the longissimus thoracis). • Anterior spinal muscles (psoas major, quadratus lumborum). • Small segmental muscles (interspinales, intertransversarii). • Respiratory or intercostal muscles. • Abdominal wall muscles (intertransversarii, internal and external oblique, rectus abdominis). • Superficial trunk muscles (broad muscles including the rhomboids, latissimus dorsi, pectoralis, trapezius, transversus abdominis). The greatest risk of injury occurs during maximum muscle tension. For example, when tripping and during the subsequent fall, a reflex triggers the forcibly and involuntarily lengthened erector spinae muscle to contract. Following deceleration of the forward movement, the erector spinae muscle will eventually accelerate the spine back into an extended posture. Muscle injury most likely occurs during such forcible lengthening while the muscle is maximally activated (ie, eccentric muscle contraction).

43

44

4

Biomechanics of the spine

2.5

LIGAMENTS

Spinal ligaments, like most soft tissues of the body, are viscoelastic in nature with nonlinear elastic responses. Ligaments connect between adjacent vertebrae (intersegmental), but individual fibers may well stretch over multiple levels. They transmit tensile loads only and, as a result, they specifically limit excessive motion. Intersegmental ligaments are: • • • • •

Anterior and posterior longitudinal ligaments Yellow ligament (ligamentum flavum) Interspinous and supraspinous ligaments Intertransverse ligaments Facet joint capsules

Spinal ligaments do not enjoy the same margins of fail safety as bones do, as they physiologically may operate under conditions relatively close to their failure strength. Ligaments mechanically fail at approximately 10 to 20 MPa of tensile stress (translates to approximately 180 N failure load for the posterior longitudinal and 340 N failure load for the anterior longitudinal ligament). For ultimate bending postures, strains in ligaments farthest from the axis of rotation can reach 20%. Hormonal concentrations can affect ligament laxity. For instance, pregnancy systemically increases ligament laxity.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

2.6

SPINAL CORD, NERVE ROOTS

The spinal cord runs down the foramen magnum of the skull base to the sacrum, protected in a bone delimited canal formed by the stacked neural arches and the vertebral body posterior walls. Paired nerve roots exit from between vertebrae through the intervertebral foramen. The dura further protects neural structures that are floating inside this sheet in cerebrospinal fluid. Vertebral fractures, disc ruptures, or bony impingements can all potentially affect neural performance, resulting in pain and/or paralysis.

4.1

General biomechanics of the spinal motion segment and the spinal organ

3

SPINAL MOTION

Kinematics is the study of the motion of bodies, regardless of the cause. Kinetics or dynamics, on the other hand, are concerned with the effects of forces on the motion of objects. Motion patterns of the entire spine are complex. Again, to simplify matters, motion is meaningfully and best described for isolated motion segments. Nonlinear motion behavior, as for most soft tissues, is characteristic for all regions of the spine. The spine also exhibits viscoelastic behavior, due to viscoelastic qualities of its tissue constituents. The hysteresis, expressed as deviate paths of the load-displacement curve for the forward and reverse directions, is a direct result of viscoelasticity.

Forces (expressed in Newton = N) lead to linear displacements; moments (expressed in Newton × meter = Nm) lead to angular displacements. Forces and moments are both called loads. The segmental flexibility for each degree of freedom can be characterized with a load-displacement relationship. Three parameters have been particularly effective in characterizing the typically nonlinear load-displacement relation ( Fig 4.1-7 ) of motion segments: neutral zone (NZ), elastic zone (EZ), and range of motion (ROM). For the biomechanical testing of implant performance, the three parameters are often compared between test groups.

Segmental flexibility

The spinal motion segment is, compared to other joints in the body, relatively unconstrained, exhibiting relevant motion in all six degrees of freedom (DOF). Motion can be adequately described by specifying the angular (rotational) and linear (translational) relative displacements for three orthogonal axes ( Table 4.1-2 ). Rotational or angular

Translational or linear

displacement description

displacement description

Plane

Rotation in plane

A xis out of plane

Translation along axis

Sagit tal

Flexion/ ex tension

Transverse

Lef t/right lateral shear

Coronal

Lef t/right side bending

Frontal

Anterior/ posterior shear

Transverse

Lef t/right axial rotation

A xial

Compression/ distrac tion

Table 4.1-2 Nomenclature for angular and linear displacements for the spine.

Fig 4.1-7 Schematic sample for typical segmental load-displacement curve. The forward and reverse paths show a hysteresis. Neutral zone (NZ), elastic zone (EZ), and range of motion (ROM) are indicated with double arrows.

45

46

4

Biomechanics of the spine

If a load is applied to a motion segment (or similarly to the entire spine), the segment quickly displaces from a neutral position to a position where an appreciable resistance is fi rst encountered. This initial displacement is called NZ, comparable to the “toe-in region” generally seen in soft-tissue elastic responses. Within the NZ, movements of a few degrees or millimeters are observed with practically no muscular effort. An extended NZ is indicative of abnormal joint laxity, also known as clinical instability. When moving beyond the NZ, stiffening of the motion segment is encountered. Loads and resulting displacements are largely proportional. This region is called EZ, a stiffness value is given by the slope of the curve. Eventually, respecting physiological magnitudes of load, a maximum displacement is reached. The span covered between minimum and maximum loads is called ROM.

For a healthy motion segment, the centrode path, even for full ROM segmental movements in the principal directions, is confi ned to a relatively small area usually overlaying the inferior vertebral body. In the degenerated or unstable spine [21], the centrode area can enlarge dramatically ( Fig 4.1-8 ). Despite

Kinematic description of multidirectional motion

Motion rarely involves a single degree of freedom only. Typically, the spine exhibits complex motions when external loads change. Complex motions are simultaneous displacements in multiple degrees of freedom (translation and rotation). Complex motions, however, are still not unpredictable, and are sometimes referred to as a motion pattern. It describes a typical “motion path” a vertebral body follows under external load changes (eg, a person sitting up from a chair or climbing stairs). The instantaneous center of rotation (ICR) is an easy kinematic notation useful in describing complex motion. Any motion in a plane (eg, in the sagittal plane for flexion/extension) can always be expressed as an angular displacement around a center point. This point, denoted for some instant in time (ie, for an infi nitesimally small displacement) is called the ICR. All ICR points connected for a larger movement form the centrode path or area.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

a

b

Fig 4.1-8 Instantaneous center of rotation paths (centrode areas) for a specimen of the lumbar spine in lateral (a) and anteroposterior projection (b). The red area represents a healthy motion segment. The blue area depicts the much larger centrode for a degenerative motion segment. Larger centrode areas are indicative of segmental instability.

4.1

General biomechanics of the spinal motion segment and the spinal organ

the ICR and centrode path being a very effective criterion in identifying abnormal segmental motion, clinicians so far have not adopted it. The wide error margin associated with tracking skeletal landmarks in subsequent radiographic views seriously impedes the reconstruction of the centrode path with reasonable fidelity.

Segmental instability

Segmental instability has been described in many different ways. White and Panjabi [24] described it as “the loss of the ability of the spine under physiological loads to maintain its pattern of displacement so that there is no initial or additional neurological deficit, no major deformity, and no incapacitating pain”.

More complex kinematic descriptions to numerically analyze motion of a rigid body in 3-D space, either use Cardanian or Eulerian sequences, direction cosign matrices, or helical axis descriptions. Complex 3-D kinematic descriptions are mostly used for body animation or 3-D visualization of joint kinematics. Coupled motion

Coupling refers to motion about or along axes secondary to those of the axis of applied load ( Fig 4.1-9 ) [22]. For example, in the middle and lower cervical spine, left lateral bending produces a concomitant left axial rotation due to the orientation of the articulating surfaces of the facets. Coupling in the lumbar spine is more complex. In the normal spine left lateral bending causes right axial rotation in the upper lumbar segments, but it causes left axial rotation in the lumbosacral joint. The L4/5 segment constitutes a transitional level [23]. In kinematic measurements conducted by the author for asymptomatic individuals walking on a treadmill, most subjects had axial rotation and lateral bending movements in the same direction, but about one third of the subjects were in opposite directions ( Fig 4.1-10 ) [22]. Seemingly, coupled motions have considerable variability between subjects, but they may perhaps be altered as a result of structural changes leading to increased or decreased segmental laxity, or causing segmental instability. It is also possible that coupled motion patterns may be due to specific contractile patterns of intrinsic and extrinsic paraspinal muscle groups.

Fig 4.1-9 L3/4 segmental kinematics analyzed for three cycles of voluntary lateral bending (full range) recorded in a healthy subject. Invasive measurements were obtained with a 3-D electromagnetic motion tracking system affixed to the spinous processes [22]. Solid lines represent angular displacement, dashed lines linear displacement. Coupled motion can be seen in axial rotation—and to a lesser degree in the other motion directions.

47

48

4

Biomechanics of the spine

Spinal stability refers to a state in which there is adequate control and support between two adjacent vertebral segments in the spine. This is accomplished by both, passive structures (ie, disc, ligaments, joint capsules) and active structures (ie, muscles). When there is a breakdown in passive structures, usually resulting in pain, it is essential that the muscles ( active structures) are trained to compensate. This can best

Fig 4.1-10 Segmental kinematics for the L4/5 motion segment were recorded while walking on a treadmill at 2.5 mph. An invasive tracking method was used [22]. The angular displacement curves normalized for a single gait cycle are overlaid for 23 asymptomatic subjects measured in this study. The left curves display lateral

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

be accomplished with abdominal muscle training, spinal stabilization exercises administered by a physiotherapist, or by similar means. Segmental instability might be caused by congenital anomalies such as spondylolisthesis/-lysis, but it also could be the result of an acquired disease, such as the early to mid stages of

bending, which was fairly uniform across the measured population. Interesting is that the coupling in axial rotation (right curves) was determined to be toward the contralateral direction for 16 subjects, and toward the ipsilateral direction for 7 subjects. This disperse result illustrates intersubject variability for coupling patterns.

4.1

49

General biomechanics of the spinal motion segment and the spinal organ

degeneration ( Fig 4.1-11), neoplasm, or infection with destruction of passive structures. Instability can also be a complication of spinal trauma or it can be iatrogenic resulting from spine surgery. Any surgery cutting or removing passive structures (ie, facetectomy, laminectomy, discectomy) can potentially lead to segmental instability. To diagnose segmental instability clinicians early on used increased range of motion manifestations such as for anteroposterior shear [25] or flexion/extension. But, segmental instability cannot be comprehensively described when using static end range of motion x-rays (ironically called “dynamic

x-rays”). Losses in segmental stiffness [26] as well as changes in coupled motion patterns were equally described [27] as clinical manifestations of segmental instability. Segmental instability is probably best identified under increased axial loading, provoking abnormal segmental motion [28]. Sudden changes in load may result in unanticipated or unpredictable intersegmental motion. Even during continual and wide trunk movements, particularly at low ligament pretension, abrupt segmental motion might occur. The manifestation of instability is more likely to be observed in the mid ranges of spinal motion, and can have a wide range of

Fig 4.1-11 Transversal disc map showing internal disc displacements during flexion-compression measured for the posterocentral (PC), posterolateral (PL), centrolateral (CL), anterolateral (AL), and anterocentral (AC) annular regions, as well as for the central nuclear region. The internal displacements were obtained from fine wire markers inserted into the disc, followed with sequential high resolution xrays while loading the specimen [3]. Discs with a Thompson degeneration grade I and II (line arrow) are compared with discs featuring a grade III or IV (solid arrow). The larger nuclear displacement during loading is evident for the more progressed degeneration stages. It is also indicative of segmental instability.

50

4

Biomechanics of the spine

manifestations ( Fig 4.1-12). Kaigle et al [29] have demonstrated midrange segmental instability motion in a porcine instability model as well as in symptomatic patients. A specific manifestation of instability is due to axial load. A linked and tall structure such as the ligamentous spine is inherently unstable and may buckle already at very low vertical loads [30]. Instability or buckling occurs when a displacement perturbation from an equilibrium position results in a force tending to increase the displacement. But, the spine cannot be analyzed as an independent structure. In vivo, the spine does not collapse easily because the trunk (ribs, connective tissue, passive musculature) stiffens the ligamentous spine already by about a factor of ten. Dynamic muscle actions stabilize the spine further, which is particularly important for the cervical spine. Muscle activation alone can increase cervical spine stiffness by a factor of five.

Fig 4.1-12 A mechanical model for segmental instability is presented. The drawing on the left shows a bowl with a rolling ball placed inside, coming automatically to rest at its lowest position. When slightly tilting the bowl, the ball will find a new position. Clearly defined ball positions that are reached after a relatively small excursion depict a stable situation. Undetermined positions and relatively large excursions of the ball are typical for instability. From top to bottom drawings, instability is increasing.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

4.1

General biomechanics of the spinal motion segment and the spinal organ

4

BIBLIOGRAPHY

Steffen T, Tsantrizos A, Aebi M (2000) Effect of implant design and endplate preparation on the compressive strength of interbody fusion constructs. Spine ; 25(9):1077–1084. 2. Noren R, Trafimow J, Andersson GB, et al (1991) The role of facet joint tropism and facet angle in disc degeneration. Spine ; 16(5):530–532. 3. Tsantrizos A, Ito K, Aebi M, et al (2005) Internal strains in healthy and degenerated lumbar intervertebral discs. Spine ; 30(19):2129–2137. 4. Urban JP, McMullin JF (1985) Swelling pressure of the inervertebral disc: infl uence of proteoglycan and collagen contents. Biorheology ; 22(2):145–157. 5. Nachemson A (1975) Towards a better understanding of low-back pain: a review of the mechanics of the lumbar disc. Rheumatol Rehabil ; 14(3):129–143. 6. Wilke HJ, Neef P, Caimi M, et al (1999) New in vivo measurements of pressures in the intervertebral disc in daily life. Spine ; 24(8):755–762. 7. McNally DS, Adams MA (1992) Internal intervertebral disc mechanics as revealed by stress profi lometry. Spine ; 17(1):66–73. 8. McNally DS, Adams MA, Goodship AE (1993) Can intervertebral disc prolapse be predicted by disc mechanics? Spine ; 18(11):1525–1530. 9. Steffen T, Baramki HG, Rubin R, et al (1998) Lumbar intradiscal pressure measured in the anterior and posterolateral annular regions during asymmetrical loading. Clin Biomech (Bristol, Avon) ; 13(7):495–505. 10. Ohshima H, Tsuji H, Hirano N, et al (1989) Water diffusion pathway, swelling pressure, and biomechanical properties of the intervertebral disc during compression load. Spine ; 14(11):1234–1244. 11. Ogata K, Whiteside LA (1981) 1980 Volvo award winner in basic science. Nutritional pathways of the intervertebral disc. An experimental study using hydrogen washout technique. Spine ; 6(3):211–216. 1.

12. Urban JP, Holm S, Maroudas A, et al (1977) Nutrition of the intervertebral disk. An in vivo study of solute transport. Clin Orthop Relat Res ; (129):101–114. 13. Broberg KB (1993) Slow deformation of intervertebral discs. J Biomech ; 26(4–5):501–512. 14. Urban JP, Maroudas A (1981) Swelling of the intervertebral disc in vitro. Connect Tissue Res ; 9(1):1–10. 15. Adams MA, McMillan DW, Green TP, et al (1996) Sustained loading generates stress concentrations in lumbar intervertebral discs. Spine ; 21(4):434–438. 16. Brinckmann P, Frobin W, Hierholzer E, et al (1983) Deformation of the vertebral end-plate under axial loading of the spine. Spine ; 8(8):851–856. 17. Adams MA, Hutton WC (1982) Prolapsed intervertebral disc. A hyperfl exion injury 1981 Volvo Award in Basic Science. Spine ; 7(3):184–191. 18. Adams MA, Hutton WC (1985) Gradual disc prolapse. Spine ; 10(6):524–531. 19. Osti OL, Vernon-Roberts B, Fraser RD (1990) 1990 Volvo Award in experimental studies. Anulus tears and intervertebral disc degeneration. An experimental study using an animal model. Spine ; 15(8):762–767. 20. Fazzalari NL, Costi JJ, Hearn TC, et al (2001) Mechanical and pathologic consequences of induced concentric anular tears in an ovine model. Spine ; 26(23):2575–2581. 21. Seligman JV, Gertzbein SD, Tile M, et al (1984) Computer analysis of spinal segment motion in degenerative disc disease with and without axial loading. Spine ; 9(6):566–573. 22. Steffen T, Rubin RK, Baramki HG, et al (1997) A new technique for measuring lumbar segmental motion in vivo. Method, accuracy, and preliminary results. Spine ; 22(2):156–166. 23. Pearcy MJ, Tibrewal SB (1984) Axial rotation and lateral bending in the normal lumbar spine measured by three-dimensional radiography. Spine ; 9(6):582–587.

51

24. Panjabi MM, Thibodeau LL, Crisco JJ 3rd, et al (1988) What constitutes spinal instability? Clin Neurosurg ; 34:313–339. 25. Dupuis PR, Yong-Hing K, Cassidy JD, et al (1985) Radiologic diagnosis of degenerative lumbar spinal instability. Spine ; 10(3):262–276. 26. Pope MH, Panjabi M (1985) Biomechanical defi nitions of spinal instability. Spine ; 10(3):255–256. 27. Pearcy MJ (1985) Stereo radiography of lumbar spine motion. Acta Orthop Scand Suppl ; 212:1–45. 28. Friberg O (1987) Lumbar instability: a dynamic approach by traction-compression radiography. Spine ; 12(2):119–129. 29. Kaigle AM, Holm SH, Hansson TH (1997) 1997 Volvo Award winner in biomechanical studies. Kinematic behavior of the porcine lumbar spine: a chronic lesion model. Spine ; 22(24):2796–2806. 30. Crisco JJ 3rd, Panjabi MM (1991) The intersegmental and multisegmental muscles of the lumbar spine. A biomechanical model comparing lateral stabilizing potential. Spine ; 16(7):793–799.

52

4 4.2

BIOMECHANICS OF THE SPINE BIOMECHANICS OF SPINAL STABILIZATION

1

Introduction …………………………………………………………………………………………………… 53

2

Posterior stabilization ……………………………………………………………………………………… 54

3

Load-sharing characteristics of stabilized spinal segments …………………………………………… 56

4

Anterior stabilization ………………………………………………………………………………………… 58

5

Intervertebral cages ………………………………………………………………………………………… 60

6

Adjacent segment effects …………………………………………………………………………………… 62

7

Dynamic stabilization ……………………………………………………………………………………… 62

8

Interspinous process distraction

9

Arthroplasty of the spine …………………………………………………………………………………… 64

10

Summary ……………………………………………………………………………………………………… 66

11

Bibliography …………………………………………………………………………………………………… 66

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

………………………………………………………………………… 63

53

Stephen J Ferguson, Thomas Steffen

4

BIOMECHANICS OF THE SPINE

4.2

1

BIOMECHANICS OF SPINAL STABILIZATION

INTRODUCTION

The following chapter is intended to provide a brief overview of current biomechanical concepts in spinal stabilization. It would not be possible in such a short review to cover the myriad of spinal stabilization devices and techniques available to the surgeon today. Indeed, whole volumes have been written about clinical instability of the spine and the biomechanics of spinal stabilization [1, 2], and the reader is encouraged to explore these resources, not only for a more in-depth treatment of the subject, but also for the historical perspective that can be gained into the rapid development of the field over the last two decades. Each surgical procedure for spinal stabilization has its own unique structural and biomechanical characteristics. It is important that one chooses the appropriate implant and technique, by understanding the specific nature of each case.

With the exception of recent developments in spinal arthroplasty, spinal stabilization is a means to achieve the end goal of solid bony fusion. The goals of conventional spinal arthrodesis are: • To support the spine when its structural integrity has been severely compromised (to reestablish clinical stability). • To maintain correction following mechanical straightening of the spine (scoliosis, kyphosis, osteotomy). • To prevent progression of deformity (scoliosis, kyphosis, spondylolisthesis). • To alleviate or eliminate pain by stiffening a region of the spine (diminishing movement between various spinal segments).

54

4

Biomechanics of the spine

2

To choose a stabilization method which will best achieve these goals, the surgeon requires an understanding of the mechanics of load transfer through the spine, how this load transfer is altered by the specific injury or pathology, and the relative merits of each particular surgical technique. There is no single solution for spinal stabilization, and in the following sections, the biomechanical aspects of some of the more common techniques will be discussed. These include: posterior stabilization, anterior stabilization, and intervertebral cages, as well as newer concepts in nonrigid stabilization.

POSTERIOR STABILIZATION

Pedicle screws have dramatically improved the outcomes of spinal fusion. Short segment surgical treatment using pedicle screws and rigid connecting plates or rods has proven to be safe and effective for the treatment of neoplastic, developmental, congenital, traumatic, and degenerative conditions [3]. The stabilizing potential of posterior spinal fi xation systems has been demonstrated in many biomechanical studies. For example, a comparison of the internal fi xator and the USS [4] has shown that motion of the stabilized spinal segment is reduced by up to 85% in flexion, 52% in extension, 81% in lateral bending, and 51% in axial rotation. Additional stability can be achieved by adding cross-links [4]. Similar results have been reported in other studies of the stabilizing potential of different posterior fi xation systems [5, 6]. Posterior systems derive their stability from a solid anchorage in the pedicle and the inherent rigidity of the connecting instrumentation. The pullout strength of pedicle screws is directly related to the bone density [7]. It is possible to achieve an increase in pullout strength by choosing convergent screw trajectories ( Fig 4.2-1) and by the addition of cross-links. Furthermore, it has been shown that with parallel pedicle screws in short-segment constructs, an unstable “four-bar” mechanism can result in the absence of adequate anterior column support ( F i g 4 . 2-2 ) ; therefore, triangulation of pedicle screws is recommended for better stability. The same rationale applies for cross-linking the rods. Here, diagonal cross-linking is preferable to the horizontal configuration in terms of rotational stability [8]. The stiffness of the fi xator construct depends heavily on the diameter of the connecting rods. Compared with a system using 7 mm rods, a 10 mm rod has a 4.1 times higher bending stiffness, while a 3 mm rod has a nearly 30 times lower bending stiffness [9]. Should these systems be made as rigid as possible? An increase in rod

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

4.2

55

Biomechanics of spinal stabilization

diameter provides a more stable construct, but at the same time it produces higher internal loads in the implant, on the clamping device, and on the pedicle screws, and thus a higher risk of screw breakage [9]. Therefore, a compromise between an absolutely rigid fi xation and a minimal risk of implant failure must be achieved, and this compromise is reflected in current implant designs which provide stable fi xation [4] with a proven service life. While pedicle screws have been accepted as a reliable and safe method for stabilizing the thoracolumbar spine, their use in the mid and upper thoracic spine is more complicated, due to the smaller overall dimensions of the thoracic pedicles, and greater variation in pedicle morphology. An alternative to standard intrapedicular screw placement is the extrapedicular screw trajectory ( Fig 4.2-3 ), fi rst described by Dvorak et al [10].

Greater pullout strength has been measured for extrapedicular screws, likely due to the greater angulation possible with this technique, the longer screw length, and the perforation of up to four strong cortices. The overall 3-D stability of thoracic spinal segments stabilized with extrapedicular screws has been shown to be equivalent to that of the conventional intrapedicular technique, and with no additional risk of loosening through fatigue [11]. Thus, this technique offers an alternative, which provides greater safety due to the increased distance between the screws and the spinal canal, while avoiding any compromise in the rigidity and strength of the construction. The use of simple lamina hooks in the thoracic spine is safe with respect to damage of neural structures. However, hook disengagement has been reported in scoliosis correction surgery [12]. To achieve a higher resistance to the complex 3-D

a

a

b

Fig 4.2-1a–b The use of convergent screw trajectories (right) increases the pullout strength and overall stability of pedicle screw constructs, in comparison with parallel screw insertion (a).

b

c

Fig 4.2-2a–c a–b The use of conventional parallel pedicle screws and rods (a) for spinal segments with diminished anterior integrity may be inadequate. Displacement of the stabilized segment by rotation of the pedicle screws—a so-called “four-bar” mechanism—may result (b). c Further stability can be achieved by the use of convergent screw trajectories or the addition of a cross-link.

56

4

Biomechanics of the spine

3

forces, pedicle hooks with additional supporting screws have been developed [13, 14]. Biomechanical pullout tests have shown that a significant increase of failure load can be achieved with the use of screw-augmented hooks [15].

a

b Fig 4.2-3 In contrast to the standard intrapedicular screw insertion (a), an extrapedicular screw insertion (b) allows a greater margin of safety with respect to the spinal canal in the thoracic spine, and may offer greater pullout strength and stability.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

LOAD-SHARING CHARACTERISTICS OF STABILIZED SPINAL SEGMENTS

Spinal implant constructs and the stabilized spinal segment together form a mechanical system. Loads and moments are shared between the natural anatomy and the stabilizing hardware. Recent in vivo measurements, using a telemetric device, have provided valuable insight into the 3-D loading of an internal (posterior) fi xator during daily physiological loading [16]. However, measurements of fi xator loads alone provide no information about the overall force flow, ie, how much of the total load is transferred by the implant and how much by the spine. Cripton et al [5], in a combined experimental and analytical study, have quantified the load sharing in stabilized spinal segments. By simultaneously measuring intradiscal pressure and the forces in a modified internal fi xator during physiological loading, analysis of the load distribution within the instrumented spinal construct was possible by applying principles of force and moment equilibrium. The results of this study provide valuable insight for the design of spinal implants, and also for the evaluation of surgical indications. In the intact stabilized spine, it has been demonstrated that, for flexion and extension, spinal loads are carried predominantly by equal and opposite forces in the disc and the fi xator, a force couple. Only a small portion of the total loading is transferred directly by bending of the implant or through the posterior elements. For side bending, the majority of loading is transferred through equal and opposite forces in the fi xator rods. For torsional loading, loading is distributed more or less evenly between implant forces, torsional resistance of the disc, and forces acting on the posterior elements. Therefore, the anterior structures play a crucial role in the overall loadbearing function of the stabilized spine. The load-sharing in instrumented spinal segments is summarized in Fig 4.2-4 .

4.2

57

Biomechanics of spinal stabilization

In the case of severe anterior column injury, all loads must be carried by the implant itself. Based on the in vivo measurements of implant loading by Rohlmann et al, and the force flow analysis in the study of Cripton et al, global moments of up to 30 Nm may act through the spine [5]. This may exceed the safe limit for many implants. Therefore, in the case of very unstable anterior column injuries, additional support of the anterior column is critical to prevent failure of the instrumentation. The importance of effective load sharing between the anterior and posterior spinal columns is further reinforced by

the work of Polly et al [17], in which it was shown that the overall stiffness of the stabilized spine increases by a factor of three, as an interbody graft is moved within the disc space from the posterior toward a more mechanically advantageous anterior position. Further work is required to characterize the force flow through instrumented spinal segments, including the load transfer through intervertebral devices and anterior stabilization constructs.

Fig 4.2-4 Predicted load sharing between a standard posterior stabilizing implant and the anatomical structures of the spine. The integrity of the anterior column is crucial for successful load bearing in the spine, even with rigid metallic implants. Adapted from Cripton et al [13].

58

4

4

Biomechanics of the spine

ANTERIOR STABILIZATION

The importance of an intact anterior column for the loadbearing function of the spine has been demonstrated. In certain cases, spinal stabilization can be effectively achieved using only anterior implants. Anterior cervical plating offers several advantages for stabilizing spinal constructs, including good visibility of the fusion site, decreased rate of graft expulsion, and increased fusion rate in multilevel constructs. Anterior cervical plates act as a tension band during spinal extension and as a buttress plate during flexion. Constrained cervical systems have a rigid, angle-stable connection between the plate and screws (eg, cervical spine locking plate, CSLP), whereas unconstrained systems rely on friction generated by compression of the plate against the anterior cortex for stability (eg, H-plate) ( Fig 4.2-5 ). In extended biomechanical testing, constrained systems have shown a greater rigidity, whereas unconstrained plates can lose a significant amount of stability over time [18]. However, it has also been shown that the capability of the CSLP to stabilize the spine after a threelevel corpectomy is significantly reduced after fatigue loading [19], whereas no difference in stability was noted for a single-level stabilization. Therefore, the demands of the surgical indication heavily influence the performance of the implant. The surgeon has the option of selecting systems with monocortical or bicortical screw fi xation, often with the same plate. In general, no significant differences in stability have been found between monocortical and bicortical fi xation [20], however, further improvements in stabilization have been shown using monocortical locking expansion screws [21]. Bicortical screw fi xation still has specific indications, eg, for multilevel stabilization, poor bone quality, or realignment of kyphotic deformities, but it also has the potential to abut on the spinal cord. Another concern in the cervical spine, with its inherent mobility and relatively low compressive forces, is delayed or nonunion (pseudarthrosis) due to possible

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

stress shielding of the graft. This is particularly true for the latest generation of constrained (locking) plates with which it is more difficult to set the graft under compression. For this reason dynamic (semiconstrained) anterior plates may be preferable. Reidy et al have shown in a cadaver corpectomy model that axial load is transmitted preferentially to the graft with the dynamic cervical plate, in comparison to a static plate, especially when the graft is undersized [22]. Several systems have been developed for anterior stabilization of the thoracolumbar spine, including the Ventrofi x and the Kaneda SR, used mostly for reconstruction in trauma, tumor, and posttraumatic kyphosis. The advantages of anterior fi xation are better decompression of the spinal canal and reconstruction of the anterior column, combined with excellent visibility.

Fig 4.2-5 Constrained cervical fixation systems rely on an angle-stable connection between the plate and screws for the efficient transfer of load; monocortical screw fixation is possible. Conventional plates rely on friction to transfer load, and usually require bicortical screw insertion.

4.2

59

Biomechanics of spinal stabilization

Anterior stabilization devices transfer load through a combination of compressive or tensile loading along the length of the implant and bending or torsion of the implant. Due to the low profi le of these implants, and their position directly on the anterior column, bending forces are much lower than for posterior pedicle screw systems, but their stabilizing potential is also lower, due to a shorter effective lever arm. The relative effectiveness of anterior, posterior, and combined anteroposterior fi xation in a corpectomy model has been addressed in a study by Wilke et al [23]. It was shown that typical posterior pedicle screw systems provide excellent stability in flexion and lateral bending, but not in extension or axial rotation; however, this stability is dependent on adequate support of the anterior column, with an interbody graft if necessary. In extension, motion is restricted only by the stiffness of the posterior implant itself. Likewise, anterior fi xation provides stabilization in flexion and lateral bending, but not in extension or axial rotation. In lateral bending, the implants provide better stabilization when the spine is bending away from the implant side, as the devices act as a tension band. Anterior double-rod systems provide better stabilization than single-rod systems, and systems which use locking head screws are stiffer than those without ( Fig 4.2-6 ). The addition of a transverse element further increases the stability of a double-rod construction. In all loading directions, combined anteroposterior fi xation provides better stabilization than posterior or anterior stabilization alone. Therefore, in cases of severe destruction of the vertebral body or gross fracture dislocation, combined anteroposterior fi xation would be warranted.

Fig 4.2-6 Anterior double-rod fixation systems provide increased resistance to torsional loading (eg, Ventrofix). The addition of transverse elements or locking head screws further increases the stiffness of such implants. In cases of severe destruction of the vertebral body or gross fracture dislocation, combined anteroposterior fixation would be warranted.

60

4

5

Biomechanics of the spine

INTERVERTEBRAL CAGES

Intervertebral cages have been developed to augment spinal arthrodesis by reconstructing the anterior column to restore the height of the intervertebral space, thereby stabilizing the affected segment and providing containment for cancellous bone graft. A variety of cage designs are available for insertion using an anterior or posterior approach [24, 25]. These include: bilateral threaded, cylindrical implants (eg, BAK, Ray TFC), bilateral box-shaped implants (eg, contact fusion cage), and single, open-box or ring-shaped implants (eg, Syncage). Intervertebral cages were originally proposed as stand-alone devices for anterior lumbar interbody fusion (ALIF) or posterior lumbar interbody fusion (PLIF). The clinical success of stand-alone intervertebral cages is beyond the scope of this brief review; however, the mechanical requirements for successful stand-alone devices are substantial. Axial compressive loads in the spine range from 400 N to more than 7000 N during heavy lifting [2]. Intervertebral cages must be strong enough to bear these loads without failure of the implant itself, however, the bone graft around and within the cage must be stressed and strained sufficiently to evoke the biological signals (release of cytokines) for bone formation [26, 27]. In this context it is proposed that extensive stressshielding may lead to delayed union or nonunion. This confl ict is reflected in most current cage geometries and materials, but further work is required to fully understand the underlying mechanobiology [28]. These devices must also resist penetration or subsidence into the underlying cancellous bone of the vertebral body. The subchondral bone of the vertebral end plate provides the necessary strength for cage support. Removal of the end plate to provide a bleeding cancellous bone bed may compromise this support, especially for devices with a limited contact area, as the resistance to implant subsidence then depends on the quality of underlying trabecular bone [29]. However, the strength of the end plate has been shown to be

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

greatest at its periphery in the posterolateral corners [30–32], and removal of only the central end plate only does not compromise the strength of the cage-bone interface, especially for implants with a large, peripheral contact area, such as the Syncage [33]. Thus, an effective compromise between the biological and biomechanical requirements for fusion can be achieved. The 3-D stabilizing potential of anterior stand-alone cages has been critically evaluated in several biomechanical studies [24, 34]. While most ALIF cage designs improve stability of the instrumented spinal segment in flexion and lateral bending, the stability in extension and axial rotation may not be adequate [34]. Comparison of anterior implantation and lateral implantation has shown that resection of the anterior annulus is not responsible for this lack of stability [35], which has led to the conclusion that the lack of rigidity may be associated with distraction of the facet joints during cage insertion ( Fig 4.2-7 ). Although this contradicts the original concept of “distraction compression” by Bagby [36], whereby the distracted annulus imparts a compressive force on the interbody cage, stabilization of the intervertebral space due to pretensioning of the annular fibers is likely only temporary, due to the viscoelastic nature of this tissue [37]. More significantly, biomechanical testing of PLIF devices has shown that, as a stand-alone device, cages inserted from a posterior approach do not provide adequate stabilization [25, 38]. Instability in all three principal motion directions has been demonstrated for PLIF devices with varying designs, most likely as a result of the necessary destruction of the facet joints and posterior annulus [25]. In this case, box-sectioned cages may be preferable in order to achieve maximum height for distraction, while minimizing the width of the required approach.

4.2

61

Biomechanics of spinal stabilization

The use of additional posterior instrumentation improves the stability of ALIF devices, and would appear to be critical in ensuring success with PLIF devices. Supplemental translaminar or pedicle screw fi xation significantly reduces the critical extension and axial rotation motion of spinal segments stabilized with intervertebral cages [25, 34]. The use of translaminar screws may be preferred, as these can be placed in a minimally invasive fashion. Without posterior stabilization, instability in extension could lead to nonunion, loosening or migration of the cage. A potential alternative for the above-mentioned combined instrumentation is the recent development of a novel “standalone” device which merges the principle of the interbody cage with the anterior tension band instrumentation (Synfi x). Cain et al have compared the stabilizing properties of this screw-cage construct with conventional 360° instrumentation using cage and pedicle screws or translaminar screws. Motion a nalysis

a

b

demonstrated a significant increase of segmental stiffness with the Synfi x compared to cage/translaminar screw instrumentation in flexion/extension and rotation [39]. However, testing was nondestructive and included only a few cycles. For a defi nite judgment the comparative biomechanical behavior under repetitive loading (fatigue) as well as clinical results and fusion rates have to be evaluated. In the cervical spine, in contrast to the lumbar spine, standalone interbody cages (or structural bone grafts) are used routinely after one-level discectomy and have demonstrated near 100% fusion rates. In a comparative biomechanical in vitro study, cervical segmental stability has been assessed after implantation of interbody cages and structural bone grafts. After single-level discectomy, physiological segmental stability could be reestablished with both techniques, with the cage tending to result in slightly higher stiffness [40].

c

Fig 4.2-7a–c The use of intervertebral cages as a stand-alone implant for spinal fusion may be limited by the poor stabilization in extension. Extension is limited in the normal spine partly by the interaction of the facet joints (a). Following the insertion of a stand-alone cage, the facet joints are distracted (b) and the spinal segment is more mobile (c).

62

4

6

Biomechanics of the spine

ADJACENT SEGMENT EFFECTS

Due to the proven rigidity of current spinal internal fi xators, it has often been suggested that degeneration of the disc adjacent to a fused spinal segment is the result of increased biomechanical stress on this motion segment. Shono et al [41] have shown in a biomechanical in vitro study that the displacement of the adjacent motion segment is increased following fusion. In these experiments, a fi xed displacement was applied to the entire spinal specimen, and it is therefore logical that, as the motion of the fused segment decreases due to its increased stiffness, motion at the adjacent segment must increase to produce the total displacement. On the other hand, Rohlmann et al [9] have demonstrated with a simplified analytical model that the influence of rigid instrumentation on the adjacent discs is minor. In their analysis, controlled loads were applied to a spinal model. This seemingly contradictory result may also be reasonable, as the response of the mobile disc to a given load is determined only by its own inherent stiffness, which is not altered by the adjacent fusion. Nevertheless, small but significant increases in adjacent segment mobility have been shown in vitro when controlled loads were applied to spinal segments [42]. Is “adjacent segment disease”, therefore, the result of altered biomechanical stresses? This depends on whether adjacent segment motion in vivo is increased following fusion. The animal study of Dekutowski et al [43] provides some support for increased adjacent segment motion, however, the overall incidence of adjacent segment degeneration would likely be much higher if its cause were purely mechanical. It is well accepted that disc degeneration is a multifactorial disease with genetic and environmental factors [44]. To which extent mechanical factors contribute to the disease likely also determines whether or not disc degeneration is initiated or aggravated adjacent to a fused segment.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

7

DYNAMIC STABILIZATION

Nonrigid posterior stabilization of the spine is a relatively new concept for the treatment of spinal pathologies. Ligamentoplasty was introduced in 1992 by Graf. This posterior dynamic stabilization system consists of pedicle screws connected via elastic polymer elements [45]. The underlying philosophy is to maintain physiological lordosis while restricting flexion/ extension motion to unload and “protect” the respective disc. In vitro studies have demonstrated that flexibility is reduced in all principal directions with the Graf ligamentoplasty [46], however, the clinical success of this device has been controversial [47, 48]. Currently, the most advanced and frequently used device is the dynamic neutralization system (Dynesys) for the spine. Dynesys is a nonfusion pedicle screw system composed of titanium pedicle screws joined by polycarbonate urethane spacers containing pretensioned polyester cords. With such a system, affected segments can be restored to their proper anatomical position, and motion in all planes can be effectively controlled. However, by their design motion is not absolutely prevented, in contrast to solid fusion implants. It has been shown, in a cadaveric model of the destabilized spine, that Dynesys is able to improve stability in all principal anatomical directions, however, axial rotation was poorly controlled while in flexion and bending the system is potentially as stiff as a conventional internal fi xator [49]. Freudiger et al [50] have demonstrated that the Dynesys limits shear translation of unstable spinal segments under much higher levels of physiological loading, and reduces bulging of the posterior annulus, which may relieve pain. Due to the compliance of the construction, overloading of adjacent segments may be prevented. Furthermore, as the development of these devices

4.2

63

Biomechanics of spinal stabilization

8

continues, the compliance of the connecting elements may be optimized to partially restore the normal kinematics of the stabilized segment. However, the efficacy of such a system depends heavily on the condition of the anterior column. Furthermore, the long-term performance of such a device may be limited by material fatigue or screw loosening, as the instrumentation must continue to bear load throughout the whole life of the patient.

INTERSPINOUS PROCESS DISTRACTION

The principle of implanting a spacer between adjacent spinous processes was already used by Knowles in the late 1950s to unload the posterior annulus in patients with disc herniation, thereby achieving pain relief [51]. In recent years various systems have entered the market, such as the Interspinous “U”, the Diam, the Wallis, and the X-Stop. All devices aim to limit motion in extension. Biomechanical testing has shown that extension motion is diminished while flexion, axial rotation, and lateral bending are maintained [52]. Restricting extension is thought to reduce narrowing of the spinal canal and buckling of the yellow ligament [53]. Furthermore, an unloading of the facet joint has been demonstrated in an in vitro cadaver study [54]. The resulting increase of segmental kyphosis is likely compensated by the adjacent segments, and how this may affect the sagittal profi le and balance in the long term needs to be evaluated in the future. However, although there is limited clinical follow-up data available, for patients with spinal stenosis which improves in flexion, the interspinous device is a feasible option which causes limited trauma with implantation.

64

4

9

Biomechanics of the spine

ARTHROPLASTY OF THE SPINE

Functional disc replacement is a logical progression in the treatment of degenerative disorders of the intervertebral joint. Arthroplasty in the spine has several possible advantages: preservation of function, decrease of adjacent level degeneration, and no requirement for the harvesting of bone graft. An excellent historical review of arthroplasty of the spine by Szpalski et al [55] highlights the many design concepts to date. Such a device must not only possess adequate strength to withstand the considerable compressive and shear loads transmitted through the spinal column, but must also respect the complex kinematics of intervertebral motion. The evolution of total joint prostheses in diarthrodial joints has been toward devices which emulate physiological motion patterns. Mobile bearing knee prostheses, for example, employ a large, conforming polyethylene plate, which is not fi xed to the tibia as in the conventional total knee joint, but rather moves on the surface of a highly polished metallic tray which is affi xed to the tibia. Theoretically, this design should allow a more natural motion pattern and a larger range of movement.

Due to its conformity throughout the full range of motion, stresses transmitted through the polyethylene and into the bone should be lower, reducing polymer wear and prosthesis loosening. A similar design philosophy is apparent in many current disc prostheses. Motion of the natural intervertebral joint cannot be compared to a simple ball-and-socket joint. The major motions of an intervertebral segment in flexion and extension are a combination of sagittal rotation plus translation. The center of rotation constantly changes throughout the full range of motion ( Fig 4.2-8 ). The Bryan cervical disc system is comprised of a low-friction elastic nucleus located between titanium end plates, which allow free rotation in all directions. A flexible membrane surrounds the articulating nucleus. Using a sliding polyethylene core between two fi xed metallic end plates, the Charité artificial disc allows a stable articulation with a physiological motion pattern determined by the interaction of the prosthesis, surrounding soft tissue, and facet joints. In contrast, the Prodisc and Maverick artificial disc are

L&R F

L

E

a

b

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

R

c

Fig 4.2-8a–c The kinematics of the intervertebral joint are complex. The center of rotation moves during flexion/extension (a), left and right side bending (b), and left and right torsion (c). Future designs for intervertebral prostheses or dynamic stabilization systems must respect this unique characteristic of spinal motion. F center of rotation in flexion E center of rotation in extension L/R center of rotation in left and right bending/rotation

4.2

65

Biomechanics of spinal stabilization

constrained devices with a single articulation, allowing free rotation in all directions around a fi xed center of rotation. Unconstrained devices allow a greater range of motion and theoretically prevent excessive facet loads in extreme motion. In contrast, constrained disc arthroplasties may reduce shear force on the posterior elements [56], however, anteroposterior misplacement of the device may lead to motion restriction and component liftoff [57]. Only comparative prospective clinical trials can conclusively show if any of these concepts is advantageous for the patient.

As with other total joint prostheses, the stability of the prosthesis and the motion segment likely depends on well balanced ligaments and surrounding soft tissues. Therefore, precise operation technique with retention of stabilizing tissue is essential for a good outcome. Wear of prosthesis components, as in other arthroplasties likely occurs, however, the histocompatibility for titanium and polyethylene particles has been tested in animal models and an absence of strong inflammatory host responses was shown [58, 59]. Finally, the kinematics of each new device must be verified against representative motion patterns of the normal spine [60]. In one study, spinal kinematics before and after implantation of a cervical disc prosthesis (Prodisc) was compared with spondylodesis. Using a displacement-controlled protocol, with the prosthesis in place almost no alteration in motion patterns could be recorded compared to the intact state, unlike in the

fusion case where the adjacent segments compensated for the fused level to achieve full motion [61]. This is in agreement with Puttlitz et al who demonstrated the establishment of an approximately physiological kinematics in all 6º of freedom with cervical disc arthroplasty [62]. In another biomechanical in vitro study, Cunningham et al compared the Charité disc prosthesis with an interbody fusion device (BAK) with and without posterior instrumentation. Unlike interbody fusion, also in the lumbar spine, the disc prosthesis exhibited a near physiological segmental motion pattern in all axes except rotation which was increased [63]. Long-term data are still scarce for the life time of disc prostheses, preservation of motion, and long-term patient satisfaction. Therefore, total disc replacement still has to establish its advantages compared to conventional spondylodesis. In contrast to total disc arthroplasty, replacement of only the degenerated or excised nucleus pulposus is an option offered by the Prosthetic Disc Nucleus (PDN). The PDN is a hydroactive implant which mimics the natural fluid exchange of the nucleus by swelling when unloaded and expressing water when under a compressive load. Wilke et al [64] have shown that the PDN implant can restore disc height and range of motion to normal values after nucleotomy. There is, however, little data on the long-term biomechanical and biological behavior of such implants in the intervertebral disc space, and the overall effectiveness of replacing only the nucleus pulposus in a degenerated disc.

66

4

10

Biomechanics of the spine

SUMMARY

Current spinal stabilization techniques are designed to limit segmental motion in order to promote bone formation and achieve solid fusion. Even rigid metallic implants rely on the principle of load sharing between the anatomical structures of the spine and the implant itself to maintain stability. The integrity of the anterior column of the spine is especially critical for a successful outcome. In the case of severe anterior injury, support of the anterior column is essential to prevent failure of the instrumentation. Combined anteroposterior spinal fi xation is more stable than either a single anterior or posterior procedure. Intervertebral cages are especially effective for restoring disc height, but may not offer adequate stability as stand-alone devices. Cages with supplemental posterior fi xation provide full 3-D spinal stabilization. New concepts in less-rigid and fully-dynamic stabilization of the spine, which are intended to restore the function of degenerated spinal segments, are being introduced and merit further study.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

11

1.

BIBLIOGRAPHY

Benzel EC (2001) Biomechanics of spine stabilization . 1st ed. Rolling Meadows:

American Association of Neurological Surgeons. 2. White AA, Panjabi MM (1990) Clinical biomechanics of the spine. 2nd ed. Philadelphia: JB Lippincott Company. 3. Gaines RW Jr (2000) The use of pedicle-screw internal fi xation for the operative treatment of spinal disorders. J Bone Joint Surg Am ; (10):1458–1476. Review. 4. Arand M, Wilke HJ, Schultheiss M, et al (2000) Comparative stability of the “internal fi xator” and the “universal spine system” and the effect of cross-linking transfi xating systems. A biomechanical in vitro study. Biomed Tech (Berl) ; 45(11):311–316. 5. Cripton PA, Jain GM, Wittenberg RH, et al (2000) Load-sharing characteristics of stabilized lumbar spine segments. Spine ; 25(2):170–179. 6. Panjabi MM, Abumi K, Duranceau J, et al (1988) Biomechanical evaluation of spinal fi xation devices: II. Stability provided by eight internal fi xation devices. Spine ; 13(10):1135–1140. 7. Halvorson TL, Kelley LA, Thomas K A, et al (1994) Effects of bone mineral density on pedicle screw fi xation. Spine ; 19(21):2415–2420. 8. Valdevit A, Kambic HE, McLain RF (2005) Torsional stability of cross-link confi gurations: a biomechanical analysis. Spine J ; 5(4):441–445. 9. Rohlmann A, Calisse J, Bergmann G, et al (1999) Internal spinal fi xator stiffness has only a minor infl uence on stresses in the adjacent discs. Spine ; 24(12):1192–1195. 10. Dvorak M, MacDonald S, Gurr KR, et al (1993) An anatomic, radiographic, and biomechanical assessment of extrapedicular screw fi xation in the thoracic spine. Spine ; 18(12):1689–1694.

4.2

67

Biomechanics of spinal stabilization

11. Morgenstern W, Ferguson SJ, Berey S, et al (2003) Posterior thoracic extrapedicular fi xation: a biomechanical study. Spine ; 28(16):1829–1835. 12. Guidera K J, Hooten J, Weatherly W, et al (1993) Cotrel-Dubousset instrumentation. Results in 52 patients. Spine ; 18(4):427–431. 13. Aebi M, Thalgott JS, Webb JK (1998) AO Principles in Spine Surgery. Berlin Heidelberg: Springer-Verlag. 14. Laxer E (1994) A further development in spinal instrumentation. Technical commission for spinal surgery of the ASIF. Eur Spine J ; 3(6):347–352. 15. Berlemann U, Cripton P, Nolte LP, et al (1995) New means in spinal pedicle hook fi xation. A biomechanical evaluation. Eur Spine J ; 4(2):114–122. 16. Rohlmann A, Bergmann G, Graichen F, et al (1995) Telemeterized load measurement using instrumented spinal internal fi xators in a patient with degenerative instability. Spine ; 20(24):2683–2689. 17. Polly DW Jr, Klemme WR, Cunningham BW, et al (2000) The biomechanical signifi cance of anterior column support in a simulated singlelevel spinal fusion. J Spinal Disord ; 13(1):58–62. 18. Spivak JM, Chen D, Kummer FJ (1999) The effect of locking fi xation screws on the stability of anterior cervical plating. Spine ; 24(4):334–338. 19. Isomi T, Panjabi MM, Wang JL, et al (1999) Stabilizing potential of anterior cervical plates in multilevel corpectomies. Spine ; 24(21):2219–2223. 20. Pitzen T, Wilke HJ, Caspar W, et al (1999) Evaluation of a new monocortical screw for anterior cervical fusion and plating by a combined biomechanical and clinical study. Eur Spine J ; 8(5):382–387.

21. Richter M, Wilke HJ, Kluger P, et al (1999) Biomechanical evaluation of a newly developed monocortical expansion screw for use in anterior internal fi xation of the cervical spine. In vitro comparison with two established internal fi xation systems. Spine ; 24(3):207–212. 22. Reidy D, Finkelstein J, Nagpurkar A, et al (2004) Cervical spine loading characteristics in a cadaveric C5 corpectomy model using a static and dynamic plate. J Spinal Disord Tech ; 17(2):117–122. 23. Wilke HJ, Kemmerich V, Claes LE, et al (2001) Combined anteroposterior spinal fi xation provides superior stabilization to a single anterior or posterior procedure. J Bone Joint Surg Br ; 83(4):609–617. 24. Tsantrizos A, Andreou A, Aebi M, et al (2000) Biomechanical stability of fi ve standalone anterior lumbar interbody fusion constructs. Eur Spine J ; 9(1):14–22. 25. Tsantrizos A, Baramki HG, Zeidman S, et al (2000) Segmental stability and compressive strength of posterior lumbar interbody fusion implants. Spine ; 25(15):1899–1907. 26. Carlisle E, Fischgrund JS (2005) Bone morphogenetic proteins for spinal fusion. Spine J ; 5(6 Suppl):240S–249S. 27. Sato M, Ochi T, Nakase T, et al (1999) Mechanical tension-stress induces expression of bone morphogenetic protein (BMP)-2 and BMP-4, but not BMP-6, BMP-7, and GDF-5 mRNA, during distraction osteogenesis. J Bone Miner Res ; 14(7):1084–1095. 28. Epari DR, Kandziora F, Duda GN (2005) Stress shielding in box and cylinder cervical interbody fusion cage designs. Spine ; 30(8):908–914. 29. Jost B, Cripton PA, Lund T, et al (1998) Compressive strength of interbody cages in the lumbar spine: the effect of cage shape, posterior instrumentation and bone density. Eur Spine J ; 7(2):132–141.

30. Grant JP, Oxland TR, Dvorak MF (2001) Mapping the structural properties of the lumbosacral vertebral end plates. Spine ; 26(8):889–896. 31. Lowe TG, Hashim S, Wilson LA, et al (2004). A biomechanical study of regional end plate strength and cage morphology as it relates to structural interbody support. Spine ; 29(21):2389–2394. 32. Oxland TR, Grant JP, Dvorak MF, et al (2003) Effects of end plate removal on the structural properties of the lower lumbar vertebral bodies. Spine ; 28(8):771–777. 33. Steffen T, Tsantrizos A, Aebi M (2000) Effect of implant design and end plate preparation on the compressive strength of interbody fusion constructs. Spine ; 25(9):1077–1084. 34. Lund T, Oxland TR, Jost B, et al (1998) Interbody cage stabilization in the lumbar spine: biomechanical evaluation of cage design, posterior instrumentation and bone density. J Bone Joint Surg Br ; 80(2):351–359. 35. Nydegger T, Oxland TR, Hoffer Z, et al (2001) Does anterolateral cage insertion enhance immediate stabilization of the functional spinal unit? A biomechanical investigation. Spine ; 26(22):2491–2497. 36. Bagby GW (1988) Arthrodesis by the distraction-compression method using a stainless steel implant. Orthopedics ; 11(6):931–934. 37. Kettler A, Wilke HJ, Dietl R, et al (2000) Stabilizing effect of posterior lumbar interbody fusion cages before and after cyclic loading. J Neurosurg ; 92(1 Suppl):87–92. 38. Brodke DS, Dick JC, Kunz DN, et al (1997) Posterior lumbar interbody fusion. A biomechanical comparison, including a new threaded cage. Spine ; 22(1):26–31. 39. Cain CM, Schleicher P, Gerlach R, et al (2005) A new stand-alone anterior lumbar interbody fusion device: biomechanical comparison with established fi xation techniques. Spine ; 30(23):2631–2636.

68

4

Biomechanics of the spine

40. Greene DL, Crawford NR, Chamberlain RH, et al (2003) Biomechanical comparison of cervical interbody cage versus structural bone graft. Spine J ; 3(4):262–269. 41. Shono Y, Kaneda K, Abumi K, et al (1998) Stability of posterior spinal instrumentation and its effects on adjacent motion segments in the lumbosacral spine. Spine ; 23(14):1550–1558. 42. Bastian L, Lange U, Knop C, et al (2001) Evaluation of the mobility of adjacent segments after posterior thoracolumbar fi xation: a biomechanical study. Eur Spine J ; 10(4):295–300. 43. Dekutoski MB, Schendel MJ, Ogilvie JW, et al (1994) Comparison of in vivo and in vitro adjacent segment motion after lumbar fusion. Spine ; 19(15):1745–1751. 44. Battie MC, Videman T, Parent E (2004) Lumbar disc degeneration: epidemiology and genetic infl uences. Spine ; 29(23):2679–2690. 45. Graf H (1992) Lumbar instability. Rachis ; 412:123–137. 46. Strauss PJ, Novotny JE, Wilder DG, et al (1994) Multidirectional stability of the Graf system. Spine ; 19(8):965–972. 47. Gardner A, Pande KC (2002) Graf ligamentoplasty: a 7-year follow-up. Eur Spine J ; 11(Suppl 2):S157–163. 48. Markwalder TM, Wenger M (2003) Dynamic stabilization of lumbar motion segments by use of Graf‘s ligaments: results with an average follow-up of 7.4 years in 39 highly selected, consecutive patients. Acta Neurochir (Wien) ; 145(3):209–214. 49. Schmoelz W, Huber JF, Nydegger T, et al (2003) Dynamic stabilization of the lumbar spine and its effects on adjacent segments: an in vitro experiment. J Spinal Disord Tech ; 16(4):418–423. 50. Freudiger S, Dubois G, Lorrain M (1999) Dynamic neutralization of the lumbar spine confi rmed on a new lumbar spine simulator in vitro. Arch Orthop Trauma Surg ; 119(3–4):127–132.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

51. Whitesides TE Jr (2003) The effect of an interspinous implant on intervertebral disc pressures. Spine ; 28(16):1906–1907. 52. Lindsey DP, Swanson KE, Fuchs P, et al (2003). The effects of an interspinous implant on the kinematics of the instrumented and adjacent levels in the lumbar spine. Spine ; 28(19):2192–2197. 53. Senegas J (2002) Mechanical supplementation by non-rigid fi xation in degenerative intervertebral lumbar segments: the Wallis system. Eur Spine J ; 11(Suppl 2):S164–169. 54. Wiseman CM, Lindsey DP, Fredrick AD, et al (2005) The effect of an interspinous process implant on facet loading during extension. Spine ; 30(8):903–907. 55. Szpalski M, Gunzburg R, Mayer M (2002) Spine arthroplasty: a historical review. Eur Spine J ; 11(Suppl 2):S65–84. 56. Huang RC, Girardi FP, Cammisa FP Jr, et al (2003) The implications of constraint in lumbar total disc replacement. J Spinal Disord Tech ; 16(4):412–417. 57. Ferguson SJ, Tolkmitt F, Nolte LP (2004) Kinematic analysis of intervertebral disc prostheses. NL-‘s-Hertogenbosch: Proceedings of the 14th Conference of the European Society of Biomechanics. 58. Anderson PA, Rouleau JP, Bryan VE, et al (2003) Wear analysis of the Bryan Cervical Disc prosthesis. Spine ; 28(20):S186–194. 59. Chang BS, Brown PR, Sieber A, et al (2004) Evaluation of the biological response of wear debris. Spine J ; 4(6 Suppl):239S–244S. 60. Cunningham BW (2004) Basic scientifi c considerations in total disc arthroplasty. Spine J ; 4(6 SuppI):219S–230S. 61. DiAngelo DJ, Foley KT, Morrow BR, et al (2004) In vitro biomechanics of cervical disc arthroplasty with the ProDisc-C total disc implant. Neurosurg Focus ; 17(3):E7.

62. Puttlitz CM, Rousseau MA, Xu Z, et al (2004) Intervertebral disc replacement maintains cervical spine kinetics. Spine ; 29(24):2809–2814. 63. Cunningham BW, Gordon JD, Dmitriev AE, et al (2003) Biomechanical evaluation of total disc replacement arthroplasty: an in vitro human cadaveric model. Spine ; 28(20):S110–117. 64. Wilke HJ, Kavanagh S, Neller S, et al (2002) [Effect of artifi cial disk nucleus implant on mobility and intervertebral disk high of an L4/5 segment after nucleotomy.] Orthopade ; 31(5):434–440.

69

Norbert Boos

BIOLOGY OF THE SPINE

5

73

Norbert Boos

5

BIOLOGY OF THE SPINE

INTRODUCTION

In the past two decades, spine surgery has made enormous progress. This progress became possible with the advent of second and third generation spinal implants. These implants allowed for a short segmental and angle-stable fi xation, correction of severe deformities, and facilitated postoperative treatment by early mobilization. Very recently motionpreserving implants, such as dynamic instrumentation systems, disc arthroplasty, and interspinous process implants, have added to the armamentarium of the spine surgeon. Vertebroplasty and kyphoplasty have come into widespread use for the treatment of osteoporotic fractures. Bone substitutes and cytokines are used even more frequently to achieve spinal fusions. The biggest risk of these technology-driven developments is that we forget about the biology of the spine and basic biological principles. In this context, a disc prosthesis will ultimately fail if we do not understand that the surgery is

addressing not only the intervertebral disc, but the whole motion segment including facet joints, muscles, and ligaments. Our understanding of the basic process leading to back pain is still very limited and we still have almost no clue about the underlying molecular mechanisms. The classic concept of fusion, ie, eliminating painful motion within segments is being increasingly challenged by the clinical success of motion-preserving implants. We still do not understand the weak correlation between morphological alterations of the intervertebral disc and pain. Similar tissue alterations can cause pain in one individual and remain asymptomatic in another. The molecular mechanisms of disc degeneration are still being unraveled, and the potential for a more biological repair remains unexplored. With the age of the population increasing in developed countries, osteoporosis and related fractures are becoming an increasing challenge for the spine specialist. Knowledge of baseline therapy and a potential prophylaxis of osteoporosis are necessary in dealing with

74

5

Biology of the spine

this pathology. The development of bone substitutes and bone morphogenetic proteins will substantially influence spine surgery in the near future. The following chapters will provide an overview of biological principles related to the treatment of spinal disorders. The understanding of these biological principles is a prerequisite for the selection of appropriate treatment modalities. Some of the most fundamental biological principles are as simple as they are obvious, but clinical practice continuously demonstrates that they are being jeopardized. • The spine is a complex organ consisting of a functional chain of motion segments. Surgery to one part also affects adjacent parts. • The kinematics of the spinal motion segment and the intrinsic and extrinsic effect of the spinal muscles are very complex and cannot easily be mimicked by implants. • Disc degeneration starts as early as the second decade of life and is related to age rather than pathology. • The vast majority of disc alterations which are addressed by surgery exhibit only a week correlation to pain. • Osteoporosis is a systemic disease and changing the stiffness in one segment (eg, by vertebroplasty) will have an impact on the adjacent segments. • Cement augmentation does not solve the problem of the diseased bone. • Fusion implants will eventually fail when solid fusion does not occur. • Bone substitutes and cytokines cannot compensate for insufficient fusion techniques.

AOSPINE MANUAL—CLINICAL APPLICATIONS

The spine specialist, and in particular the spine surgeon, must be aware of the aforementioned problems, challenges, and biological principles when choosing treatment modalities and caring for patients.

75

76

5 5.1

BIOLOGY OF THE SPINE BIOLOGY OF THE MOTION SEGMENT

1

Introduction …………………………………………………………………………………………………… 77

2

Intervertebral disc …………………………………………………………………………………………… 79

3

Facet joint dysfunction and osteoarthritis ……………………………………………………………… 79

4

Degeneration of the motion segment …………………………………………………………………… 81

5

Alterations of the posterior support structures and pain ……………………………………………… 83

6

Bibliography …………………………………………………………………………………………………… 84

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

77

Norbert Boos

5

BIOLOGY OF THE SPINE

5.1

1

BIOLOGY OF THE MOTION SEGMENT

INTRODUCTION

The basic functions of the human spine are to support the body, protect the spinal cord and nerve roots, and allow for movements of the trunk. Thus, the spine simultaneously provides enough stability to keep an upright posture, but at the same time allows for enough mobility of the trunk. The spine can further absorb energy and thereby protect itself against impact. The spine consists of seven cervical vertebrae, twelve thoracic vertebrae, five lumbar vertebrae, five fused sacral vertebrae, and three to four partially fused vertebrae forming the coccyx. The unfused vertebrae are separated by the intervertebral discs in the front and the bilateral zygapophyseal joints (so-called facet joints) in the back. The vertebrae are further connected by spinal ligaments, facet joint capsules, and segmental muscles. The spinal ligaments consist of interspinous, supraspinous intertransverse, yellow, anterior, and posterior longitudinal ligaments. In contrast to the extrinsic muscles, the intrinsic muscles span only two

vertebrae and consist of splenius, erector spinae, transversospinal, and segmental muscles. The smallest anatomical unit of the spine, which exhibits the basic functional characteristics of the entire spine, is called motion segment or functional spinal unit (FSU) ( Fig 5.1-1). The term motion segment was fi rst coined by Schmorl and Junghanns in 1968 [1]. Normal spinal function largely depends on the integrity of these different components and their coordinated interplay. Any alteration of these components or their interplay may result in dysfunction fi nally leading to back pain, deformity, and neurological compromise. Kirkaldy-Willis introduced the term “the three-joint complex” to highlight the importance of normal interaction between the three joints (ie, intervertebral disc and facet joints) for a healthy spine. However, in this complex the role of the back muscles is not incorporated, although undoubtedly essential for the normal functioning of the motion segment.

78

5

Biology of the spine

Our present understanding of the underlying mechanisms of back pain is still poor. Progress in the understanding of back pain and also of the development of spinal deformity depends on basic knowledge of the biology and the interplay of the functional spinal units from the level of gross anatomy to the level of cells and extracellular matrix components. While increasing knowledge has been gained on the biology of the intervertebral disc, little is known about the biology of facet joint dysfunction and its clinical relevance. The gross morphology of the spinal muscles and their basic kinematics has been extensively studied. However, only sparse information is available with regard to spinal muscle function on the level of the motion segment and its relation to the development of degenerative changes. The objective of this chapter is to provide a short overview on the current concepts of the biology of the spinal motion segment which, however, is hampered by limited knowledge in this area.

Fig 5.1-1 The functional spinal unit (FSU)

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

5.1

Biology of the motion segment

2

INTERVERTEBRAL DISC

Since its fi rst description by Mixter and Barr in 1934 [2], alterations of the intervertebral disc have been the focus of research exploring its association with back pain. Because of the importance of the intervertebral disc, the authors will separately cover biology in health and disease in chapter 5.2 Aging and pathological degeneration.

79

3

FACET JOINT DYSFUNCTION AND OSTEOARTHRITIS

The facet joints are also called zygapophyseal joints and resemble the features of synovial joints. They are an essential part of the posterior support structures of the spine consisting of: pedicles, lamina, and spinous and transverse processes. Adams and Hutton [3] found that the facet joints resist most of the intervertebral shear force. The posterior annulus is protected in torsion by the facet surfaces and in flexion by the capsular ligaments. The posterior elements also serve as anchors for the spinal muscles which stabilize the spinal column. The biomechanics of the functional motion segment is covered in chapter 4.1 General biomechanics of the spinal motion segment and the spinal organ, and will not be further addressed here. The significance of menisci in the lumbar zygapophyseal joints remains questionable. Bogduk [4, 5] described these menisci as rudimentary fibrous invaginations of the dorsal and ventral capsule. They are basically fat-fi lled synovial reflections, some of which contain dense fibrous tissue, which probably arises as a result of mechanical stress. Bogduk outlined that meniscal entrapment is probably an overstated cause of those forms of “acute locked back” that respond to manipulation [4]. Despite the recognition of the association between facet joint arthropathy and back pain by Goldthwait [6] in 1911, the facet joint was ignored until Mooney and Robertson [7] initiated a revival of the so-called “facet joint syndrome”. Nevertheless, data on the pathogenesis of facet joint osteoarthritis are still very sparse in contrast to the knowledge gathered on synovial joint osteoarthritis [8–12]. As in large synovial joints, it has been assumed that malalignment is a predisposing factor for the development of osteoarthritis (OA). Fujiwara et al [13] examined the association between orientation and OA of the lumbar facet joints in 107 consecutive patients who underwent plain

80

5

Biology of the spine

radiography and magnetic resonance imaging. A significant association was found between sagittal orientation and osteoarthritis of the lumbar facet joints, even in patients without degenerative spondylolisthesis. The authors concluded that facet joint osteoarthritis, rather than spondylolisthesis, is the pathoanatomical feature that is associated with sagittal orientation of the facet joints in patients with degenerative spondylolisthesis. Vernon-Roberts [14] outlined that, in contrast to osteoarthritis in large synovial joints (eg, hip joint), an intact covering of hyaline cartilage is frequently retained by the articular surfaces even when large osteophytes have formed. He hypothesized that this preservation of articular cartilage may result from changing joint stresses. However, Swanepoel et al [15] found that the apophyseal cartilage of the facet joint surfaces show a greater extent and prevalence of cartilage fibrillation than knee, hip, or ankle, with significant damage in specimens younger than 30 years. In late stages of OA, the facet joints also demonstrate the classic features, ie, complete loss of articular cartilage, cysts and pseudocysts in the bone, dense bone sclerosis, and large osteophyte formation. At this stage, end-plate fractures can occur which resemble breaches in the subchondral bone plate with protrusion of a portion of the articular cartilage into the subarticular bone [16]. Of importance is the notion that spontaneous fusion of the facet joints is very rare in the absence of ankylosing spondylitis or ankylosing hyperostosis [14]. Taylor and Twomey [17] investigated the degenerative changes of zygapophyseal joints in relation to biomechanical function. Their results indicated that articular cartilage and subchondral bone of the anterior, coronally oriented third of the joint show changes that are likely to be related to loading of this part of the joint in flexion. The posterior, sagittally

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

oriented two-thirds of the joint show different age changes, which may reflect shearing forces imparted to the articular cartilage through the fibrous capsule, from insertion of some fibers of multifidus into the fibrous capsule. Swanepoel et al [15] explored the extent and location of fibrillated areas of the apophyseal cartilage of the joint surfaces of 29 lumbar motion segments. They found that damage was predominantly located peripherally, superiorly, and posteriorly in the concave superior apophyseal surfaces, and was predominantly peripheral and posterior in the inferior surfaces, with a tendency to be located inferiorly. The pattern of damage to the inferior surfaces lends some support to the hypothesis that their apices impact the laminae of the lumbar vertebra inferior to them, the result of degeneration and narrowing of the associated intervertebral disc. The predominantly peripheral location of fibrillation of both superior and inferior surfaces may be associated with inadequate mechanical conditioning of marginal joint areas [15]. In 24 (51%) of 47 facet joints analyzed histologically by Gries et al [18], the concave (or superior) facet was altered more severely than the convex (inferior) facet. In ten (21%) specimens, the convex facet showed more advanced changes. No difference was found in the remaining specimens. Gries et al [18] reported that the most severe changes were located in the posterior one-third of the joint in 16 (34%), in the anterior one-third in seven (15%), and equally distributed between anterior and posterior one-third in the remaining 14 (30%) specimens. Fujiwara et al [19] investigated the effect of both disc degeneration and facet joint osteoarthritis on lumbar segmental motion. They reported that with cartilage degeneration of the facet joints, the axial rotational motion increased, whereas the lateral bending and flexion motion decreased in female segments. In male segments, however, motion in all directions increased with moderate cartilage damage and decreased

5.1

81

Biology of the motion segment

4

with severe cartilage degeneration. Subchondral sclerosis significantly decreased the motion, and the severity of osteophytes had no significant association with the segmental motion. However, the authors were not able to clarify whether facet joint OA differs between genders and how facet joint OA affects the stability of the spinal motion segment. There is a void of information in the literature with regard to a comparison of the osteoarthritic changes in large weightbearing synovial joints and facet joints on a histological and molecular level.

DEGENERATION OF THE MOTION SEGMENT

Based on the early work of Schmorl and Junghanns, KirkaldyWillis [20] proposed the concept of a close interaction between the two facet joints and the intervertebral disc in a so-called “three-joint complex” ( Fig 5.1-2 ). Thus, degenerative alteration of the intervertebral disc will fi nally affect the facet joints and vice versa. According to the Kirkaldy-Willis’ concept [20] progressive degenerative changes in the posterior joints lead to marked destruction and instability. Similar changes in the disc can result in herniation, internal disruption, and resorption. Combined changes in the posterior joint and disc sometimes produce entrapment of a spinal nerve in the lateral

DEGENERATION OF THE THREE-JOINT COMPLEX facet joint degeneration synovial reactions

disc degeneration facet syndrome

circumferential tears

DYSFUNCTION cartilage degeneration

disc herniation

radial tears

capsular laxity

dynamic lateral stenosis

internal disruption

subluxation

degenerative spondylolisthesis

disc narrowing

osteophyte formation

fixed lateral stenosis

osteophyte formation

INSTABILITY STABILIZATION facet and lamina enlargement

central stenosis

vertebral body enlargement

multisegmental spondylosis

Fig 5.1-2 The degeneration of the so-called three-joint complex according to Kirkaldy-Willis [20] (modified).

82

5

Biology of the spine

recess, central stenosis at one level, or both of these conditions. Changes at one level often lead, over a period of years, to multilevel spondylosis and/or stenosis. Developmental stenosis is an enhancing factor in the presence of a small herniation or moderate degenerative stenosis ( Fig 5.1-2 ). Debate has continued on the temporal relationship between degenerative changes of the intervertebral disc and facet joints. Several studies have provided some evidence that disc degeneration usually precedes facet joint OA [21–24]. Although degeneration of the intervertebral disc results in decreased disc height, which alters the load distribution in the motion segment, the relation is not simple [3]. Gries et al [18] have shown that microscopic changes occur in discs and facet joints at an early age. However, the authors were not able to establish a close correlation with age, neither for the intervertebral disc nor for the facet joints at the same level. More interestingly, they reported that grades of disc degeneration did not correlate with those for the facet joints. From their results the authors concluded that it is not possible to demonstrate that pathological changes progress more rapidly in one element relative to the other in the young adult spine [18]. Notwithstanding its usefulness as a model for the study of the spine, Yeong-Hing and Kirkaldy-Willis [25] have outlined the limitations of the three-joint complex with regard to the effect of intrinsic and extrinsic action of the trunk muscles. At the level of the motion segment, there is only sparse data on the muscle function. Goel at al [26] have reported a threedimensional, nonlinear, fi nite element model of a ligamentous L3/4 motion segment for the predictions of stresses in the

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

motion segment. From their model the authors were able to show that the muscles imparted stability to the ligamentous motion segment. The presence of muscles also led to a decrease in stresses in the vertebral body, the intradiscal pressure, and other mechanical parameters of importance. However, the load bearing of the facets increased compared to the ligamentous model. Thus, facet joints play a significant role in transmitting loads in a normal intact spine. These results provide quantitative data on the stabilizing effects of muscles on the mechanics of a ligamentous spine. The results also provide a scientific explanation in support of the “degenerative cascade” concept proposed in the literature. The model predictions, in conjunction with the degenerative cascade concept, also support the observation that the osteoarthritis of facets may follow disc degeneration. There is a lack of studies exploring the concept of the so-called degenerative cascade in other regions of the spine, particularly the cervical region. A conclusive and comprehensive assessment of the behavior of the muscles on the motion segment cannot be achieved with in vitro models alone, but requires in vivo models. In an animal model, Kaigle et al [27] studied the in vivo kinematics of a degenerated lumbar motion segment. From their results, the authors concluded that the lumbar paraspinal muscles are less efficient overall in providing stability during flexionextension when chronic lesions are made in the intervertebral disc and facet joints. This is due possibly to altered mechanisms in the neuromuscular feedback system in the degenerated motion segment and, consequently, in the lumbar spine as a whole. These neuromuscular feedback mechanisms need to be incorporated in any model when we want to better understand the function of the motion segment in health and disease.

5.1

Biology of the motion segment

5

ALTERATIONS OF THE POSTERIOR SUPPORT STRUCTURES AND PAIN

The prerequisite for the transmission of pain is a neural innervation of the structure. Bogduk provided a comprehensive review of the neural innervation of the lumbar spine [4], highlighting the importance of the innervation of the posterior support structures, which are innervated by branches of the dorsal rami. The facet joint capsule is richly innervated with C and A delta fi bers [28] and exhibit a neural innervation arising from the same level but also from the level above [29]. Cavanaugh [28] has outlined that under normal conditions pain fibers (nociceptors) have high mechanical thresholds. Under pathological conditions, such as inflammation, chemical mediators sensitize the nerve endings and they can begin to fi re spontaneously, ie, with levels of physiological stress and strain. Neuromediators that contribute to the transmission process, particularly substance P, have been observed in the facet joint capsules [30, 31]. Based on the results of neurophysiological and neuroanatomical studies in anesthetized New Zealand white rabbits, Cavanaugh [28, 32] summarized the evidence in support of facet pain: • An extensive distribution of small nerve fibers and endings in the lumbar facet joint. • Nerves containing substance P. • High threshold mechanoreceptors in the facet joint capsule. • Sensitization and excitation of nerves in facet joint and surrounding muscle when the nerves were exposed to inflammatory or algesic chemicals.

83

An acute traumatic strain in the facet joint capsule could lead to inflammation and subsequently prolonged nociceptor excitation. The incorporation of the mechanisms of pain generation, transmission, and modulation in the concept of back pain resulting from degenerative disorders of the motion segment is still at its beginning. Based on the results of imaging studies in asymptomatic individuals [33–35], degenerative changes may not per se be painful. A research approach focusing on pain generation and modulation is therefore more sensible and more likely to be successful in the vast majority of the patients where the ultimate clinical goal is to relieve pain.

84

5

Biology of the spine

6

BIBLIOGRAPHY

1.

Schmorl CG, Junghanns H (1968) Die gesunde und die kranke Wirbelsäule in Röntgenbild und Klinik. 5th ed. Stuttgart: Thieme.

2.

Mixter WJ, Barr JS (1934) Rupture of the intervertebral disc with involvement of the spinal canal. N Engl J Med; 211:210–215. Adams MA, Hutton WC (1983) The mechanical function of the lumbar apophyseal joints. Spine; 8(3):327–330. Bogduk N, Engel R (1984) The menisci of the lumbar zygapophyseal joints. A review of their anatomy and clinical signifi cance. Spine; 9(5):454–460. Review. Engel R, Bogduk N (1982) The menisci of the lumbar zygapophysial joints. J Anat; 135(4):795–809. Goldwaith JE (1911) The lumbo-sacral articulation: An explanation of many cases of lumbago, sciatica, and paraplegia. Boston Med Surg J; 164:365–372. Mooney V, Robertson J (1976) The facet syndrome. Clin Orthop; (115):149–156. Fukui N, Purple CR, Sandell LJ (2001) Cell biology of osteoarthritis: the chondrocytes response to injury. Curr Rheumatol Rep; 3(6):496–505. Review. Goldring MB (2000) Osteoarthritis and cartilage: the role of cytokines. Curr Rheumatol Rep; 2(6):459–465. Review. Martel-Pelletier J, Alaaeddine N, Pelletier JP (1999) Cytokines and their role in the pathophysiology of osteoarthritis. Front Biosci; 4:D694–703. Review. Poole AR (1999) An introduction to the pathophysiology of osteoarthritis. Front Biosci; 4:D662–670. Review. Poole AR, Rizkalla G, Ionescu M, et al (1993) Osteoarthritis in the human knee: a dynamic process of cartilage matrix degradation, synthesis and reorganization. Agents Actions Suppl; 39:3–13. Review. Fujiwara A, Tamai K, An HS, et al (2001) Orientation and osteoarthritis of the lumbar facet joint. Clin Orthop; (385):88–94.

3.

4.

5.

6.

7. 8.

9.

10.

11.

12.

13.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

14. Vernon-Roberts B (1992) Age-related and degenerative pathology of intervertebral discs and apophyseal joints. Jayson MIV (ed), The Lumbar Spine and Back Pain. Edinburgh: Churchill Livingstone, 17–41. 15. Swanepoel MW, Adams LM, Smeathers JE (1995) Human lumbar apophyseal joint damage and intervertebral disc degeneration. Ann Rheum Dis; 54(3):182–188. 16. Vernon-Roberts B, Pirie CJ (1977) Degenerative changes in the intervertebral discs of the lumbar spine and their sequelae. Rheumatol Rehabil; 16(1):13–21. 17. Taylor JR, Twomey LT (1986) Age changes in lumbar zygapophyseal joints. Observations on structure and function. Spine;11(7):739–745. 18. Gries NC, Berlemann U, Moore RJ, et al (2000) Early histologic changes in lower lumbar discs and facet joints and their correlation. Eur Spine J; 9(1):23–29. 19. Fujiwara A, Lim TH, An HS, et al (2000) The effect of disc degeneration and facet joint osteoarthritis on the segmental fl exibility of the lumbar spine. Spine; 25(23):3036–3044. 20. Kirkaldy-Willis WH, Wedge JH, Yong-Hing K, et al (1978) Pathology and pathogenesis of lumbar spondylosis and stenosis. Spine; 3(4):319–328. 21. Butler D, Trafimow JH, Andersson GB, et al (1990) Discs degenerate before facets. Spine; 15(2):111–113. 22. Fujiwara A, Tamai K, Yamato M, et al (1999) The relationship between facet joint osteoarthritis and disc degeneration of the lumbar spine: an MRI study. Eur Spine J; 8(5):396–401. 23. Gotfried Y, Bradford DS, Oegema TR Jr (1986) Facet joint changes after chemonucleolysis-induced disc space narrowing. Spine; 11(9):944–950. 24. Oegema TR Jr, Bradford DS (1991) The inter-relationship of facet joint osteoarthritis and degenerative disc disease. Br J Rheumatol; 30:16–20. Review.

25. Yong-Hing K, Kirkaldy-Willis W (1990) The three-joint complex. Weinstein JN, Wiesel

SW, International Society for the Study of the Lumbar Spine (eds), The Lumbar Spine. Philadelphia: WB Saunders Company, 80–87. 26. Goel VK, Kong W, Han JS, et al (1993) A combined fi nite element and optimization investigation of lumbar spine mechanics with and without muscles. Spine; 18(11):1531–1541. 27. Kaigle AM, Holm SH, Hansson TH (1997) 1997 Volvo Award winner in biomechanical studies. Kinematic behavior of the porcine lumbar spine: a chronic lesion model. Spine; 22(24):2796–2806. 28. Cavanaugh JM, Ozaktay AC, Yamashita HT, et al (1996) Lumbar facet pain: biomechanics, neuroanatomy and neurophysiology. J Biomech; 29(9):1117–1129. Review. 29. Bogduk N (1983) The innervation of the lumbar spine. Spine; 8(3):286–293. 30. Giles LG, Harvey AR (1987) Immunohistochem ical demonstration of nociceptors in the capsule and synovial folds of human zygapophyseal joints. Br J Rheumatol; 26(5):362–364. 31. Gronblad M, Korkala O, Konttinen YT, et al (1991) Silver impregnation and immunohistochemical study of nerves in lumbar facet joint plical tissue. Spine; 16(1):34–38. 32. Cavanaugh JM, Ozaktay AC, Yamashita T, et al (1997) Mechanisms of low back pain: a neurophysiologic and neuroanatomic study. Clin Orthop; (335):166–180. Review. 33. Boden SD, Davis DO, Dina TS, et al (1990) Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am; 72(3):403–408. Review. 34. Boden SD, McCowin PR, Davis DO, et al (1990) Abnormal magnetic-resonance scans of the cervical spine in asymptomatic subjects: a prospective investigation. J Bone Joint Surg Am; 72(8):1178–1184.

5.1

Biology of the motion segment

35. Boos N, Rieder R, Schade V, et al (1995) 1995 Volvo Award in Clinical Sciences. The diagnostic accuracy of magnetic resonance imaging, work perception and psychosocial factors in identifying symptomatic disc herniations. Spine; 20(24):2613–2625.

85

86

5 5.2

BIOLOGY OF THE SPINE AGING AND PATHOLOGICAL DEGENERATION

1

Introduction …………………………………………………………………………………………………… 87

2

Morphology of the degenerated intervertebral disc …………………………………………………… 88

3

Innervation of intervertebral disc ………………………………………………………………………… 89

4

Vascular changes during disc degeneration ……………………………………………………………… 90

5 5.1 5.2 5.3

Changes of the extracellular matrix during disc degeneration ……………………………………… Collagens ……………………………………………………………………………………………………… Proteoglycans ………………………………………………………………………………………………… Noncollagenous proteins ……………………………………………………………………………………

6

Proteolytic activity in degenerated discs ………………………………………………………………… 93

7

Evidence for enhanced oxidative stress in degenerating disc tissue ………………………………… 95

8

Phenotypic changes of disc cells ………………………………………………………………………… 96

9

Modulation of disc cells by cytokines and growth factors …………………………………………… 97

10

A current concept of painful disc degeneration ………………………………………………………… 98

11

Bibliography …………………………………………………………………………………………………… 99

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

91 91 92 93

87

Mauro Alini, Keita Ito, Andreas G Nerlich, Norbert Boos

5

BIOLOGY OF THE SPINE

5.2

1

AGING AND PATHOLOGICAL DEGENERATION

INTRODUCTION

The intervertebral discs separate the vertebrae of the spine where they facilitate the bending and twisting motion to which the spine is subjected and they also cushion compressive loading due to impact or gravity. These functions are endowed by the unique structure of the intervertebral disc, which is initially established during embryonic development and evolves throughout life. The vertebral column develops in the embryonic mesoderm at about 4 weeks gestational age in humans [1]. Vertebral bodies mature under the combined influence of the notochord and neuronal tube. Discs grow initially in an environment that contains few blood vessels and are surrounded by a perichondral layer, the future longitudinal vertebral ligaments. Between the vertebrae, the notochord expands as local aggregations of cells (the notochordal cells) within a proteoglycan matrix, forming the gelatinous center of the intervertebral disc, the nucleus pulposus. The nucleus is later surrounded by the circularly arranged fibrous lamellae of the anulus fibrosus, which are derived from the perichordal mesenchyme. The rapid increase

in notochordal nucleus pulposus volume in fetuses occurs at the expense of the inner annular region. At this point, the cells populating the nucleus pulposus are a mixture of notochordal and chondrocyte-like cells, while the rest of the intervertebral disc contains fibroblast-like cells. At the junction with the notochordal sheath, the cells of the inner annulus are closer in shape to the chondrocyte-like cells found within the nucleus. The exact role and interaction of notochordal cells with the other disc cells is unknown. The lumbar intervertebral disc undergoes very extensive destructive changes with age and degeneration [2]. The degree of this tissue destruction is closely linked to age, however, different components of the disc undergo more extensive alterations than others [3]. Although several investigations confi rm the notion that disc degeneration is extensively seen in advanced age individuals, previous personal investigation has shown that initial degenerative alterations on the histological level are even visible in infantile discs [4]. Therefore,

88

5

Biology of the spine

2

substantial individual differences can be observed in the sense that young individuals exhibit the discs of an elderly person and vice versa. Because of the extensive destructive changes that ultimately lead to an ankylosed motion segment [4, 5], many clinicians and researchers believe that the intervertebral disc is a predominant source of low back pain. From a clinical point of view, differentiating “normal” age-related (ie, asymptomatic) from “pathological” degenerative (ie, painful) changes would be sensible. However, this task is extremely difficult due to the lack of a reference standard for painful disc degeneration. So far, the best reference standard is provocative discography. However, the role and potential benefit of this diagnostic procedure is controversially discussed in the literature [6].

MORPHOLOGY OF THE DEGENERATED INTERVERTEBRAL DISC

There exist several studies on the macro- and histomorphology of the intervertebral disc [3–5, 7] that will not be reviewed here in more detail. In summary, all these analyses indicate that there is a progressive destruction of the disc tissue ( Fig 5.2-1), mainly starting in the nucleus and extending to the annulus, that is accompanied by reactive cellular proliferation ( Fig 5.2-2 ) and granular and mucoid matrix degeneration ( Fig 5.2-3, Fig 5.2-4 ). Finally, the regular disc structure is lost and frequently a replacement of the disc by a scar-like tissue is seen. These changes are associated with increasing age.

For the purpose of this review, the focus will be on those morphological and molecular alterations which may be involved in the development of a painful degenerated disc. However, the authors do not imply that the alterations described within this chapter are exclusively found in individuals with a painful degenerated disc. Therefore, the term “degenerated” disc is not synonymous with painful disc.

Fig 5.2-1 Histomorphological features of disc degeneration. Here, extensive formation of tissue clefts within the nucleus pulposus suggests major disruption of the matrix (22-year-old person, original magnification × 250).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

89

5.2

Aging and pathological degeneration

3

INNERVATION OF INTERVERTEBRAL DISC

It is obvious that innervation of the disc may be crucial for the development of discogenic pain. In this regard, it is of significant interest that the disc is one of the largest aneural tissues of the human body. There is good evidence that discs are particularly free of nerve endings within their central portions [8, 9]. Only adjacent to capillary vessels in the outermost zone of the anulus fibrosus (AF) are small immunohistochemically detectable nerves identifiable [8].

Fig 5.2-2 Histological signs of chondrocyte proliferation. Frequently, cluster-like proliferations of chondrocytes are present in disc tissue suggesting an attempt to restore the disc function (28-year-old person, original magnification × 400).

As yet, we have no information whether these are only small nerves for vessel wall innervation or if they contain sensory components. There is still an open, undecided debate whether discal structures contain neurosecretory granules that may diffuse to juxta-discal receptors and thereby induce any signalling. This also holds true for previous studies [9, 10] that suggested the ingrowth of veritable nerves into degenerated discs.

Fig 5.2-3 Histological signs for granular matrix degeneration with the local deposition of amorphous eosinophilic debris (21-year-old person, original magnification × 400).

Fig 5.2-4 Histological features of mucoid matrix degeneration. This is characterized by the deposition of extensive amounts of mucoid matrix material (62-year-old person, original magnification × 250).

90

5

Biology of the spine

4

VASCULAR CHANGES DURING DISC DEGENERATION

Besides innervation the intervertebral discs are unique in that they lack any significant vascular supply [3, 4, 9, 11]. Accordingly, it has been assumed that the vascular supply of the intervertebral disc seems to be of most significance for the development of age-associated and degenerative alterations. While fetal and infantile discs contain vascular loops, the juvenile, adolescent, and adult disc is nonvascularized [12]. Therefore, the adult disc is the largest avascular tissue structure of the human body. Nutritional supply is provided predominantly by diffusion with a maximum diffusion distance of up to 1 cm (from the vertebral bone marrow to the center of the nucleus pulposus, NP). The only exception is the nutritional supply of the outermost zone of the AF, where small capillary blood vessels extend from the adjacent longitudinal ligaments between the lamellae of the AF [12] ( Fig 5.2-5 ). All of the studies that have investigated the relative importance of these two sources of nutrition have supported the general consensus that the central region of the end plate is the predominant route of transport for metabolic processes of the disc [13–15]. For small solutes, it has been shown that diffusion is the predominant molecular transport mechanism from the end plate to the nucleus [16]. It has also been suggested that convective solute transport may play a role in nourishing the disc [17–19], but it is believed to be of importance only for the transport of larger solutes [16, 20]. Worthy of note, reduced permeability of end plates is generally seen in association with disc degeneration and age related changes [21]. Both of these

changes may be due to calcification of the end plates and occlusion of the marrow contact channels observed with disease and age [22–24]. Hence, with decreased nutrient and metabolite transport, it is believed that the cells are unable to maintain the matrix necessary for the normal function of the disc, resulting in degeneration [24].

Fig 5.2-5 Immunolocalization of small blood vessels in the outer anulus fibrosus by staining of endothelial cells (CD 31). Normal discs show few small capillaries only in the outer zone of the anulus fibrosus (anti-CD 31, original magnification × 400).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

5.2

91

Aging and pathological degeneration

5

In degenerated discs, the ingrowth of small capillary blood vessels is seen in those areas that present with a disruption of the matrix and the formation of clefts and tears. This, however, is only seen in annular lesions, but not in nuclear clefts. The presence of blood vessels is particularly seen in disc herniation ( Fig 5.2-6 ).

CHANGES OF THE EXTRACELLULAR MATRIX DURING DISC DEGENERATION

The structure and composition of the extracellular matrix are of fundamental significance for the biomechanical properties of the intervertebral disc.

5.1

COLLAGENS

The main structural component of the discal extracellular matrix is represented by collagen which is seen in the different anatomical subsettings with a variable composition of isoforms [11, 25, 26]. Several different collagen types have been identified in normal intervertebral discs (ie, types I, II, III, V, VI, IX, and XI) with some differences between the various anatomical regions. While the overall collagen content in the NP remains fairly constant over the years, that of the AF decreases with advancing age. In addition to these quantitative changes in the collagen content, major alterations occur with the tissue distribution of various collagen isoforms. This affects either those collagen types that are normally present within disc structures but which may reveal a quantitative or qualitative change, or may affect collagen types that are not normally present within the disc tissue.

Fig 5.2-6 Increased formation of small blood vessels in herniated disc material. The immunolocalization of blood vessels in herniated disc tissue reveals the occurrence of small capillary buds close to clefts (anti-CD 31, original magnification × 250).

The most obvious example for an “abnormally” occurring collagen molecule within aging (and degenerating) discs is the appearance of collagen type X in those discs of individuals of advanced age. This collagen molecule—“physiologically” expressed by hypertrophic chondrocytes during the growth period of long bones and therefore present in the notochord of fetal/infantile discs and in the growth zone of the end plate [27]—is completely absent in discs from individuals aged between 20 and 60 years, but reoccurs focally (and in very low amounts) associated with degenerative disc lesions of old-age individuals [28, 29]. Similarly, the authors observed the abortive expression of fragments of basement membrane

92

5

Biology of the spine

collagen type IV in discs from individuals between 20 and 50 years of age. This collagen molecule which normally is exclusively seen in epithelial, endothelial, or pericellular basement membranes seems to indicate partial changes in disc cell differentiation [11]. In addition, there is a significant change in the distribution of collagens with disc degeneration. Likewise, collagen type I is significantly expressed in NP tissue along with enhanced deposition of collagen types III and VI. In parallel, the content of collagen type II seems to be reduced in NP tissue, while the AF also contains more collagen III and VI in the aged and more degenerated situation. Similarly, EP tissue reveals the “abnormal” expression of collagen type III and VI which also parallels morphological signs for tissue disarrangement [11, 25].

5.2

PROTEOGLYCANS

With aging and degeneration comes a marked decrease in proteoglycan content in the nucleus ( Fig 5.2-7 ) and significant alterations in proteoglycan structure [30–32]. This particularly affects aggrecan ( Fig 5.2-8 ), which is the most abundant proteoglycan, and versican, another proteoglycan with the ability to form aggregates with hyaluronate [33]. As a consequence of these changes, the ability of the nucleus tissue to imbibe water and to radially and homogeneously distribute the compressive forces will progressively diminish with increased degeneration [34].

Involvement of aggrecan in disc degeneration

Degeneration

CS 2

CS 1

KS IGD

Fig 5.2-7 Changes in the proteoglycan amount, measured as glycosaminoglycan content, related to age within the different intervertebral disc regions.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

G2 G1

LP

Fig 5.2-8 Aggrecan is represented as a core protein with three globular domains (G1, G2, and G3). The G1 and G2 domains are separated by an interglobular domain (IGD) and the G2 and G3 domains are separated by glycosaminoglycan-attachment domains bearing predominantly keratan sulfate

chains (KS domain) or chondroitin sulfate chains (CS1 and CS2 domains). The CS1 domain possesses a variable number of tandem repeats (polymorphism), so that individuals may have aggrecan core proteins that can be short (left molecule) or long (middle left molecule). Disc degeneration involves proteolytic cleavage of the aggrecan core protein, often within the CS2 domain or IGD, resulting in fragments enriched in the CS1 domain (middle right molecule) or the G1 domain (right molecule). These fragments and the intact aggrecan are localized in the tissue via their interaction with hyaluronan (HA) and stabilized by link proteins (LP). (Courtesy of Dr Peter Roughley, Shriners Hospital for Children, Montreal, Canada.)

5.2

93

Aging and pathological degeneration

6

5.3

NONCOLLAGENOUS PROTEINS

Noncollagenous proteins represent up to 45% of the dry weight of the nucleus pulposus and 25% of the anulus fibrosus in human discs [35]. Several of the identifiable noncollagenous proteins include fibronectin, thrombospondin, and elastin. Preliminary personal analyses provide evidence that there exist age- and degeneration-related changes in at least some of these noncollagenous proteins of the disc. Thus, the authors have shown by immunohistochemistry and nonradioactive in situ hybridization that fibronectin is synthesized in enhanced amounts in those areas of morphological degeneration ( Fig 5.2-9 ). Therefore, it can be assumed that these noncollagenous proteins may be involved in the age- and degenerationassociated matrix remodeling. The specific role of these proteins, however, remains to be elucidated.

Fig 5.2-9 Immunolocalization of fibronectin in degenerated disc material showing a widespread and enhanced deposition of this matrix molecule (antifibronectin; original magnification × 100).

PROTEOLYTIC ACTIVITY IN DEGENERATED DISCS

A major hallmark of discs is the loss of height during disc degeneration [36, 37]. In addition, as already indicated above, the occurrence of clefts and tears is seen to a greater extent in areas of disc degeneration in adults [4]. These observations indicate that matrix molecules are degraded. A perturbation of the turnover of collagen molecules and proteoglycans in turn explains a loss of biomechanical stability and a weakening of the functional properties of affected discs. It is generally accepted that proteinases play a major role in this process [32, 38, 39]. The primary proteinases thought to be involved in the direct destruction of the disc tissue are the matrix metalloproteinases (MMPs) [32, 38, 39]. Within this family of proteolytic enzymes, there exist various groups that differ in their substrate specificity and thus in their proteolytic capacity. Intact interstitial collagen molecules, such as collagen types I, II, or III can only be degraded by the interstitial collagenases, with MMP-1 being the most important and widespread enzyme [32, 38, 39]. The other collagenases that can degrade intact interstitial collagen molecules are synthesized by polymorphonuclear leukocytes (MMP-8) or are not yet analyzed in disc material (eg, MMP-13). PMN-leukocytes are usually not present in disc material. Denatured collagen molecules can be further cleaved by the two gelatinases (MMP-2 and MMP9) while the stromelysins (MMP-3, MMP-10, and MMP-11) degrade both denatured collagen and noncollagenous proteins, such as fibronectin and others. Proteoglycans are also cleaved by the stromelysins, including aggrecan and versican. Once activated, for example, by an enzymatic conversion of the proenzyme to the active enzyme, the MMPs cleave their substrate until they are inhibited by specific tissue-inhibitorsof-matrix-metalloproteinases (TI MPs) which exist in the human body with three isoforms. The balance between MMPs and TIMPs therefore control the level of proteolytic activity.

94

5

Biology of the spine

Under normal conditions, MMPs are synthesized in human lumbar disc tissue on a low, basal level [32, 39]. Previous extensive studies on the occurrence of MMPs (by immunohistochemistry) and their mRNA (by in situ hybridization) suggest that this synthesis is up-regulated in those areas with morphological signs of tissue degeneration, such as cleft formation ( Fig 5.2-10 ). Besides the presence of enhanced amounts of various MMPs, the authors [39] and others [38] have shown by in situ zymography that enhanced foci of tissue proteolysis are indeed associated with cleft formation and disc tissue disruption ( Fig 5.2-11). These observations clearly indicate that the up-regulation and activation of various MMPs is an important step in the degradation of the disc matrix finally leading to a disruption of the disc tissue. However, as yet no information is available on any changes of TIMP levels and it may be speculated that an enhanced synthesis of those TIMPs could represent a therapeutic option to prevent tissue disruption. In addition to the MMPs, two further relevant enzymes have become a focus of disc tissue research. Thus, it has been described that the two aggrecanases (which are members of the ADAMTS-enzyme family) are also found in degenerated disc tissue [32, 38]. Both have similar substrate specificity and their activation may contribute significantly to the proteolysis of proteoglycans. The loss of proteoglycans in turn leads to a reduced water-binding capacity of the affected discs and thereby to a loss of function.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 5.2-10 Immunohistochemical identification of matrix metalloproteinase MMP-1 positive cells in degenerated nuclear disc tissue. This collagenase is expressed in areas with evidence for tissue degeneration in enhanced amounts (anti-MMP-1; original magnification × 400).

Fig 5.2-11 In situ localization of proteolytic activity by in situ zymography of degenerative altered NP material. The dissolution of an underlying gelatine matrix (light areas) suggests the presence and activation of matrix-degrading enzymes (in situ zymography, original magnification × 250).

5.2

95

Aging and pathological degeneration

7

EVIDENCE FOR ENHANCED OXIDATIVE STRESS IN DEGENERATING DISC TISSUE

Recent studies provide increasing evidence that oxidative waste products occur in disc tissue in association with degenerative alterations. This is even obvious by the accumulation of brown products in degenerated disc material [40]. These metabolic products may play a significant role in the aforementioned induction of synthesis and/or activation of cytokines and matrix metalloproteinases in discs.

Fig 5.2-12 Immunolocalization of the oxidation end product carboxymethyllysin (CML) in degenerated nuclear disc material. This metabolic product is the result of accumulated oxidative stress and is normally not present in disc material (anti-CML, original magnification × 250).

Likewise, previous analyses revealed the deposition of a stable and irreversible end product of oxidative reactions, the specific amino acid modification carboxymethyllysin (CML) [41], to be increasingly present in discs with histological degenerative lesions [11] ( Fig 5.2-12 ). The deposition of CML occurred as early as 16 years of age when initial significant histological signs of degeneration are seen. The CML-formation increased steadily in the nucleus pulposus from approximately the 20th to the 85th year of age and was also seen slightly later in annular disc areas and end-plate cartilage [11]. The occurrence of this marker modification indicative of oxidative stress strongly suggests that the intradiscal accumulation of oxidation products is an early and increasing reaction, obviously on the grounds of local hypoxia of disc cells, which in turn may trigger subsequent events of tissue destruction and disarrangement. Although the fi nal proof for a direct role of CML as an inductor for cytokine production has not been accomplished, there is increasing evidence that the formation of CML-modifications of long-living matrix molecules enhances biosynthetic pathways of transmitter substances, matrix molecules and matrix-degrading enzymes [41]. The further investigation of the occurrence and formation of CML-modifications will provide potential insight into the initial metabolic changes that may induce or enhance disc degeneration.

96

5

Biology of the spine

8

PHENOTYPIC CHANGES OF DISC CELLS

Since the matrix is actively remodeled by the disc cells, the aforementioned changes in the composition of the extracellular matrix are induced, controlled, and executed by specific cells. In this regard, it is of interest that within the various anatomical settings significant differences in the phenotype can be assumed. However, the exact defi nition of the discal cellular phenotype is at present very difficult to establish.

When using the complete interstitial matrix composition as a defi nition criterion, the chondrocytes of the EP can be regarded as being closely related to articular chondrocytes, while those of the (inner) AF are ranked into the group of fi brochondrocytes, such as seen in discs or menisci of other body regions. The phenotype of the NP “chondrocytes” is somehow intermediate between the two.

Although all disc cells have previously been classified as chondrocytic cells—with the exception of fibroblastic cells of the outer AF—there seem to be major differences between those “chondrocytes” within various settings. Since there is no general marker that defines chondrocytes, a phenotypic classification is difficult and at present still insufficient. A generally accepted, but not specific marker molecule (which is also present on certain other cell types, such as some types of macrophages, adipocytes, melanocytes, and a few other cell types) is the S100-protein [42]. Using this molecule as a defi ning criterion of disc cells as “chondrocytes”, almost all cells of the NP and EP and most of the cells of the inner AF can be classified as “chondrocytes” which is evidenced by positive cellular labelling.

There exists only indirect evidence for phenotypic changes of disc cells as evidenced from the above-described changes in the pericellular matrix. In addition, a recent personal analysis provides evidence that some disc “chondrocytes” undergo a specific phenotypic change during degeneration. This has been shown by the immunohistochemical analysis of the expression pattern of the CD-68 molecule, a lysosomal protein which is typically found in a series of cell types that all share phagocytic properties. In this study [44], the authors provided evidence that CD-68 positive phagocytic cells occur exclusively in disc tissue with morphological signs of disc degeneration and that those cells are seen mainly adjacent to areas with tissue disruption. These observations suggest that part of the disc cells undergo a phenotypic “switch” to phagocytic cells which may be the result of altered environmental conditions. A further phenotypic analysis of disc cells seems to be required in order to identify further cellular alterations during aging and disc degeneration.

Recently, Sive et al [43] used a panel of the three “marker molecules” collagen type II, aggrecan and Sox 9 to defi ne chondrocytes in disc material. Using these markers and in situ hybridization techniques to identify mRNA-expression, the authors describe the presence of all three markers in normal NP material, but a reduction of aggrecan-mRNA expression in degenerated nuclear samples suggesting some changes in the cellular phenotype during degeneration.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Furthermore, recent studies [45, 46] suggest that at sites of herniation inflammatory cells locally invade the disc. These are mainly macrophages, but also mast cells that are assumed to liberate factors that may promote sensory irritation.

5.2

Aging and pathological degeneration

9

MODULATION OF DISC CELLS BY CYTOKINES AND GROWTH FACTORS

Until now more and more information has been gathered about the composition of the disc during aging and degeneration, and certain important alterations in the associated cellular and molecular features have been identified. However, only very little is known about the underlying modulating factors, their potential source and effects. This is mainly due to a technically problematic identification and quantitation of those factors within disc tissue material because the tissue concentrations may be very low and the turnover rate of those factors may be high. Therefore, only a few studies are available that use either isolated cells cultivated under in vitro conditions or tissue samples for (immuno-) morphological analysis [47–49]. In this system it can be assumed that a multitude of potential factors is available that may influence the growth pattern and biosynthetic capacity of disc cells. Out of this bulk of factors, some may be selected which may have a significant influence on disc metabolism. With regard to this, the authors previously suggested that TGF-ß may be of interest, since this cytokine plays an important role in controlling cell proliferation, MMP synthesis, and production of matrix molecules, and all three parameters seem to be disturbed during disc degeneration. Although only limited information is available, in vitro studies revealed an enhanced responsiveness of disc cells to TGF-ß1 [50, 51] suggesting increased sensitivity of isolated disc cells to this factor. Recently, the authors detected enhanced amounts of TGF-ß1 producing cells within discs showing an increasing degree of morphological degeneration. This was confi rmed by immunohistochemistry (for TGF-ß1 protein) and nonradioactive in situ hybridization (for TGF-ß1-mRNA).

97

Thereby, the notion was corroborated that an enhanced TGFß-expression may contribute to increased matrix protein synthesis and thereby may promote matrix remodelling. Although this study clearly showed that TGF-ß1 is synthesized by local disc cells, we do not know the underlying stimulus for its production. Furthermore, initial data are available about the temporospatial distribution of a further cytokine within the disc which the authors assume to play a major role in the enhancement of any “inflammatory” reaction [49]. Similarly, data has been accumulated that support such an “inflammatory” reaction. Therefore, TNF-α has been investigated immunomorphologically in various human lumbar disc samples and has also been shown to be increasingly expressed in association with disc degeneration. Since TNF-α is a potent proinflammatory cytokine, this enhanced expression may indicate an inflammatory reaction of local disc cells that may stimulate the liberation of further, pain-inducing neurotropic factors. Ultimately, the activation of the TNF-α pathway may lead to pain induction in the juxta-discal spaces that are well innervated. The confi rmation of these data and the corroboration of this hypothesis seem to be of utmost importance for the understanding of discogenic pain induction.

98

5

Biology of the spine

10

A CURRENT CONCEPT OF PAINFUL DISC DEGENERATION

Taking all these observations and experimental information together, the authors feel that a current concept of degenerative disc pathogenesis and the generation of discogenic pain can be developed. Therefore, the following sequence of events may be crucial in inducing disc degeneration: Beginning in early childhood, the closure of the blood vessels that enter the discs leads to a progressive “malnourishment” of disc cells [4]. This is significantly aggravated during the period of longitudinal growth of the body during puberty and it may also be worsened by structural disarrangements of the EP including calcifications, etc. These structural changes in the EP may lead to an occlusion of the marrow contact channels which in turn further diminishes nutritional supply. Finally, the distance for the supply of the disc cells by diffusion extends to such a length that nuclear disc cells suffer from severe and irreversible hypoxic damage leading to a progressive collapse of those cells. This may be reflected by the accumulation of oxidative waste products such as the CMLmodification of long-living proteins [11]. In order to adapt to these novel conditions of impaired nutritional supply, part of those cells may change their phenotype to a phagocytic cell type which is able to produce enhanced amounts of matrix degrading MMPs [38, 44]. Then, the cleavage of collagen molecules leads to a disruption of the disc structure and the formation of major discal clefts and tears. Although these clefts and tears may enhance the influx of nutritional supply, the functional integrity of the disc structure is lost. The mechanical properties of the disc become insufficient so that fi nally the complete motion segment loses its function. This phenotypic change along with the synthesis of proinflammatory cytokines, eg, TNF-α, leads to the induction of an “inflammatory” reaction and it can be assumed that the extensive clefts and tears may promote the rapid spread

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

of those cytokines to sensory nerve endings in the wellinnervated regions surrounding the spinal canal. Thereby, significant pain may be induced leading to disabling of the affected individual. However, the high prevalence of asymptomatic disc alterations indicates that additional factors are required which ultimately turn an asymptomatic condition into painful disc degeneration. Although several aspects of this concept are still speculative, the authors believe that additional studies on this issue may further support this hypothesis. Only the elucidation of those presumed pathomechanisms may provide a causal option for a therapeutic interference.

5.2

11

BIBLIOGRAPHY

Humzah MD, Soames RW (1988) Human intervertebral disc: structure and function. Anat Rec; 220(4):337–356. Review. 2. Buckwalter JA (1995) Aging and degeneration of the human intervertebral disc. Spine; 20(11):1307–1314. Review. 3. Coventry MB, Ghormley RK, Kernohan JW (1945) The intervertebral disc: Its microscopic anatomy and pathology. Part II. Changes in the intervertebral disc concomittant with age. J Bone Joint Surg; 27A:233–247. 4. Boos N, Weissbach S, Rohrbach H, et al (2002) Classifi cation of age-related changes in lumbar intervertebral discs: 2002 Volvo Award in basic science. Spine ; 27(23):2631–2644. 5. Vernon-Roberts B (1988) Disc pathology and disease states. Gosh P (ed) The Biology of the Intervertebral Disc. Boca Raton, Florida: CRC Press, 73–119. 6. Carragee EJ, Hannibal M (2004) Diagnostic evaluation of low back pain. Orthop Clin North Am; 35(1):7–16. 7. Vernon-Roberts B (1992) Age-related and degenerative pathology of intervertebral discs and apophyseal joints. Jayson MIV (ed) The Lumbar Spine and Back Pain. Edinburgh: Churchill Livingstone, 17–41. 8. Melrose J, Roberts S, Smith S, et al (2002) Increased nerve and blood vessel ingrowth associated with proteoglycan depletion in an ovine annular lesion model of experimental disc degeneration. Spine; 27(12):1278–1285. 9. Johnson WE, Evans H, Menage J, et al (2001) Immunohistochemical detection of Schwann cells in innervated and vascularized human intervertebral discs. Spine; 26(23):2550–2557. 10. Freemont AJ, Peacock TE, Goupille P, et al (1997) Nerve ingrowth into diseased intervertebral disc in chronic back pain. Lancet; 350(9072):178–181. 11. Nerlich AG, Schleicher ED, Boos N (1997) 1997 Volvo Award winner in basic science studies. Immunohistologic markers for age-related changes of human lumbar intervertebral discs. Spine; 22(24):2781–2795. 1.

99

Aging and pathological degeneration

12. Boos N, Nerlich AG (2005) Immunolocalization of blood vessels in human intervertebral discs. (Unpublished data.) 13. Holm S, Moroudas A, Urban JP, et al (1981) Nutrition of the intervertebral disc: solute transport and metabolism. Connect Tissue Res; 8(2):101–119. 14. Ogata K, Whiteside LA (1981) 1980 Volvo award winner in basic science. Nutritional pathways of the intervertebral disc. An experimental study using hydrogen washout technique. Spine; 6(3):211–216. 15. Crock HV, Goldwasser M (1984) Anatomic studies of the circulation in the region of the vertebral end-plate in adult Greyhound dogs. Spine; 9(7):702–706. 16. Maroudas A, Stockwell RA, Nachemson A, et al (1975) Factors involved in the nutrition of the human lumbar intervertebral disc: cellularity and diffusion of glucose in vitro. J Anat; 120(1):113–130. 17. Urban JP, Holm S, Maroudas A, et al (1982) Nutrition of the intervertebral disc: effect of fl uid fl ow on solute transport. Clin Orthop; (170):296–302. 18. Malko JA, Hutton WC, Fajman WA (1999) An in vivo magnetic resonance imaging study of changes in the volume (and fl uid content) of the lumbar intervertebral discs during a simulated diurnal load cycle. Spine; 24(10):1015–1022. 19. Ayotte DC, Ito K, Tepic S (2001) Directiondependent resistance to fl ow in the endplate of the intervertebral disc: an ex vivo study. J Orthop Res; 19(6):1073–1077. 20. Holm S, Maroudas A, Urban JP, et al (1981) Nutrition of the intervertebral disc: solute transport and metabolism. Connect Tissue Res; 8(2):101–119. 21. Nachemson A, Lewin T, Maroudas A, et al (1970) In vitro diffusion of dye through the end-plates and the anulus fi brosus of human lumbar inter-vertebral discs. Acta Orthop Scand; 41(6):589–607.

22. Roberts S, Menage J, Eisenstein SM (1993) The cartilage end plate and intervertebral disc in scoliosis: calcifi cation and other sequelae. J Orthop Res; 11(5):747–757. 23. Roberts S, Urban JP, Evans H, et al (1996) Transport properties of the human cartilage endplate in relation to its composition and calcifi cation. Spine; 21(4):415–420. 24. Horner HA, Urban JP (2001) 2001 Volvo Award Winner in Basic Science Studies: Effect of nutrient supply on the viability of cells from the nucleus pulposus of the intervertebral disc. Spine; 26(23):2543–2549. 25. Roberts S, Menage J, Duance V, et al (1991) 1991 Volvo Award winner in basic science. Collagen types around the cells of the intervertebral disc and cartilage end plate: an immunolocalization study. Spine; 16(9):1030–1038. 26. Kääpä E, Holm S, Han X, et al (1994) Collagens in the injured porcine intervertebral disc. J Orthop Res; 12(1):93–102. 27. Nerlich AG, Kirsch T, Wiest I, et al (1992) Localization of collagen X in human fetal and juvenile articular cartilage and bone. Histochemistry; 98(5):275–281. 28. Boos N, Nerlich AG, Wiest I, et al (1997) Immunolocalization of type X collagen in human lumbar intervertebral discs during ageing and degeneration. Histochem Cell Biol; 108(6):471–480. 29. Aigner T, Greskötter KR, Fairbank JC, et al (1998) Variation with age in the pattern of type X collagen expression in normal and scoliotic human intervertebral discs. Calcif Tissue Int; 63(3):263–268. 30. Antoniou J, Steffen T, Nelson F, et al (1996) The human lumbar intervertebral disc: evidence for changes in the biosynthesis and denaturation of the extracellular matrix with growth, maturation, ageing, and degeneration. J Clin Invest; 98(4):996–1003. 31. Sztrolovics R, Alini M, Mort JS, et al (1999) Age-related changes in fi bromodulin and lumican in human intervertebral discs. Spine; 24(17):1765–1771.

100

5

Biology of the spine

32. Sztrolovics R, Alini M, Mort JS, et al (1997) Aggrecan degradation in human intervertebral disc and articular cartilage. Biochem J; 326:235–241. 33. Melrose J, Ghosh P, Taylor TK (2001) A comparative analysis of the differential spatial and temporal distribution of the large (aggrecan, versican) and small (decorin, biglycan, fi bromodulin) proteoglycans of the intervertebral disc. J Anat; 198:3–15. 34. Roughley P, Alini M, Antoniou J (2002) The role of proteoglycan in aging, degeneration and repair of the intervertebral disc. Biochem Soc Trans; 30:869–874. Review. 35. Melrose J, Ghosh P (1988) The noncollagenous proteins of the intervertebral disc.

Ghosh P (ed), The Biology of the Intervertebral Disc, vol 2. Boca Raton: CRC Press, 190–237. 36. Thompson JP, Pearce RH, Schechter MT, et al (1990) Preliminary evaluation of a scheme for grading the gross morphology of the human intervertebral disc. Spine; 15(5):411–415. 37. Brinckmann P, Grootenboer H (1991) Change of disc height, radial disc bulge and intradiscal pressure from discectomy. Spine; 16(6):641–646. 38. Roberts S, Caterson B, Menage J, et al (2000) Matrix metalloproteinases and aggrecanase: their role in disorders of the human intervertebral disc. Spine; 25(23):3005–3013. 39. Weiler C, Nerlich AG, Zipperer J, et al (2002) 2002 SSE award competition in basic science: Expression of major matrix metalloproteinases is associated with intervertebral disc degeneration and resorption. Eur Spine J; 11(4):308–320. 40. Ishii T, Tsuji H, Sano A, et al (1991) Histochemical and ultrastructural observations on brown degeneration of human intervertebral disc. J Orthop Res; 9(1):78–90.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

41. Schleicher ED, Wagner E, Nerlich AG (1997) Increased accumulation of the glycoxidation product N(epsilon)-(carboxymethyl)lysine in human tissues in diabetes and aging. J Clin Invest; 99(3):457–468. 42. Kahn HJ, Marks A, Thom H, et al (1983) Role of antibody to S100 protein in diagnostic pathology. Am J Clin Pathol; 79(3):341–347. 43. Sive JI, Baird P, Jeziorsk M, et al (2002) Expression of chondrocyte markers by cells of normal and degenerate intervertebral discs. Mol Pathol; 55(2):91–97. 44. Nerlich AG, Weiler C, Zipperer J, et al (2002) Immunolocalization of phagocytic cells in normal and degenerated intervertebral discs. Spine; 27(22):2484–2490. 45. Habtemariam A, Gronblad M, Virri J, et al (1998) A comparative immunohistochemical study of infl ammatory cells in acute-stage and chronic-stage disc herniations. Spine; 23(20):2159–2165. 46. Freemont AJ, Jeziorska M, Hoyland JA, et al (2002) Mast cells in the pathogenesis of chronic back pain: a hypothesis. J Pathol; 197(3):281–285. 47. Kang JD, Georgescu HI, McIntyre-Larkin L, et al (1996) Herniated lumbar intervertebral discs spontaneously produce matrix metalloproteinases, nitric oxide, interleukin-6, and prostaglandin E2. Spine; 21(3):271–277. 48. Miyamoto H, Saura R, Harada T, et al (2000) The role of cyclooxygenase-2 and infl ammatory cytokines in pain induction of herniated lumbar intervertebral disc. Kobe J Med Sci; 46:13–28. 49. Burke JG, Watson RW, McCormack D, et al (2002) Intervertebral discs which cause low back pain secrete high levels of proinfl ammatory mediators. J Bone Joint Surg Br; 84(2):196–201. 50. Gruber HE, Stasky AA, Hanley EN (1997) Characterization and phenotypic stability of human disc cells in vitro. Matrix Biol; 16(5):285–288.

51. Alini M, Li W, Markovic P, et al (2003) The potential and limitations of a cell-seeded collagen/hyluronan scaffold to engineer an intervertebral disc-like matrix. Spine; 28(5):446–454.

101

102

5 5.3

BIOLOGY OF THE SPINE BIOLOGY OF THE OSTEOPOROTIC SPINE

1

Epidemiology of osteoporosis …………………………………………………………………………… 103

2

Definition of osteoporosis ……………………………………………………………………………… 104

3 3.1 3.2

Etiology ……………………………………………………………………………………………………… 105 Primary osteoporosis……………………………………………………………………………………… 105 Secondary osteoporosis ………………………………………………………………………………… 105

4 4.1 4.2 4.3

Diagnostic assessment …………………………………………………………………………………… 108 Imaging studies …………………………………………………………………………………………… 108 Laboratory workup ………………………………………………………………………………………… 111 Bone biopsy ………………………………………………………………………………………………… 112

5

Treatment ……………………………………………………………………………………………………

113

6

Bibliography …………………………………………………………………………………………………

114

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

103

Maximilian A Dambacher, Norbert Boos

5

BIOLOGY OF THE SPINE

5.3

1

BIOLOGY OF THE OSTEOPOROTIC SPINE

EPIDEMIOLOGY OF OSTEOPOROSIS

The proportion of older people in the total population is increasing with rising life expectancy and therefore, the importance of osteoporosis, a disease that manifests itself with ageing, is also increasing [1]. In 1900 the life expectancy of a newborn girl was less than 50 years; by 1999 it was 83 years. The number of women over the age of 65 will double by the year 2040. Today, osteoporosis is the most common skeletal disorder. Osteoporosis is so prevalent that vertebral fractures occur in 30% of all postmenopausal women and 20% of men over 50 [2–5]. Osteoporosis has become an enormous sociomedical problem. Unfortunately, only about 20–30% of the patients are treated. Management of the osteoporosis problem will only be possible with a successful diagnosis of osteoporosis, for example, with quantitative methods before the disease becomes evident in conventional x-rays, and then starting an appropriate “prophylactic” therapy.

The main problem in future will be to identify patients at risk, ie, patients who start to develop osteoporosis after menopause. Women at risk today include: those with premature menopause, a diet that is deficient in calcium/vitamin D3, a lack of physical activity, a family history, and smokers. In the EPOS (European prospective osteoporosis study), which used over 7,000 men and women, it was found that after 3.6 years subjects who had a lower bone density (T-score less than –2.5 SD) at the beginning of the study had osteoporotic fractures of the spine 1.4 times more often than men and women with normal bone density [6–8]. The study complies with the requirements of evidence-based medicine in all criteria. This means that patients with a fracture risk can certainly be identified and then provided with an effective treatment. In 1998 the European Parliament passed the resolution that osteodensitometry must be covered by the national health services and made available to the population in order to identify women with an osteoporosis risk.

104

5

Biology of the spine

2

DEFINITION OF OSTEOPOROSIS

The human skeleton consists (in about equal parts) of basic substance and hydroxyapatite. In osteopenia and osteoporosis (in contrast to osteomalacia) this ratio is more or less preserved, but total bone mass is reduced.

In contrast to earlier years, the focus is now more on the pathological changes in structure, eg, how the trabeculae are linked, especially because they can now be made visible and measured not only in vitro ( Fig 5.3-1a–b ), but also in vivo.

The Consensus Conferences in Copenhagen 1990, Hong Kong 1993, and Amsterdam 1996, defined osteoporosis: “Osteoporosis is a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue with a consequent increase in bone fragility and susceptibility to fracture. The magnitude of peak bone mass and the rate of duration of bone loss determine the likelihood of developing osteoporosis.”

In addition, the WHO quantifies osteoporosis based on a densitometric bone density assessment ( Fig 5.3-2 ):

This defi nition contains three key elements of osteoporosis [9]: • Bone mass (how much is still left) • Loss of bone mass (how much is lost) • Microstructural changes (how is the bone structured)

a

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

• Normal: T-score • Osteopenia: T-score • Osteoporosis: T-score

+ or –1 SD –1 to –2.5 SD – 2.5 or less

The authors defi ne a state without fractures as preclinical osteoporosis and a state with fractures as manifest osteoporosis (–2.5 SD or less). In general, osteoporosis develops in episodes. In postmenopause, a high bone turnover (increased formation and resorption) is identical with rapid bone loss and is referred to as “fast-bone-loser”. Although rapid trabecular bone loss occurs with an annual rate of more than 3.5%, after

Fig 5.3-1a–b Normal (a) and osteoporotic (b) bone structures (cancellous bone, vertebra) (µCT 20, Scanco Medical AG, Zürich).

5.3

105

Biology of the osteoporotic spine

the onset of menopause (in terms of the total group). Only 34% of women are at risk to develop osteoporosis. Vice versa, stability in severe age-related osteoporosis (formerly referred to as senile osteoporosis) has proved erroneous. In this form of osteoporosis, a fast-loser state is found in about 75% of the patients.

3

ETIOLOGY

3.1

PRIMARY OSTEOPOROSIS

The etiology of primary osteoporosis is still not completely understood. The most frequent form is postmenopausal osteoporosis (90%). Pathogenetic factors for the development of osteoporosis are: • Estrogen deficiency • Reduced calcium resorption • Vitamin D deficiency

BMD (g/cm²)

In contrast to women, the cause of osteoporosis in men is identified substantially less frequently with lack of testosterone being the most common cause. 1.4 1.3 +2 SD 1.2 +1 SD 1.1 1.0 –1 SD 0.9 –2 SD 0.8 0.7 0.6 T-score = –3 SD 0.5 = osteoporosis 0.4 20 40

3.2 Z-score = –2 SD = normal 60

80

100

Age (years) BMD: bone mineral density

Fig 5.3-2 Definition of T-score and Z-score in osteodensitometry. T-score expresses the deviation of a measurement from the mean value of healthy women aged 20–45 (peak bone mass) in mean ± SD. Z-score expresses the deviation of a measurement from the mean average bone density of a peer population in standard deviation (± SD). This Z-score is hardly used any more today.

SECONDARY OSTEOPOROSIS

In contrast to primary osteoporosis, secondary osteoporosis ( Table 5.3-1 ) is secondary to a preexisting disease (eg, hypogonadism in women, anorexia nervosa, osteoporosis in professional dancers and athletes, or drug treatment) [10]. Corticosteroid osteoporosis

The most important secondary osteoporosis is caused by steroid treatment and is similar to the osteroporosis caused by Cushing syndrome. The likelihood of developing osteoporosis due to steroid treatment is related to the dose and the time period of use. However, there exits a substantial individual variation; not every patient treated with steroids develops osteoporosis. Some evidence shows that steroid treatment in doses less than 7.5 mg per day over 1 year is not a risk factor [11].

106

5

Biology of the spine

Synopsis of secondary osteoporosis Inflammatory disease

Synopsis of secondary osteoporosis

Rheumatoid ar thritis

Miscellaneous

Pregnanc y/lac tation Ank ylosing spondylitis Hypercalciuric nephrolithiasis

Drugs

Alcohol

Inflammator y bowel disease Cystic fibrosis Bone marrow disorders

Multiple myeloma Mastoc y tosis

C af feine

Leukemia

Anticonvulsant s

Disorders associated

Parkinson disease

with immobilization

Poliomyelitis Cerebral palsy

Methotrexate Heparin Cyclosporin

Paraplegia Defective synthesis of

Osteogenesis imper fec ta

connective tissue

Mar fan syndrome Homoc ystinuria

Disorders associated with hypogonadism

Athletic amenorrhea (marathon runner osteoporosis) Hemochromatosis Turner syndrome Klinefelter syndrome Postchemotherapy Hypopituitarism

Disorders associated with

Anorexia ner vosa

low body weight

Diabetes mellitus t ype I

Disorders associated

Coeliac disease

with malabsorption

Postgastrec tomy Liver disease Total parenteral nutrition

Endocrinological disorders

Thyreotoxicosis Hyperparathyroidism

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Table 5.3-1 Secondary osteoporosis as a consequence of a preexisting disease.

5.3

107

Biology of the osteoporotic spine

Hypogonadism in women

Anorexia nervosa

Apart from postmenopausal osteoporosis, accelerated bone loss may also occur in women after ovarectomy with estrogen deficiency. Remember that an estrogen deficiency may be present after a hysterectomy, even if the ovaries are left surgically intact, depending on the surgical technique (iatrogenic compromise of ovarian blood supply). Therefore, standard practice should determine the estrogen and gonadotropin levels when clinical signs of hypogonadism are present, even if the patient denies that the ovaries were removed at the hysterectomy.

Women with sustained anorexia nervosa frequently present a very severe, predominantly cancellous osteoporosis. The treatment of these seriously underweight patients with special nutrition often results in a further marked loss of cancellous bone, the cortical bone being less involved. In this context, it should be emphasized that esophageal/gastrointestinal symptoms are relevant side effects of bisphosphonates in daily practice, which in turn could lead to further decreased food intake. Marathon runner osteoporosis

Turner syndrome

This genetic anomaly causes a congenital form of hypogonadism in women (gonadal dysgenesis). These patients have normal female genitals, but rudimentary gonads without any function. Radiologically a coarse bone dystrophy with kyphosis and hypostosis can be found. If the syndrome is diagnosed late, eg, in adulthood, the authors frequently fi nd that estrogen replacement, which would be the most obvious treatment, is not recommended in order to avoid undesirable psychological and physical perturbations.

This functional disorder of the ovaries that causes hyperprolactinemic amenorrhea is commonly found in dancers and top athletes. Estrogen replacement therapy is necessary if such an amenorrhea lasts more than 6 months. Surprisingly, this problem is refuted or played down to an extent. This form of secondary osteoporosis is closely associated with anorexia nervosa and has similar psychological behavior patterns, including physical hyperactivity (best described as “being driven”). Corticosteroid osteoporosis, osteoporosis associated with anorexia nervosa, and marathon runners osteoporosis, are characterized by a loss of cancellous bone rather than a loss of cortical bone.

108

5

Biology of the spine

4

DIAGNOSTIC ASSESSMENT

4.1

IMAGING STUDIES

Standard x-rays

Decreased bone density results in enhanced radiation permeability and cannot be detected radiologically until the loss of substance reaches a level of about 30–50%. Thereby, the patients’ physical health (eg, obesity) and the radiological technique also play a role. In addition, the assessment of a reduced radiological shadow density may vary considerably. Therefore, the criterion “reduced radiological shadow density” should no longer be used. The diagnosis of osteoporosis using standard x-rays has thus been surpassed by modern densitometry methods, which are considerably more sensitive and reliable. However, standard x-rays still play a major role in the assessment of spinal deformity and of fractures.

is affected, while in primary hyperparathyroidism cortical bone is mainly affected. In hyperprolactinemic amenorrhea in young athletes (“marathon runner osteoporosis”) a total loss of cancellous bone may occur. The rate of cancellous bone loss in menopause is about 1% per year in healthy women, 1–3.5% in “slow-loser” patients and more than 3.5% in “fastloser” patients. Thus, quantitative densitometry methods must

DXA

Measurement sites

Lumbar spine, proximal femur, radius

Radius, tibia, hand

Parameters

Integral cor tical with cancellous bone

Elec tive cancellous and cor tical bone struc ture parameters

Dimension

g /cm 2 (sur face value)

mg /m 3 (volume value)

Reproducibilit y

± 1–2% (young healthy subjec t s)

± 0.3% (mixed collec tive)

Accurac y

3– 6 %

< 1%

Exposure ( mSv)

< 0.05

< 0.1

Time/site ( min )

approximately 10 min

Densitometry

Two methods are widely used, ie, dual x-ray absorptiometry (DXA) [10, 12] and high-resolution quantitative computed tomography hrpQCT [13]. A comparison of these two techniques, namely DXA and hrpQCT in multi-layer technique, is presented in Table 5.3-2 . This demonstrates the great differences between the individual methods, both in terms of reproducibility, exposure, and location of measurement. The thin and multi-layer technique hrpQCT is the most sensitive. hrpQCT allows cancellous and cortical bone to be measured either together or separately at peripheral sites (radius and tibia) with minimum radiation exposure and with a low, thus optimal reproducibility. This is important because cancellous and cortical bone represent two different systems that may change in different ways and rates, both with regard to the development of osteoporosis and the therapy. For example, in steroid osteoporosis and osteoporosis associated with anorexia nervosa mainly the cancellous and less of the cortical bone

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

hrpQCT multi-layer/ thin-layer

4 slices/8 min 16 slices/16 min

Table 5.3-2 Comparison between the dual x-ray absorptiometry (DXA) and high-resolution peripheral quantitative computed tomography (hrpQCT) using thin- and multi-layer technique.

5.3

109

Biology of the osteoporotic spine

have a very good reproducibility to measure these differences and provide useful information for therapy decisions. The cancellous bone measured at the distal radius correlates well with the cancellous bone of the lumbar spine. The indications for densitometry are summarized in Table 5.3-3.

Indications for densitometry Confirmed indications

Manifest osteoporosis with fracture Long-term glucocorticoid treatment

Of particular interest is the reproducibility of the measurements to assess the progression of the disease (eg, slow vs fast loser) and the treatment effect (eg, change in medication). The long-term reproducibility determines the minimum assessment intervals. The reproducibility data provided by the manufacturer are normally attained by a highly qualified investigator in healthy subjects, at short intervals, and under laboratory conditions; this is why it often deviates considerably from the long-term reproducibility in clinical practice. The importance of reproducibility is illustrated by the following example: A patient with severe osteoporosis (eg, who has already lost more than 50% of bone mass) is examined using an osteodensitometry method with a long-term reproducibility in healthy subjects of ± 2% (eg, DXA). In this case, the question arises as to which time interval should be chosen, eg, if a minimum change of ± 3% per year is to be detected with 95% times certainty ( Table 5.3-4 ). One needs to wait 45 months before a change of ± 3% can be detected with certainty (95% confidence). In a clinical setting this is unacceptable for obvious reasons. If a method has a reproducibility of ± 0.3% (eg, hrpQCT) than one only has to wait 7 months under the same conditions as above, thereby facilitating therapeutic decision making ( Table 5.3-4 ).

Hypogonadism Anorexia Chronic gastrointestinal disorders (eg, Crohn disease, malabsorption) Primary hyperparathyroidism (unclear surgical indication, bone involvement) Organ transplant (especially heart, lung, liver) Osteogenesis imperfecta Evaluation of therapy success Identification of slow-loser and fast-loser patients Possible indications

Family history of osteoporotic fractures Estrogen deficiency syndrome Menopause before the age of 45 Primary and secondary amenorrhea Fractures after inadequate trauma Radiological signs of osteoporosis (conventional x-ray)

Table 5.3-3 These indications vary from country to country, depending in particular on the health authorities and national health services.

110

5

Biology of the spine

Relative bone density

Reproducibility ± 0.3%

± 0.5%

± 1.0%

± 1.5%

± 2.0%

± 2.5%

± 3.0%

± 3.5%

± 4.0%

± 4.5%

± 5.0%

120

3

5

9

14

19

24

28

33

38

42

47

110

3

5

10

15

21

26

31

36

41

46

51

100

3

6

11

17

23

28

34

40

45

51

57

90

4

6

13

19

25

31

38

44

50

57

63

80

4

7

14

21

28

35

42

50

57

64

71

70

5

8

16

24

32

40

49

57

65

73

81

60

6

9

19

28

38

47

57

66

75

85

94

50

7

11

23

34

45

57

68

79

91

102

113

40

8

14

28

42

57

71

85

99

113

127

142

Table 5.3-4 Minimum time intervals (in months) depending on bone density and reproducibility for identifying bone loss with a magnitude of ± 3% on the 95% confidence level.

Bone scan

Magnetic resonance imaging (MRI)

A bone scan is indicated in cases where a generalized bone disease, tumor, or infection is suspected. Although this method has a high sensitivity, its specificity is low. On the other hand, about twice as many metastases can be identified with scanning as with x-ray. The major advantage of a bone scan is that the relevant parts of the body (such as the spine, pelvis, skull, ribs, and the proximal tibia) can be imaged in a single examination [14].

MRI is more sensitive than bone scans in the assessment of metastasis. A further advantage is the sensitivity to scintigraphically negative lesions, such as multiple myeloma, and the more precise anatomical demonstration of any infective or tumorous alterations [15, 16]. A fluid sensitive sequence (eg, STIR) can be helpful in deciding whether a vertebral compression fracture is acute, subacute, or already healed.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

5.3

Biology of the osteoporotic spine

4.2

LABORATORY WORKUP

The laboratory work primarily aims to rule out secondary osteoporosis and is summarized in Table 5.3-5 . A further detailed differentiation of the bone metabolism can be investigated using metabolic products and enzymes formed by the bone cells, and products of the bone matrix released into the serum, mainly during bone resorption [17].

111

of rapid degradation, the samples must be processed very quickly. Increased levels are found in renal failure and during treatment with calcitriol. Procollagen/propeptide

As mentioned above, the organic matrix consists of about 90% collagen type I. During integration in the bone matrix, aminoterminal, and carboxy-terminal fragments are separated from the procollagen type I molecule and secreted into the serum.

The following parameters for bone formation are available in clinical practice: Alkaline phosphatase

Alkaline phosphatase is not only found in bone, but also in the liver, kidneys, intestine, and placenta (alkaline phosphatase, isoenzymes). The amino acid sequence is identical, except differences in the tertiary structure. Alkaline phosphatase of the bone is localized in the membranes of the osteoblasts and it also plays a role in the mineralization of the osteoid. The enzyme has no circadian rhythm and is relatively stable after drawing blood. Raised serum levels are found in the presence of a increased bone turnover rate or mineralization disorder. In osteoporosis the values are usually within the normal range or slightly raised. Osteocalcin

Synthesis of osteocalcin is controlled by calcitriol. Osteocalcin is 10–20% of the noncollagen proteins in the matrix. The precise function is still unknown. This probably plays a role in the mineralization of the osteoid. Osteocalin is integrated in the bone matrix, and about 20–30% is released into the serum. It can be quantified with specific immunoassays. While the half-life of alkaline phosphatase is 1–2 days, the half-life of osteocalcin is 4 minutes. Osteocalcin has a circadian rhythm with a maximum in the early hours of the morning. Because

Level 1

Level 2

Level 3

Exclusion of

Clinical suspicion

Dynamics of

secondary

of secondary

bone metabolism

osteoporosis

osteoporosis

C a, P,

25(OH) D3 (malabsorption),

alkaline phosphatase,

parathyroid hormone,

osteocalcin,

T4,

creatinine,

TSH,

bilirubin,

testosterone,

SGOT/SGPT,

1,25(OH)2D3 (renal osteodystrophy)

er y throc y te sedimentation rate,

Osteocalcin (bone formation parameter), desox ypyridinoline/ creatinine ratio (bone resorption parameter)

serum immuno elec trophoresis, blood count, urine status

Table 5.3-5 Laboratory workup in patients with suspected osteoporosis.

112

5

Biology of the spine

The carboxy- and amino-terminal fragments can be measured in the serum using immunoassays. Thus, they represent the osteoblast collagen synthesis. There is a circadian rhythm, but the stability after taking the sample is greater than that of osteocalcin. The clinical value has not yet been fully explored. The following parameters for bone resorption are available in clinical practice:

4.3

BONE BIOPSY

Since bone biopsy requires a surgical intervention and the processing of the biopsy is very complex, especially where the diagnosis of the activity of osteoporosis is concerned, this comes quite late in the diagnostic workup [18]. Recently, fewer bone biopsies have been performed because of the diagnostic accuracy of CT, MRI, and FDG-PET. A bone biopsy is indicated:

Hydroxyproline

Hydroxyproline is no longer used as a marker for bone resorption, since it requires a 3-day proline-free diet for measurement and the collection of 24-hour urine. Pyridinoline cross-links

Pyridinoline and desoxypyridinoline are bone-specific and exhibit a circadian rhythm with the highest levels found in the early morning and the lowest in the afternoon. These substances are released during bone resorption and eliminated as free amino acids or as telopeptides. A specific diet prior to the urine sampling period is not required. The analysis method in urine is very complex. It may be expected that the pyridinoline cross-links in the serum will be determined more often in future. The advantage of this cross-link determination method in the serum is that a single blood sample could also be used to measure osteocalcin as a formation parameter, and cross-links as a resorption parameter. Tartrate-resistant acid phosphatase

This enzyme is released in the osteoclasts, the prostate, and the hematopoietic system. Tartrate-resistant acid phosphatase is very unstable and samples must be processed immediately.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

• If a bone scan or MRI indicates “malignant” growth. • If a hematological disorder is suspected. • If the previous tests did not allow a clear distinction between osteoporosis and osteomalacia. • In all cases of “unusual” osteoporosis, eg, in young women who are still menstruating or in young men. The prerequisite for a morphometric evaluation of bone biopsies is, however, that the removed biopsy specimen is large enough and has not been destroyed for the evaluation of bone structures. When processing the samples in order that the tetracycline marker for identifying the mineralization front remains visible, they must not be decalcified. Only preparations that have not been decalcified will allow one to distinguish whether osteoidosis or true osteomalacia is present (tetracycline marker present or diffuse). Osteoidosis is found, for example, with high bone turnover (fluoride therapy), osteomalacia with malabsorption and maldigestion. The tetracycline marker is imperative for a correct interpretation of the bone biopsy. Preparations that have not been decalcified allow the calculation of morphometric structure parameters (if microCT/microtomography is not available), and in par-

5.3

113

Biology of the osteoporotic spine

5

ticular the quantitative measurement of the osteoblasts and osteoclasts. These parameters can then be used later for a specific therapy and show whether the bone loss shown objectively by quantitative computed tomography is due to osteoblast insufficiency or an increase in osteoclasts.

Drugs influencing bone mass Substances that stimulate bone formation

Fluorides Anabolics Estrogens at high doses (eg, implant s) D -hormone preparations PTH injec tions Substances that inhibit bone resorption

Estrogens C alcitonin Bisphosphonates Anabolics (anticatabolics) D -hormone preparations C alcium/vitamin D

Table 5.3-6

TREATMENT

Guidelines for the treatment of osteoporosis are far beyond the scope of this chapter; therefore, the authors only briefly review some general principles ( Table 5.3-6 ). Obviously treatment of osteoporosis with bone stimulating and antiresorptive substances must be differentiated. The osteoporosis of each patient must be examined individually and therapeutic measures should be based on the dynamics of the disease [19]. In general, osteoblasts can be stimulated with fluorides and PTH (parathormone) given subcutaneously, while osteoclasts can be inhibited with estrogens, calcitonin, bisphosphonates, and D-hormone metabolites. An appropriate “prophylactic” therapy includes estrogens, SERMs (selective estrogen receptor modulators), estrogen-like substances, bisphosphonates, and calcium/vitamin D. In osteoporosis, treatment is dependent on the bone metabolism. Accordingly, in slow-loser patients drugs that promote formation and in fast-loser patients that inhibit resorption are predominantly used ( Table 5.3-6 ).

114

5

Biology of the spine

6

BIBLIOGRAPHY

1.

2.

3.

4.

5.

6.

7.

Kanis JA (2002) Diagnosis of osteoporosis and assessment of fracture risk. Lancet; 359:1929–1936. Anonymous (2000) Osteoporosis prevention, diagnosis, and therapy. NIH Consensus Statement; 17:1–45. Anonymous NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis, and Therapy (2001) Osteoporosis prevention, diagnosis, and therapy. Jama. 285:785–795. Walker-Bone K, Dennison E, Cooper C (2001) Epidemiology of osteoporosis. Rheum Dis Clin North Am; 27:1–18. Review. Wolf RL, Zmuda JM, Stone KL, et al (2000) Update on the epidemiology of osteoporosis. Curr Rheumatol Rep; 2:74–86. Review. Anonymous (2002) Incidence of vertebral fracture in Europe: results from the European Prospective Osteoporosis Study (EPOS). J Bone Miner Res; 17:716–724. Ismail AA, O’Neill TW, Cockerill W, et al (2000) Validity of self-report of fractures: results from a prospective study in men and women across Europe. EPOS Study Group.

European Prospective Osteoporosis Study Group. Osteoporos Int; 11:248–254. Roy DK, O’Neill TW, Finn JD, et al (2003) Determinants of incident vertebral fracture in men and women: results from the European Prospective Osteoporosis Study (EPOS). Osteoporos Int; 14:19–26. 9. Radspieler H, Dambacher MA, Kissling R, et al (2000) Is the amount of trabecular bone-loss dependent on bone mineral density? A study performed by three centres of osteoporosis using high resolution peripheral quantitative computed tomography. Eur J Med Res; 5:32–39. 10. Kroger H, Reeve J (1998) Diagnosis of osteoporosis in clinical practice. Ann Med; 30:278–287. Review. 11. van Staa TP, Geusens P, Pols HA, et al (2005) A simple score for estimating the long-term risk of fracture in patients using oral glucocorticoids. QJM; 98(3):191–198. 8.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

12. Leib ES, Lenchik L, Bilezikian JP, et al (2002) Position statements of the International Society for Clinical Densitometry: methodology. J Clin Densitom; 5(Suppl):5–10. Review. 13. Dambacher MA, Neff M, Kissling R, et al (1998) Highly precise peripheral quantitative computed tomography for the evaluation of bone density, loss of bone density and structures. Consequences for prophylaxis and treatment. Drugs Aging; 12:15–24. 14. Gosfield Ed, Alavi A, Kneeland B (1993) Comparison of radionuclide bone scans and magnetic resonance imaging in detecting spinal metastases. J Nucl Med; 34:2191–2198. 15. Rahmouni A, Divine M, Mathieu D, et al (1993) Detection of multiple myeloma involving the spine: Effi cacy of fat-suppression and contrast-enhanced MR imaging. AJR Am J Roentgenol; 160(5):1049–1052. 16. Sanal SM, Flickinger FW, Caudell MJ, et al (1994) Detection of bone marrow involvement in breast cancer with magnetic resonance imaging. J Clin Oncol ; 12:1415–1421. 17. de Vernejoul MC (1998) Markers of bone remodelling in metabolic bone disease. Drugs Aging; 30:278–287. Review. 18. Recker RR (1994) Bone biopsy and histomorphometry in clinical practice. Rheum Dis Clin North Am; 20(3):609–627. 19. Stephen AB, Wallace WA (2001) The management of osteoporosis. J Bone Joint Surg Br; 83(3):316–323. Review.

115

116

5 5.4

BIOLOGY OF THE SPINE BIOLOGY OF FUSION WITH BONE AND BONE SUBSTITUTES

1

Introduction …………………………………………………………………………………………………

117

2 2.1 2.2 2.3 2.4 2.5

Biology of spinal fusion ………………………………………………………………………………… Autogenous bone graft …………………………………………………………………………………… Allograft …………………………………………………………………………………………………… Ceramics and bone substitutes ………………………………………………………………………… Demineralized bone matrices …………………………………………………………………………… Osteoinductive growth factors …………………………………………………………………………

118 119 120 122 124 124

3

Conclusion ………………………………………………………………………………………………… 125

4

Bibliography ………………………………………………………………………………………………… 126

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

117

Dante G Marchesi

5

BIOLOGY OF THE SPINE

5.4

1

BIOLOGY OF FUSION WITH BONE AND BONE SUBSTITUTES

INTRODUCTION

Major factors affecting bone graft healing Bone graft material

During the last two decades the number of spinal surgery procedures has been steadily increasing and vertebral fusions even more so. In North America alone over 200,000 spinal arthrodeses are presently performed each year. Globally, the number is expected to show continued growth. Despite modern technological progress, the rate of failure to achieve a solid bony union (ie, pseudarthrosis or nonunion) is reported to vary between 5% and 35% with single-level fusions and to increase even more when multilevel arthrodeses are attempted [1]. The outcome of spinal fusion depends on a complex process influenced primarily by the type of graft material used and many local and systemic factors that affect the healing response of an arthrodesis ( Table 5.4-1) [1]. The biomechanical aspect of the specific type of fusion (ie, posterolateral intertransverse process, anterior interbody, anteroposterior combined), the level of the arthrodesis, and the efficacy of spinal immobilization (internal or external) after surgery

Type Quantit y Sterilization technique

Local factors—biological

Preparation of recipient site Vascular supply of sof t tissue Local bone disease (tumor, marrow infiltration disease) Radiation

Local factors—mechanical

Biomechanical stabilit y Biomechanical loading

Systemic factors

Nutrition Hormones, grow th fac tors Drugs Osteoporosis Infec tions Smoking

Table 5.4-1

118

5

Biology of the spine

2

will require special attention. Considerable progress in the technology of spinal instrumentation has almost maximized the benefits of mechanical stabilization, but the limiting step to arthrodesis remains the biological ability to form osseous consolidation between adjacent vertebral segments. Recent advances in minimally or less invasive surgical techniques and the potential for biological regulation or manipulation of bone formation make it important to reexamine our understanding of the biology of bone graft materials and the spinal fusion process. This chapter will not review the multiplicity of local and systemic factors affecting spinal arthrodesis, but it will address the clinical applications of mineralized and demineralized bone graft preparations in spine surgery, reviewing the basic science and clinical experience supporting the use of these substitutes.

BIOLOGY OF SPINAL FUSION

A bone graft material is any implanted material that alone or in combination with other materials promotes a bone healing response by osteogenic, osteoconductive, or osteoinductive activity at a local site ( Fig 5.4-1): • Graft material that is osteogenic contains viable cells at some stage of osteoblastic differentiation that are, or potentially can be, capable of directly forming bone. • Osteoconductive materials provide a biocompatible physical structure or scaffold that supports new bone formation [2, 3]. • Osteoinductive graft material contains cytokines capable of inducing differentiation of our undetermined osteoprogenitor stem cell to an osteogenic cell type. The ideal bone graft material possesses three distinct properties—osteogenesis, osteoconduction, osteoinduction—with optimal immunological response in the host and without risk of disease transmission. The osteogenic potential of a graft material is directly derived from its cellular content. It must contain viable cells that can form bone (osteogenic precursor cells). These cells participate in the early stages of the healing process to unite the graft to the host bone and their viability must be ensured during the grafting procedure. Fresh autogenous bone and bone marrow are the best known graft materials. Osteoconductivity is the physical property of a graft material that allows the ingrowth of neovascularization and infi ltration of osteogenic precursor cells. Appropriate osteoconduction is provided by direct apposition between host bone and implant. Host bone must be viable, the host-bone-implant interface

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

5.4

119

Biology of fusion with bone and bone substitutes

must be stabilized (no macromotion) and the implant needs a structure (porosity) allowing new bone ingrowth. Optimal interconnection between pores (connective porosity) and a pore size superior to 100 µm have been demonstrated by biological studies to be essential in this process. Osteoinduction is the process by which some factors or substances stimulate the undetermined osteoprogenitor stem cell (responding cells) to induce the osteoblastic pathway and to differentiate into an osteogenic cell type (osteocyte). Osteoinductive properties have been found in demineralized bone matrix (DBM) and different morphogenetic proteins (BMP, TGF- , PDGF, EGF, etc) ( Fig 5.4-1).

2.1

AUTOGENOUS BONE GRAF T

Autogenous bone contains all of these three properties which promote bone fusion and it is considered as the gold standard among graft materials against which all others are compared. Corticocancellous bone, usually harvested from the iliac crest, has been the most common and most successful graft material in spinal fusion surgery. It is thought to contain both determined and inducible osteogenic precursor cells (osteogenic property), noncollagenous bone matrix proteins including growth factors (osteoinductive property), and bone mineral and collagen (osteoconductive property). Recent studies have also shown that all osteotropic growth factors known to be

BIOLOGY OF SPINAL FUSION osteoblastic pathway

stem cells

osteoprogenitors resting

pro-osteoblasts

osteoblasts

proliferation mineralization

BMP—PDGF—EGF—TGF- —… Fig 5.4-1

osteocytes

matrix

120

5

Biology of the spine

sequentially involved in the spinal fusion process are present in iliac crest bone autograft of all age groups but with a higher variation in older people (osteoinductive property) [4]. In addition, autogenous bone has the advantage of being nonimmunogenic and nonpathogenic.

hyperesthesia, dysesthesia, or diminished sensitivity in the cutaneous nerve territory. The quantity of bone available from the iliac crest, the increased operative time and its related cost, the blood loss, and the possible need for transfusion are additional problems to be considered.

Cancellous bone contains greater osteogenic potential because of the large number of surviving cells in the marrow and because it is favorable to vascular ingrowth and exposure of inductive proteins due to the large trabecular surface area and interconnected spaces. Cortical bone offers greater mechanical strength compared to cancellous bone, but is less effective for the following reasons. There is less or no marrow, and consequently fewer osteogenic cells; its structure is less favorable for new bone formation and is more resistant to vascular ingrowth and remodeling [3].

For these reasons, more and more attention has been directed in recent years to the development and use of bone graft substitutes or extenders. In spine surgery, the ideal bone graft substitute should be osteogenic, biocompatible, bioabsorbable, easy to use, and cost effective, it should also provide structural support. However, success in achieving these goals depends upon the material and its biological properties, as well as the particular host environment into which it is placed.

2.2

Osseous spinal fusion remains a cornerstone of surgical treatment for severe spinal disorders. However, the success rate is still debated and difficult to assess in the presence of metal implant material. A general failure rate of autogenous bone graft arthrodesis has been reported as higher than 30% in some series and, although progress in spinal instrumentation has decreased this sequela, the incidence of nonunion has remained unacceptably high [1]. Moreover, the morbidity related to harvesting bone graft from the iliac crest for lumbar spinal fusion can sometimes be more problematic than the primary surgical procedure itself. Major complications such as pelvic fractures, vascular injuries, and deep infection have been reported in as many as 9% of patients, while minor complications including chronic donor site pain and superficial infection have been observed in up to 25% of the cases. The most common minor complication is the alteration in sensation over the donor site area, manifested as chronic pain,

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

ALLOGRAF T

Historically, allograft has been the most common substitute for autogenous bone. It is highly osteoconductive, weakly osteoinductive, and not osteogenic, because cells do not survive transplantation. For these reasons, there has been concern as to whether allograft can reliably produce spinal fusion. Allografts may be available in a reasonable quantity and are quite versatile, in that the shape, contour, and mineral density of the graft can be defi ned by the specific part of the skeleton used to obtain the material and the machining that is performed. Major concern exists among clinicians as well as the public about the possibility of infectious disease transmission, despite meticulous screening and serological testing of the donor [5]. Allografts are processed and preserved in ways that affect the osteoinductive and osteoconductive capacity of the material as

5.4

121

Biology of fusion with bone and bone substitutes

well as its immunogenicity. Preservation is obtained by either fresh-freezing or freeze-drying, both of which allow extended storage but reduce immunogenicity of the graft and may alter its mechanical strength and leave the worrisome risk of viral disease transmission. Further sterilization with high-dose gamma irradiation or ethylene oxide gas is used, but both methods may further reduce osteoinductivity. Ethylene oxide sterilization is believed by most clinicians to prevent viral infection. However, studies have shown that the gas fails to penetrate cortical bone. Furthermore, several sterilization methods have been investigated, including ethylene oxide, irradiation, hydrochloric acid decalcification, dimethyl sulfoxide, and freeze-drying for their ability to destroy the feline leukemia virus in the donor bone. All methods of sterilization failed to eliminate the virus. This is a significant finding because the feline leukemia virus is a retrovirus similar to human immunodeficiency virus. Allograft is available in many preparations, however, the majority are composed primarily of cancellous or cortical bone. Cortical allografts provide immediate significant mechanical stability and structural support, while cancellous bone lends little mechanical stabilization on implantation but has a faster rate of incorporation. Cancellous allograft and particulate allograft preparations (cancellous or cortical) incorporate with new bone forming on the surface of trabeculae, with a large surface area available for new bone formation [5]. In contrast, cortical incorporation occurs slowly via a process of periosteal new bone formation around the allograft as an external callus derived from the host bone. Particulate and structural grafts demonstrate significant differences in the histology of incorporation. Particulate grafts show more rapid and complete revascularization than structural

grafts. Particulate bone remodels completely with time, while cortical bone remains a mixture of necrotic and viable bone. The process of creeping substitution also differs significantly between these forms of allograft with new bone formation occurring appositionally followed by resorption in cancellous bone. This process is reversed in cortical bone. These differences in biological capacity between graft types lead to significant differences in optimal clinical application. Various clinical reports examining the performance of allograft for spinal fusion have been presented in the literature, but only a few have been prospectively designed and well conducted. The most favorable data are reported for one-level interbody fusion in the cervical spine with low rates of graft subsidence or resorption, but the union rate drastically drops in multilevel procedures. In the lumbar spine, cortical allograft is generally used for structural support (femoral rings) in combination with autogenous bone graft, showing only rare pseudarthrosis [2]. During recent years, the use of machinethreaded cortical allograft bone dowels or allograft interbody cages obtained from midshaft or diaphyseal bone have gained considerable popularity for anterior arthrodesis in the lumbar spine. Outcome data using these particular allografts are currently being collected. Several published reports have also addressed the use of allograft alone or mixed with autograft for posterior spinal fusion. When used in an instrumented thoracic spine the results are reported as favorable, in the lumbar spine it shows a lower fusion rate and higher resorption when compared with autograft.

122

5

Biology of the spine

2.3

CERAMICS AND BONE SUBSTITUTES

Because of the previously discussed problems associated with autograft and allograft material, there has recently been increased interest in biodegradable osteoconductive ceramic bone graft substitutes, which would be available in unlimited quantity, without donor-site complications or infectious risks. For synthetic implants to be useful in vivo they must have certain properties: • • • • • • •

Compatibility with surrounding tissues. Chemical stability in body fluids. Compatibility of mechanical and physical properties. Ability to be produced in functional shapes. Ability to withstand the sterilization process. Reasonable cost of manufacturing. Reliable quality control [5].

The most common ceramic preparations that have been used in spinal reconstructive surgery include hydroxyapatite (HA) and tricalcium phosphate (TCP) [5]. They provide a biocompatible osteoconductive surface for bone regeneration and may contribute limited structural support. Other advantages of ceramic matrices include low immunogenicity and toxicity, stability at physiological pH levels, and the ability to withstand sterilization procedures without loosing structural integrity. Ceramics present with specific porosity, which may be artificially created and acts as a scaffold for further osteoblastic ingrowth. The unnatural pathway of a ceramic matrix with poorly interconnected porosity affects the ingrowing bone and retards the normal rate of bone healing and the remodeling process required to attain optimal mechanical strength.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

The remodeling process depends on the biodegradability of the ceramic: nonresorbing materials may interfere with remodeling, be the locus of a mechanical stress riser, and impede the accretion of strength of the fusion mass. The various calcium phosphate ceramics differ with regard to their bioresorbability characteristics. Hydroxyapatite is relatively inert and biodegrades poorly, which may hinder remodeling, prolong the strength deficiency of new bone, and leave permanent stress risers in the fusion mass. Conversely, ceramic TCP undergoes biodegradation within the fi rst 4–8 weeks of implantation, possibly too early for optimal fusion mass healing. Natural ceramics derived from sea coral (Porites asteroides) are reported to have ideal pore size, interconnective porosity, and are structurally similar to cancellous bone. Composed of 97% calcium carbonate in the form of aragonite, coral undergoes a thermal reaction where calcium carbonates are transformed into HA. Coral is extremely biocompatible and has yielded promising results when it has been used to replace or augment autogenous bone graft [5], or as part of a composite with an osteoinductive bone protein. Despite these properties, the poor bioresorbability of the HA also applies to these natural ceramics resulting in poor bone remodeling. While the organic phase of bone confers bone stiffness and compressive strength, ceramics are inherently brittle and susceptible to fracture with elasticity moduli significantly higher than those of cortical and cancellous bone and with low tensile strength. The overall crystalline structure and composition of bone apatite is similar, but not identical to that of HA, and this may explain differences observed in remodeling and resorption of ceramic preparations. Overall, more crystallization and higher mineral density yield greater

5.4

123

Biology of fusion with bone and bone substitutes

mechanical strength and lasting stability. In contrast, an amorphous simple preparation of calcium phosphate and calcium sulfate may also provide an osteoconductive matrix useful as a bone graft expender in spinal fusion, while retaining a rate of resorption that equals the rate of formation. Optimal remodeling of the spinal fusion mass is dependant upon biodegradability of the ceramic, which, depending on the crystalline structure and composition, may take from several months to several years. This problem seems to have been at least partially resolved by altering the processing of the coral with a partial thermoreaction where only 20% of the calcium carbonate is converted into HA [6]. Several preparations of ceramic matrices are currently available and they present different biological and mechanical characteristics [5]. Calcium sulfate and calcium phosphate are purely osteoconductive, with resorption properties closely matching the rate at which new bone is deposited. These materials are replaced by host bone through a process of creeping substitution. They exist in many different preparations including powders, pellets, putty, and injectable cement. Tricalcium phosphate has been used to fi ll bone defects. It has advantages similar to those of HA because it is biocompatible and bioabsorbable, but it is brittle and has very low impact resistance. Porous TCP has compressive and tensile strengths similar to, but lower than those of cancellous bone. The efficacy of these different ceramics in obtaining spinal arthrodesis has been studied in animal models and in selective clinical studies [7, 8]. Anterior interbody fusion in the thoracic spine of dogs has been analyzed using autologous tricortical iliac crest graft, HA ceramics, calcium carbonate, and a mixture of HA and TCP. Autogenous graft was shown to be the most effective material tested in these comparative studies [9]. Also, when combined with internal fixation, autogenous bone was significantly better than calcium carbonate ceramics.

Posterolateral intertransverse lumbar fusion was analyzed mainly in sheep and dogs [10–12]. Some authors demonstrated better results with autologous bone when compared with different ceramics, other authors showed similar results in terms of fusion rate when using coral porites (calcium carbonate) or a combination of HA and TCP. Clinical data on the use of ceramics alone or combined with autogenous bone are limited. High fusion rates were reported in patients who had undergone cervical spinal fusion with interbody titanium cages filled with coral HA [13]. French authors have advocated the use of ceramics as graft extenders for autogenous bone in long instrumented fusions for deformities [14–16]. In a study of 12 adolescent patients, Passuti et al used a combination of HA and TCP with autogenous bone and found all patients to be clinically and radiologically fused [16]. Histology of specimens obtained from two of these patients indicated de novo bone ingrowth into the ceramic pores. However, these results must be interpreted with caution. In the cases of adolescent scoliosis treated with spinal fusion, the ceramic was used as a bone graft extender to supplement the local bone used as graft material. In addition, the patient population studied in this trial has a high propensity to healing, even without the addition of bone graft [16]. Although selective data from both animal and clinical studies seem favorable, the role of ceramic implants is still not well defi ned. Further discussion of their use as a complementary agent in composite form with osteoinductive growth factors are presently under investigation and are discussed in the next section.

124

5

Biology of the spine

2.4

DEMINERALIZED BONE MATRICES

The osteoinductive factors of bone are contained within the organic phase. While mineralized matrices have minimal osteoinductive activity, demineralized preparations have demonstrated a potent effect on the differentiation of osteoprogenitor cells into osteoblasts. A bone morphogenetic substance was fi rst identified by Marshall Urist in his pioneering studies using soluble extracts from demineralized bone [17]. The demonstration of neoosteogenesis in response to ectopic submuscular implants of demineralized bone was a cornerstone in the further identification and cloning of bone morphogenetic proteins (BMPs). The capacity of demineralized bone matrix (DBM) to induce new bone formation is now well established. The primary osteoinductive component of DBM consists of small amounts of glycoproteins in the organic phase of bone, the most important of which are the BMPs. The major pathway of neoosteogenesis induced by DBM is endochondral in subcutaneous and submuscular implants, and by direct induction of resident mesenchymal stem cells to osteoblasts and direct formation of bone without a cartilaginous intermediate in calvarial defects. This difference indicates the importance of the host environment in the process of osteogenesis induced by DBM. Despite animal data suggesting a positive effect of DBM on spinal fusions, the clinical utilization of DBM in spinal arthrodesis has not demonstrated similar efficacy [18]. DBM preparations may be effective as graft extenders in the setting of limited autografts and as graft enhancers when comparing fusion quality to that of autograft alone, but results are equivocal. It is important to note that the osteogenic activity of a DBM preparation is highly dependent upon the type and

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

specific preparation of bone used. The differing efficacy of DBM in spinal fusion demonstrated in the literature is likely to be a result of the different DBM preparations used [5]. In summary, despite good evidence for osteogenic activity in DBM, there is little evidence suggesting its effectiveness as a substitute for autogenous bone graft. There is an important variability in the osteoinductive capacity of different commercial demineralized bone graft preparations and this may contribute to variability in clinical experience. DBM offers no structural or mechanical stability independently of its carrier and does not appear to be a reliable substitute for autogenous bone graft. The material may have a role as a graft extender or as a supplement in hosts with compromised bone forming capacity.

2.5

OSTEOINDUCTIVE GROW TH FACTORS

Advances in cellular and molecular biology have led to the identification of specific cytokines that are active in mediating cellular activities including mitogenesis, anabolic activity, and cell differentiation. The ability to control cellular activity is a potentially powerful instrument in the management of orthopedic disorders and surgical reconstructions. Many growth factors and other cytokines have been shown to be osteoinductive in animal models. The growth factors that may enhance bone formation in vitro include insulin-like growth factor (IGF-1), acidic and basic fibroblast growth factors (aFGF, bFGF), platelet-derived growth factor (PDGF) and transforming growth factor beta (TGF- ) [5, 19, 20]. Bone morphogenetic proteins are a subset of the TGFfamily and are the only cytokines that demonstrated a

5.4

125

Biology of fusion with bone and bone substitutes

3

capacity to induce new bone formation in vivo. Recent developments in recombinant techniques allow isolation of BMP in pharmacological quantities, in contrast to isolation through demineralization, in which less than 20 µg of osteoinductive material was extracted from 10 kg of bovine cortical bone. Clinical experience with BMP in spinal fusion studies sug gested a valuable role as bone graft supplements or substitutes [21]. Composite grafts allow the combination of the osteoinductive and osteogenic capacities of growth factors or autogenous bone with the structural capacity of mineralized matrices. The ideal carrier for bone growth morphogenetic proteins has not been determined, but it would have reversible affinity to glycoprotein, structural characteristics, possibly including malleability or mechanical rigidity, limited immunogenicity and toxicity, and resorbability to permit complete replacement by bone. Inorganic carriers of BMP that have shown efficacy in promoting spinal arthrodesis include true bone ceramic (TBC), derived from sintered bovine bone, and hydroxyapatite TCP. Organic carriers include polylactic acid polymers (PLA), collagen and noncollagenous protein carriers, mineralized or demineralized bone matrix, and autograft. Their advantages include the capacity for chemical bonding to growth factors, and the provision of a biodegradable environment for new bone formation and graft incorporation. However, many organic carriers are weakly immunogenic and lack the osteoconductive function of inorganic bone cements. The structural capacity of inorganic cements is a further advantage of this carrier. Composite grafts offer potential for the design of bone graft substitutes that are specific for the structural and biological demands of the host, and it is likely that very different composites will be used for anterior interbody arthrodesis than for instrumented posterior fusion.

CONCLUSION

An understanding of the biological and structural characteristics of mineralized and demineralized bone matrix is necessary for their effective clinical application. The existing preparations have clear limitations in their clinical efficacy. Grafting materials and composites will continue to evolve for specific applications and their choice will be determined by the properties of the local host environment. The development of growth factors and other cytokines functioning as potent induction agents for neoosteogenesis offers tremendous potential for the design of composite materials providing osteoinduction, osteoconduction, and structural functions. Advances in tissue engineering and gene therapy will add a further osteogenic capacity to future bone graft substitutes.

126

5

Biology of the spine

4

BIBLIOGRAPHY

Boden SD (1998) The biology of posterolateral lumbar spinal fusion. Orthop Clin North Am; 29(4):603–619. Review. 2. Boden SD, Sumner DR (1995) Biologic factors affecting spinal fusion and bone regeneration. Spine; 20(24 Suppl):102S–112S. Review. 3. Sandhu HS, Boden SD (1998) Biologic enhancement of spinal fusion. Orthop Clin North Am; 29(4):621–631. Review. 4. Lind M, Bünger C (2001) Factors stimulating bone formation. Eur Spine J; 10(Suppl 2): S102–S109. Review. 5. Berven S, Tay BK, Kleinstueck FS, et al (2001) Clinical applications of bone graft substitutes in spine surgery: consideration of mineralized and demineralized preparations and growth factor supplementation. Eur Spine J; 10(Suppl 2):S169–177. Review. 6. Steffen T, Marchesi D, Aebi M (2000) Posterolateral and anterior interbody spinal fusion models in the sheep. Clin Orthop; 371:28–37. 7. Emery SE, Fuller DA, Stevension S (1996) Ceramic anterior spinal fusion: Biologic and biomechanical comparison in a canine model. S pine; 21(23):2713–2719. 8. Zerwekh JE, Kourosh S, Scheinberg R et al (1992) Fibrillar collagen-biphasic calcium phosphate composite as a bone graft substitute for spinal fusion. J Orthop Res; 10(4):562–572. 9. Fuller DA, Stevenson S, Emery SE (1996) The effects of internal fi xation on calcium carbonate. Ceramic anterior spinal fusion in dogs. Spine; 21(18):2131–2136. 10. Baramki HG, Steffen T, Lander P, et al (2000) The effi cacy of interconnected porous hydroxyapatite in achieving posterior lumbar fusion in sheep. Spine; 25(19):1053–1060. 11. Guigui P, Plais PY, Flautre B, et al (1994) Experimental model of posterolateral spinal arthordesis in sheep. Part 1. Experimental procedures and results with autologous bone graft. Spine; 19(24):2791–2797. 1.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

12. Guigui P, Plais PY, Flautre B, et al (1994) Experimental model of posterolateral spinal arthordesis in sheep. Part 2. Application of the model: evaluation of vertebral fusion obtained with coral (Porites) or with a biphasic ceramic (Triosite). Spine; 19(24):2798–2803. 13. Thalgott JS, Fritts K, Giuffre JM, et al (1999) Anterior interbody fusion of the cervical spine with coralline hydroxyapatite. Spine; 24(13):1295–1299. 14. Delecrin J, Takahashi S, Gouin F, et al (2000) A synthetic porous ceramic as a bone graft substitute in the surgical management of scoliosis: a prospective, randomized study. Spine; 25(5):563–569. 15. Heise U, Osborn JF, Duwe F (1990) Hydroxyapatite ceramic as a bone substitute. Int Orthop; 14(3):329–338. 16. Passuti N, Daculsi G, Rogez JM, et al (1989) Macroporous calcium phosphate ceramic performance in human spine fusion. Clin Orthop; 248:169–176. 17. Urist MR (1965) Bone: formation by autoinduction. Science; 150(698):893–899. 18. An HS, Simpson JM, Glover JM, et al (1995) Comparison between allograft plus demineralized bone matrix versus autograft in anterior cervical fusion. Spine; 20(20):2211–2216. 19. Mundy GR (1996) Regulation of bone formation by bone morphogenetic proteins and other growth factors. Clin Orthop; 324:24–28. Review. 20. Solheim E (1998) Growth factors in bone. Int Orthop; 22(6):410–416. 21. Schimandle JH, Boden SD (1994) Spine update. The use of animal models to study spinal fusion. Spine; 19(17):1998–2006.

127

Serge Nazarian

SURGICAL ANATOMY OF THE SPINE

6

131

Serge Nazarian

6

SURGICAL ANATOMY OF THE SPINE

INTRODUCTION

Spine surgery, as much if not more than other fields of surgical practice, requires sound knowledge of morphology and topography. Considered to be the axial skeleton of the neck and the trunk, the spine belongs to several different anatomical areas of the body. The spine’s normal function, dysfunction, and clinical symptoms, as well as its surgical management and postoperative care are heavily influenced by its morphology and topography. In some areas, the topographical relationships are very complex, making the surgical approach and exposure relatively difficult. For these reasons, the specific morphological and topographical details related to a given surgical management should be well known by all those who practice spine surgery.

This section does not aim to be a textbook of spinal anatomy, but it should provide the reader with an overview of the essential anatomical knowledge required to perform the procedures described in section 7 Spinal instrumentation. This section is divided into seven chapters, all of which correspond to anatomical areas: the upper cervical spine, the lower cervical spine, the cervicothoracic junction, the thoracic spine, the thoracolumbar junction, the lumbar spine and lumbosacral junction, and the sacrum.

132

GENERAL ANATOMY

The spine is the axial skeleton and support of the neck and the trunk. The main anatomical characteristic of the spine is the fact that it contains and protects the spinal cord and the nerves which connect the spinal cord to the peripheral organs. Stability and mobility are the main functional properties on which its anatomical attributes are based. Except the coccyx, the spine is made of 25 bony pieces (7 cervical bones, 12 thoracic bones, 5 lumbar bones, and the sacrum), which are joined to each other by the so-called “mobile segment” of Junghans. Each mobile segment, except except the occipitoaxial complex (C0–2), includes three joints: the interbody joint and the two zygapophyseal joints. The anterior joint between the bodies is mainly made of the intervertebral disc which is comprised of the nucleus and the anulus fibrosus surrounding it. The posterior joints are synovial joints. The elementary intervertebral angular mobility is reduced to some degree, to protect the neural content of the spinal canal. The extensive global mobility of the spine can be attributed to the addition of the elementary intervertebral motions related to the multisegmental organization of its skeleton. The vertical weight-bearing system is organized according to anatomical three-column system: one is anterior, made of vertebral bodies and discs, and two are posterior, made of zygapophyseal masses.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

The intervertebral stability is provided to a great extent by ligamentous elements: • The intervertebral discs extend between the adjacent end plates. • The anterior and posterior longitudinal ligaments extend from the occipital bone to the coccyx, and along the anterior and posterior aspects of the vertebral bodies and discs. These ligaments are mainly attached to the anterior and posterior rims of the vertebral end plates and to the discs. • The ligamenta flava (yellow ligaments) extend from the lower half of the anterior aspect of the overlying lamina to the upper border of the underlying lamina; they also extend laterally on the anterior aspect of the zygapophyseal joint capsule. • The zygapophyseal joint capsule. • The interspinous ligaments extend between the adjacent spinous processes. • The supraspinous ligament. The anterior longitudinal ligament and the anterior halves of the discs act as an anterior tension band protecting the mobile segments from hyperextension. The ligamentous elements located behind the nucleus act as a posterior tension band protecting the mobile segments from hyperflexion.

6

133

Surgical anatomy of the spine

SOURCE BOOKS

In the upright position, the vertebrae are kept vertically piled up by the harmonious and correctly balanced support of the muscles, which act as active tension bands. • The erector muscles, by means of their sagittal and symmetric bracing, pull the spine backward to an upright position and maintain the lordotic curves in the lumbar and cervical segments of the spine; any loss of strength in these muscles may induce the loss of the lordosis and lead to a kyphotic deformity, which is associated with an imbalance in the sagittal plane. • The symmetric support of the lateral tension bands keeps the spine strictly vertical in the coronal plane; any rightleft asymmetry in the muscle tone, as well as a pelvic tilt in the coronal plane due to congenital asymmetry or a leg length discrepancy, may induce an abnormal curve in the coronal plane associated with an axial rotation.

Agur AM, Lee MJ (1999) Grant‘s Atlas of Anatomy. 10th ed. Philadelphia: Lippincott Williams & Wilkins. Louis R (1983) Surgery of the Spine. Surgical Anatomy and Operative Approaches . Berlin Heidelberg New York: Springer-Verlag. Rohen JW, Yokochi C, Lütjen-Drecoll E (2002) Color Atlas of Anatomy. 5th ed. Stuttgart: Schattauer.

The section editor has heavily relied on these textbooks throughout section 6.

134

6 6.1

1 1.1 1.2 1.3 1.4 1.5

SURGICAL ANATOMY OF THE SPINE UPPER CERVICAL SPINE

Anatomy related to posterior procedures……………………………………………………………… Anatomical basis for C1/2 fusion using wires and bone graft …………………………………… Anatomical basis for C1 lateral mass screw insertion ……………………………………………… Anatomical basis for C2 pedicle screw insertion …………………………………………………… Anatomical basis for C1/2 fusion using transarticular screws ……………………………………… Anatomical basis for occipitocervical fixation using screws associated with plates or rods …………………………………………………………………………

135 136 138 139 140 141

2 2.1 2.2

Anatomy related to anterior procedures ……………………………………………………………… 142 Anatomical basis for transoral approach ……………………………………………………………… 142 Anatomical basis for odontoid process screw fixation ……………………………………………… 144

3

Suggested reading ………………………………………………………………………………………… 145

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

135

Serge Nazarian, Cyril Solari

6

SURGICAL ANATOMY OF THE SPINE

6.1

1

UPPER CERVICAL SPINE

ANATOMY RELATED TO POSTERIOR PROCEDURES

A posterior approach of the upper cervical spine may be required for atlantoaxial or occipitocervical fusion. The patient is placed in a prone position. If a headrest is used, care should be taken to keep the eyes free of any contact in order to avoid any ophthalmological complication. Any posterior surgical procedure requires that the hair on the back of the neck is clipped up to the level of the external occipital protuberance. The nuchal skin is usually thick and displays transverse folds which partly disappear by pulling down the arms along the body and placing the head in a slight flexion.

The neck is very thick here and the spine is located in the middle of its sagittal diameter. The posterior tubercle of C1 and the underlying spinous processes are attached to the cervical aponeurosis by the thick fibrous nuchal septum, which is the cervical equivalent of the supraspinous ligament at the lower levels. Carrying out the posterior incision through this fibrous septum significantly reduces bleeding. The fi rst bony element to be palpated at the bottom of the incision is the spinous process of C2, under the inferior aspect of the occipital bone. Immediately above, on the midline, and 1–2 cm anteriorly, lies the posterior tubercle of C1. A thick layer of the neck’s muscles covers these bony elements. The cranial insertion of these muscles should be detached from the inferior aspect of the occipital bone in order to expose the posterior arch of C1 and C2.

136

6

1.1

Surgical anatomy of the spine

ANATOMICAL BASIS FOR C1/2 FUSION USING WIRES AND BONE GRAF T

The insertions for the suboccipital muscles are the posterior arch of C1, the spinous process, and the laminae of C2 ( Fig 6.1-1). From the posterior tubercle of C1 these muscles are detached mediolaterally from the posterior arch using a thin rasp, a curette, or a clamp-held nut. Beyond the fi rst 12–15 mm of this detachment care should be taken concerning:

• The vertebral artery, which turns around the lateral mass from lateral to medial, joining the upper aspect of the posterior arch ( Fig 6.1-2 ). • The 1st cervical nerve, between the occipital bone and C1. • The 2nd cervical nerve (Arnold nerve), between C1 and C2. • The voluminous suboccipital venous plexus, behind the C1/2 interspace ( Fig 6.1-3 ).

1 1 2

4

1 2

2

3 anterior

cranial cranial right

Fig 6.1-1 Posterior view of the craniocervical junction showing the suboccipital muscles. 1 rectus capitis posterior minor 2 superior oblique 3 inferior oblique 4 rectus capitis posterior major

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

a

right

b

Fig 6.1-2a–b a Posterior view of the lateral mass of C1 and its relationship with the vertebral artery. 1 vertebral artery 2 posterior arch of C1 b Cross section showing the relationship of the vertebral artery and the lateral mass of C1 (inferior side of the slice). 1 lateral mass of atlas (C1) 2 vertebral artery (partially calcified in this specimen)

right

6.1

137

Upper cervical spine

The detachment should be continued laterally using the clampheld nuts in order to avoid any severe neurovascular injury. The flat tendon of the transversospinales (multifidus) muscle should be detached from its spinous and laminar insertions using thin scissors, and then pushed away from the underlying lamina with a curette. Under the muscular layer the bony elements are attached to each other by flat ligaments which can be detached from the bone using a thin curve curette:

• The foramen magnum and the cranial border of the posterior arch of the atlas are attached by the atlantooccipital membrane. • The inferior border of the posterior arch of the atlas and the superior border of the lamina of the axis are attached by the atlantoaxial membrane. • The fi rst yellow ligament (ligamentum flavum) attaches from the anterior aspect of the lower half of the laminae of C2 to the upper border of the laminae of C3; it is usually completely covered by the inferior border of the lamina of C2.

3 1

2

a Fig 6.1-3a–b a Posterior view of the craniocervical junction. 1 suboccipital venous plexus 2 2nd cervical nerve 3 1st cervical nerve b Parasagittal cut showing the suboccipital venous plexus (arrow).

cranial

b

anterior

138

6

Surgical anatomy of the spine

Serge: 1.2 = new subchapter by VA.

1.2

ANATOMICAL BASIS FOR C1 LATERAL MASS SCREW INSERTION

The lateral masses of C1 have recently been described for a C1 screw insertion. The entry point for screw insertion into the lateral mass of C1 is reached by dissecting the inferior aspect of the C1 lamina with a Penfield elevator. Dissection is carried out on the same vertical line that is used for the C2 pedicle screw insertion. The C2 nerve root and the posterior root ganglion are retracted distally.

a

The screw should be oriented in a slight ascending direction parallel to the C1 ring, and slightly medial. Anteriorly, the screw should be inserted bicortically and perforate the anterior cortex of the C1 lateral mass. The screw tip is separated from the pharyngeal cavity by the prevertebral muscle layers. The danger is on the lateral side where the internal carotid artery is located.

b

Fig 6.1-4a–c a Posterior view for screw insertion into the lateral masses of C1. b Axial inferior view—the screw is inserted slightly convergent to avoid the internal carotid artery. c Parasagittal view—the direction of the screw is parallel to the C1 arch.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c

139

6.1

Upper cervical spine

1.3

ANATOMICAL BASIS FOR C2 PEDICLE SCREW INSERTION

From an anatomical standpoint the axis is a very special vertebra. Its upper and lower halves are very different, but they are nevertheless in harmonious continuity, which charac terizes the transitional phenomenon. Its lower half resembles a middle cervical vertebra (Fig 6.1-5, see Fig 6.1-7). The obliquity and the transverse diameter of the pedicle are variable ( angulation with respect to the sagittal plane: 25–35°; transverse diameter: 2–5 mm). The best way to identify this is to detach the atlantoaxial membrane and to palpate the medial border of the isthmus with a Penfield retractor because the screws must be as close as possible to the medial border of the isthmus. Immediately lateral to the pedicle lies the foramen transversarium (transverse foramen). At this level the vertebral artery forms a mediolateral loop below the superior

articular process, at middistance between its lateral and medial borders. In the case of a very narrow pedicle, there is a high risk of vascular lesion to the vertebral artery by violation of the foramen transversarium. The axial line of the pedicle projects posteriorly halfway between the upper and the lower articular surfaces, on the vertical line bisecting the zygapophyseal column (articular mass). From this point, the drill should be oriented 25° cranially and 25–35° medially (this latter angle can be reduced to 15° if the position of the artery is more medial). Before any procedure requiring a pedicle screw in C2, it is highly recommended to perform a CT scan or a vertebral angiogram associated with coronal and sagittal reformatting, to analyze the relationship of the pedicle with the vertebral artery ( Fig 6.1-6 ).

cranial anterior

1 2

a Fig 6.1-5 Inferior view of the axis (C2). 1 left foramen transversarium (transverse foramen) 2 left pedicle

b

Fig 6.1-6a–b a AP view of an angiogram showing the loop of the vertebral artery. b Lateral view of the loop of the vertebral artery.

140

6

Surgical anatomy of the spine

The pedicle of C2 is defi ned as follows: its superior border extends with the superior articular process. The medial border of the pedicle is limited by the spinal canal. Its lateral border is closely related to the vertebral artery. The vertebral artery penetrates the bone at a level projecting approximately to the middle of the transverse diameter of the superior articular process; it hits the inferior aspect of the superior articular process, turns laterally and then cranially, forming a loop before leaving the foramen transversarium. The posterior end of the pedicle is included in the so-called pars interarticularis or isthmus. The caliber of the vertebral artery and the transverse diameter of the pedicle are variable.

1.4

ANATOMICAL BASIS FOR C1/2 FUSION USING TRANSARTICULAR SCREWS

The screw has to cross the C1/2 joint, ending its course in the lateral mass of C1 without crossing the path of the vertebral artery. It has to penetrate the articular mass of C2 at the medial border of the inferior articular process (at its junction with the lamina) and about 3 mm above its lower border. From this entry point the drill should be oriented sagitally, and enough cranially to perforate the articular surface of C2 near its posterior border with a strong purchase in the lateral mass of C1 ( Fig 6.1-7 ).

P

T

P T

3 mm

a Fig 6.1-7a–d a Posterior view of the entry points and pathways for pedicle screw fixation (P, left side) and transarticular C1/2 screw fixation (T, right side). For the pedicle screw insertion, the entry point is located middistance between the superior and inferior articular processes. The entry point for the transarticular screw fixation is approximately located at the medial border of the inferior articular process of C2.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b b Superior view of the path of the screw in the transverse plane. The direction of the screw is anteromedial for pedicle screws (P) and sagittal for transarticular screw (T) fixation.

6.1

141

Upper cervical spine

In obese patients and in patients requiring cervical lordosis to reduce an anterior subluxation, direct aiming is not possible through the approach incision. An additional distal percutaneous stab incision through the skin and muscles should be carried out to allow for the drilling, taping, and inserting of the screws in the adequate direction. The diameter of the drill should not exceed 2.5 mm.

1.5

ANATOMICAL BASIS FOR OCCIPITOCERVICAL FIXATION USING SCREWS ASSOCIATED WITH PLATES OR RODS

The screws should be inserted in the lower surface of the occipital bone. This area is located between the superior nuchal line cranially and the foramen magnum caudally. It is made of cortical bone. The thickness of the cortical bone is variable according to the individual and, in the same individual, according to the location. The maximal thickness is near the midline and along the superior nuchal line, at the level of the external occipital protuberance (inion). The plates or rods should be curved toward the midline to have their most cranial screws inserted in the strongest area of the occipital bone. The length of these occipital screws usually ranges from 10 to 16 mm in the midline, but can be as short as 5 mm if inserted laterally to the midline.

T c

P d

Fig 6.1-7a–d c Lateral view of the pathway in the sagittal plane. For transarticular C1/2 screw fixation (T), the screw is oblique cranially and anteriorly. It should aim at the posterior part of the superior articular surface of C2, in order to avoid damaging the loop of the vertebral artery. d Lateral view of the pathway in the sagittal plane for C2 pedicle screw fixation (P). The screw follows the long axis of the pedicle and is directed 25° upward.

142

6

Surgical anatomy of the spine

2

ANATOMY RELATED TO ANTERIOR PROCEDURES

2.1

ANATOMICAL BASIS FOR TRANSORAL APPROACH

This approach allows the exposure of the anterior aspect of the craniocervical junction which is comprised of the anterior border of the foramen magnum, the anterior arch of the atlas, the body and the odontoid process of the axis, the medial half of the atlantoaxial joints, the intervertebral disc of C2/3, and the upper part of the body of C3.

All these elements are located behind the oral cavity and the nasal fossae ( Fig 6.1-8 ). Although these cavities are well known for their septicity, in well-trained centers, the septic complications are not more frequent than for other exposures, owing to the high blood flow and an adequate pre-, intra-, and postoperative care.

1 2 3 4

9

1

9

2

5 8 6

3

8

4 7

5

anterior

7

a

cranial

left

Fig 6.1-8a–b Anterior and posterior relationship of the superior cervical spine. a Transverse cut of C1 (inferior view). 1 soft palate 2 pharyngeal cavity 3 anterior arch of the atlas 4 odontoid process 5 transverse ligament 6 vertebral artery behind the lateral mass of the atlas 7 cerebellum (right hemisphere) 8 spinal cord 9 lateral mass of the atlas (C1)

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

6

b b Sagittal cut. 1 anterior arch of the atlas 2 soft palate 3 axis and odontoid process 4 epidural space 5 tongue 6 epiglottis 7 spinous process of C2 8 posterior arch of the atlas 9 cerebellomedullary cistern

anterior

6.1

143

Upper cervical spine

The anterior relationship of the craniocervical junction is represented by the prevertebral muscles and the posterior wall of the pharynx. The exposure of this area requires the full opening of the mouth, cranial retraction of the soft palate, and caudal retraction of the tongue and epiglottis. The pharyngeal wall includes the pharyngeal mucosa, fascia, and muscles. The retropharyngeal elements include the muscles attached to the vertebrae (rectus capitis anterior and lateralis, longus colli and longus capitis muscles).

6

1

5

2

4 3

anterior left

Fig 6.1-9 The infraparotid space on a transverse cut at the level of the anterior arch of C1. 1 pharyngeal cavity 2 posterior wall of the pharynx and prevertebral muscles 3 anterior arch and lateral mass of the atlas 4 internal jugular vein 5 internal carotid artery 6 posterior inferior subparotid space and its vasculonervous content

The posterior relationships include the spinal canal and its contents ( Fig 6.1-8 ). The closest element is the posterior longitudinal ligament. The canal is wide because the range of motion is important in the upper levels of the cervical spine. There is an average space of 3 mm between the anterior wall of the canal and the spinal cord, sufficiently large to accept the tip of a screw. The closest lateral relationships include the posterior subparotid space and the vascular and nervous elements passing through it ( Fig 6.1-9 ): the internal jugular vein, the internal carotid artery, the superior cervical ganglion (vegetative), and the four lowest cranial nerves. Injury to these elements is prevented by detaching the prevertebral muscle layers directly from the bony surface and not extending the detachment more than 2 cm from the midline.

144

6

2.2

Surgical anatomy of the spine

ANATOMICAL BASIS FOR ODONTOID PROCESS SCREW FIX ATION

The odontoid process is in vertical alignment with the vertebral body of the axis in the coronal and sagittal planes. The principle of the screw fi xation is to introduce one or two lag screws from the rostrum or the anterior part of the end plate of C2, in the direction of the odontoid process through the fracture. From an anatomical standpoint (Table 6.1-1), the feasibility of the procedure depends on the following points: • Intraaxial drilling direction permitted by the AP topography of the manubrium, the sagittal diameter of the cervical opening of the thoracic cavity, the length of the neck in flexion/extension, and the sagittal curvature of the neck ( Fig 6.1-10 ). • Diameter of the “neck” of the odontoid process. • Approach and retraction of the soft tissues.

Favorable for screw fixation

Unfavorable for screw fixation

Narrow sagit tal diameter of the cer vicothoracic junc tion

Large sagit tal diameter of the cer vico thoracic junc tion (barrel chest)

Long neck

Shor t neck

Lordotic cur ve of the cer vical spine

Kyphotic cur ve of the cer vical spine

Table 6.1-1 Anatomical conditions for screw fixation.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

The drill bit should penetrate at the rostrum. It is not easy to insert an awl into this part of C2 without excising the sharp angle between the anterior cortex and the anterior part of the lower end plate. The number of screws (one or two) is dependant on the transverse diameter of the odontoid’s neck as measured on the CT scan. The tip of the screw(s) should aim at the tip of the odontoid process or the area just behind it. The screw should cross the cortex to take a strong purchase into the odontoid process. The anatomical specificities related to the approach and exposure are described in chapter 6.2 Lower cervical spine.

6.1

145

Upper cervical spine

3

cranial posterior

Fig 6.1-10 The feasibility of screw insertion into the odontoid process is dependant upon several conditions such as the length of the neck, the level of the manubrium, the sagittal diameter of the cervicothoracic junction, and the curve of the cervical spine. This case is ideal: long neck, manubrium at T2, short sagittal diameter of the cervicothoracic junction, and lordotic cervical curve.

SUGGESTED READING

Doherty BJ, Heggeness MH (1995) Quantitative anatomy of the second cervical vertebra. Spine ; 20(5):513–517. Ebraheim NA, Lu J, Biyani A, et al (1996) An anatomic study of the thickness of the occipital bone. Implications for occipitocervical instrumentation. Spine ; 21(15):1725–1730. Ebraheim NA, Xu R, Ahmad M, et al (1998) The quantitative anatomy of the vertebral artery groove of the atlas and its relation to the posterior atlantoaxial approach. Spine ; 23(3):320–323. Gebhard JS, Schimmer RC, Jeanneret B (1998) Safety and accuracy of transarticular screw fi xation C1-C2 using an aiming device. An anatomic study. Spine ; 23(20):2185–2189. Jun BY (1998) Anatomic study for ideal and safe posterior C1-C2 transarticular screw fi xation. Spine ; 23(15):1703–1707. Lu J, Ebraheim NA, Yang H, et al (1998) Anatomic considerations of anterior transarticular screw fi xation for atlantoaxial instability. Spine ; 23(11):1229–1236. Lu J, Ebraheim NA (1998) Anatomic considerations of C2 nerve root ganglion. Spine ; 23(6):649–652. Nucci RC, Seigal S, Merola AA, et al (1995) Computed tomographic evaluation of the normal adult odontoid. Implications for internal fi xation. Spine ; 20(3):264–270. Roberts DA, Doherty BJ, Heggeness MH (1998) Quantitative anatomy of the occiput and the biomechanics of occipital screw fi xation. Spine ; 23(10):1100–1108. Xu R, Nadaud MC, Ebraheim NA, et al (1995) Morphology of the second cervical vertebra and the posterior projection of the C2 pedicle axis. Spine ; 20(3):259–263. Zipnick RI, Merola AA, Gorup J, et al (1996) Occipital morphology. An anatomic guide to internal fi xation. Spine ; 21(15):1719–1730.

146

6 6.2

SURGICAL ANATOMY OF THE SPINE LOWER CERVICAL SPINE

1 1.1 1.2

Anatomy related to posterior procedures …………………………………………………………… 147 Anatomical basis for muscular detachment …………………………………………………………… 147 Anatomical basis for zygapophyseal mass screws …………………………………………………… 149

2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8

Anatomy related to anterior procedures ……………………………………………………………… Muscular, vascular, and visceral relationships ……………………………………………………… Neurovascular bundle of the neck ……………………………………………………………………… Vertebral artery …………………………………………………………………………………………… Sympathetic trunk ………………………………………………………………………………………… Phrenic nerve …………………………………………………………………………………………… Inferior laryngeal nerve …………………………………………………………………………………… Anterior anatomy of the cervical spine related to disc resection, interbody fusion, and disc replacement …………………………………………………………………………………… Anterior landmarks of the vertebral levels ……………………………………………………………

3

Suggested reading ………………………………………………………………………………………… 159

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

150 150 152 153 154 155 155 157 157

147

Serge Nazarian, Cyril Solari

6

SURGICAL ANATOMY OF THE SPINE

6.2

LOWER CERVICAL SPINE

1

ANATOMY RELATED TO POSTERIOR PROCEDURES

1.1

ANATOMICAL BASIS FOR MUSCULAR DETACHMENT

The nuchal fascia and muscles

The cervical spine is covered posteriorly by a very thick mass of muscles. They are spread out on each side in four layers ( Fig 6.2-1): • First layer—trapezius muscle. • Second layer—splenius muscle and levator scapulae muscle. • Third layer—semispinalis capitis muscle and longissimus capitis muscle. • Fourth layer—deep muscles lying against the vertebrae, transversospinales muscle ( Fig 6.2-2 ).

The right and left layers are separated by the thick fibrous septum nuchae, which is attached anteriorly to the spinous processes and interspinous ligaments, and posteriorly to the cervical fascia. The posterior midline incision should be carried out through and along this septum, from the cervical fascia to the spinous processes. Transversospinales muscle ( Fig 6.2-2 )

This muscle is the deepest layer; it lies against the laminae and spinous processes. Its fan-shaped fibers extend from the transverse process of a given level to the three to four adjacent overlying laminae and spinous processes. Using a Cobb elevator to detach them can be disastrous, leading to hematoma, calcification, and neck stiffness. The less invasive method of detaching them involves cutting their insertion along the spinous process and the lamina, using thin dissection scissors, before pushing them gently to the lateral side, using a curette or a clamp-held nut.

148

6

Surgical anatomy of the spine

1 1

2

2 3 4

11

5 6

10 9

7 8 anterior right

Fig 6.2-1 Transverse cut of the neck through C4/5. 1 thyroid cartilage of the larynx 2 common carotid artery 3 internal jugular vein 4 sternocleidomastoid muscle 5 transversospinales muscle 6 levator scapulae muscle 7 septum nuchae 8 trapezius muscle 9 splenius capitis muscle 10 semispinalis capitis muscle 11 splenius cervicis muscle

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

3

cranial left

Fig 6.2-2 A left transversospinales muscle in the cervical area. 1 left lamina 2 spinous process 3 left transversospinales muscle

149

6.2

Lower cervical spine

1.2

ANATOMICAL BASIS FOR ZYGAPOPHYSEAL MASS SCREWS

C6 ( Fig 6.2-3c ). Its sagittal diameter ranges from 12 to 18 mm according to the individual and the level.

Zygapophyseal mass screws are used in the cervical vertebrae from C3 to C7. Each zygapophyseal mass is a short segment of the posterolateral column of the spine. In the cervical spine it is a vertical cylindroid, slightly fl attened anteroposteriorly, limited by the pedicle anteromedially, the posterior root of the transverse process anteriorly, and the lamina medially ( Fig 6.2-3a–b ). It is usually in a direct relationship with the nerve root at each level and the vertebral artery from C3 to

In order to improve the bony purchase, the screws are introduced and driven in the bone as close as possible to the superior end plate and the underlying dense subchondral bone. To avoid any injury to the nerve roots and vertebral artery, the point of introduction lies 2 mm medially from the vertical midline of the lateral mass and its direction is anterior and lateral (chapter 7.2.3. Middle and lower cervical spine).

a

b

Fig 6.2-3a–c The zygapophyseal masses of a cervical vertebra. a Superior view. b Posterior view. c Transverse cut at C5/6 level. 1 vertebral artery 2 6th cervical nerve 3 zygapophyseal mass

1 2 3

anterior

c

right

150

6

2

Surgical anatomy of the spine

ANATOMY RELATED TO ANTERIOR PROCEDURES

The shape of the neck varies according to the individuals and their corpulence. The bony limits of the neck include the sternum and clavicle at its lower end, as well as the mandible and mastoid bone at its upper end. The cervical spine is located approximately in the middle of the neck in the sagittal plane. The hyoid bone can be identified by palpation; it is found approximately at the level of C4. The cricoid cartilage is located at the level of the C6/7 disc. Due to the anterior and inferior obliquity of the 1st rib, T1 and T2 usually extend above the level of the manubrium ( Fig 6.2-4 ).

1 6 5 4 2 3

cranial posterior

Fig 6.2-4 Sagittal cut showing the topography of T1 and T2, usually above the manubrium. In this specimen the cranial border of the manubrium is at the level of T2. The thyroid cartilage and the hyoid bone are not clearly visible on this photo. 1 C2 vertebral body 2 T1 vertebral body 3 cranial border of the sternal manubrium 4 cricoid cartilage 5 thyroid cartilage (internal aspect) 6 hyoid bone

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

2.1

MUSCULAR, VASCULAR, AND VISCERAL RELATIONSHIPS

Under the skin and the thin layer of subcutaneous fat, lies the platysma muscle running from the mandible to the clavicle, obliquely downward and backward. Its thickness is about 1–3 mm, according to the individual. It is perforated by some neurovascular bundles going to and from the skin. By a blunt dissection, using clamp-held nuts, the skin, platysma, and fat pad are separated from the superficial layer of the cervical fascia, and so the transverse skin incision gives access to a longitudinal approach, along the anterior border of the sternocleidomastoid muscle. The sternocleidomastoid muscle ( Fig 6.2-5 ) runs obliquely downward and forward, from its mastoid and occipital superior attachments to its inferior sternal and clavicular attachments. This muscle divides the anterolateral aspect of the neck into three areas: the anterior triangle, from the midline to the anterior border of the sternocleidomastoid muscle; the sternocleidomastoid region in the oblique area of the muscle itself; and the posterior triangle, from the posterior border of the sternocleidomastoid to the anterior border of the trapezius muscle. Several neurovascular elements run through these areas: the external jugular vein, several branches of the superficial cervical plexus, and the anterior jugular vein. Most of these elements can be preserved by performing a presternocleidomastoid approach. The superficial layer of the cervical fascia is, depending on the individual, more or less thick and strong. If its division is not progressively obtained by blunt dissection, it can be started with scissors, carefully in order to avoid any vascular or visceral injury.

6.2

151

Lower cervical spine

Under this fi rst layer lies a second fibrous layer, the pretracheal layer of the cervical fascia (middle cervical aponeurosis). It envelops the subhyoid muscles (sternohyoid, sternothyroid, and omohyoid muscles). The omohyoid muscle ( Fig 6.2-6 ) runs along the lateral border of this pretracheal layer.

Cervical viscerae are located in front of the spine. They include the pharynx and larynx from C1 to C6, and the trachea and esophagus from C6 to the lower end of the neck. Thyroid and parathyroid glands are attached to the anterior and lateral aspects of the trachea and larynx. An overretraction of these elements during the exposure of the spine can lead to dysphonia and dysphagia.

1 4 1 2 3 2

cranial

3 anterior

Fig 6.2-5 Right anterolateral view of the neck showing the sternocleidomastoid muscle. The skin, subcutaneous fat, and superficial layer of the cervical fascia have been detached and everted laterally in this cadaveric dissection. 1 sternocleidomastoid muscle 2 clavicle 3 pectoralis major muscle

c r a n ia

l

anter

io r

Fig 6.2-6 Right anterolateral view of the middle cervical aponeurosis and omohyoid muscle. 1 omohyoid muscle 2 sternohyoid muscle 3 pretracheal layer of the cervical fascia 4 sternocleidomastoid muscle

152

6

2.2

Surgical anatomy of the spine

NEUROVASCULAR BUNDLE OF THE NECK

The neurovascular bundle of the neck includes the internal jugular vein laterally, the carotid arteries medially, the pneumogastric nerve posteriorly, and the descending branch of the hypoglossal nerve anteriorly. These elements are surrounded by a sheath of connective tissue, the carotid sheath. They are covered by the sternocleidomastoid muscle ( Fig 6.2-5, 6.2-7 ) which must be retracted to expose them.

1 2

Internal jugular vein

3

The internal jugular vein ( Fig 6.2-7 ) is the most superficial and lateral element of the bundle. It receives the so-called thyrolinguofacial trunk and the middle thyroid vein(s).

4 5 6

Common carotid artery

The common carotid artery ( Fig 6.2-7 ) is the medial element of the bundle. It bifurcates 1 cm above the cranial border of the thyroid cartilage into the external and internal carotid arteries. Just before this bifurcation, the common carotid artery displays a dilatation, the carotid sinus ( Fig 6.2-7 ), which contains the sensitive carotid baroceptors. The external carotid artery is medial and anterior to the internal carotid artery at its origin; it progressively runs lateral, crossing the internal carotid artery anteriorly. Vagus nerve

The vagus nerve ( Fig 6.2-7 ) runs in a posterior dihedral angle between the carotid artery and the internal jugular vein.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

cranial anterior

Fig 6.2-7 The internal jugular vein, carotid artery, vagus nerve, and descending branch of the 12th cranial nerve. 1 descending branch of the 12th cranial nerve, giving off branches to the infrahyoid muscles 2 carotid sinus 3 common carotid artery 4 vagus nerve (it is pulled anteriorly between the carotid artery and the jugular vein after dissection and separation of these elements) 5 internal jugular vein 6 sternocleidomastoid muscle

6.2

153

Lower cervical spine

The neurovascular bundle is crossed by two nerves of high surgical relevance:

artery or the internal jugular vein in order to allow the medial retraction of the visceral axis.

• The superior laryngeal nerve originates from the lower end of the plexiform ganglion of the vagus nerve. Running downward and forward, medial to the jugulocarotid bundle, it appears below the origin of the lingual artery. It divides near the tip of the greater horn of the hyoid bone, giving rise into the superior and inferior terminal branches which penetrate the larynx through the thyrohyoid and cricothyroid membranes respectively. The superior laryngeal nerve transmits sensory impulses from the larynx and base of the tongue, and motor impulses to the cricothyroid muscle. • The hypoglossal nerve penetrates the upper part of the anterior cervical region, running downward and outward behind the internal carotid artery, and then downward and forward between the common carotid artery medially and the internal jugular vein laterally. Running along the lower border of the posterior belly of the digastric muscle, it follows a curve with superior concavity, above the greater horn of the hyoid bone. It fi nally penetrates the base of the tongue passing medially to the tendon of the stylohyoid muscle. When it passes behind the external carotid artery, it gives rise to its descending branch which then supplies motor rami to the subhyoid muscles.

The omohyoid muscle, running along the lateral border of the pretracheal layer of the cervical fascia, must sometimes be divided through its anterior belly in order to facilitate the approach to the lowest cervical vertebrae and discs.

These two nerves are prone to injury in anterior approaches to the cervical spine as a result of an overretraction. The surgical access to the prevertebral region passes between the visceral axis medially and the neurovascular bundle and sternocleidomastoid muscle laterally. It sometimes requires the ligature of the collateral branches of the external carotid

2.3

VERTEBRAL ARTERY

In relation to the lower cervical spine, the vertebral artery ( Fig 6.2-8 ) is divided into a pretransverse and a transverse segment. The mean diameter of the artery is 4.5 mm. The rightleft symmetry of these segments is observed in only 40% of individuals. Unilateral hypoplasia is found in 9% and one of the two arteries is lacking in 2.5% of individuals. The vertebral artery is usually the fi rst collateral branch of the subclavian artery. But it can originate from the aorta, the common carotid artery, or by trifurcation of the brachiocephalic trunk. The pretransverse segment extends from its origin to its penetration into the transverse canal. Its length varies according to its origin and the level of its entry in the transverse canal. In 90% of cases it enters at C6 whereas in 10% it penetrates at C5, C7, or even at C4 or C3. The transverse segment extends from its entrance in the transverse canal to the level C2/3. The canal is alternately osseous and soft. This segment is practically straight but a few loops may exist and constitute a danger in vertebral surgery. It is accompanied by several veins anastomosing in a periarterial plexus of which the hemostasis is often critical. In this segment the artery is in a special relationship with the spinal nerve lying behind it and the sympathetic plexus around it.

154

6

Surgical anatomy of the spine

2.4

SYMPATHETIC TRUNK

The sympathetic trunk ( Fig 6.2-9 ) lies in the prevertebral fascia covering these muscles.

a

b

Fig 6.2-8a–b AP and lateral views showing the usual situation of the vertebral artery. (Courtesy of Prof Nadine Girard.)

7 6

1

5

2

4

b

3 cranial left

Fig 6.2-9 Right anterolateral view of the cervical spine after medial retraction of the viscerae and lateral retraction of the vessels and sternocleidomastoid muscle revealing the sympathetic trunk. 1 esophagus, retracted medially 2 cervical spine covered by the anterior longitudinal ligament 3 sternocleidomastoid muscle 4 internal jugular vein (collapsed) 5 right sympathetic trunk 6 longus colli muscle 7 right common carotid artery

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

6.2

Lower cervical spine

2.5

PHRENIC NERVE

The phrenic nerve lies laterally. It runs in the fascia covering the anterior scalenus muscle.

2.6

INFERIOR LARYNGEAL NERVE

The inferior laryngeal nerve (recurrent laryngeal nerve) originates from the vagus nerve at the level of the arterial element embryologically derivated from the 4th aortic arch. On the left side, this arterial derivate is a part of the aortic crossa, and on the right side, a part of the subclavian artery. The recurrent nerve crosses anteroposteriorly and lateromedially the vessel developed from the 4th aortic arch, then runs cranially and medially to join and follow the vertical dihedral angle between the trachea and the esophagus. So, on the left side ( Fig 6.2-10 ), the recurrent nerve, starting from a thoracic level, has already reached the tracheoesophageal angle when it arrives at the base of the neck; the retraction of the visceral axis to the opposite side of the neck does not submit it to excessive tension. On the contrary, on the right side, the recurrent nerve, starting from the base of the neck, has not yet reached the visceral axis ( Fig 6.2-11). It is more anterior, more lateral, and less vertical than its left homologous nerve. This difference in depth and obliquity explains its higher vulnerability to retraction in lower cervical spine approaches. The retraction of the visceral axis to the opposite side exposes the nerve to an overdistraction. The incidence of vocal paralysis reported is 2–12%. They mainly occur in right approaches. 1–2% of paralysis cases do not recover.

155

10

1 2

9

3

8

4 5

6 cranial

7 left

Fig 6.2-10 Left recurrent nerve. Left anterolateral view of the cervicothoracic junction after a midline sternotomy and lateral retraction of each half. The left lobe of the thyroid gland is pulled to the right side to expose the esotracheal angle. 1 C6/7 osteophyte bulging into the retroesophageal space 2 left common carotid artery 3 prevertebral (retroesophageal) space: the esophagus, trachea, and recurrent nerve are retracted to the right side 4 left border of the esophagus 5 left recurrent laryngeal nerve isolated on a blue vessel loop 6 left brachiocephalic vein 7 aorta 8 right common carotid artery 9 trachea 10 thyroid gland (left lobe)

156

6

Surgical anatomy of the spine

1

12

1

8

11 10

2

9

3 4 5 6

7

2

6 5

3

7

c ra

cranial

a

left

8

Fig 6.2-11a–b Right recurrent nerve. a Right anterolateral view of the cervicothoracic junction after resection of the clavicle. 1 subhyoid muscles covering the thyroid gland 2 right internal carotid artery 3 right vagus nerve 4 right recurrent nerve 5 right subclavian artery 6 trachea 7 brachiocephalic artery and its division into right common carotid and subclavian arteries 8 left brachiocephalic vein 9 right sternocleidomastoid muscle (detached and retracted laterally) 10 right internal jugular vein 11 right sympathetic trunk 12 longus colli muscle

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

nia

le f

4 l

t

b Right anterolateral view of the cervicothoracic junction after resection of the clavicle and division of the right subclavian artery. The right common carotid artery has been resected. The proximal and distal segments of the subclavian artery are pulled medially and laterally to show the vagus nerve and the infrasubclavian loop of the inferior laryngeal nerve detaching from it. 1 thyroid gland 2 trachea 3 proximal segment of the right common carotid artery 4 left brachiocephalic vein 5 proximal segment of the right subclavian artery 6 right inferior laryngeal nerve (recurrent laryngeal nerve) 7 distal segment of the right subclavian artery 8 right vagus nerve

6.2

Lower cervical spine

2.7

ANTERIOR ANATOMY OF THE CERVICAL SPINE RELATED TO DISC RESECTION, INTERBODY FUSION, AND DISC REPLACEMENT

The vertebral bodies are attached together anteriorly by the anterior longitudinal ligament, which plays the role of an anterior tension band; in the anterior approach to the disc this ligament must be interrupted in front of the disc by two transverse incisions at the level of each end plate. The resection of the disc and posterior and lateral osteophytes, as well as any interbody implantation (bone graft, cage, or artificial disc) require a wide anterior opening of the disc space. The anterior annulus is partly hidden by the anterior inferior rostrum of the overlying vertebra. This rostrum should be resected at the beginning of the procedure, as soon as the anterior longitudinal ligament and the anterior annulus have been resected. The anterior exposure of the discs and vertebral bodies, from the left to the right uncus, requires the detachment of the longus colli muscle. Care should be taken to detach only the muscle and not the longitudinal ligament, in order to avoid ectopic calcifications. This detachment requires the use of a sharp rasp. In the cases requiring anterior plating, care should be taken not to detach the anterior longitudinal ligament from the adjacent overlying and underlying discs; care should also be taken not to cover, even partially, one of the adjacent discs with the plate. A set of modular plates is strongly recommended.

157

2.8

ANTERIOR LANDMARKS OF THE VERTEBRAL LEVELS

The hyoid bone can be identified by palpation; it is approximately at the level of C4. The cricoid cartilage is at the level of the C6/7 disc. Due to the anterior and inferior obliquity of the 1st rib, the vertebrae T1 and T2 usually extend above the level of the manubrium (see Fig 6.2-4 ). The neck being in slight extension, the vertebrae C3 and C4 project in the upper third of the neck, C5 and C6 in the middle third, and C7 and T1 in the lower third. The following transverse cuts at the superior cervical level C3/4 ( Fig 6.2-12 ), the middle cervical level C6 ( Fig 6.2-13 ), and the lower cervical level C7 ( Fig 6.2-14 ) show the anterior anatomical relationship of the cervical spine.

158

6

Surgical anatomy of the spine

1

1 2 3 4

9 2 3 4 5

8 7

8 7

5

6

6 anterior right anterior right

Fig 6.2-12 Cross section of the neck at the C3/4 level (upper side of the slice). 1 tongue 2 pharyngeal cavity 3 carotid arteries 4 internal jugular vein 5 vertebral artery 6 spinal cord 7 C4 vertebra 8 C3/4 disc

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 6.2-13 Cross section of the neck at the C6 level (upper side of the slice). 1 cricoid cartilage 2 common carotid artery 3 internal jugular vein 4 longus colli muscle 5 C5/6 zygapophyseal joint 6 spinal cord in the spinal canal 7 C6 vertebra 8 vertebral artery 9 sternocleidomastoid muscle

6.2

159

Lower cervical spine

3

1 2 3

10 9

4 5

8 7

6

anterior right

Fig 6.2-14 Cross section of the neck at the level C7 (upper side of the slice). 1 right thyroid lobe 2 internal jugular vein 3 common carotid artery 4 scalenus anterior muscle 5 longus colli muscle 6 C7/T1 zygapophyseal joint 7 posterior nerve root ganglion of C7 8 left vertebral artery 9 esophagus 10 trachea

SUGGESTED READING

Curylo LJ, Mason HC, Bohlman HH, et al (2000) Tortuous course of the vertebral artery and anterior cervical decompression: a cadaveric and clinical case study. Spine ; 25(22):2860–2864. Ebraheim NA, An HS, Xu RX, et al (1996) The quantitative anatomy of the cervical nerve root groove and the intervertebral foramen. Spine ; 21(14):1619–1623. Ebraheim NA, Xu R, Knight T, et al (1997) Morphometric evaluation of lower cervical pedicle and its projection. Spine ; 22(1):1–6. Jones EL, Heller JG, Silcox DH, et al (1997) Cervical pedicle screws versus lateral mass screws. Anatomic feasibility and biomechanical comparison. Spine ; 22(9):977–982. Karaikovic EE, Daubs MD, Madsen RW, et al (1997) Morphologic characteristics of human cervical pedicles. Spine ; 22(5):493–500. Kowalski JM, Ludwig SC, Hutton WC, et al (2000) Cervical spine pedicle screws: a biomechanical comparison of two insertion techniques. Spine ; 25(22):2865–2867. Miller RM, Ebraheim NA, Xu R, et al (1996) Anatomic consideration of transpedicular screw placement in the cervical spine. An analysis of two approaches. Spine ; 21(20):2317–2322. Panjabi MM, Shin EK, Chen NC, et al (2000) Internal morphology of human cervical pedicles. Spine ; 25(10):1197–1205. Sodeyama T, Goto S, Mochizuki M, et al (1999) Effect of decompression enlargement laminoplasty for posterior shifting of the spinal cord. Spine ; 24(15):1527–1532. Uematsu Y, Tokuhashi Y, Matsuzaki H (1998) Radiculopathy after laminoplasty of the cervical spine. Spine ; 23(19):2057–2062. Xu R, Ebraheim NA, Nadaud MC, et al (1995) The location of the cervical nerve roots on the posterior aspect of the cervical spine. Spine ; 20(21):2267–2271. Xu R, Kang A, Ebraheim NA, et al (1999) Anatomic relation between the cervical pedicle and the adjacent neural structures. Spine ; 24(5):451–454.

160

6 6.3

SURGICAL ANATOMY OF THE SPINE CERVICOTHORACIC JUNCTION

1 1.1 1.2 1.3

Anatomy related to posterior procedures …………………………………………………………… Muscular relationships …………………………………………………………………………………… Landmarks for pedicle screw fixation in C7 …………………………………………………………… Landmarks for pedicle screw fixation in T1, T2, and T3 ……………………………………………

161 161 161 162

2 2.1 2.2 2.3 2.4

Anatomy related to anterior procedures ……………………………………………………………… Sternum …………………………………………………………………………………………………… Muscles ……………………………………………………………………………………………………… Vascular elements ………………………………………………………………………………………… Nervous elements …………………………………………………………………………………………

164 164 164 166 168

3

Suggested reading ………………………………………………………………………………………… 171

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

161

Serge Nazarian, Cyril Solari

6

SURGICAL ANATOMY OF THE SPINE

6.3

CERVICOTHORACIC JUNCTION

1

ANATOMY RELATED TO POSTERIOR PROCEDURES

1.1

MUSCULAR RELATIONSHIPS

The cervicothoracic junction is covered posteriorly by four muscular layers. • Trapezius muscle—its medial insertions are on the spinous processes from C7 to T12. • Rhomboid muscles. • Serratus posterior inferior and serratus posterior superior muscles. • “Spinal” muscles—they extend from the spinous processes to the transverse processes and posterior angles of the ribs. They form a 6–8 cm wide muscular band on each side of the midline. They insert on all the underlying bony elements.

At each level and on each side, a neurovascular bundle rises from the intercostal vessels and the nerve, runs backward below the transverse process, and reaches the muscular layers. These bundles should be respected as much as possible.

1.2

LANDMARKS FOR PEDICLE SCREW FIX ATION IN C7

Pedicle screws are usually not used for posterior fi xation at the cervical levels because the diameter of these pedicles is not large enough. Their close relationship with the spinal cord and the vertebral artery does not allow any error in their entry point or orientation.

162

6

Surgical anatomy of the spine

However in C7, the diameter of the pedicle is larger and the vertebral artery has not yet penetrated the transverse canal. The point of entry is located at the inferior border of the superior articular process of C7, on the vertical line crossing the middle of the articular surface ( Fig 6.3-1). The direction is oriented in the sagittal plane, at 90° with the local curve of the spine, and in the horizontal plane, at 15–45°, depending on how medial/lateral the entry point is. The entry point can be identified more securely after resection of the inferior half of the overlying inferior articular process (C6) to expose partially the lowest part of the superior articular surface of C7.

1.3

The length of the screw is about 25–30 mm and the recommended diameter is 3.5–4.0 mm.

The mediolateral situation can be identified by the vertical line passing at the lateral border of the inferior articular pro-

a

LANDMARKS FOR PEDICLE SCREW FIX ATION IN T1, T2, AND T3

Insertion of pedicle screws in T1, T2, and T3 is easier than in T5, T6, or T7, because the diameter of the pedicles is usually larger (chapter 6.4 Thoracic spine). The entry point can be identified according to several techniques. The craniocaudal topography is the same in all cases—it is located on the upper ridge of the transverse process or immediately above it ( Fig 6.3-2a ).

b

Fig 6.3-1a–b a Posterior view of vertebra C7 with landmarks for entry points for pedicle screw fixation. The axial line of the pedicle crosses the posterior cortex below the superior articular surface, on the vertical line passing by the middle of this surface. b Superior view of vertebra C7 with entry points and two directions of screws. The direction of the screw following the axis of the pedicle makes an angle of 15–45° with the sagittal plane. The value of this angle decreases if the entry point is more medial (right side).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

6.3

163

Cervicothoracic junction

cess of T1, T2, or T3—the penetration of the screw should be performed at the point where this vertical line crosses the bony ridge, which continues medially to the upper border of the transverse process ( Fig 6.3-2b ). Alternatively, the mediolateral situation can be identified by the vertical line passing through the middle of the upper articular surface of T1, T2, or T3. The inferior third of the overlying inferior facet—ie, the inferior facet of C7, T1, or T2—should be resected in order to identify the middle of the superior facet of T1, T2, and T3 ( Fig 6.3-2c ). The screw should be introduced at the point where this vertical line crosses the bony ridge, which continues medially to the upper border of the transverse process.

The direction of the screw makes a 15–30° angle with the sagittal plane, depending on the vertebral level and the more or less medial situation of the entry point ( Fig 6.3-2d ). None of these procedures has an accuracy of 100%. The techniques of identification of the entry point give a useful approximation which must be checked by an AP fluoroscopic control ( Fig 6.3-2e ) after the introduction of the tip of a K-wire through the cortex of the vertebrae to be instrumented. The K-wires are introduced by 2–3 mm and their topography related to the pedicle is checked by an AP fluoroscopic control. The use of a radiolucent operating table is necessary. If this kind of table is not available, a small laminotomy is recommended to localize the inner border of the pedicle, using the tip of a Penfield retractor.

1/2 1/2

cranial

a Fig 6.3-2a–e Landmarks allowing the identification of entry points for pedicle screws in T1, T2, and T3. a The upper ridge of the transverse process. The axis of the pedicle projects approximately at the level of this ridge or immediately above it.

b

c b The vertical line by the lateral border of the inferior articular process. It crosses the upper ridge of the transverse process at a point corresponding to the axis of the pedicle (especially valid for T1 and T2).

1/2 1/2

right

c The middle of the transverse diameter of the upper articular surface. A vertical line drawn from it crosses the upper ridge of the transverse process. The clinical photograph shows the situation after resection of the inferior third of the inferior facet of the overlying vertebra.

6

Surgical anatomy of the spine

2

ANATOMY RELATED TO ANTERIOR PROCEDURES (Fig 6.3-3, Fig 6.3-4, Fig 6.3-5, Fig 6.3-6)

The anterior part of the cervicothoracic junction is characterized by: • Its deepness, increasing from cranial to caudal. • Two major obstacles which are the sternum (knowing that its upper border projects at a level which varies from C7 to T3, according to the individual) and the main vessels of the superior mediastinum and the base of the neck, which must be retracted and sometimes ligatured to allow the anterior approach to the spine.

d

The management of these anterior obstacles, which are comprised of the sternum, the viscerae, the vessels, and the nerves lying in the upper mediastinum, is a major concern when performing this approach. 1/3 1/3

164

e Fig 6.3-2a–e d The direction of the screw according to the more or less medial situation of the entry point. e On the AP view, the tip of the inserted K-wire should be in the middle or in the lateral third of the oval area of the pedicle.

2.1

STERNUM

The splitting of the manubrium or the whole sternum is not always necessary. The horizontal projection of the suprasternal notch is located in 40% of cases in T2, in 50% in T3, and in 7% in T4. The splitting of the manubrium becomes necessary for the anterior management of upper thoracic spinal lesions, when the lesions are located below the suprasternal notch of the sternum.

2.2

MUSCLES

To achieve maximum exposure from the lower part of the cervical approach, inferior insertions of the sternocleidomastoid and infrahyoid muscles (omohyoid, sternohyoid, and sternothyroid muscles) should be divided. The inferior insertion of the sternocleidomastoid muscle is comprised of

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

6.3

165

Cervicothoracic junction

two heads ( F ig 6 . 3 -3a ); the sternal one is attached to the anterior aspect of the manubrium, and the clavicular one is attached to the middle third of the clavicula. These two heads should be divided in order to get access to the underlying elements and to retract them medially ( Fig 6.3-3b ). The omohyoid muscle is the most lateral element of the infrahyoid muscle layer; it is the lateral end of the middle cervical aponeurosis by which it is attached to the sternocleidohyoid muscle. The fi rst obstacle when approaching the cervicothoracic junction is to divide and retract this muscle. The middle cervical aponeurosis is perforated and crossed by the anterior and lateral jugular veins which must be divided before

cutting the other infrahyoid muscles. The inferior insertion of the sternocleidohyoid muscle is on the posterior aspect of the medial end of the clavicle, on the posterior sternoclavicular ligament, and on the adjacent part of the manubrium. The sternothyroid muscle lies behind the sternohyoid muscle, and its inferior attachment is on the posterior aspect of the manubrium and the 1st costal cartilage. The division of the attachments of these two medial muscles allows the medial retraction of the visceral axis and gives access to the anatomical elements of the upper mediastinum, mainly vascular, which are interposed on the approach to the cervicothoracic spine.

cranial

c ran

anterior

ial le f t

1 2 3

1 2

a Fig 6.3-3a–b a The inferior insertion of the sternocleidomastoid muscle. 1 sternal head 2 clavicular head

b b Anterolateral view. 1 omohyoid muscle 2 sternocleidohyoid muscle 3 middle cervical aponeurosis

166

6

2.3

Surgical anatomy of the spine

VASCULAR ELEMENTS

Veins ( Fig 6.3-4 )

The venous layer is in front of the arterial layer. The internal jugular vein enters into the thorax behind the two heads of the sternocleidomastoid muscle and anastomoses with the subclavian vein forming the brachiocephalic vein, behind the sternoclavicular articulation. The left brachiocephalic vein, slightly oblique, downward, and to the right side, has an almost transverse course in front of the arterial branches rising from the aorta and behind the manubrium. It receives on its superior edge the inferior thyroid veins. The right brachiocephalic vein is more vertical and stays on the right side, running downward and medially to join the left brachiocephalic vein, and so gives rise to the superior vena cava. Arteries ( Fig 6.3-4 )

The aortic arch runs obliquely backward and to the left in the direction of the left aspect of the spine. Its highest point is at the level of T4. It lies behind and below the sternal manubrium and the left brachiocephalic vein. The convexity of this arch gives rise to the three main branches of the aortic arch, successively, from the front to the back and from the right to the left. The arterial brachiocephalic trunk rises obliquely to the right, behind the manubrium, and divides behind the sternoclavicular articulation into the right common carotid artery and the subclavian artery. The left common carotid artery rises obliquely to the left behind the left sternoclavicular articulation. The left subclavian artery also rises behind the left sternoclavicular joint.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

The subclavian arteries display three portions: • The prescalenic portion, medial to the scalenus muscles, crosses anteriorly by the internal jugular vein, the subclavian vein, and the brachiocephalic venous trunk; this portion gives rise to the vertebral artery, which joins the foramen transversarium (transverse foramen) of C6; the internal thoracic artery, which runs down behind the anterior thoracic wall, at 15 mm from the lateral border of the sternum; the superior intercostal artery supplying the three fi rst intercostal spaces; and the thyrobicervicoscapular artery, which gives rise to the inferior thyroid artery after ascending to the transverse process of C6. This artery runs medially and passes between the vascular axis in the front and the vertebral artery behind; doing so, it describes a sinuous curve downward and then upward to join the inferior pole of the thyroid lobe, and then it divides into its terminal branches. In this very short last portion the inferior thyroid artery runs along the tracheoesophageal axis and crosses the inferior laryngeal nerve. • The interscalenic portion runs between the anterior and middle scalenus muscles. • The postscalenic portion runs through the costoclavicular corridor, behind the subclavian vein, and in front of the brachial plexus. Lymphatic vessels

The thoracic duct appears at the superior opening of the thorax at the left side of the esophagus. Then it runs behind the left subclavian artery. At the level of C7, it makes a loop and then runs anteriorly until its end at the posterior aspect of the left jugulosubclavian venous confluent. It can be injured during left inferior cervical approaches.

6.3

167

Cervicothoracic junction

cranial

c ra

cran ial

left

left

nia

l

le f t

1 8 1

8 7

2

1

2

2

8

3

7 6

7

4

3

6 5

a Fig 6.3-4a–c a Left anterolateral view—the innominate vein is intact. 1 retroesophageal space 2 left common carotid artery 3 innominate vein (left brachocephalic vein) hiding the origin of the branches of the aortic arch 4 aortic arch 5 divided and retracted sternum 6 right brachiocephalic artery 7 trachea 8 left border of the esophagus

3

6

4

5

5

5 4 5

b b Left anterolateral view—the innominate vein has been divided. 1 retroesophageal space 2 left common carotid artery 3 origin of the left subclavian artery 4 retrosternal fat and cardiac nerves 5 divided and retracted sternum 6 aortic arch 7 right brachiocephalic artery 8 trachea

c c Right anterolateral view—the sternum has been divided and retracted. 1 longus colli muscle and retroesophageal space 2 right recurrent nerve 3 trachea 4 right brachiocephalic artery 5 left brachiocephalic (innominate) vein 6 right internal jugular vein 7 right subclavian artery 8 right common carotid artery

168

6

2.4

Surgical anatomy of the spine

NERVOUS ELEMENTS

Vagus nerve

The vagus nerve runs downward in the vascular sheath, behind the vessels. Then it gives off the inferior laryngeal nerve (recurrent nerve) which crosses anteroposteriorly and lateromedially the vessel developed from the fourth aortic arch. This arterial derivate is represented on the left side by a part of the aortic arch and on the right side by the subclavian artery.

1

9

2

8

3 4

7 6

5

Recurrent nerve

The recurrent nerve (chapter 6.2 Lower cervical spine, Fig 6.2-10 and Fig 6.2-11) runs upwardly and medially to join the dihedric tracheoesophageal angle. So, on the left side, starting from the level of the fourth thoracic vertebra, the recurrent nerve has already reached the tracheoesophageal angle when it arrives at the superior opening of the thorax. The retraction of the visceral axis to the opposite side does not submit it to excessive tension. On the contrary, on the right side, the recurrent nerve, starting from the base of the neck, has not yet reached the visceral axis. It is more anterior, more lateral, and less vertical than its left homologous nerve. This difference in depth and obliquity explains its higher vulnerability to distraction in lower cervical spinal approaches. The retraction of the visceral axis to the opposite side exposes the nerve to overdistraction. The incidence of vocal paralysis reported is 2–12%. They mainly occur when approaching the cervicothoracic junction from the right side. 1–2% of paralysis cases do not recover.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

cranial

a Fig 6.3-5a–d a Upper side of a transverse cut through C6. 1 cricoid cartilage 2 internal jugular vein 3 longus colli muscle 4 vertebral artery 5 C5/6 zygapophyseal joint 6 spinal cord in the spinal canal 7 C6 vertebra 8 sternocleidomastoid muscle 9 common carotid artery

right

6.3

169

Cervicothoracic junction

1

1 2

3

4 5

14 13 12

4

6

11

6 7 8

3 10 9

2 15

5

8 7

10

9

cranial

cranial

b Fig 6.3-5a–d b Upper side of a transverse cut through C7. 1 trachea 2 left thyroid lobe 3 internal jugular vein 4 common carotid artery 5 scalenus anterior muscle 6 longus colli muscle 7 C7/T1 zygapophyseal joint 8 C7 nerve 9 left vertebral artery 10 esophagus

right

c

right

c Upper side of a transverse cut through T1. 1 sternohyoid and sternothyroid muscles 2 thyroid gland 3 right common carotid artery and internal jugular vein 4 right recurrent nerve 5 anterior scalenus muscle 6 middle scalenus muscle 7 sympathetic trunk 8 longus colli muscle 9 T1 vertebra 10 1st rib 11 C7/T1 disc 12 brachial plexus 13 esophagus 14 left recurrent nerve 15 trachea

170

6

Surgical anatomy of the spine

1

17

2

16 15

3 4

14 13

5

12 11

6

10

7

9

8

cranial left

d Fig 6.3-5a–d d Lower side of a transverse cut through T1/2. 1 trachea 2 left clavicle 3 left common carotid artery 4 left subclavian vein 5 left subclavian artery 6 T1 vertebra 7 left lung 8 left 2nd rib 9 T2 vertebra

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

10 11 12 13 14 15 16 17

T1/2 disc right lung esophagus right subclavian artery and vein right clavicle right internal jugular vein right common carotid artery sternocleidomastoid muscle

6.3

171

Cervicothoracic junction

3

SUGGESTED READING

Sharan AD, Przybylsky GJ, Tartaglino L (2000) Approaching the upper thoracic vertebrae without sternotomy or thoracotomy: a radiographic analysis with clinical application. Spine ; 25(8):910–916. Cauchoix J, Binet JP (1957) Anterior surgical approaches to the spine. Ann R Coll Surg Engl; 21(4):234–243. Ebraheim NA, Tremains M, Xu R, et al (2000) Anatomic study of the cervicothoracic spinal nerves and their relation to the pedicles. Am J Orthop; 29(10):779–781.

1 2

3 4 6 5

cranial posterior

Fig 6.3-6 Sagittal midline cut. 1 larynx, cricoid cartilage 2 spinal cord 3 esophagus 4 trachea 5 manubrium 6 left brachocephalic vein (innominate vein)

172

6 6.4

SURGICAL ANATOMY OF THE SPINE THORACIC SPINE

1 1.1 1.2 1.3

Anatomy Anatomy Anatomy Anatomy

related related related related

to to to to

posterior procedures……………………………………………………………… lamina hook insertion …………………………………………………………… pedicle hook insertion …………………………………………………………… pedicle screw insertion …………………………………………………………

173 174 175 177

2 2.1 2.1.1 2.1.2 2.2 2.3 2.4 2.5 2.5.1 2.5.2

Anatomy related to anterior procedures ……………………………………………………………… Thoracic walls ……………………………………………………………………………………………… Anterolateral thoracic wall ……………………………………………………………………………… Lateral thoracic wall ……………………………………………………………………………………… Right thoracic cavity ……………………………………………………………………………………… Left thoracic cavity ……………………………………………………………………………………… Close relationships of the thoracic spine ……………………………………………………………… Anatomical variations of thoracic vessels …………………………………………………………… Variations of the azygos venous system ……………………………………………………………… Cardiovascular variations …………………………………………………………………………………

182 182 182 183 184 185 186 188 188 189

3

Suggested reading ………………………………………………………………………………………… 190

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

173

Serge Nazarian, Cyril Solari

6

SURGICAL ANATOMY OF THE SPINE

6.4

1

THORACIC SPINE

ANATOMY RELATED TO POSTERIOR PROCEDURES

Posterior approaches to the thoracic spine are required to expose the posterior aspect of the thoracic vertebrae, spinous processes, laminae, zygapophyseal joints, transverse processes, and costovertebral joints. The posterior approach requires a midline skin incision and a detachment of the muscle insertions from the vertebrae. The skin and the bony palpable landmarks

The palpable bony landmarks consist mainly of the tips of the spinous processes and the scapula. The posterior arches of the ribs cannot be numerically identified. The inferior angle of the scapula corresponds approximately to the 7th rib.

Muscular layers

Posteriorly, four muscular layers cover the thoracic spine. • Trapezius muscle—its medial insertions are on the spinous processes from C7 to T12. • Rhomboid muscles • Serratus posterior superior and inferior muscles • “Spinal” muscles Vasculonervous bundles

At each level, a neurovascular bundle composed of a branch from the intercostal artery, a branch to the intercostal vein, and the posterior ramus of the intercostal nerve, runs backward below the transverse process and supplies the aforementioned muscles.

174

6

1.1

Surgical anatomy of the spine

ANATOMY RELATED TO LAMINA HOOK INSERTION

In the thoracic spine lamina hooks are used in a supralaminar position: • At the upper end of an instrumentation, to make a claw with an ipsilateral pedicle hook placed on the underlying vertebra, or to secure the purchase of a pedicle screw. • At the middle part of a construct, when a slight distraction is required or when the use of a pedicle screw is anatomically impossible. The anatomical details for hook insertion are described in Fig 6.4-1.

1 cranial

a

cranial

right

Fig 6.4-1a–e a The lamina of a thoracic vertebra is hidden by the spinous process and lamina of the overlying vertebra. To expose the superior border of the lamina, the overlying spinous process should be cut and removed.

b

right

b Posterior view after section of the spinous process and the lower part of the laminae. Once the spinous process is removed, the lamina appears; its upper border is hidden by the insertion of the yellow ligament (1 ligamentum flavum). This ligament shows thin vertical striations.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

cranial

c

right

c Posterior view of the lamina after resection of the yellow ligament. Using a thin curette, the inferior attachment of the yellow ligament is delicately detached. This detachment exposes the superior border of the thoracic lamina which usually describes a concave line (arrow).

6.4

175

Thoracic spine

1.2

ANATOMY RELATED TO PEDICLE HOOK INSERTION

Pedicle hooks are inserted between the inferior and superior facet of two contiguous thoracic vertebrae. Their bifid tip is made to straddle the inferior border of the pedicle with its bifurcated cranial end ( Fig 6.4-2b–c ). cranial right

d

*

6 mm

cranial

e

right

Fig 6.4-1a–e d The lamina is sometimes irregularly delineated due to the partial calcification of the yellow ligament. e Posterior view of the lamina after preparation for hook placement (asterisk). The width of the lamina hook is 3–5 mm. Protecting the dura with a thin dissector and using a 2 mm thin Kerrison rongeur, the upper border of the lamina is cut out in order to obtain a 6 mm wide, straight lined, horizontal border, which fits perfectly to the blade of the lamina hook.

The inferior zygapophyseal facet is roundly shaped from its medial to lateral aspect. It is partly covered by the capsule. This capsule should be removed to clearly identify the limits of the facet. After removal of the capsule, the inferior end of the facet is cut out to fit the hook perfectly ( Fig 6.4-2a ). The medial limit should be correctly identified. It is located at the point of junction between the lamina and the facet. A pedicle hook fi nder, which has the same breadth as the hook, is introduced between the superior and inferior facets. The pedicle lies in the middle of the facet, in approximate cranial alignment. The fi nder should be moved cranially, parallel to the middle axis of the zygapophyseal column and aiming for the middle of the overlying facet. Care should be taken not to deviate medially, in order to avoid any injury to the dural sac or spinal cord. Once the inferior border of the pedicle is reached by the tip of the fi nder, its bifurcated end fits the inferior border of the pedicle, thereby avoiding any possibility of lateral or medial displacement. The stability of the hook can be improved to some extent by slightly impacting the hook in the inferior border of the pedicle.

176

6

Surgical anatomy of the spine

* cranial

a

right

Fig 6.4-2a–c a The inferior border of the inferior articular process is round shaped (arrow). To be mechanically able to receive and withstand the vertical strains transmitted by a pedicle hook, it must be reshaped by a strictly transverse cut at its junction with the lamina. A short vertical cut using a thin chisel is fi rst required medially, at the junction between the joint and the lamina, to mark the limit of the medial extent of the transverse cut. Then the transverse cut is carried out using the same chisel, strictly perpendicular to the vertical cut and the vertical axis of the zygapophyseal mass. (Asterisk = preparation for pedicle hook.)

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

cranial

b b The hook fi ts the inferior border of the facet.

c

anterior

c Its bifurcated tip straddles the inferior border of the pedicle.

6.4

Thoracic spine

1.3

ANATOMY RELATED TO PEDICLE SCREW INSERTION

The main concerns in this procedure are in identifying the correct entry point and in drilling in the accurate direction. The screws, obliquely introduced toward the midline, should be driven into the vertebral body as far as possible without perforating the cortex. Due to the anatomical variability and the close relationship of the pedicles with the spinal cord and nerve roots, there is no technique with a reliability of 100% reported today which allows the positioning of pedicle screws without any control device. Pedicle screw fi xation should be addressed only if a navigational system or AP and lateral fluoroscopic monitoring are available, or if an open laminar technique is being applied.

177

Identification of entry points

The second concern is in identifying the entry point with reference to reliable and reproducible anatomical landmarks, such as: the middle of the transverse diameter of the superior facet of the vertebra, the lateral border of the superior facet, the inferior and lateral borders of the overlying inferior facet, the superior border of the transverse process, the midline of the transverse process, and the upper ridge of the transverse process. The anatomical landmarks should be easy to fi nd and identify. Distances cannot be taken as reference values because a given distance is not similarly applicable to any vertebra of any individual. The anatomical landmarks described here are related to the surgical technique.

The anatomical landmarks of the entry points and the direction of the screws, from the entry point into the vertebral body, vary according to the level and the individual anatomy. Evaluation of pedicles

The fi rst hurdle is fi nding out whether the use of pedicle screws is at all possible or not. This depends on the size of the pedicle, and especially, its transverse diameter; the vertical diameter is not a major issue. For a surgeon who is familiar with the technique and the AP x-rays, a single AP and a lateral view are enough. Another technique of evaluation and preoperative planning consists of making a CT scan of the respective pedicles at each level and measuring their transverse diameter ( Fig 6.4-3 ).

a

b

Fig 6.4-3a–b Preoperative evaluation of the transverse diameter of the pedicles. a Using a plain x-ray. b Using a CT scan.

178

6

Surgical anatomy of the spine

An anatomical analysis of thoracic vertebrae from 20 spines showed that craniocaudally the middle axis of the pedicle is near the inferior border of the superior articular surface, between the inferior border of the facet and the upper ridge of the transverse process. These landmarks are easy to fi nd for any thoracic vertebra from T1 to T11 ( Fig 6.4-4a ). Mediolaterally, the most reliable landmark is the superior facet; the middle axis of the pedicle runs below the middle of the facet or more often laterally to it; the middle of the superior facet or the point at the junction of the middle and lateral thirds of the facet can be taken as a primary landmark ( Fig 6.4-4b ). Combining these data, the entry point of the thoracic pedicles, from T1 to T11, can be placed at the intersection of a transverse line passing by the superior edge of the transverse process and a vertical line passing by the middle of the superior 1

4

2 3

facet or by a point between its middle and lateral thirds. This point is 2–3 mm lateral from the bottom of the posterior concavity of the lamina ( Fig 6.4-4c ). The overlying inferior facet is not a reliable landmark; its distal half can be resected with a thin chisel to disclose the underlying superior facet (except at the upper end of a construct). 1 1/3 1/3 1/3 1/2

1/2

2

b

c

b Other landmarks. 1 border between the middle and lateral thirds 2 middle of the superior facet

c Bottom of the posterior concavity of the lamina.

cranial

1

a

1

right

Fig 6.4-4a–d The landmarks. a The classic bony landmarks. 1 lateral border of the superior facet 2 upper ridge of the transverse process 3 middle line of the transverse process 4 middle of the transverse diameter of the superior facet

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

2 cranial

d

right

d The mammillary process of the 12th thoracic vertebra. 1 mammillary process 2 accessory process

6.4

179

Thoracic spine

The 12th thoracic vertebra (T12) is a special vertebra, characterized by the absence of transverse processes. However, it displays a relevant anatomical landmark ( Fig 6.4-4d ) which is the anatomical equivalent of the “mammillary process” in the lumbar vertebrae. Another, smaller process lies lateral and inferior to it. The entry point of the pedicle can be primarily placed between these two processes or at the tip of the “mammillary” process. Once all the entry points are identified, a 1.6 mm K-wire is introduced superficially (approximately 2 mm) through the cortex of the vertebra ( Fig 6.4-4c ). A fluoroscopic control picture is made; the AP view is of major relevance. A radiolucent table should be used for such a surgery. The tip of the K-wire should lie in the middle of the oval area of the pedicle, or between its middle and its lateral border ( Fig 6.4-5 ). If not, the wire should be replaced and a new control performed, if necessary.

cranial right

Fig 6.4-5 The tip of the K-wire should lie in the area of the pedicle, preferably in its middle or its lateral half.

Trajectory of drilling

The third concern is the direction of the drilling. The screw should enter through the pedicle and penetrate the vertebral body, as far as possible. The pedicle feeler is the most appropriate instrument for performing the drilling. It allows progressive penetration of the pedicle as it adapts its direction to the inner contour of the cortical walls of the pedicle, thus reducing the risk of their perforation. In the sagittal plane for a monoaxial screw, the drilling direction is perpendicular to the spinal curve at the drilling level ( Fig 6.4-6 ). In the transverse

cranial anterior

90°

Fig 6.4-6 For a monoaxial screw the drilling direction in the sagittal plane is perpendicular to the spinal curve at the drilling level.

180

6

Surgical anatomy of the spine

plane, the drilling direction in relation to the sagittal plane is theoretically 20 –25° in T1, 15–20° in T2, 10 –15° in T3, and 5–10° from T4 to T12 ( Fig 6.4-7 ). In case of a polyaxial screw, the direction of the drilling may be oriented 10–20° downward. When the use of pedicle screws is impossible due to the narrowness of the pedicle, a pedicle hook can be used instead, which

can easily fit the vertical rod because it is in the same pedicular alignment as the neighboring screws. The inconvenience of the screw-hook combination is that the hook is bulkier and more superficial than the screws. Another alternative consists of using extrapedicular screws. These screws obliquely penetrate the tip of the transverse processes in the transverse plane, in the direction of the

T1

T1

T1

T2 T2

T2 a T8

T8

b

T8

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 6.4-7a–d The landmarks of the entry point of the pedicle screws at the different levels in the thoracic spine. a T1 and T2, posterior and superior views, AP x-ray. The point of entry in T1, T2, and T3 is at the intersection of the superior ridge of the transverse process and a vertical line from the lateral border of the inferior facet or the middle of the superior facet. b T8, posterior view and superior views, AP x-ray. The point of entry in T4–9 is at the intersection of the superior edge of the transverse process and a vertical line from the middle or midlateral thirds of the superior facet. It is 2–3 mm lateral from the bottom of the posterior concavity of the lamina.

6.4

181

Thoracic spine

vertebral body and the midline ( Fig 6.4-8 ). This construct can be useful, although it is bulky, under the laterovertebral muscles. This technique also involves the risk of violating the pleural cavity.

T11

T11

T11

c

T12 T12 d

Fig 6.4-7a–d c T11, posterior and superior views, AP x-ray. The entry point in T10 and T11 is at the level of a little tubercle located at the superior ridge of the transverse process. d T12, posterior view, AP x-ray. The entry point is at the level of the mammillary process or between the mammillary process and the accessory process (a remnant of the transverse process).

Fig 6.4-8 Extrapedicular screws—superior view.

182

6

Surgical anatomy of the spine

2

ANATOMY RELATED TO THE ANTERIOR PROCEDURES

2.1

THORACIC WALLS

From the standpoint of the surgical approach, the distinction between an anterolateral thoracic wall and a purely lateral thoracic wall is important.

Branches of the superficial cervical plexus and superficial cervical vessels

These vessels reach the region passing in front of the clavicle. Anterior branches of the intercostal vessels and nerves

These structures lie 1–2 cm lateral to the sternum. 2.1.1 ANTEROLATERAL THORACIC WALL

Branches of the external thoracic vessels and the lateral branches of the intercostal nerves

The limits of the region are represented by the clavicle above, the common costal cartilage below, the sternum on the midline, and the midaxillary line laterally.

They enter the region along the anterior axillary line. Numerous hemostases are required during the surgical approach due to the transection of these vessels.

Skin, subcutaneous tissue, mammary gland

Thoracic wall

The breast extends in average from the 3rd to the 7th rib, its inferior limit is the submammary fold usually used in wide anterolateral thoracotomies. The mammary gland, the skin, and subcutaneous tissue are easily mobilized on the fascia of the pectoral muscle.

The thoracic wall itself is composed of the anterior arch of the ribs and the costal cartilages extending to the lateral border of the sternum. The costal cartilages are 4–5 cm long. The fi rst costal cartilage is under the medial end of the clavicle, the second lies at the level of the sternal angle, the third, fourth, and fi fth are at the lateral border of the sternal body, and the sixth and seventh are on the basis of the xiphoid process. The costal cartilages and ribs are counted from the second one, which is easily identified at the lateral border of the sternal angle.

Superficial muscular layer

The superficial muscular layer is composed of the pectoralis major and pectoralis minor muscles anteriorly, the serratus anterior muscle medially, and the superior digitations of the abdominal muscles. The pectoralis major muscle inserts on the clavicle, the sternum, the ribs, and the anterior layer of the sheath of the rectus abdominis muscle. The fibers converge laterally toward the anterior aspect of the humeral metaphysis. The pectoralis minor muscle extends its digitations from the tip of the coracoid process to the 3rd, 4th, and 5th ribs. The serratus anterior muscle is a large flat muscle extending from the medial border of the scapula to the anterior arches of the fi rst ten ribs; it curves around along the lateral thoracic wall; its nerve supply is provided by the long thoracic nerve which runs down along the posterior axillary line.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Content of the intercostal space

The content of the intercostal space includes five layers: the external intercostal muscle; the external fibrocellular layer; the middle intercostal muscle; the middle intermuscular cellular tissue containing the intercostal vein, artery, and nerve in its upper part, in the costal groove (located at the inferior border of the overlying rib); and the internal intercostal muscle.

6.4

183

Thoracic spine

The intercostal region is covered on its deeper surface by the endothoracic fascia and the parietal pleura. Deeper than the anterior part of the intercostal space, 10–15 mm from the lateral border of the sternum, lie the internal thoracic artery and vein. Right anterolateral approach

The right anterolateral approach allows access to the six fi rst intercostal spaces, and especially to the fourth and fi fth. The opening of the fi fth intercostal space by an additional section of the fi fth costal cartilage allows access to the thoracic cavity. The thoracic cavity is thus clearly exposed, showing the diaphragmatic cupula below and the spine posteriorly, covered by the intercostal bundles, the azygos vein, and the mediastinal pleura. The exposure of the spine requires the anterior retraction of the lung. The thoracic vertebrae from T4 to T10 can easily be approached. The approach to T2, T3, T11, and T12 necessitates special care.

Superficial layer—skin and fat tissue

The skin is slightly adherent to the deeper layers in the axillary fossa. The anterior wall of the axillary fossa is the pectoralis major muscle, the posterior margin is the latissimus dorsi muscle. The topography of the anterior axillary line is given by the pectoralis major muscle. The posterior axillary line passes by the latissimus dorsi muscle. The midaxillary line lies between the anterior and posterior lines. Muscular layer

The muscular layer is composed of the lateral part of the pectoralis major muscle, anteriorly; the lateral part of the latissimus dorsi muscle covering the scapula, posteriorly; the serratus anterior muscle medially against the rib cage; its anterior and inferior digitations mix with those of the external oblique muscle of the abdomen. At the top of the middle axillary line are the upper insertions of the medial muscles of the arm: the coracobrachialis muscle, biceps muscle, and triceps muscle.

2.1.2 LATERAL THORACIC WALL Axillary veins and artery, nerves of the upper limb

Its exposure requires the abduction and elevation of the upper limb. The limits of the lateral wall are represented by: • The top of the axillary fossa, cranially. • The lateral border of the pectoralis major muscle, anteriorly. • The lateral border of the latissimus dorsi muscle, posteriorly. • The inferior border of the rib cage, caudally. The lateral thoracic wall is composed of three parietal layers.

These vessels and nerves lie between the brachial muscles. The collateral branches of these neurovascular elements, supplying the thoracic wall, can be injured during the approach. The lateral thoracic pedicle lies posterior to the pectoralis major muscle. The collateral cutaneous branches of the intercostal nerves, divide into anterior and posterior branches to the subcutaneous tissue. The thoracodorsal artery and vein arise from the subscapular vessels and run down the middle part of the axilla. The long thoracic nerve lies behind the thoracodorsal vessels under the latissimus dorsi muscle; it should be preserved in order to avoid altering the respiratory function of the serratus anterior muscle. Several lymph nodes lie along the vessels of the axillary fossa, embedded in the fat tissue.

184

6

Surgical anatomy of the spine

Lateral thoracic wall

The lateral wall itself is composed of the ribs and the intercostal layers. The lateral part of the costal arches and the intercostal spaces are positioned obliquely downward and forward. The five layers of muscles and cellular tissue are the same as those found in the anterolateral wall.

the spine. It can be pushed down using a fan retractor. Costovertebral joints are visible through the parietal pleura. The mediastinal elements can be exposed by opening the parietal pleura. Three mediastinal levels can be identified with respect to the arch of the azygos vein.

Lateral approach

The lateral approach allows access to the spine through the intercostal spaces. The patient is placed in a right or left lateral decubitus position. The upper and middle thoracic vertebrae are more easily approached by the right side, whereas the thoracolumbar vertebrae are approached by the left side. The thoracic cavity is thus clearly exposed, showing the diaphragmatic cupula below and the spine posteriorly. The thoracic spine is covered by the intercostal bundles, the azygos vein on the right side, the aorta on the left side, and the mediastinal pleura. The exposure of the spine requires an anterior retraction of the lung and an inferior retraction of the diaphragmatic cupula. The approach to the thoracic vertebrae from T4 to T10 is easily performed from the right side. The approach to T2 and T3 require special care. T11, T12, L1, and L2 can be approached from the left.

2.2

1 2 3 4 12

11 10 5 6

9

RIGHT THORACIC CAVIT Y

The thoracic spine is located on the midline behind the mediastinum ( Fig 6.4-9 ). Its exposure requires the retraction of the lung and the diaphragmatic cupula, and a longitudinal incision of the parietal pleura. The exposure of the spine is possible if the right lung is collapsed and retracted anteriorly and medially. The diaphragmatic cupula is also an obstacle to

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

8

cranial

7 anterior

Fig 6.4-9 Right lateral view of the mediastinum and spine. 1 brachiocephalic artery 2 trachea 3 vagus nerve 4 azygos arch 5 heart, right atrium 6 esophagus 7 diaphragm 8 azygos vein 9 intercostal bundle (artery and veins) 10 right sympathetic trunk 11 thoracic spine 12 rib heads

6.4

185

Thoracic spine

Supraazygos level—corresponding to T2 and T3

In the anterior part of this level lie the brachiocephalic vein, the internal thoracic artery anterior to it, and the phrenic nerve along its right flank. In the posterior part lie the brachiocephalic artery, the trachea, and the right vagus nerve crossing them laterally; the esophagus is the most posterior element in contact with the spine. The 2nd, 3rd, and 4th intercostal arteries and their corresponding veins run obliquely between the spine and the esophagus; the sympathetic ganglions lie laterally to the heads of the ribs. Arch of the azygos vein

The azygos arch lies at the level of T4. It runs from the spine posteriorly to the posterior aspect of the superior vena cava anteriorly. During this course, it crosses the right flank of the esophagus, the right vagus nerve, the trachea, and the brachiocephalic artery. Its posterior superior angle receives the intercostal veins from the right upper intercostal spaces.

2.3

LEF T THORACIC CAVIT Y

The left side of the thoracic spine becomes visible after collapsing and retracting the left lung ( Fig 6.4-10 ). The left diaphragmatic cupula should also be retracted downward using a fan retractor. The opening of the mediastinal pleura along the costovertebral joints allows disclosure of the three mediastinal levels with respect to the aortic arch.

1 14 13

2

12

Infraazygos level

The infraazygos level extends from T5 to T10/11. From the front to the back, this area displays the right pulmonary hilus, the right atrium covered by its pericardium, the esophagus and the right vagus nerve at its flank, and the azygos vein. The intercostal vascular bundles lie between the azygos vein and the intercostal spaces. The intercostal arteries coming from the thoracic aorta cross the midline, in front of the spine, at the left of the azygos vein. The sympathetic ganglions and splanchnic nerves run down between the azygos vein and the costovertebral joints. A loose connective tissue surrounds the mediastinal viscerae, usually allowing an easy dissection.

3

11

4 5

10 9 6

8 cranial anterior

7

Fig 6.4-10 Left lateral view of the mediastinum and spine. 1 thoracic spine 2 rib head 3 accessory hemiazygos vein 4 intercostal bundle 5 left sympathetic trunk 6 hemiazygos vein 7 diaphragm 8 thoracic aorta 9 left vagus nerve 10 heart, left atrium 11 vagus nerve 12 thoracic duct 13 left subclavian artery 14 esophagus

186

6

Surgical anatomy of the spine

Supraaortic level corresponds to T2 and T3

From the front to the back, the following elements are found: the brachiocephalic vein with the left phrenic nerve along its left flank, the left common carotid artery, the left vagus nerve along its left flank, the left subclavian artery, the thoracic duct, and the esophagus.

2.4

CLOSE RELATIONSHIPS OF THE THORACIC SPINE

The azygos venous system

The azygos venous system is shown in Fig 6.4-11 and it is composed of the following veins. Azygos vein

Aortic arch

The aortic arch lies at the level of T4. From the front to the back, the aortic arch is in relation with the phrenic nerve, the vagus nerve, the loop of the left inferior laryngeal nerve, and the left superior intercostal arteries along the lateral aspect of the spine. The upper intercostal veins form the accessory azygos vein. Infraaortic level

The infraaortic level is composed of the left lung and hilus, and the left ventricle and atrium covered with their pericardium. Behind the heart, the esophagus appears as well as the left vagus nerve along its left anterior surface. The anterior surface of the spine is hidden by the descending thoracic aorta. The intercostal arteries arise from the posterior lateral aspect of the aorta with the upper arteries running obliquely upward, the middle arteries running horizontally, and the lower arteries running slightly downward. The intercostal vascular bundles lie against the flank of the spine, and the hemiazygos and accessory hemiazygos vein cross over their left surfaces. These veins pass posterior to the thoracic aorta at the T6/7 level and join the azygos vein. The sympathetic trunk and splanchnic nerves lie anterior to the heads of the ribs.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

The azygos vein originates in front of the 12th thoracic vertebra from the confluence of a medial vein arising from the inferior vena cava and a lateral vein formed by the anastomosis of the twelfth intercostal vein and the ascending lumbar vein. From this origin, the azygos vein runs upward along the right lateral aspect of the spine until the level of T4. Then it leaves the spine and runs anteriorly, passing above the left pulmonary pedicle to join and fi nish its course in the posterior aspect of the superior vena cava. During its course, the azygos vein is joined by the hemiazygos vein and the accessory hemiazygos vein on its left side and the intercostal veins on its right side. Hemiazygos vein

The hemiazygos vein is formed by the confluence of three veins (a medial vein coming from the left renal vein and a lateral vein formed by the anastomosis of the twelfth intercostal vein with the ascending lumbar vein). It runs upward along the left aspect of the spine until the T6/7 level. Then it crosses the midline behind the thoracic aorta and joins the azygos vein. Accessory hemiazygos vein

The accessory hemiazygos vein results from the confluence of the upper intercostal veins. It runs down along the left aspect of the thoracic spine until T6/7. At this level it crosses the spine from left to right to join the azygos vein.

6.4

187

Thoracic spine

1 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

left jugular vein left subclavian vein left superior intercostal vein accessory hemiazygos vein hemiazygos vein 12th left intercostal vein left renal vein left ascending lumbar vein left iliac vein right iliac vein right ascending lumbar vein inferior vena cava right renal vein 12th right intercostal vein azygos vein right superior intercostal vein superior vena cava right brachiocephalic vein

2 3

17

The thoracic duct penetrates the inframediastinal space behind the aorta. It runs cranially and medially to the level of T4. Then it runs progressively to the left side of the spine and enters the anterior cervical region. Intercostal arteries

16

6

The intercostal arteries arise from the posterior lateral surface of the aorta. The fi rst four to six intercostal arteries run from T4 toward the upper adjacent vertebrae very obliquely, almost vertically, over the anterior aspect of the spine. The preservation of these arteries necessitates an oblique or vertical surgical incision to approach to the spine. The middle intercostal arteries run anterior to the spine in a transverse direction. The lower intercostal arteries from T9 to T12 run obliquely downward, anterior to the spine.

7

Sympathetic trunk

4

15

5

14

13 12

8

11

9 10

Thoracic duct

cranial left

Fig 6.4-11 The azygos venous system in its typical morphology.

The sympathetic trunks (see Fig 6.4-9, Fig 6.4-10 ) run along the lateral aspects of the spine. They are in a lateral position with respect to the heads of the ribs in the upper half of the thoracic spine and in a medial position in its lower half. The splanchnic nerves rise from the level of T6 and then run caudally in a medial position with respect to the sympathetic trunks. Anterior longitudinal ligament

The anterior longitudinal ligament covers the anterior surface of the spine. It adheres very strongly to the intervertebral discs and the adjacent end plates. Its detachment is not easy to perform, and it is not recommended except in the case of a perivertebral infection or pediatric spinal deformities.

188

6

Surgical anatomy of the spine

Intervertebral foramen

The intervertebral foramina are located at the level of the rib heads and intervertebral discs. They contain the intercostal nerve and its spinal ganglion, the spinal artery and the foraminal veins. These veins are particularly numerous and often form a foraminal plexus. The spinal arteries rise from the intercostal arteries and penetrate the spine through the foramina to supply the neuromeningeal elements. The intercostal nerves run in close relationship with the medial aspect and the lower border of the upper pedicle of each foramen. In the middle and lower foramina the intercostal nerves run transversely through the middle of the foramen or obliquely toward the border of the lower pedicle. Spinal canal

The spinal canal contains the epidural fat, the epidural venous plexuses, the dura mater, the spinal cord, and the metameric emergences of the spinal nerve roots. The cord is attached to the dural sac on each side by the denticulate ligament.

2.5

ANATOMICAL VARIATIONS OF THORACIC VESSELS

2.5.1 VARIATIONS OF THE A ZYGOS VENOUS SYSTEM

Nine different types are distinguished, whereby type I was described above. Type 2

The hemiazygos vein is a short trunk joining the azygos vein at the level of T10. The overlying veins are small trunks which anastomose directly with the azygos vein. Type 3

The hemiazygos vein does not exist. The twelfth intercostal vein joins the inferior vena cava directly. The other intercostal veins anastomose directly with the azygos vein. The accessory hemiazygos vein does exist. Type 4

The azygos system is reduced to a median azygos vein. It receives the upper lumbar veins and the right and left intercostal veins. The midline venous axis can be preserved during the approach. Right and left superior intercostal veins anastomose with the arch of the azygos vein. Type 5

This is a symmetrical azygos system composed of two azygos veins. The right one joins the superior vena cava, the left one joins the brachiocephalic venous trunk.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

6.4

189

Thoracic spine

Type 6

The azygos and hemiazygos veins fuse at the level of T7 on the left flank of the spine and form a single paramedian venous axis receiving the intercostal and lumbar veins on both sides. At the level of T11, the azygos and the accessory azygos vein display a typical position. The inferior part of the azygos vein is lacking from T7 to T11.

2.5.2 CARDIOVASCULAR VARIATIONS

Overlooking some of the following anomalies in the cardiovascular system may result in catastrophic outcomes. Situs inversus

There is an inversion of the position of the great supracardiac vessels.

Type 7

The origin of the azygos vein is in the lumbar region. It receives the lumbar veins on both sides and the intercostal veins on the left side only. The two hemiazygos veins anastomose and form a common trunk anastomosing the azygos vein.

Coarctation of the aorta proximal to a ductus arteriosus

Type 8

Coarctation of the aorta distal to a ductus arteriosus

Absence of the hemiazygos vein associated with a hypertrophy of the accessory azygos vein.

The arterial ligament remains collapsed without the presence of an aortopulmonary shunt.

Type 9

Duplication of the aortic arch

Absence of the suprahepatic part of the inferior vena cava compensated by a very large azygos vein joining the superior vena cava. During surgery this anomaly requires the preservation of the azygos venous axis.

This creates an arterial loop around the visceral axis of the neck. It hinders the surgical mobilization. The duplication must not be mistaken for a large vein which may be ligatured and divided.

A shunt persists between the descending aorta and the pulmonary artery. This results in arterial hyperpressure in the vessels arising from the aortic arch.

Retroesophageal right subclavian artery

This artery arises from the left side of the thoracic aorta and runs behind the esophagus to reach the right upper limb. The ligature of this artery must be avoided.

190

6

Surgical anatomy of the spine

3

Abnormal origin of the left common carotid artery

This artery arises from a common trunk with the right brachiocephalic trunk. Care should be taken not to confuse this artery with a left brachiocephalic venous trunk. Its ligature may produce severe brain damage. Abnormal termination of the superior vena cava

The superior vena cava lies to the left of the aorta and left atrium, and it terminates directly in the coronary sinus. The superior vena cava should not be mistaken for a left brachiocephalic venous trunk and ligatured. Duplication of the superior vena cava

One vein lies on the right side but does not receive the left brachiocephalic venous trunk. The other is the vertical left brachiocephalic trunk passing to the left of the aorta and left atrium, and terminating directly in the coronary sinus. Abnormal return of the pulmonary veins

The left brachial systemic venous trunk ends in the left atrium. The interatrial shunt leads to an overloading of the right atrium. Ligation of the brachiocephalic trunk on the midline can induce venous stasis of the left hemiface.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

SUGGESTED READING

Cinotti G, Gumina S, Ripani M, et al (1999) Pedicle instrumentation in the thoracic spine. A morphometric cadaveric study for placement of screws. Spine ; 24(2):114–119. Ebraheim NA, Jabaly G, Xu R, et al (1997) Anatomic relations of the thoracic pedicle to the adjacent nervous structures. Spine ; 22(14):1553–1556. Ebraheim NA, XU R, Ahmad M, et al (1997) Projection of the thoracic pedicle and its morphometric analysis. Spine ; 22(3):233–238. Husted DS, Yue JJ, Fairchild TA, et al (2003) An extrapedicular approach to the placement of screws in the thoracic spine: an anatomic and radiographic assessment. Spine ; 28(20):2324–2330. Marchesi D, Schneider E, Glauser P, et al (1988) Morphometric analysis of the thoracolumbar and lumbar pedicles: anatomo-radiologic study. Surg Radiol Anat ; 10(4):317–322. McLain RF, Ferrara L, Kabins M (2002) Pedicle Morphometry in the upper thoracic spine: limits to screw placement in older patients. Spine ; 27(22):2467–2471. Myles RT, Fong B, Esses SI, et al (1999) Radiographic verifi cation of pedicle screw pilot hole placement using Kirschner wires versus beaded wires. Spine ; 24(5):476–480. Panjabi MM, O‘Holleran JD, Crisco JJ, et al (1997) Complexity of the thoracic spine pedicle anatomy. Eur spine J ; 6(1):19–24. Vaccaro AR, Rizzolo SJ, Allardyce TJ, et al (1995) Placement of pedicle screws in thoracic spine. Part I: Morphometric analysis of the thoracic vertebrae. J Bone Joint Surg Am ; 77(8):1193–1199. Vaccaro AR, Rizzolo SJ, Allardyce TJ, et al (1995) Placement of pedicle screws in thoracic spine. Part II: An anatomical and radiographic assessment. J Bone Joint Surg Am ; 77(8):1193–1199. Xu R, Ebraheim NA, Ou Y, et al (1998) Anatomic considerations of pedicle screw placement in the spine. Roy-Camille technique versus open-lamina technique. Spine ; 23(9):1065–1068.

191

192

6 6.5

SURGICAL ANATOMY OF THE SPINE THORACOLUMBAR JUNCTION

1

Anatomy of the diaphragm ……………………………………………………………………………… 193

2

Surgical anatomy of the thoracoabdominal approach (thoracophrenolumbotomy) …………… 197

3 3.1 3.2 3.3

Surgical anatomy of the less invasive thoracoabdominal approaches …………………………… Button-hole diaphragmatic split ………………………………………………………………………… Sagittal diaphragmatic split ……………………………………………………………………………… Partial thoracoabdominal approach (partial thoracophrenotomy) ………………………………

4

Suggested reading ………………………………………………………………………………………… 203

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

200 200 201 201

193

Serge Nazarian, Cyril Solari

6

SURGICAL ANATOMY OF THE SPINE

6.5

1

THORACOLUMBAR JUNCTION

ANATOMY OF THE DIAPHRAGM

The diaphragm is a large, flat musculotendinous transverse septum between the thoracic and abdominal cavities. Its peripheral insertions attach to the fi rst lumbar vertebrae, the inner surface of the four lower ribs and six lower costal cartilages, and the posterior surface of the xiphoid process. The costal insertions are continous with the fibers of the transversus abdominis muscle. The diaphragm displays two lateral cupulae and a central depression ( Fig 6.5-1). The right cupula lies on the liver, the left one on the gastric tuberosity and the spleen; the median depression contains the heart on its upper surface. Between the plane passing by its parietal insertions and another one passing by the top of the cupulae, a “thoracoabdominal region” is individualized ( Fig 6.5-1). It contains in its peripheral area the lower part of the costodiaphragmatic recesses, the pleural cavities and the lungs, and at its center, the upper part of the intraabdominal viscerae projecting into the thoracic area.

The diaphragm is made of two parts, the cloverleaf shaped central tendon, and the peripheral part made of fleshy fibers radiating from the central tendon to the parietal insertions ( Fig 6.5-2 ). The central tendon is a flat tendon that looks like a three-leaf clover, with an anterior leaf lying at a short distance from the sternum and two posterior lateral leaves. This tendon is made of two flat superposed fibrous bundles, the so-called semilunar bands. The narrow middle part of these bands surrounds the caval opening, a large smooth angled quadrilateral orifice, projecting at the level of the 9th thoracic vertebra. The fleshy fibers of the diaphragm insert on the thoracoabdominal wall, directly or by means of a fibrous arch at those places where the wall is made by vertical and moving elements (muscles, vessels, or nerves) running from the thorax to the abdomen or vice versa.

194

6

Surgical anatomy of the spine

10

1 2

9

3 4 8

1 2

10

5 6

7

3 4

9 8 7

5

cranial right

6 Fig 6.5-1 The area between the two transverse lines is the thoracoabdominal region. 1 heart 2 left lung 3 left diaphragmatic cupula 4 left colic angle 5 stomach 6 pancreas 7 gallbladder 8 liver 9 right diaphragmatic cupula 10 right lung

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 6.5-2 Superior view of the diaphragm. 1 pericardium 2 right diaphragmatic pleura 3 caval opening of the diaphragm 4 right phrenic nerve 5 aortic hiatus 6 internal aspect of the thoracic wall covered with its parietal pleura 7 diaphragm peripheral attachment 8 esophageal hiatus 9 left diaphragmatic pleura 10 left phrenic nerve

anterior right

6.5

195

Thoracolumbar junction

The posterior part of the diaphragm inserts on the spine, on the median arcuate ligament, on the medial arcuate ligament, and on the lateral arcuate ligament ( Fig 6.5-3 ). A direct spinal insertion is achieved by two tendons, the right and left crura. The right crus is the longest one and it inserts on the first three or four lumbar vertebrae, whereas the left crus inserts on the fi rst two or three lumbar vertebrae. They are made of a thick fibrous bundle covering the anterior lateral aspect of the vertebral bodies and discs to which they adhere very closely. The right and left crura run anteriorly and medially and eventually meet on the midline and form the median arcuate ligament which surrounds the aortic hiatus. The median arcuate ligament and the right crus give rise to two flat fleshy bundles which surround the esophageal hiatus at the level of the 10th thoracic vertebra. Muscular fibers rising from the left and right crura join the lateral leaves of the central tendon. An auxiliary crus may be found rising laterally from the higher lumbar vertebrae. The medial arcuate ligament (or psoas arch) extends from the lateral part of the vertebral body of L1 (or L2) and passes in front of the psoas muscle to the tip of the costal process of L1 (or L2). The lateral arcuate ligament (arch of the quadratus lumborum muscle) extends from the tip of the costal process of L1 to the tip of the 12th rib, or the 11th rib when the 12th is short or lacking. The fleshy fibers from the lateral arcuate ligament also run to the lateral leaves of the central tendon. More laterally, the muscular fi bers insert on the tip of the 11th rib and on the fibrous band between the tip of the 11th and the 12th costal cartilage. The costal part of the diaphragm inserts on the inner aspect of the four lower ribs and six lower costal cartilages. Its fibers are in the same direction as those of the transverse muscle and seem to be in continuity with them. The sternal insertions of the diaphragm attach on the posterior surface of the xiphoid process.

1 10

2 3 4

9 8 7

6

Fig 6.5-3 Posterior attachments of the diaphragm. 1 central tendon 2 esophageal hiatus, abdominal esophagus, and vagus nerves 3 inferior diaphragmatic arteries 4 aortic hiatus, aorta, and thoracic duct, limited anteriorly by the median arcuate ligament 5 left crus 6 sympathetic trunk 7 lateral arcuate ligament 8 medial arcuate ligament 9 right crus 10 caval opening, vena cava, and right phrenic nerve

5 6

196

6

Surgical anatomy of the spine

Several vessels, nerves, and viscera cross the diaphragm between its musculotendinous bundles. The aorta and the thoracic duct cross the diaphragm through the aortic hiatus. The inferior vena cava and the right phrenic nerve pass through the caval opening. The greater splanchnic nerve and the medial component of the azygos vein on the right side and the hemiazygos vein on the left cross the diaphragm through a hiatus between the median arcuate ligament and auxiliary crus. The sympathetic trunk, the splanchnic and the inferior splanchnic nerves cross through a hiatus between the auxiliary crus and the medial arcuate ligament. The psoas muscle and the ascending lumbar vein cross the medial arcuate ligament. The arterial supply is provided by the superior and inferior diaphragmatic arteries. The superior diaphragmatic arteries arise from the internal thoracic arteries and join the diaphragm along with the branches of the phrenic nerves. The inferior right and left diaphragmatic arteries ( Fig 6.5-3 ) arise as the fi rst branches of the abdominal aorta; running upward on each side of the esophagus, they supply anterior and posterior branches to the lateral parts of the central tendon and fleshy fibers detaching from it. The inferior diaphragmatic veins run with the corresponding arteries and fi nally end in the inferior vena cava, the gastric veins, or exceptionally the azygos vein. The nerve supply is provided by the right and left phrenic nerves which arise from the anterior branch of the 4th and 5th cervical nerves, less commonly from the 3rd. The right phrenic nerve (see Fig 6.5-2 ) joins the diaphragm on the right side of the inferior vena cava, dividing into three branches. The anterior branch supplies the anterior sternal

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

and costal fleshy fibers and gives a retroxiphoid anastomosis to the left phrenic nerve. The lateral rami of the nerve supply the lateral costal part of the diaphragm. The posterior branch goes through the caval opening and runs along the inferior diaphragmatic artery. The posterior terminal branches of the right phrenic nerve join the diaphragmatic plexus and provide some branches to the right semilunar ganglion of the solar plexus. The left phrenic nerve (see Fig 6.5-2 ) reaches the diaphragm slightly posterior to the apex of the heart and divides into three branches which pass through the central tendon. An anterior branch anastomoses with the right phrenic nerve, a lateral branch provides some rami to the left costal fibers, and a posterior branch innervates the diaphragmatic plexus and supplies other rami to the solar plexus. Consequently, the diaphragm should be incised in its peripheral zone along its costal insertions because the branches of the phrenic nerve do not cross this part. A radiate incision of the diaphragm is not allowed because it may cause injury to one or several branches of the phrenic nerves, depriving the muscle of its motor supply. Both sides of the diaphragm are covered by serous membranes (see Fig 6.5-2 , Fig 6.5-4 ). The superior aspect is covered with the pericardium on the anterior leaf and the parietal pleura on the cupulae. These membranes, especially the pleura, fi rmly adhere to the diaphragm and contribute to the solidity of the suture after peripheral transection of the diaphragm. The inferior aspect is covered with the parietal peritoneum and its adhesion to the diaphragm is fi rm only on the central tendon, so the detachment of the peritoneal serosa is easy from the fleshy, peripheral parts of the diaphragm.

6.5

197

Thoracolumbar junction

2

1 9

2

8

3 4 5

7 6 anterior

1 left

Fig 6.5-4 Superior view of the diaphragm after removal of thoracic viscerae. 1 internal aspect of the thoracic wall covered with its parietal pleura 2 right diaphragmatic pleura 3 caval opening of diaphragm 4 right phrenic nerve 5 aorta 6 left diaphragmatic pleura 7 esophageal hiatus 8 left phrenic nerve 9 pericardium

SURGICAL ANATOMY OF THE THORACOABDOMINAL APPROACH (THORACOPHRENOLUMBOTOMY)

This incision is used for procedures that are extended to the several levels crossing the thoracolumbar junction. It includes a simultaneous thoracic and retroperitoneal approach to the spine, anatomically separated by the diaphragm. This transverse muscular partition between the thoracic and abdominal cavities hides the thoracolumbar junction of the spine. The full and continuous access to the thoracolumbar junctional vertebrae requires a simultaneous thoracotomy, lumbotomy, and hemicircumferential division of the diaphragm, from the anterolateral thoracoabdominal parietal incision to the aortic hiatus of the diaphragm. The parietal incision is centered using those levels which are to be reached. It usually lies along the 8th intercostal space. The abdominal incision is continous with the thoracic one, down to the lateral border of the rectus abdominis muscle. After the incision of the subcutaneous fat tissue, the external oblique muscle is divided along the direction of its fibers, giving access to the 9th rib. The thoracic and lumbar parts of the incision are made fi rst separately and then secondarily joined by the section of the 9th costal cartilage and the circumferential division of the diaphragm and the overlying pleura. The thoracotomy is made through the 8th intercostal space by detaching the lower attachments of the intercostal muscles from the upper border of the 9th rib, to the costal cartilage anteriorly. This detachment does not give direct access to the thoracic cavity because it usually leaves the parietal pleura intact. This pleural sheet should then be divided along the upper ridge of the rib, using scissors. Then the lateral aspect of the left diaphragmatic cupula usually appears.

198

6

Surgical anatomy of the spine

The lumbotomy is carried out by fi rst dividing the external oblique muscle in the direction of its fibers. This maneuver exposes the underlying internal oblique muscle. Its fibers are perpendicular to the direction of the incision; the fibers are cut transversely, in the same direction as the division of the external oblique muscle, thus exposing the transverse muscle. This muscular layer is easily divided along the direction of its fibers to reach the transversalis fascia, the preperitoneal fat, and the anterior parietal peritoneum. Using blunt dissection with clamp-held nuts, these preperitoneal soft tissues are detached from the posterior aspect of the transverse muscle and, progressing backward and upward, from the inferior aspect of the diaphragm, the fibers of which are continous with those of the transverse muscle. The crucial step of the approach is the division of the anterior part of the 9th costal cartilage, which provides access to the muscular septum between the thorax and the abdomen. Using curve dissecting forceps, the cartilaginous arch is isolated at its anterior end and cut using a scalpel. The adjacent intercostal vessels should be quickly coagulated or ligated. The last fibers of the transverse muscle and fi rst fibers of the diaphragm are divided using scissors or an electric knife. From this point the diaphragm is progressively divided from the front to the back, using an electric knife, after having detached the peritoneum from its inferior aspect by blunt dissection. This division is made at a 10–15 mm distance from the costal insertion. It simultaneously cuts the muscular fibers and the overlying diaphragmatic pleura, which is the only strong tissue allowing a reliable suture at the end of the procedure. The division of the diaphragm is made along the thoracic wall from the front to the back. Simultaneous hemostasis of the muscular edges using bipolar electrocoagulation is required. Stay sutures are placed on either side of the line of division to facilitate the diaphragmatic repair at the end of the procedure.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

A “chest” retractor is then introduced and progressively opened. Near the spine, once the retroperitoneal soft tissues are detached from the psoas muscle and the left crus of the diaphragm, the aortic hiatus can be individualized from the fat tissue and the aorta by blunt dissection. The last step of the approach is the division of the left crus of the diaphragm, in medial continuity with the circumferential division of the diaphragm. To perform this step correctly, it is necessary to start on the thoracic side and divide the parietal pleura vertically along the vertebral bodies of T11 and T12. The diaphragmatic pleura is divided on the left side of the spine, in the continuity of the circumferential line of detachment of the diaphragm ( Fig 6.5-5 ). An angled dissector is then introduced into the aortic hiatus, close to the medial edge of the left crus and posterior to the left greater splanchnic nerve, which was previously identified along the lower thoracic spine by blunt dissection. The left crus is then divided after identifying and marking its distal edge with a stay suture. Once the left crus has been detached, the anterior and lateral aspect of the spine is exposed by incompletely detaching the tendon of the left crus, pushing and retracting the prespinal vessels slightly to the right side, coagulating or ligating the segmental vessels, and implanting the retractors. If necessary, the spine can then be exposed from T9 to L4 ( Fig 6.5-6 ). The closure starts with the suture of the crus. Then the repair of the whole circumference can be performed from posterior to anterior.

6.5

199

Thoracolumbar junction

1

4 2

anterior

3

cranial

Fig 6.5-6 Thoracoabdominal approach—view from posterior. The diaphragm is detached along its periphery in a circular approach (dotted line). 1 costal cartilage 2 diaphragm 3 parietal pleura 4 psoas muscle anterior left

Fig 6.5-5 Surgical anatomy of the thoracoabdominal approach. The diaphragm is detached 10–15 mm from its costal insertion (superior view).

200

6

Surgical anatomy of the spine

3

SURGICAL ANATOMY OF THE LESS INVASIVE THORACOABDOMINAL APPROACHES

3.1

BUT TON-HOLE DIAPHRAGMATIC SPLIT

The presence of the liver and inferior vena cava on the right side makes the left sided approach preferable ( Fig 6.5-7 ). The patient is prepared in a right lateral decubitus position. The parietal projection of the vertebra to be treated is identified using an image intensifier. The thoracic approach is performed from the left thoracic wall with an oblique incision 6–10 cm long, following an intercostal space. After opening the intercostal space one of the ribs can be divided in order to facilitate the spreading of the space. A segmental resection of a rib can be performed in some cases allowing a larger exposure. The aorta and the spine should fi rst be identified behind the parietal pleura. The parietal pleura is then divided vertically along the lateral aspect of the last thoracic vertebral bodies. The

incision should be performed slightly and superficially using the tip of scissors in order to avoid injury to the segmental vessels running laterally to the vertebral bodies. The diaphragm is then pushed down using a fan retractor and the pleural incision is continued on the superior aspect of the diaphragm along 2–3 cm. A 2–3 cm button-hole split is made in the diaphragm in line with the pleural incision. Blunt dissection using clampheld nuts is useful in exposing the lateral aspect of the 11th and 12th vertebrae and then, cutting the adjacent left edge of the aortic hiatus, in exposing the lateral aspect of L1 and L2 if necessary. The retraction of soft tissues around the operating field is carried out using sharp Hohmann retractors inserted in the distal vertebral bodies and connected to a ring retractor or Steinmann pins ( Fig 6.5-8 ).

3 1 anterior cranial

Fig 6.5-7 Surgical anatomy of the button-hole diaphragmatic split.

cranial anterior

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

2

Fig 6.5-8 Thoracoabdominal approach with a button-hole split in the diaphragm. A Hohmann retractor is inserted in the vertebral body of L2 to allow direct view of the spine from T10 to L1. 1 diaphragm 2 parietal pleura 3 L1 vertebra

201

6.5

Thoracolumbar junction

3.2

SAGIT TAL DIAPHRAGMATIC SPLIT

This is a limited anterior approach ( Fig 6.5-9 ) to the thoracolumbar vertebrae, leaving the left crus intact and allowing a lower extension to L3—the split in the diaphragm is sagittal. The fi rst steps are identical to those previously described. After a limited lateral transthoracic approach, the left cupula of the diaphragm is retracted caudally and the lung is retracted cranially and anteriorly. The parietal pleura is divided vertically along the left aspect of the lowest thoracic vertebrae. Its edges are separated by blunt dissection allowing the exposure of the segmental vessels running transversely along the lateral groove of the vertebral bodies, and the splanchnic nerves. The splanchnic nerves are preserved if possible, the segmental vessels are tied or coagulated according to their size. The

medial edge of the left crus is identified but not divided. A sagittal split is made, using scissors, through the left cupula, 5 cm lateral to the aortic hiatus. The medial edge of the split is held by dissecting forceps and the underlying fat tissue is detached from its inferior aspect by blunt dissection in the direction of the aortic hiatus and the T12–L1 vertebral bodies. The dissection is then extended caudally down to L2 and L3 if necessary. The exposure is improved by the identification and division of the segmental vessels, the detachment and retraction of the left crus from the spine, and the retraction of surrounding soft tissues using Hohmann retractors connected to a ring retractor or Steinmann pins anchored in the vertebral bodies adjacent to the injury. This approach allows instrumentation from the thoracic to the abdominal spine and preserves the continuity of the left crus, preventing transdiaphragmatic herniation. In this technique the closure of the diaphragm is simplified because the diaphragm has only partially been cut.

3.3

PARTIAL THORACOABDOMINAL APPROACH (PARTIAL THORACOPHRENOTOMY)

If a more comfortable exposure is required, use the posterior limited thoracophrenotomy ( Fig 6.5-10 ). A suitable less invasive approach is the posterior limited phrenotomy.

cranial anterior

Fig 6.5-9 Surgical anatomy of the sagittal diaphragmatic split.

The design of the diaphragmatic incision is the same as for the thoracoabdominal approach except for the fact that it is limited to the posterior half of the thoracic attachment of the diaphragm. The procedure requires only a small thoracic incision. The lumbar component of the incision and approach is not required.

202

6

Surgical anatomy of the spine

Once the approach is carried out and the diaphragm is exposed, a circumferential incision is started, at the junction of the lateral and posterior insertions of the diaphragm, 1.5 cm from the thoracic wall, in the direction of the spine. This incision is gently made with electrocautery and hemostasis is achieved step-by-step. Using blunt dissection, the underlying retroperitoneal fat tissue is detached from the underlying aspect of the diaphragm and the posterior abdominal wall. Approaching the aortic hiatus, the parietal pleura covering the spine is divided vertically along the vertebral bodies and its edges detached by blunt dissection. The segmental vessels are identified and ligatured or coagulated. The splanchic nerves are usually retracted. The medial edge of the left crus is identified using angled dissecting forceps introduced in a

cranial

a

anterior

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

craniocaudal direction through the aortic hiatus. Once the pathway is opened, the forceps are introduced from the medial end of the diaphragmatic incision to the medial edge of the aortic hiatus. The left crus is then divided. The infradiaphragmatic vertebral bodies are cleared from the retroperitoneal fat tissue which is detached by blunt dissection from the tendon of the left crus and the anterior insertion of the psoas muscle. The left crus and the anterior insertions of the psoas muscle are in turn detached and retracted backward. The segmental vessels are coagulated or tied. Hohmann retractors or Steinmann pins implanted into the adjacent vertebral bodies allow an excellent exposure of the junctional vertebrae.

Fig 6.5-10a–b Surgical anatomy of the posterior limited thoracophrenotomy.

6.5

Thoracolumbar junction

4

SUGGESTED READING

Costa MM, Pires-Neto MA (2004) Anatomical investigation of the esophageal and aortic hiatuses: physiologic, clinical and surgical considerations . Anat Sci Int ; 79(1)21–31. Kawahara N, Tomita K, Baba H, et al (1996) Cadaveric vascular anatomy for total en bloc spondylectomy in malignant vertebral tumors. Spine ; 21(12):1401–1407. Khoo LT, Rhim SC, Fessler RG (1999) Complications during anterior surgery of the lumbar spine: an anatomically based study and review. Neurosurg Focus ; 7(6):e9.

203

4

A 1 C i h c K C s

S S U S O S F s b

204

6 6.6

SURGICAL ANATOMY OF THE SPINE LUMBAR SPINE AND LUMBOSACRAL JUNCTION

1 1.1 1.2 1.3 1.4

Anatomy Anatomy Anatomy Anatomy Anatomy

2 2.1 2.1.1 2.1.2 2.1.3 2.1.4 2.2 2.3 2.4 2.5 2.6 2.7

2.8.1 2.8.2

Anatomy related to anterior procedures ……………………………………………………………… Anterolateral wall of the abdomen …………………………………………………………………… Arterial supply ……………………………………………………………………………………………… Venous drainage …………………………………………………………………………………………… Nervous supply …………………………………………………………………………………………… Three types of incision …………………………………………………………………………………… Iliopsoas muscle ………………………………………………………………………………………… Lumbar plexus and its branches ………………………………………………………………………… Prevertebral vessels ……………………………………………………………………………………… Sympathetic trunk ………………………………………………………………………………………… Superior hypogastric plexus ……………………………………………………………………………… Anatomical variations of the prevertebral vessels and their management during the anterior approach to the spine …………………………………………………………… Management of the peritoneal sac in the anterior approach to the lumbar and lumbosacral spine …………………………………………………………………………………… Anatomical basis for transperitoneal approach and exposure of the lumbosacral spine ……… Anatomical basis for retroperitoneal approach and exposure of the lumbosacral spine ………

3

Suggested reading ………………………………………………………………………………………… 233

2.8

related related related related related

to to to to to

posterior procedures……………………………………………………………… 205 laminectomy ……………………………………………………………………… 207 supralaminar hooks ……………………………………………………………… 207 infralaminar hooks ……………………………………………………………… 210 the pedicle screw insertion …………………………………………………………211

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

216 216 219 220 220 220 221 221 222 224 226 229 230 230 232

205

Serge Nazarian, Cyril Solari

6

SURGICAL ANATOMY OF THE SPINE

6.6

1

LUMBAR SPINE AND LUMBOSACRAL JUNCTION

ANATOMY RELATED TO POSTERIOR PROCEDURES

External morphology

The midline depression corresponds to the deep insertion of the lumbar and sacral fascia along the spinous processes; this depression continues caudally as the intergluteal fold. The erector spinae muscles lie on each side of this depression. The posterior superior iliac spines are easily palpable landmarks. The transverse line joining these posterior superior spines lies at the level of the L5/S1 interspinous space and the L5/S1 zygapophyseal joints. Lumbar fascia

The thoracolumbar fascia is a thick and strong fibrous sheet which is attached to the tip of the lumbar spinous processes and sacral crest on the midline and to the iliac crest laterally.

Correct reattachment of this fascia to the spinous processes reestablishes the efficacy of the erector spinae muscles. Lumbar muscles

On each side of the spinous processes lie the erector spinae muscles. It is not possible to expose the posterior aspect of the vertebrae without detaching these muscles from the vertebrae. This has a deleterious effect on the efficacy of these muscles. If the spinous processes are left intact, a temporary postoperative muscular weakness is normal until the solid biological reattachment of these muscles occurs. If the spinous processes have been removed, the lever arm effect of the erector spinae muscles is significantly reduced. This leads to excessive effort and possible lumbar pain, and in some cases lumbar kyphosis and sagittal imbalance.

206

6

Surgical anatomy of the spine

From medial to lateral, six muscle groups are found: • The interspinales muscles (interspinalis thoracis and interspinalis lumborum) lie between the spinous processes. • The two spinales muscles (spinalis thoracis and spinalis lumborum) originate at L2 and extend cranially to the flanks of the spinous processes. • The transversospinalis muscles are formed of numerous bands originating from the accessory processes; these bands usually insert on the three overlying spinous processes and laminae. • The longissimus lumborum and thoracis muscles originate from the common mass of the spinales muscles and extend up to the transverse processes and the posterior angles of the ribs. • The iliocostalis lumborum and thoracis muscles originate from the common mass of the spinalis muscles and extend upward to the posterior angles of the ribs. • The intertransversarii and quadratus lumborum muscles extend coronally and form the boundary between the posterior and anterior lumbar region.

internal fi xation devices should not extend too far laterally, beyond the lateral edge of the articular processes. These excessive exposures are the main cause of postoperative deterioration of erector spinae muscles and the ensuing sagittal imbalance of the spine. Unfortunately, it is not always possible to adequately preserve the blood and nerve supply to these muscles. Their permanent retraction should also be released for 10–15 minutes each hour, in order to avoid long-lasting ischemia and nerve stretching.

3

1

2

Vasculonervous bundles ( Fig 6.6-1)

The lateral edge of the middle part of the zygapophyseal column, between the superior and inferior articular processes, shows a slight depression at the level of the so-called isthmus or pars interarticularis. This depression on each side gives passage to the posterior neurovascular bundle including for the most part an artery, a vein, and a nerve to the erector spinae muscles. This zone should be respected during dissection of the muscles by avoiding an extension of the midline exposure beyond the lateral margin of the isthmus; it is also advisable to avoid extensive multilevel coagulation. Bone grafting and

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

cranial left

Fig 6.6-1 A left segmental vasculonervous bundle after detachment and retraction of the erector muscles (posterior view). 1 lateral border of the left isthmus (pars interarticularis) 2 left inferior zygopophyseal joint 3 segmental vasculonervous bundle (arrows)

6.6

Lumbar spine and lumbosacral junction

1.1

ANATOMY RELATED TO LAMINECTOMY

A laminectomy ( Fig 6.6-2 ) is the resection of the spinous process(es) and medial part of the lamina(e), which aims to open the spinal canal. It can be a part of a major vertebral resection or the fi rst step of a decompression procedure in spinal stenosis.

207

One important issue in this procedure is the preservation of the continuity of the posterior zygapophyseal columns. So it is necessary, when performing a laminectomy, to preserve at least 5 mm of bone on each side from the lateral border of the pars interarticularis (isthmus).

1.2

5–6 mm

1 2

ANATOMY RELATED TO SUPRALAMINAR HOOKS

Supralaminar hooks are used in the lumbar spine in order to apply distraction/compression or to make a claw. The application of a supralaminar hook requires an adequate preparation of the interlaminar space. The upper edge of the lamina is not very broad ( Fig 6.6-3 ) and the space available is usually reduced laterally by the inferior articular process of the overlying vertebra. The inferior border of the overlying lamina hides more or less the superior edge of the vertebra which is to be instrumented. This edge is additionally covered by the inferior insertion of the yellow ligament which must be detached. Overlying lamina

The inferior edge of the overlying lamina (see Fig 6.6-6 ) hides more or less completely the underlying lamina and the yellow ligament (ligamentum flavum), especially on the concave side of a scoliosis where the hypertrophy of the articular process laterally decreases the size of the interlaminar space. Fig 6.6-2 Posterior view of the lumbar laminectomy. A 5–6 mm distance must be respected between the lateral limit of the laminectomy and the lateral edge of the isthmus to avoid any fracture of the zygapophyseal column. 1 lateral edge of the isthmus (pars interarticularis) 2 lateral limit of the laminectomy

In such cases, the inferior edge of the overlying lamina and the medial part of its articular process should fi rst be “thinned out” using a high-speed burr or a thin chisel.

208

6

Surgical anatomy of the spine

Yellow ligament (ligamentum flavum)

The yellow ligament ( Fig 6.6-4 ) is the interlaminar ligament. It inserts from the inferior half of the anterior aspect of the overlying lamina above to the upper border of the underlying lamina below. It extends laterally to the anterior aspect of the zygapophyseal joint. Once the yellow ligament is exposed, it is detached from the upper edge of the lamina, which is to be instrumented, using a thin curette. The complete removal requires its detachment from the anterior aspect of the lower half of the overlying lamina and from the anterior aspect of the medial part of the zygapophyseal joint.

a

1 2

3

b Fig 6.6-3a–b Posterior and posterolateral views of the lamina of a lumbar vertebra. The laminar area is narrow. It is limited by the spinous process medially and the zygapophyseal column laterally. 1 lamina 2 zygapophyseal column 3 spinous process

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

1

1

1

Fig 6.6-4 Parasagittal cut. The yellow ligament attaches to the superior aspect of the inferior lamina, and on the inferior half of the superior lamina. 1 yellow ligament (ligamentum flavum)

6.6

209

Lumbar spine and lumbosacral junction

Upper edge of the lamina

After the complete removal of the yellow ligament, the upper edge of the lamina ( Fig 6.6-5 ) should be prepared. The natural form of this edge is usually curved and concave; additionally, ligamentous calcifications make its contour irregular. The upper edge of the lamina should be reshaped in order to fit the hook. Additional osteophytes should fi rst be removed. The medial end of the articular process should be resected, perpendicular to the upper edge of the lamina, in order to offer enough room for the placement of the hook. This preparation is made using a Kerrison rongeur in order to create a stable horizontal base for the hook. Care should be taken to remove a minimal quantity of cortical bone.

a

b Fig 6.6-5a–b The upper edge of the lamina. a Before preparation. b After preparation in order to insert a supralaminar hook (arrows).

210

6

1.3

Surgical anatomy of the spine

ANATOMY RELATED TO INFRALAMINAR HOOKS

The infralaminar hooks can be used for compression/distraction or to create claws. The lower edge of the lamina is narrow, thick, and oblique ( Fig 6.6-6 ). The upper end of the yellow ligament inserts on its anterior aspect and with it forms a short and narrow transverse groove. A notch at the inferior aspect of the lamina should be cut out to receive the blade of the lamina hook. Its anterior blade should be introduced in front of the inferior half of the lamina after having detached the yellow ligament using a rasp or a curette. The yellow ligament can be very thick at this level and the spinal canal can be narrow. The complete removal of the yellow ligament is then recommended. The inferior border of the lamina is not usually as straight transversely as the hook is. If such is the case, it should be reshaped using a thin rongeur in order to perfectly adapt to the hook. In the case of a large inferior articular process narrowing the transverse dimension of the inferior edge of the lamina, the medial edge of the articular process can be gently removed in order to enlarge the base for the implantation of the hook.

a

b Fig 6.6-6a–b The inferior edge of the lamina is very narrow and oblique. a A lateral posterior inferior view of the lower edge of the laminae. b Posterolateral view of the left lamina. The red area shows the notch to be created for insertion of the lamina hook.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

6.6

Lumbar spine and lumbosacral junction

1.4

ANATOMY RELATED TO PEDICLE SCREW INSERTION

211

Pedicle screws are now common implants used to link the vertebrae to supporting devices such as plates or rods. They were previously designed for a purely pedicular implantation, associated with plates. The current fi xation procedures, more often associated with reduction maneuvers, require a stronger implantation of the screws, deep through the pedicle into the vertebral body. So the recommended direction for the screws, from their entry point, is slightly oblique toward the midline ( Fig 6.6-7 ). Screw implantation can be significantly facilitated by the use of a navigation system (chapter 8 Computer-assisted surgery). But this device is not available in every center and knowledge of the vertebral anatomy is essential even for those who are used to working with computerized navigation.

1

2

a

The procedure described below requires AP and lateral fluoroscopic intraoperative control. The parameters for screw implantation are: • Diameter of the pedicle • Point of entry • Direction in the transverse plane • Direction in the sagittal plane • Length of the screws • Density of the bone

b

1

2

Fig 6.6-7a–b Pedicle screws. Purely pedicular, with a sagittal orientation (1). Extented into the vertebral body, with an anteromedial oblique orientation (2).

212

6

Surgical anatomy of the spine

Diameter of the pedicle

The diameter of the pedicle is the main decision-making parameter when determining whether or not the use of screws is allowed. The diameter can be assessed on a CT scan, or even on plain x-rays by a physician used to working with plain x-rays ( Fig 6.6-8 ). The smallest screw diameter is 4.2 mm for thoracolumbar implants. In small children 3.5 mm cervical screws can be used with caution. As a rule of thumb the diameter of the pedicle should be more than 5.0 mm at its narrowest point. Point of entry

The entry point should be chosen so that the screw is able, from this point, to run obliquely through the pedicle and fi nish

its path in the anterior part of the vertebral body, without violating the cortex of the bone. In lumbar vertebrae the entry point is located at the superior articular process, immediately lateral to the posterior end of the articular surface, at the level of the mammillary process or 2–3 mm lateral, between the mammillary and accessory processes. Due to individual anatomical variations, this point is located more or less cranially on the superior articular and mammillary processes ( Fig 6.6-9 , Fig 6.6-10 , Fig 6.6-11). For screw insertion into S1 the point of entry is at the caudal and lateral angle of the articular process ( Fig 6.6-12 ).

a

a

b

Fig 6.6-8a–b The transverse diameter of the pedicles. a AP x-ray of a pedicle which can be instrumented with a screw. b AP x-ray of a pedicle which is too narrow to be instrumented with a screw.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

Fig 6.6-9a–b Entry point in L1 and L2. a Posterior view. b AP x-ray.

6.6

213

Lumbar spine and lumbosacral junction

a

a

a

b

b

b

c

c

Fig 6.6-10a–b Entry point in L3. a Posterolateral view. b AP x-ray.

Fig 6.6-11a–c Entry point and orientation of the drilling in L4 and L5. a Posterior view. b Superior view. c AP x-ray.

Fig 6.6-12a–c Entry point and orientation of the screw in S1. a Posterior view. b Superior view. c AP x-ray.

214

6

Surgical anatomy of the spine

Care should be taken to have the pedicle screw entry points in harmonious alignment ( Fig 6.6-13 ), in order to allow the easy insertion of the rod into the screws. If alignment is not respected, two things may happen: Either the screw cannot be attached to the rod, or the bone-screw interface may loosen. Polyaxial screws have partially solved this problem.

The correct location of the entry point is checked under C-arm fluoroscopy. A small K-wire is inserted in the fi rst 2–3 mm of the entry point. On the lateral view, the tip of the wire should be at the middle height of the pedicle ( Fig 6.6-14 ). On the AP view the tip of the wire should project in the middle of the oval area of the pedicle, or between the middle and the lateral thirds ( Fig 6.6-15 ).

*

*

Fig 6.6-13 Alignment of the pedicular entry points in the coronal plane. Harmonious alignment on the right side. Disharmonious alignment on the left side (wrong entry points at L4 and L5 level, asterisks).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 6.6-14 On the lateral view, the inserted screw should lie at the middle height or at the upper half of the pedicle.

Fig 6.6-15 On the AP view, the tip of the K-wire should lie in the middle of the oval area of the pedicle or in its lateral half.

6.6

215

Lumbar spine and lumbosacral junction

Direction in the transverse plane

Direction in the sagittal plane

This direction follows the axis of the pedicle. It is slightly oblique toward the midline (10–15° on average). Its angular value is variable according to the individual and according to the level (it increases from L1 to L5). The obliquity toward the sagittal plane can be preoperatively evaluated on a CT scan.

Several values have been reported in the literature. The rule is to stay, as much as possible, parallel to the superior end plate. If the pedicle feeler is inserted perpendicular to the spine curvature, it is usually parallel to the superior end plate ( Fig 6.6-16 ).

The best way to avoid any error is to drill the path using a pedicle feeler and let the feeler follow its way through the cancellous bone of the pedicle.

For S1, the direction in the sagittal plane should not be parallel to the end plate, but it should be slightly ascending toward the superior S1 end plate in order to have a strong purchase in the dense subchondral bone ( Fig 6.6-17 ). Screw length

The length of the screw varies: 40–45 mm for L1, 45–50 mm for L5, 35–40 mm for S1, and 30–35 mm for S2. Bone density

It is important to consider bone density for the purchase of the screws. If necessary, in an osteoporotic vertebra, the pedicle screw can be augmented by injecting bone cement into the screw hole before inserting the screw.

Fig 6.6-16 The pedicle screw is inserted perpendicular to the spinal curvature, its direction is approximately parallel to the superior end plate.

Fig 6.6-17 Pedicular position of a screw in S1. The direction in the sagittal plane is ascending toward the superior S1 end plate.

216

6

Surgical anatomy of the spine

2

ANATOMY RELATED TO ANTERIOR PROCEDURES

The approach and the exposure of the anterior part of the lumbar vertebrae (ie, the vertebral bodies and intervertebral discs) require: • • • •

Division of the anterior or anterolateral abdominal wall. An intra- or retroperitoneal approach. Retraction of the surrounding vessels and nerves. Careful retraction of the surrounding vessels, nerves, and ureters.

2.1

ANTEROLATERAL WALL OF THE ABDOMEN

The anterolateral abdominal wall is the most superficial anatomical element lying in front of the lumbar spine. It consists on each side of musculoaponeurotic elements which comprise the muscles and their fascias. Medially, the rectus abdominis muscle—a multigastric longitudinal muscle—extends from the xyphoid process and 5th, 6th, and 7th costal cartilages to the pubic crest and pubic symphysis. Laterally, lie three large, flat, superimposed muscles, extending to the midline by three flat aponeurotic sheets forming the sheath of the rectus abdominis muscle on each side.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

• The external oblique muscle originates from the external surface of the eight lower ribs, and extends obliquely downward and medially; its posterior fibers insert on the anterior part of the iliac crest, its intermediate and anterior fibers give rise to an aponeurotic sheet which contributes to the inguinal ligament and the sheath of the rectus. • The internal oblique muscle originates from the thoracolumbar fascia, the iliac crest, and the iliac fascia; it extends cranially and medially; its posterior fibers end at the level of the three lowest ribs, its anterior fibers give rise to an aponeurotic sheet which splits into two layers to enclose the rectus abdominis muscle. Medially, the same aponeurosis contributes to the linea alba. The posterior layer disappears a few centimeters below the umbilicus, forming the arcuate line with the aponeurotic sheet coming from the transversus abdominis muscle. • The transversus abdominis muscle originates from the iliac fascia, the thoracolumbar fascia, and the medial aspect of the six lowest costal cartilages. The upper fibers run medially while the middle and inferior fibers run medially and in an increasingly oblique downward direction. The fleshy fibers give rise medially to an aponeurotic sheet which joins the posterior layer of the sheath of the rectus abdominis muscle ( Fig 6.6-18 ).

6.6

217

Lumbar spine and lumbosacral junction

1 2 3

10

1 2 3

9 8 7

4

4 5

a

5

4 5

b

2 3 4 5 6 9 2

c

6

1 2

Fig 6.6-18a–c a Transverse section of the anterior lateral wall above the arcuate line. b Transverse section of the anterior lateral wall below the arcuate line. c Parasagittal section through the rectus—the rectus abdominis sheath. 1 linea alba 2 anterior sheet (layer) of the rectus sheath 3 posterior sheet of the rectus sheath 4 transversalis fascia 5 anterior parietal peritoneum 6 arcuate line 7 transverse muscle 8 internal oblique muscle 9 external oblique muscle 10 rectus abdominis muscle

a Fig 6.6-19a–b a Anterior view of the abdominal wall (left side). The left rectus is retracted laterally, disclosing the posterior layer of the rectus abdominis sheath and its inferior limit, the arcuate line. 1 umbilicus 2 rectus abdominis muscle 3 posterior sheet of the rectus sheath (from internal oblique and transverse muscles) 4 arcuate line 5 linea alba 6 anterior sheet (layer) of the rectus sheath

218

6

Surgical anatomy of the spine

The upper and middle fibres run behind the rectus abdominis muscle and fuse with the posterior layer of the aponeurosis of the internal oblique muscle, giving rise to the posterior layer of the rectus sheath. This posterior layer extends from the thoracic insertion of this muscle to the so-called arcuate line ( Fig 6.6-19 ), its lower end, located some centimetres below the umbilicus. The lower fibres pass anteriorly to the rectus, making part of the anterior layer of the rectus sheath. Finally, the posterior aspect of the rectus abdominis muscle is covered with the posterior layer of the sheath from its thoracic insertion to the arcuate line. Below the arcuate line the posterior aspect

1 2 1 2

b Fig 6.6-19a–b b Anterior view of the arcuate line after lateral retraction of the rectus abdominis muscle in a cadaveric specimen (left) and in a patient operated retroperitoneally for disc replacement (right). 1 posterior sheet of the rectus sheath 2 arcuate line

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

of the rectus is just covered with the transversalis fascia. The peritoneum adheres closely to the fascia transversalis and the posterior layer of the rectus sheath. The main obstacle to the contralateral retraction of the peritoneal sac and its contents is the arcuate line. Its level varies according the individual. When it constitutes a serious obstacle in retracting the peritoneal sac, it must be isolated, as cranially as possible, from the peritoneum by blunt dissection, and then divided. The aponeurotic sheets join on the midline and form the linea alba. It is a narrow fibrous area corresponding to the embryological midline fusion line (median raphe) between the right and left anterior components of the abdominal wall. It includes the umbilicus, which lies in the middle of the distance between the xiphoid process and the pubic symphysis. The linea alba is the commonly used pathway of abdominal penetration in transperitoneal approaches. The surgical division must be perfectly repaired to avoid secondary disruption and visceral herniation. The segmental neurovascular bundles run obliquely between the internal oblique and the transverse muscles, providing these muscles with their arterial, venous, and nervous supply. They fi nally end, in the lateral border of the rectus ( Fig 6.6-20 ). The retroperitoneal approach should therefore pass medial to the rectus in order to spare the neurovascular bundles joining the lateral border of this muscle. This is in contradiction to the classic pararectal approach that sacrifices the innervation of the rectus, as the approach is lateral to the rectus. Our approach medial to the rectus does not cut the innervation of the rectus abdominis muscle.

6.6

219

Lumbar spine and lumbosacral junction

1 5 2 5

3

5

Fig 6.6-20a–b a The lateral border of the rectus abdominis muscle, its segmental vascular and nervous supply. 1 umbilicus 2 linea alba 3 rectus abdominis muscle 4 os pubis 5 vascular nervous bundle

4

a

3

2 1

b

b Anterior view of the innervation of the rectus abdominis muscle (right side). The rectus abdominis sheath has been divided and retracted laterally. 1 vascular nervous bundle 2 rectus abdominis muscle 3 deep aspect of the anterior layer of the rectus abdominis sheath

2.1.1 ARTERIAL SUPPLY

The arterial blood supply comes from inferior, superior, and lateral sources. The inferior sources are composed of the superficial and deep components. Superficial to the fascia of the external oblique muscle, the superficial epigastric artery runs in the direction of the umbilicus and the superficial circumflex iliac artery along the inguinal fold. In the deep part of the abdominal muscles, under the transversalis fascia, the inferior epigastric artery and the deep circumflex iliac artery originate from the external iliac artery. The inferior epigastric artery describes in men an arch below the spermatic cord before running upward along the posterior aspect of the rectus abdominis muscle; it penetrates the sheath of this muscle at the level of the arcuate line and then it progressively penetrates the rectus abdominis muscle, before anastomosing with the superior epigastric artery 5–6 cm above the umbilicus. This artery is perfectly identifiable during the anterior step of the extraperitoneal approach to the lumbosacral spine. The deep circumflex iliac artery gives rise to an ascending branch which runs between the internal oblique and the transversus muscle, before anastomosing with the lumbar arteries.

220

6

Surgical anatomy of the spine

Superior arterial source

The superior arterial blood source is the internal thoracic artery, which gives rise to the superior epigastric artery and the musculophrenic artery. The superior epigastric artery penetrates the sheath of the rectus, runs behind the muscle and then penetrates the muscle and anastomoses with the inferior epigastric artery.

2.1.3 NERVOUS SUPPLY

The lateral sources are comprised of three to four lumbar arteries. They run obliquely between the transversus and the internal oblique muscles, and give rise to branches diverging toward the lateral margin of the rectus sheath. The three arterial areas are anastomosed.

The nerve supply of the abdominal wall is carried out by the thoracoabdominal nerves arising from the 7th to the 12th intercostal nerve. These nerves run between the internal oblique and the transversus muscles before penetrating the rectus sheath and terminating into the medial half of the rectus. These nerves give rise to lateral and anterior cutaneous branches. The 8th nerve reaches the midline at a level between the xyphoid process and the umbilicus; the 10th nerve reaches the midline around the umbilicus; and the 12th reaches the midline above the pubic symphysis. The inferior part of the abdominal wall is innervated by the iliohypogastric and ilioinguinal nerves, arising from the L1 nerve.

2.1.2 VENOUS DRAINAGE

2.1.4 THREE T YPES OF INCISION

Each artery is accompanied by paired veins.

Three types of incision can be recommended, taking into account the topography of the arteries and veins of the abdominal wall:

Lateral sources

Superficial veins of the epigastric area

The superficial veins of the epigastric area run toward the umbilicus and anastomose with the veins of the portal system via the ligamentum teres hepatis. Veins of the lower abdominal wall

The veins of the lower abdominal wall are drained by the saphenous vein. Venous drainage of the lateral abdominal wall

The venous drainage of the lateral abdominal wall is carried out by the thoracoepigastric veins which anastomose with the iliofemoral and external thoracic veins.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

• Midline vertical incision. • Transverse, Pfannenstiel-like skin incision, providing access to a midline vertical incision of the underlying layers. • Lateral oblique incision, running downward and medially, following the direction of the parietal nerves.

6.6

Lumbar spine and lumbosacral junction

2.2

ILIOPSOAS MUSCLE

This muscle is the closest anterolateral relationship of the spine. The psoas muscle extends from the anterolateral aspect of T12 and the five lumbar vertebrae to the lesser trochanter. At the level of the internal iliac fossa, it is joined and reinforced by the iliac muscle. The psoas muscle itself inserts on the lateral aspect of the vertebral bodies and discs from T12 to L5, and on the costal processes of the five lumbar vertebrae. This double insertion causes the muscle to form a dihedral pattern in the transverse plane ( Fig 6.6-21). This dihedral angle looks posteromedially and faces the intervertebral foramens; it receives the anterior branches of the lumbar spinal nerves which mix their fibers and form the lumbar plexus. The function of this muscle is the flexion and external rotation of the hip. It is innervated by branches from the lumbar plexus. Infection of the spine that extends to the psoas may provoke a painful spasm of this muscle with flexion and external rotation of the hip, the so-called psoitis. The surgical exposure of the lateral aspect of the lumbar vertebrae, required for anterolateral instrumentation, is not easy due to the thickness of the muscle and its strong attachment to the discs and the adjacent vertebral bodies. The iliopsoas muscle is surrounded by an aponeurotic sheet, the iliac fascia which accompanies it from the lower thoracic spine (T12) to the upper part of the thigh, through the diaphragm (medial arcuate ligament) and the femoral ring. An abscess of the spine can migrate along this sheath and externalize at the base of the thigh.

221

2.3

LUMBAR PLEXUS AND ITS BRANCHES

The anterior branches of the lumbar spinous nerves, which contribute to the constitution of the lumbar plexus, lie in a lateral position, in the thickness of the psoas muscle between its posterior and anterior insertions (see Fig 6.6-21). Anterior branches of the L2, L3, and L4 spinal nerves contribute to the femoral nerve and obturator nerve. The posterior retraction of the muscle, required for the correct exposure of the vertebral bodies ( Fig 6.6-22 ), includes a risk of overretraction of the lumbar plexus and a subsequent palsy in the area of the branches of the plexus, especially the femoral nerve. In the lateral decubitus position, the hip on the side which is to be operated on should be flexed in order to reduce the excessive tension of the lumbar plexus during the posterior retraction of the psoas, which is required for the exposure of the lateral aspect of the vertebral bodies.

Fig 6.6-21 The iliopsoas muscle. Transverse section at the level of L3 showing the psoas muscle and its relationship with the lumbar plexus.

222

6

Surgical anatomy of the spine

The lowest branch of this plexus is the lumbosacral trunk. It lies along the posterolateral margin of the L5/S1 disc, crossing the brim of the sacral ala. The reduction of a spondylolisthesis can stretch this nerve on this brim ( Fig 6.6-23 ) and lead to a weakness of the extensors and abductors of the foot.

2.4

PREVERTEBRAL VESSELS

The prevertebral vessels include the aorta and its terminal bifurcation into the right and left common iliac arteries, the right and left common iliac veins joining themselves to form the inferior vena cava, the visceral branches of these prevertebral vessels, and the parietal branches with their segmental arteries and veins ( Fig 6.6-24 ).

Fig 6.6-22 The effect of retraction of the psoas on the lumbar plexus.

1 2

3 4 5

a Fig 6.6-23a–b The distractive effect of a spondylolisthesis reduction on the lumbosacral trunk.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

Fig 6.6-24 The prevertebral vessels. The classic type. 1 aorta 2 inferior vena cava 3 left common iliac artery 4 left common iliac vein 5 middle sacral artery and vein

6.6

223

Lumbar spine and lumbosacral junction

These vessels are topographically located in the retroperitoneal space, surrounded by a variable amount of fat tissue. The topographical arrangement of these vessels displays some variability according to the individual. The anterior aspect of the lumbosacral spine lies behind these vascular elements. Approaching the spine retroperitoneally from the lateral to the medial side, owing to the surrounding fat tissue, the aorta, the inferior vena cava and their visceral branches are easily retracted from the spine by blunt dissection. The only vascular elements remaining against the vertebrae are the segmental vessels. Segmental vessels

The segmental vessels run anteroposteriorly, between the lateral concavity of the vertebral body and the psoas muscle, and they are usually divided at the level of the spine which is to be instrumented. One of these segmental arteries, an intercostal or lumbar artery, gives rise to the artery supplying the lumbar enlargement of the spinal cord ( Fig 6.6-25 ). It is more often a left one, usually located at a level from T8 to L2, but it can also be found at a lower one. This artery (artery of Adamkiewicz), is the unique artery supplying this part of the cord. In spite of the anatomical existence of anastomoses between the segmental radiculomedullary arteries in the region of the intervertebral foramina, preoperative angiographic identification of the segmental artery which gives rise to the artery of Adamkiewicz should be helpful in order to avoid any ischemic complication to the spinal cord. In fact, if one takes into account the severity of the consequences of the ligature of the segmental artery giving rise to the artery of Adamkiewicz, the functionality of such anastomoses should not be taken as absolutely reliable.

End of the aorta and the origin of the inferior vena cava

The end of the aorta and the origin of the inferior vena cava display a reverse Y-shaped bifurcation. The aortic bifurcation lies anteriorly and to the left side, whereas the inferior vena cava lies behind the aorta and to the right side. The level of these bifurcations is variable. The end of the aorta lies in 45% of individuals at L4, 25% at L4/5, 20% at L3/4, and 5% at L3 and L5 respectively. The origin of the inferior vena cava lies in 45% at L5, in 30% at L4/5, in 20% at L4, and in 5% at L5/S1. The angle of the bifurcation ranges from 45° to 70° for the aorta and from 65° to 72° for the inferior vena cava. The neighboring collateral branches of the aorta and inferior vena cava are comprised of the middle sacral vessels and the 4th and 5th segmental vessels. The middle sacral artery arises from the posterior aspect of the aortic bifurcation; the middle sacral vein(s) join(s) either the right common iliac vein, or the venous confluent, or the left common iliac vein.

Fig 6.6-25 The artery of Adamkiewicz originating from a left segmental artery (most often between T8 and L2).

224

6

Surgical anatomy of the spine

The middle sacral vessels lie on the anterior aspect of the sacrum, near the midline in 30–35% of individuals, on the right of the anterior sacral aspect in 20–25%, and on the left in 40%. They are attached to the presacral longitudinal ligament by very thin fiber tracts. They are in close relationship to the superior hypogastric plexus from which they must be detached by blunt dissection before being tied. Each vessel must be identified and ligated separately to avoid any unlucky division of the plexus.

6

5

1

4 2

The 4th lumbar vessels display a metameric course toward the 4th lumbar vertebra. The 5th lumbar vessels may arise from the iliolumbar, the 4th lumbar, or the middle sacral arteries or from the aorta. The 5th lumbar vein can anastomose with any of the other veins.

2.5

SYMPATHETIC TRUNK

The right and left lumbar sympathetic trunks ( Fig 6.6-26 ) lie along the vertical dihedral angle formed by the anterolateral aspect of the vertebral bodies and intervertebral discs medially and the insertions of the psoas muscle on the discs and vertebral bodies laterally. This trunk is continued downward by the sacral trunk. Four sympathetic ganglia are usually found in the lumbar region. They are connected to each other by the trunk itself. The sympathetic ganglia receive white rami communicantes from the lumbar nerves and give rise to the grey rami communicantes. They also give rise to the lumbar splanchnic nerves which join the peripheral autonomic plexuses.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

3 anterior

a

cranial

Fig 6.6-26a–c a Right sympathetic trunk (arrow). (Right sided view.) 1 L3/4 disc 2 segmental lumbar vein 3 psoas muscle 4 right sympathetic trunk 5 lumbar splanchnic nerves 6 inferior vena cava

6.6

225

Lumbar spine and lumbosacral junction

7 1 6

1

6

2

5

3

2

5 4

4

3

anterior

b

cranial

Fig 6.6-26a–c b Right sympathetic trunk. The inferior vena cava is divided and bent downward. (Right sided view.) 1 inferior vena cava 2 right lumbar splanchnic nerves 3 right sympathetic trunk 4 right psoas muscle 5 L2/3 disc 6 right diaphragmatic crus 7 aorta

anterior

c

c Left 1 2 3 4 5 6

cranial

sympathetic trunk. (Left sided view.) left lumbar splanchnic nerves left sympathetic trunk left psoas muscle L3/4 disc segmental lumbar vein aorta

226

6

Surgical anatomy of the spine

2.6

SUPERIOR HYPOGASTRIC PLEXUS

The superior hypogastric plexus ( Fig 6.6-27 ) is a peripheral autonomic plexus. It is an area of exchange of nervous fibers, presenting as a strip of connective and nervous tissue, extending from the inferior mesenteric artery to a level varying from the body of L5 to S2 (L5: 5%, L5/S1: 15%, S1: 75%, S2: 5%).

It can display three morphological types, the truncal type comprised of two to three longitudinal trunks joined together by few fi ne fibers (8/20), the plexiform or reticular type formed by three to five trunks linked by numerous anastomotic fibers (9/20), and the lamellar type resembling a flat and continuous ribbon descending in front of the vessels and the spine.

6 1

1

5 2 3

4

4

2

3

4

1

2 3

cranial

a

right

Fig 6.6-27a–c The superior hypogastric plexus (anterior view). a Truncal type, as it can be seen in a transperitoneal approach. 1 posterior parietal peritoneum, divided and retracted 2 left common iliac artery 3 superior hypogastric plexus 4 right common iliac artery

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

2

3

cranial

b

right

b Truncal type shown in a cadaveric dissection. 1 left common iliac artery 2 left common iliac vein 3 superior hypogastric plexus 4 disc L5/S1 5 right common iliac artery 6 right common iliac vein

cranial

c

right

c Lamellar type. 1 left common iliac vessels 2 disc L5/S1 3 superior hypogastric plexus 4 right common iliac vessels

6.6

227

Lumbar spine and lumbosacral junction

The superior hypogastric plexus is an area of exchange of nervous fibers between three groups of afferences (two lateral and one middle). The left and right lateral afferences are the four lumbar splanchnic nerves coming from the left and right sympathetic trunks. The middle afference is the previsceral plexus connected with the coeliac, superior mesenteric, and inferior mesenteric plexuses. After a descending course in front of the vessels and the lumbosacral junction, it ends by bifurcation at a level which may vary from L5 to S2 (L5: 5%, L5/S1: 15%, S1: 75%, S2: 5%). The bifurcation gives rise to the right and left hypogastric nerves. On each side, the hypogastric nerve, together with the sacral splanchnic nerves coming from the 2nd and 3rd sacral ganglia, and the erector’s nerves from 2nd, 3rd, 4th sacral nerves, are the afferent nerves to the inferior hypogastric plexus. The superior hypogastric plexus is located in front of the discs. Its bifurcation is usually caudal to the promontorium, and therefore, it can be moved on one side to give access to the disc. It is attached by its splanchnic connections, but these connective nerves are inconsistant at this level (right splanchnic afferents are found in 33 % from L3 and 45 % from L4, left afferents in 15% from L3 and 30% from L4) and the lower splanchnic nerves have no role in urogenital function. Additionally, its fibrous adhesions are detachable from the spine by blunt dissection. The right/left exchange of fibers, before the bifurcation, provides a set of fibers coming from splanchnic nerves from both sides to each hypogastric nerve. The superior hypogastric plexus is an anatomical structure protecting the nerve supply of the

pelvic viscerae. Unilateral lesions to the lumbar splanchnic nerves are usually asymptomatic. The symptoms often appear after a lesion to the plexus itself (or to its branches on both sides). Management of the superior hypogastric plexus (SHP) in transperitoneal approaches

To reach the L5/S1 disc by a right sided approach, the SHP is progressively pushed to the left side. The anterior aspect of L5/S1 is easily exposed on the right side of the plexus, except in the case of a very broad lamellar variety. The left side approach of L5/S1 is not as easy because of the fi bers anastomosing the superior hypogastric plexus to the superior rectal plexus. The right sided approach, when exposing the L4/5 disc, is very difficult or even impossible, and not recommended. The anterior aspect of L4/5 is hidden behind the confluence of common iliac veins. The left sided approach is performed by dissecting the arterial bifurcation from its left side and progressively reclining the plexus around the arterial bifurcation by the left side and progressively reclining the plexus and the arterial bifurcation to the right. Management of the superior hypogastric plexus in retroperitoneal approaches

In retroperitoneal exposures, the right retroperitoneal approach ( Fig 6.6-28 ) is chosen if the L5/S1 disc is the only level to be operated. The superior hypogastric plexus is progressively moved to the left and the disc is clearly exposed.

228

6

Surgical anatomy of the spine

The left retroperitoneal approach is suitable for L5/S1 as well as for L4/5 or L3/4, especially in multilevel procedures. The L5/S1 disc is exposed between the vascular bifurcation. The discs L4/5 and L3/4 are exposed by progressively moving the vessels and superior hypogastric plexus to the right, together with the posterior parietal peritoneum. The superior hypogastric plexus is a part of the autonomous nervous system and it intervenes in the physiology of ejaculation through its sympathetic components which come mainly from the fi rst two lumbar splanchnic nerves. Direct injury to

the plexus should be prevented by using blunt dissection and by avoiding the use of scissors, forceps, scalpel, and electrocoagulation in retroperitoneal soft tissue. In spite of the strictest precautions, indirect injury can occur, due mainly to excessive tension on the nerves by overretraction, provoking a retrograde ejaculation (0.5–6%). This complication should be taken into consideration when an anterior procedure is indicated. The patient must be clearly informed about this risk as well as about the vascular risks.

1 2

1 2

3 3 cranial

4

4

cranial

5

a

left

Fig 6.6-28a–b The superior hypogastric plexus in a retroperitoneal approach. a The right approach to the L5/S1 disc. Right anterolateral view of the lumbosacral junction. 1 right common iliac artery 2 L5 vertebra 3 L5/S1 disc 4 S1 vertebra

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

left

b The left approach to L5/S1 and L4/5 (and overlying discs). Left anterolateral view of the lumbosacral junction. 1 vertebra L4 2 disc L4/5 3 vertebra L5 4 left common iliac artery 5 disc L5/S1

229

6.6

Lumbar spine and lumbosacral junction

2.7

ANATOMICAL VARIATIONS OF THE PREVERTEBRAL VESSELS AND THEIR MANAGEMENT DURING THE ANTERIOR APPROACH TO THE SPINE

Topographical arrangement

The topographical arrangement of the prevertebral vessels ( Fig 6.6-29 ) is variable, for the most part, according to the level of the bifurcation of the aorta and the origin of the inferior vena cava. Five types have been described:

a

Fig 6.6-29a–e Morphological and topographical variations of the prevertebral vessels. a Classic type. b Paradoxical type. c Unusual high type. d Unusual low type. e Dissociated type.

• Classic type (63.5%)—the aortic bifurcation is at the level of the L4/5 disc; the confluence of the common iliac veins lies just below. • Paradoxical type (6.4%)—the aortic bifurcation lies below the origin of the inferior vena cava. • Unusual high type (15.4%)—the aortic bifurcation is cranial to the L3/4 disc and the venous confluence is cranial to the L4/5 disc. • Unusual low type (5.1%)—the aortic division lies at the level of L5, the origin of the vein is at the level of the L5/S1 disc. • Dissociated type (9.7%)—the aortic bifurcation lies more than one vertebra above the origin of the inferior vena cava.

b

c

d

e

230

6

Surgical anatomy of the spine

The knowledge of these topographical variations enables the surgeon to choose the adequate maneuvers required to mobilize the vessels and expose the lumbosacral spine (interiliac exposure of the L5/S1 disc; interiliac mobilization with temporary collapse of the left iliac vein; left lateroaortic mobilization to approach the L4/5 and overlying discs; interaortocaval exposure, more rarely required, to expose the L3/4 and L2/3 discs).

2.8

MANAGEMENT OF THE PERITONEAL SAC IN THE ANTERIOR APPROACH TO THE LUMBAR AND LUMBOSACRAL SPINE

The management depends on whether the approach is made trans- or retroperitoneally. The transperitoneal approach is indicated when additional reduction maneuvers are required (high grade spondylolisthesis). The retroperitoneal approaches are recommended in all other situations.

Some rare anatomical variations and their management

The duplication of the inferior vena cava results from the persistence of twin embryonic venae cavae. It is characterized by the presence of a large vein on the left side of the aorta. This vein can be mobilized after ligature of the left segmental vessels. Its ligature is possible, but usually not necessary. The persistence of a left inferior renal vein in a retroaortic position needs no special care during the lumbosacral approaches. The right retrocaval ureter, due to an abnormal fusion of embryonic segments of the ureter, can be exposed in right lumbotomies and can be confused with an anastomotic vessel.

2.8.1 ANATOMICAL BASIS FOR TRANSPERITONEAL APPROACH AND EXPOSURE OF THE LUMBOSACRAL SPINE

The anterior parietal peritoneum is a thin serous sheet covering the anterior musculoaponeurotical abdominal wall. A more or less thick layer of fat tissue is interposed between this sheet and the transversalis fascia. The skin incision can be longitudinal, on the midline, along the vertical line where the abdominal hair converge, or transverse, according to Pfannenstiel, along the inferior abdominal fold. In the latter incision, the subcutaneous fat is divided vertically to joint the linea alba. After dividing the skin longitudinally along the midline, the subcutaneous fat tissue is divided using electrocautery until the linea alba is reached. The linea alba is narrow. That is the reason why, in order to lie exactly in front of it, the skin incision should be carried out along the vertical line where the abdominal hair converge. Once the fibers of the linea alba are divided, the preperitoneal fat can be divided using blunt dissection until the peritoneal sac appears. The peritoneum can then be held with two forceps before cutting it longitudinally

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

6.6

231

Lumbar spine and lumbosacral junction

with a pair of thin scissors. The enlargement of the access is made by dividing the linea alba cranially and caudally. An autostatic retractor is then positioned. The greater omentum is displaced cranially, the sigmoid colon and its mesocolon are pushed to the left side; the small bowels and the mesentery are gently pushed and maintained cranially and on the right side. The retroperitoneal space is penetrated by dividing the posterior parietal peritoneum along the midline, in front of the vessels. In slim patients, the enlargement of these vessels or their beating can sometimes help to identify them through the posterior parietal peritoneum and the retroperitoneal fat. If not, fi nger palpation can also help to identify the arteries as well as the promontorium. Thin white nervous bundles can also be seen sometimes due to transparency; they are components of the superior hypogastric plexus. Here, a single sheet of the posterior parietal peritoneum separates the peritoneal cavity from the retroperitoneal space; it is located between the root of the mesentery on its right side and the primary root of the mesosigmoid on its left side. The peritoneum is held and folded transversely using smooth vascular forceps. A small cut is made vertically on the midline, using thin scissors. It is then enlarged cranially and caudally. Long thin retractors applied to the peritoneal edges help to expose the retroperitoneal space. Approach to disc L5/S1

Blunt dissection allows identification of the confluence of the common iliac veins, usually in front of L5. The left common iliac vein and the venous confluence can sometimes be in an abnormal caudal situation, below the L5/S1 disc. In these rare cases, it is preferable to abandon the anterior procedure rather than risk dangerous vascular injuries. The longitudinal fibers of the superior hypogastric plexus run in front of the vein;

they must be preserved and gently pushed to the left side. This maneuver is not very difficult in the case of a plexiform or truncular presentation. In a very thick and broad lamellar presentation, it is recommended to abandon the anterior procedure rather than take any risk that might affect sexual function through laceration or overdistraction of the plexus. A layer of fibrous fat tissue is lying in front of the vessels and the lumbosacral disc, usually embedding the fibers of the superior hypogastric plexus. The anterior aspect of the disc should be reached at fi rst on its right side, medial to the right iliac vein. This maneuver is performed using clamp mounted nuts. The difficulty or ease with which this is done can vary depending on the density and cohesion of this fibrous and adipose tissue. The superior hypogastric plexus has no relevant links on the right side, while on the left side it is more or less attached by anastomotic branches to the rectosigmoid and inferior mesenteric plexuses, thus making the left sided approach more challenging than one from the right side. Once the anterior surface of the disc is reached (easily identifi able by its pearly and shiny vertical fibers), it is exposed by blunt dissection and the retroperitoneal soft tissue is progressively pushed cranially and laterally. The medial edge of the right and left common iliac veins and the inferior edge of their confluence are gently and progressively pushed laterally and cranially using blunt dissection. When the disc is clearly exposed, Steinmann pins or Hohmann retractors are anchored in the adjacent vertebrae in order to maintain the exposure. Approach to discs L4/5 and L3/4

The approach to the L4/5 and L3/4 discs is performed by mobilizing the aorta and its bifurcation to the right side. The superior hypogastric plexus is left in front of the aorta within

232

6

Surgical anatomy of the spine

its surrounding fat tissue. The lowest left lumbar splanchnic nerves (3rd and 4th) can be injured during this exposure without any significant consequence to sexual function. The approach to the disc on the left side of the vessels is performed using blunt dissection. Segmental vessels of various sizes may be encountered during this approach. They are located at the level of the vertebral bodies. The artery and the vein must be tied separately, using knots or clips. The ascending lumbar vein can sometimes be a serious obstacle to this approach when it is too large. It has to be carefully isolated and ligatured. Once the aorta and the inferior vena cava are separated from their segmental attachments, they can be progressively and carefully displaced to the right side, using blunt dissection. Once a significant area of the disc (2–3 cm on each side of the midline) is freed from the overlying vessels, the exposure is maintained by the appropriate retracting devices. The tips of Steinmann pins or Hohmann retactors should be introduced into the adjacent vertebral bodies, after pushing and maintaining the overlying soft tissue beyond the point where the pin will be implanted into the bone. Similarly, at the moment of the extraction of the pin, the surrounding soft tissue should be pushed away from the pin, using a clampmounted nut, in order to avoid any direct or indirect injury to the vessels. The retraction should be released each half hour, to avoid or reduce the risk of ischemia and any neurological damage due to the overdistraction.

2.8.2 ANATOMICAL BASIS FOR RETROPERITONEAL APPROACH AND EXPOSURE OF THE LUMBOSACRAL SPINE Side of approach

The L5/S1 disc can be approached either from the right or the left side. The advantage of the right sided approach is that the right half of the disc is less covered than its left side by the SHP, thus making it an easier approach. Additionally, the right common iliac vein is protected by the artery which is more medial than the vein. If the L5/S1 disc is the only disc to be approached, the right sided approach is preferable. If two or more discs, including L5/S1, have to be treated, the approach must then be carried out from the left side. Approach to disc L5/S1

From the right as well as from the left side, the retroperitoneal approach starts after the longitudinal dissection of the anterior sheet of the rectus abdominis sheath. The medial border and the posterior aspect of the rectus abdominis muscle are then progressively detached from the posterior sheet of the rectus sheath to the lower border of the arcuate line. The peritoneal sac is then detached from the anterior abdominal wall, below the arcuate line, using blunt dissection or clamp-mounted nuts— or more simply, the fi ngers. The detachment is particularly difficult from the posterior aspect of the posterior sheet of the rectus sheath. The detachment should be very progressive, using clamp-mounted nuts in order to avoid perforation of the peritoneal sac. If such a perforation occurs, it must be sutured immediately. The detachment of the peritoneal sac is easier at the lowest part of the abdominal wall, especially from the pubis and then from the lowest part of the internal iliac fossa. So, it is easier to start the detachment at this level and then continue it in a medial and cranial direction toward the promontorium.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

6.6

233

Lumbar spine and lumbosacral junction

3

The medial retraction of the peritoneal sac and its contents may be hindered by the tension of the arcuate line. The inferior border of this fibrous sheet (the posterior sheet of the rectus sheath) is not at the same level in all individuals. In cases where it hinders the medial progression of the exposure, its longitudinal division is recommended after detachment from the peritoneal sac by blunt dissection using clamp-held nuts. The inferior epigastric vessels might require ligation. The gonadal vessels in males join the spermatic duct near the inner orifice of the inguinal canal. As well as the ureter, these elements are usually easily detached from the wall with their surrounding retroperitoneal fat and pushed medially with the peritoneal sac. The median sacral artery and vein(s) should be isolated and divided separately. The last steps of the exposure are identical to those described for the transperitoneal approach. Approach to disc L4/5

The L4/5 disc is approached from the midline. It should be approached from the left side. An approach from the right side is difficult due to the vascular anatomy and should be done only in case of right convex scoliosis. The arcuate ligament needs to be divided proximally after blunt dissection and detachment from the peritoneal sac in order to have access to the L4/5 disc. The next steps are similar to those for the transperitoneal approach. The anterior aspect of the disc is identified and progressively exposed using clamp-held nuts. The retroperitoneal vessels, nerves, and the peritoneal sac are progressively detached from the anterior aspect of the spine by blunt dissection, and pushed laterally to the right side. The segmental vessels and the iliolumbar vein require their division before the fi nal exposure of the disc.

SUGGESTED READING

Barker PJ, Briggs CA (1999) Attachments of the posterior layer of lumbar fascia. Spine ; 24(17):1757–1764. Daggfeldt K, Huang QM, Thorstensson A (2000) The visible human anatomy of the lumbar erector spinae. Spine ; 25(21):2719–2725. Ebraheim NA, Lu J, Hao Y, et al (1997) Anatomic considerations of the lumbar isthmus. Spine ; 22(9):941–945. Ebraheim NA, Xu R, Darwich M, et al (1997) Anatomic relations between the lumbar pedicle and the adjacent neural structures. Spine ; 22(20):2338–2341. Lu J, Ebraheim NA, Biyani A, et al (1996) Vulnerability of great medullary artery. Spine ; 21(16):1852–1855. Marchesi D, Schneider E, Glauser P, et al (1988) Morphometric analysis of the thoracolumbar and lumbar pedicle anatomo-radiologic study. Surg Radiol Anat ; 10(4):317–322. Mirkovic SR, Schwartz DG, Glazier KD (1995) Anatomic considerations in lumbar posterolateral percutaneous procedures. Spine ; 20(18):1965–1971. Olszewski AD, Yaszemski MJ, White AA (1996)The anatomy of the human lumbar ligamentum fl avum. New observations and their surgical importance. Spine ; 21(20):2307–2312. Petraco DM, Spivak JM, Cappadona JG, et al (1996) An anatomic evaluation of L5 nerve stretch in spondylolisthesis reduction. Spine ; 21(10):1133–1139. Tribus CB, Belanger T (2001) The vascular anatomy anterior to the L5-S1 disk space. Spine ; 26(11):1205–1208. Vraney RT, Phillips FM, Wetzel FT, et al (1999) Peridiscal vascular anatomy of the lower lumbar spine. An endoscopic perspective. Spine ; 24(21):2183–2187.

234

6 6.7

SURGICAL ANATOMY OF THE SPINE SACRUM

1

External morphology ……………………………………………………………………………………… 235

2 2.1 2.2

Anatomy of facets and pedicles ………………………………………………………………………… 236 Anatomy of the facets related to translaminar L5/S1 screw fixation ……………………………… 236 Anatomy of the pedicles related to pedicle screw insertion ……………………………………… 236

3

Suggested reading ………………………………………………………………………………………… 239

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

235

Serge Nazarian, Cyril Solari

6

SURGICAL ANATOMY OF THE SPINE

6.7

1

SACRUM

EXTERNAL MORPHOLOGY

The sacrum is the base of the spine, which is anatomically and functionally included in the pelvic ring. The sacrum is the keystone of this ring, allowing the transmission of the weight to flow from the trunk to the lower limbs through the lumbosacral joint and then on to the sacroiliac and hip joints. The posterior aspect of the sacrum ( Fig 6.7-1) is most often used in spine surgery for the implantation of screws or rods. It is convex and triangular with a superior base. This posterior aspect displays three vertical crests. The median sacral crest results from the fusion of the spinous processes. More laterally, the intermediate sacral crest is the equivalent of the fused zygapophyseal columns; it is separated from the median crest by the laminar gutter. The posterior sacral foramina are located laterally between the intermediate sacral crest and the lateral sacral crest. The diameter of the posterior sacral foramina ranges from 5 to 10 mm. The lateral sacral crest is the equivalent of the tip of the fused costiform processes.

1 4

2 3

Fig 6.7-1 The posterior aspect of the sacrum. 1 median sacral crest 2 intermediate sacral crest 3 lateral sacral crest 4 posterior sacral foramina

236

6

Surgical anatomy of the spine

The anterior aspect of the sacrum ( Fig 6.7-2 ) is concave and triangular. Sacral bodies are clearly seen along the middle part of the sacrum. A small ridge, the transverse ridge, is found between each vertebral body. It corresponds to the ossification of the disc. These transverse ridges end laterally at the medial border of the adjacent foramina. The four foramina are more or less located equidistant from one another. Their average size is about 13 mm for S1 and S2, and 10 mm for those below. Their medial edge is sharp and their lateral edge is smooth, due to the orientation of the sacral nerves.

3

1

2

ANATOMY OF FACETS AND PEDICLES

2.1

ANATOMY OF THE FACETS RELATED TO TRANSLAMINAR L5/S1 SCREW FIX ATION

The coronal orientation of the L5/S1 facets means that the joint space is in more or less the same orientation as that of the laminae. The relative thickness of these laminae make the translaminar screw fi xation of L5/S1 difficult. A direct facet screw insertion (Boucher), by driving the screw in a craniocaudal and mediolateral direction into the ala of the sacrum, is feasible at this level. In both procedures the purchase is poor, due to the poor quality of the cancellous bone in the ala.

2.2

ANATOMY OF THE PEDICLES RELATED TO PEDICLE SCREW INSERTION

2

The fi rst and second pedicles of the sacrum can be used for screw fi xation. The limits of the S1 pedicle ( Fig 6.7-3 ) are the lateral recess of the canal medially, the lateral aspect of the facet laterally, the L5/S1 foramen cranially, and the first sacral foramen caudally. Fig 6.7-2 The anterior aspect of the sacrum. 1 sacral bodies 2 anterior sacral foramina 3 transverse ridge S1/2

The limits of the S2 pedicle ( Fig 6.7-4 ) are the lateral recess of the canal medially or a vertical line passing by the medial border of the fi rst and second posterior foramina, the vertical line passing by the lateral border of the fi rst and second posterior foramina laterally, the inferior limit of the fi rst foramen cranially, and the superior limit of the second foramen caudally. The S1 pedicles are the widest pedicles of the spine. In the horizontal plane, the size is about 10–15 mm measured from the bottom of the lateral recess of S1 to the lateral border of the

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

6.7

237

Sacrum

1 2

1

1

2 3 4

2 3

b c

a b Posterior view. 1 superior limit 2 right facet of S1 3 inferior limit 4 first posterior foramen

Fig 6.7-3a–c The limits of the S1 pedicle. a Superior view. 1 lateral limit 2 medial limit

1

2 3 4

Fig 6.7-4 The limits of the S2 pedicle. 1 first posterior foramen 2 superior limit 3 inferior limit 4 second posterior foramen

c Coronal-oblique cut. 1 left pedicle of S1 2 left pedicle of S2 3 left pedicle of S3

238

6

Surgical anatomy of the spine

S1 facet; the lateral aspect of the pedicle is in continuity with the ala. In the vertical plane, the average height ranges from 25 to 30 mm. The length of the S1 pedicle itself is 12 mm. This is the reason why the screws should be inserted obliquely toward the vertebral body of S1, in order to increase their purchase. Due to the shortness of the pedicle and its lateral fusion with the costal process and the ala, the orientation of the pedicle screw is not easy to defi ne. However, two factors must be kept in mind: In the transverse plane, the screw must be directed medially to reach as far as possible into the vertebral body of S1 in order to increase the extent of its purchase. In the sagittal plane, it should be located as close as possible to the end plate of S1, to increase its purchase of the dense cancellous bone lining the end plate.

a

Practically, in most cases of sacral pedicle screw usage, the overhang of the iliac crest limits the medial orientation to 15–20°. Conversely, the screw directed toward the ala can have an outward direction of up to 30–45°. In cases requiring a strong purchase, the large size of the S1 pedicle allows the insertion of two screws, one oriented medially toward the promontory and the other oriented laterally toward the ala and the sacroiliac joint. The maximum length of the screw varies according to the direction of the screw. A straightforward S1 pedicle screw may be no longer than 30 mm. A screw oriented medially and cranially toward the promontory can be 40–50 mm long. And, a screw oriented 30–45° laterally, can be 35–40 mm in length if it is parallel to the end plate, and 50–55 mm if it is oriented 45° caudally.

b

Fig 6.7-5a–c Anatomical landmarks for S1 pedicle screw insertion. Entry point of the pedicle of S1 at the posterior inferior angle of the articular process of S1. a Posterior view. b Superior view. c AP x-ray.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c

6.7

239

Sacrum

3

The anatomical landmarks for sacral pedicle screw insertion into S1 ( Fig 6.7-5 ) are the inferior and lateral angle of the sacral facet for a screw aiming medially (15–20° with respect to the sagittal plane) and cranially (10° with respect to the end plate) in the direction of the promontory. For insertion into S2, the anatomical landmarks ( Fig 6.7-6 ) are the middle of the distance between the inferior border of the fi rst posterior foramen and the superior border of the second posterior foramen, in vertical alignment with the S1 screw, practically at the inferior lateral border of the inferior facet of S1 (which is fused along the intermediate sacral crest); the medial orientation should be about 30° and the length of the screw about 30 mm. Alternatively, S2 screws can be inserted with a 30° outward direction toward the sacral ala.

1

Fig 6.7-6 Anatomical landmarks for S2 pedicle screw insertion. The pedicle screw is inserted at the inferior lateral angle of the inferior facet of S1, and oriented medially with an angle of 30°. 1 entry point of S2 pedicle screw

SUGGESTED READING

Ebraheim NA, Xu R, Biyani A, et al(1997) Morphologic considerations of the fi rst sacral pedicle for iliosacral screw placement. Spine ; 22(8):841–846. Mazda K, Khairouni A, Penneçot GF, et al (1998) The ideal position of sacral transpedicular endplate screws in Jackson‘s intrasacral fi xation. An anatomic study of 50 sacral specimens. Spine ; 23(19):2123–2126. Peretz AM, Hipp JA, Heggeness MH (1998) The internal bony architecture of the sacrum. Spine ; 23(9):971–974. Xu R, Ebraheim NA, Robke J, et al(1996) Radiologic and anatomic evaluation of the anterior sacral foramens and nerve grooves. Spine ; 21(4):407–410.

Vincent Arlet

SPINAL INSTRUMENTATION

7

243

Vincent Arlet

7

SPINAL INSTRUMENTATION

INTRODUCTION

The AOSpine principles and techniques for instrumentation in spine surgery were traditionally based upon the universal spine system (USS) and the use of trauma reconstruction plates. With these two kinds of instrumentation it was, and still is, possible to successfully address almost all the pathologies of the spine. In less than 10 years after the publication of the fi rst AO manual of the principles of spine surgery, there has been a myriad of new types of instrumentation, all supposedly better than the previous ones and often more expensive. Therefore, a totally new section was needed on spinal instrumentation so that new technological advances could be reviewed one by one with their respective advantages, disadvantages, complications, and respective indications and contraindications—keeping in mind that new is not always better. Some types of instrumentation will eventually be replaced by new ones, and only principles are here to stay. The creativity and skills of the surgeons have led to the development of new fi xation techniques in the cervical spine

such as the anterior C1/2 fi xation, the posterior C1 lateral mass screw fi xation, and the cervical pedicle screws [1–3]. In the thoracic spine, thoracic pedicle screw fi xation has now been popularized even for the treatment of complex spinal deformities [4]. Such creativity has the advantage of achieving ever stronger fi xations, and/or better corrections, but often at the price of increased potential complications. For example, lateral mass screws in the cervical spine meet the demands of cervical fi xation in 95% of the cases and rarely are cervical pedicle screws necessary. In rare instances, cervical pedicle screws may offer a definitive advantage in osteoporosis or in a long fi xation, but these benefits are dependent on very demanding techniques and neurological complications do occur more frequently than with lateral mass screws. In spinal deformities the use of thoracic pedicle screw fi xation achieves better correction of the Cobb angle, but it is defi nitely more dangerous than hooks and its superiority in clinical outcome has not been proved. In the osteoporotic spine advanced methods, such as pedicle screw fi xation augmented with cement, are becoming increasingly used and have allowed shorter

244

instrumentations, and hopefully fewer transition problems. All the tricks of the trade must, however, be mastered in order to avoid complications such as cement leakage. There has been a trend toward emphasizing the ease of application of different systems so that the instrumentation would become more user-friendly and more forgiving. This had a very positive impact on some fi xation techniques, like the polyaxial top-loading cervical fi xation. On the other hand, using polyaxial screw fi xation in the thoracolumbar spine had only limited advantages except for its ease of application, its forgiving nature, and the reduced risk of failure at the bonescrew interface in cases with osteoporosis. The disadvantages of such polyaxial fi xations are: the increased price, the higher profi le, their decreased ability to correct deformities, and their tendency toward failure at the polyaxial head which couples to the screw mechanism [5]. To some extent the use of such polyaxial systems has been very detrimental when used for fracture fi xation. The change from the fi rst generation USS to the dual-opening USS was also done in an attempt to improve screw design and to simplify the use of the system so that all implants could be loaded on either side. Screw shafts now have a double core design and a blunt tip. This results in an increase in screw purchase in the vertebral body which is important in osteoporotic bone. Simple deformity corrections are now made easier with the use of the dual-opening USS system. However, the market driven requirement of having a lower profi le system (small stature 5 mm rod system) rendered the rod capture mechanism of the USS II cap and nut far more difficult than the USS I. This is particularly true in complex deformities at the lumbosacral junction where the rod requires marked bending to match the lordosis. Today, due to this

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

problem, a lumbosacropelvic fi xation is still best addressed with USS I type instrumentation. Less invasive spine surgery with better instrumentation, tools, and retractor systems were major advances of the last 10 years. The blade retractor system (Synframe) is a typical example of a very modular system in which the same retractor system with different modules can be used to perform minimally invasive lumbar disc surgery, anterior or posterior cervical spine surgery, or very extensive thoracoabdominal or anterior cervicothoracic approaches [6]. Judicious use of such a retractor system allows, in most cases, an approach to the thoracolumbar spine without exposing the diaphragm (eg, one incision from T12 to the sacrum or from T7 to L2). Likewise, using this retractor a minimally invasive access to the lumbosacral spine can be used to insert a disc prosthesis or an expandable cage, always with perfect illumination and possible fluoroscopy control. Cage technology has seen tremendous improvements with the advent of expandable cages, which allow less invasive anterior approaches to the spine. These cages can be inserted through a very small incision or even when using a thoracoscopic approach. The success of these expandable cages has by and large surpassed what they were initially designed for, ie, insertion from the anterior aspect of the spine. However, it became obvious that in some circumstances it was possible to insert them in a posterolateral fashion, through the same incision as the posterior fixation [7]. Nevertheless, such expandable cages carry with them a major trade off that limits their use; the amount of metal in these cages restricts their ability to perform a spinal fusion. Subsidence around the end plate has been observed after the use of such cages, in some indications even after posterior fi xation. In the end,

7

245

Spinal instrumentation

BIBLIOGRAPHY

their cost may in some indications and health systems remain prohibitive, but the classic vertebral body replacement with cement can still be judiciously used for conditions such as metastases. New advances in implant manufacturing have made the cages available in polyetheretherketone (PEEK). PEEK cages have a module of elasticity very close to the cortical bone of the vertebrae and they are radiolucent. The advantage includes better MRI visibility after cage insertion, which is always an advantage in degenerative conditions. At last, some taboos which existed with good reason in the AOSpine community, like “performing an anterior stand-alone interbody fusion with cage,” are now being challenged by the success of the new stand-alone anterior cage made of PEEK and the built-in plate fi xation with locking screws (Synfi x) [8]. More recently, the so-called motion-preservation technology has gotten more attention as it is believed to represent the future of degenerative spine surgery. These advanced technologies are now being challenged by evidence-based literature as well as by third-party payers, and the future will tell if they are superior to classic fusion techniques.

CONCLUSION

Spinal instrumentation will no doubt continue to evolve, and our role as spine surgeons is not to pick “the latest instrumentation out there or the most expensive instrumentation”, but to apply the principles of spinal instrumentation and to choose or devise the most appropriate one for our patient’s needs, for our surgical skills, and for the socioeconomic environment we are working in. Otherwise we shall merit the old saying: “A fool with a tool is still a fool”.

1. Reindl R, Sen M, Aebi M (2003) Anterior instrumentation for traumatic C1/2 instability. Spine ; 28(17):E329–333. 2. Harms J, Melcher RP (1999) Posterior C1-C2 fusion with polyaxial screw and rod fi xation. Spine ; 26(22):2467–2471. 3. Abumi K, Kaneda K, Shono Y, et al (1999) One-stage posterior decompression and reconstruction of the cervical spine by using pedicle screw fi xation systems. J Neurosurg ; 90(1 Suppl):19–26. 4. Suk SI, Lee CK, Kim WJ (1995) Segmental pedicle screw fi xation in the treatment of thoracic idiopathic scoliosis. Spine ; 20(12):1399–1405. 5. Fogel GR, Reitman CA, Liu W, et al (2003) Physical characteristics of polyaxial-headed pedicle screws and biomechanical comparison of load with their failure. Spine ; 28(5):470–473. 6. Aebi M, Steffen T (2000) Synframe: a preliminary report. Eur Spine J ; 9 (Suppl 1):S44–50. 7. Hunt T, Shen FH, Arlet V (2006) Expandable cage placement via a posterolateral approach in lumbar spine reconstructions. Technical note. J Neurosurg Spine ; 5(3):271–274. 8. Cain CM, Schleicher P, Gerlach R, et al (2005) A new stand-alone anterior lumbar interbody fusion device: biomechanical comparison with established fi xation techniques. Spine ; 30(23):2631–2636.

246

7 7.1

SPINAL INSTRUMENTATION MODULARITY OF SPINAL INSTRUMENTS (SYSTEMS)

1

Introduction ………………………………………………………………………………………………… 247

2

Modularity applied to instrumentation for different spinal areas ………………………………… 248

3

Modularity for anterior and posterior use …………………………………………………………… 249

4

Modularity for different pathologies …………………………………………………………………… 249

5

Suggested reading ………………………………………………………………………………………… 250

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

247

Max Aebi

7

SPINAL INSTRUMENTATION

7.1

1

MODULARITY OF SPINAL INSTRUMENTS (SYSTEMS)

INTRODUCTION

It has been one of the major goals and later an important asset of AO to standardize principles and techniques, in addition to improving instruments and implants, in order to facilitate surgical procedures and to increase safety for the patient. Simple developments, such as the hexagonal screw driver and the imbus screw head in two different sizes (as part of the trauma instrumentation), enabled AO surgeons to perform procedures in a standardized manner, regardless of which part of the world they were practicing in. The same advantages hold true for the plate-screw concepts and intramedullary nail systems. In the 1980s, when instrumented spine surgery began to gather momentum many spine surgeons, working together with industry, started developing their own instrumentation and implants. This was done mostly out of a necessity to address specific needs when dealing with different spinal pathologies. Almost every pathology had its own instrumentation, often in several different forms, resulting in a large variety of implants and instruments. The value of this trend could certainly be

questioned. As a result, a systematic modularity of implant and instruments was developed, such as the universal spine system (USS) and the cage system. Nevertheless, today, it is sometimes no longer clear whether the constant drive to renew instrumentation and implants is determined by the needs of patient care or rather by commercial implications and the individual ambitions of the surgeon. This trend toward commercial success sometimes opposes the concept of modularity as there is less profit to be made through the sale of less complex systems with modularity and greater universal application. Different applications of modular spinal instrumentation: • Systems used in different spinal areas: cervical, thoracic, and lumbosacral spine. • Systems for anterior or posterior applications of an instrumentation. • Systems for different pathologies.

248

7

2

Spinal instrumentation

MODULARITY APPLIED TO INSTRUMENTATION FOR DIFFERENT SPINAL AREAS

The concept of posterior instrumentation from the occiput to S1 is the same: anchorage in the spine, mostly by pedicle or lateral mass screws or by hooks, longitudinal vertical bars connecting to the anchorage points, be it through a plate or a rod; and fi nally, a connector piece between the anchorage and the longitudinal bar, which may be incorporated into the screw head or hook. For the thoracolumbar spine, pedicle screws are the primary solution for an efficient fi xation of an implant to the spine. The pedicle screw is anchored in the strongest part of the vertebra and as a result not only allows bone fixation, but also the possibility of manipulating individual vertebrae, thereby correcting deformities and dislocations. The alternative in the thoracic spine is the hook, which is either used as a lamina or a pedicle hook. In order to enable surgeons accustomed to pedicle hooks to continue the use of this technique, the two senior editors of this book developed the concept of the “pedicle hook screw”. This concept is unique, since it allows for the pedicle hooks to be positioned as usual and also allows for stabilization at the same time, so that they almost behave like pedicle screws. The pedicle hook is meant to grasp the pedicle from below (exactly like the pedicle hook fi nder), forming a tight fit between the pedicle and the hook. It is obvious that a screw, which is centered exactly between the two limbs of the hook grasping the pedicle, must inevitably hit the pedicle in the center. The concept gave birth to the pedicle hook screw of the universal spine system (USS) deformity module, thus allowing safe use of the pedicle to fix a strong anchorage, even in deformed spines. The screw and hook elements are absolutely modular in their use: The connection to the longitudinal bar

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

(rod) is the same for all applications, as are the instruments needed to handle both hooks and screws. The side-opening pedicle screws and hooks in the original USS (USS I) allow for the translation of the anchorage (screw/hook) toward the vertical bar (rod) in corrective surgery. Whether the screw or hook has to be translated from the lateral or medial side does not matter because the screw can be turned by 180° to fit the rod, and the hooks are available as left-opening and rightopening. With bilaterally opening screws and hooks (USS II), the turning of a screw or the exchange of a hook is no longer necessary. The screw head in the cervical spine systems, eg, of the Axon system, is different, since the requirements differ with respect to the anatomy and the force to be applied. However, the rod can be connected to the regular USS by using a domino connector, which allows for a cervico-thoracolumbo-sacral fixation or by using a tapered rod (cephalad part = 3.5 mm diameter and caudal part = 6 mm; chapter 7.2.5 Cervicothoracic junction). The polyaxial screws were originally developed for routine lumbar and lumbosacral cases with posteriorly open screws (Click’X, top loading) or laterally open screws, eg, variable axis screws (VAS), to accommodate small deviations of the anchorage points from each other, still allowing the rod delivery either from the top or from the side. This screw-rod concept has also been applied to the cervical spine system. The instruments used to manipulate the screws, hooks, and rods follow the same principles and are reduced in number to a minimum. The introduction of the top-loading screws in the Click’X and the Axon instrumentation make very few new instruments necessary. The rod is universal and can be elongated via a domino connector, either of the same size, a smaller diameter if required for the cervical spine, or a tapered rod (3.5 –6.0 mm) can be used.

7.1

Modularity of spinal instruments (systems)

3

MODULARITY FOR ANTERIOR AND POSTERIOR USE

The instrumentation for the thoracolumbar spine can be applied anteriorly and posteriorly. Some additional elements have been introduced for better adaptation to the anterior application—creation of a rod-holding platform to the vertebral body in the Ventrofi x instrumentation, although most of the anterior pathologies could be treated by using the regular USS. The fixation of this platform to the spine is made by locking screws, a technology previously developed for the anterior cervical plates, eg, cervical spine locking plate (CSLP). The posterior scoliosis instrumentation (USS I and USS II) can also be applied for anterior corrective surgery and the only additional implants that may be required are two different types of washer. The correction follows the same principles of translation for each individual vertebra fi xed with screws to the rod, which is precontoured to the planned form of the corrected spine.

249

4

MODULARITY FOR DIFFERENT PATHOLOGIES

Application of the pedicle screw and the type of posterior instrumentation to be used is basically independent of the pathology. However, there are specific needs that may be typical for certain pathologies. For example, a short instrumentation with a stable angle between the screw and the rod is preferable for the treatment of traumatic as well as pathological fractures, and most secondary tumors. However, the ability to change the angle between rod and screw is needed—such a complex requirement is fulfi lled by the unique clamp of the internal fi xator, which is an integral part of the USS. The characteristics of this clamp allow for the correction of a deformity caused by a traumatic or pathological fracture (eg, restoring lordosis, correction of frontal and horizontal deviations) and then ensure maintenance of the corrected position at a stable angle. Such characteristics do not apply to the regular USS screws or most of the other systems available on the market. The USS is a modular system with three basic modules: • Deformity module (characteristics: pedicle hook screw and hooks). • Fracture module (characteristics: internal fi xator clamp). • Degenerative module including VAS if necessary, its most simplified version is the Click’X. In this third module it is possible to add a specific device for the reduction of spondylolisthesis. With this modular system 95% of all spinal thoracolumbar pathologies can be treated when a posterior approach is chosen. For different pathologies that may necessitate anterior column reconstruction, the Ventrofi x and anterior use of the USS are sufficiently modular and may be combined with intervertebral cages, of which a whole product family exists; standard cages (Syncage, cages made of polyetheretherketone, ie, PEEK, and

250

7

Spinal instrumentation

5

Synfix), expandable cages (Synex), or customized cages (Synmesh), which have identical footprints of different sizes. Again, the Synmesh is a modular system per se and can be applied from the cervical spine down to the sacrum. In today’s environment of health economy constraints it is mandatory to work with systems that include common features and instruments in order to reduce costs. This aim is not possible with separate implant and instrument systems for each individual problem.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

SUGGESTED READING

Laxer E (1994) A further development in spinal instrumentation. Technical Commission for Spinal Surgery of the ASIF. Eur Spine J; 3(6):347–352.

251

252

7 SPINAL INSTRUMENTATION 7.2 CERVICAL SPINE 7.2.1 MODULARITY AND EVOLUTION OF INSTRUMENTATION FOR THE CERVICAL SPINE

1

Introduction ………………………………………………………………………………………………… 253

2 2.1 2.2 2.3 2.4

Anterior cervical plating ………………………………………………………………………………… H-plate ……………………………………………………………………………………………………… Constrained or static plates ……………………………………………………………………………… Dynamic and hybrid plates ……………………………………………………………………………… Screws ………………………………………………………………………………………………………

254 254 255 257 258

3 3.1 3.2 3.3 3.4

Posterior cervical fixation ………………………………………………………………………………… Wires and cables ………………………………………………………………………………………… Hook plates ………………………………………………………………………………………………… Plate systems ……………………………………………………………………………………………… Rod-screw systems ………………………………………………………………………………………

258 259 260 260 261

4

Conclusion ………………………………………………………………………………………………… 262

5

Bibliography ………………………………………………………………………………………………… 263

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

253

Michael E Janssen, Jason C Datta, Vincent Arlet

7 7.2

SPINAL INSTRUMENTATION CERVICAL SPINE

7.2.1

MODULARITY AND EVOLUTION OF INSTRUMENTATION FOR THE CERVICAL SPINE

1

INTRODUCTION

Prior to the 1950s, cervical spine surgery was largely done posteriorly. In the mid-1950s pioneers such as Smith and Robinson, Cloward as well as Bailey and Badgley introduced the concept of anterior cervical spine surgery. Cloward used a technique involving a cylindrical graft for an anterior fusion following a discectomy [1]. In the same year, Smith and Robinson reported their experience with the use of a tricortical bone graft to achieve fusion [2]. In 1960, Bailey and Badgley reported an approach to the anterior spine and used an onlay graft for fusion [3]. The advantages of the anterior approach were that it was easy to perform and the spine could be decompressed through a wide exposure. In the 1970s, anterior cervical plating was reported by Orozco [4] and Senegas [5]. With procedures involving anterior plating it was necessary to achieve screw purchase in both cortices to prevent screw pullout. In the early 1980s, the concept of anterior plating was popularized with

the Caspar plate system [6], and in the late 1980s early 1990s, the Morscher plate with the unicortically fi xed-angle locking screw was popularized under the name cervical spine locking plate or CSLP [7]. After 2000, the concept of load sharing with dynamization was introduced and led to the development of new plate systems, thereby improving the original design of the CSLP [8].

254

7 7.2

Spinal instrumentation Cervical spine

2

ANTERIOR CERVICAL PLATING

The primary disadvantages of bone grafting without spinal instrumentation include the tendency of the bone graft to migrate anteriorly and the potential for sagittal deformity. Following bone graft placement for discectomy or corpectomy, the graft is subject to axial loading as well as secondary bending or shear forces. To stop the graft from migrating anteriorly, plating was introduced to the anterior cervical spine. Anterior cervical plates act as buttress plates. Buttress plates are placed on the side of load application and are applied to the area of the spine requiring support. These plates function to minimize compression and shear forces and also act to minimize torque forces.

a

b

c

2.1

H-PLATE

Before plates with intrinsic stability provided by locked constructs were available, the primary method for anterior cervical stabilization was a standard H-plate ( Fig 7.2.1-1). This system had very little instrumentation and worked as a buttress plate in flexion and as a tension band in extension. The plate provided protection of neurological function by enhancing stability. It was also able to maintain the overall alignment of the spine, supply mechanical stability to allow an early return to function, and it provided security to the bone graft until

d

e

Fig 7.2.1-1a–e The former H-plate principle: bicortical fixation and excentric drilling for graft compression. a After discectomy, the AP diameter of the vertebral body is measured with a depth gauge. The bone graft is then inserted. b The drill lengt is adjusted 2 mm by 2 mm until bicortical fixation is achieved. c–e Drilling and screw insertion is eccentric in order to achieve compression of the graft.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

255

7.2.1 Modularity and evolution of instrumentation for the cervical spine

fusion had occurred. The surgical technique for these plates, however, was demanding due to the need for a bicortical screw fi xation. As a result, the certainty of the sagittal diameter of the vertebral body was essential to success. The need for posterior cortex penetration increased the risk of injury to the spinal cord. If bicortical purchase (purchase in both cortices) was not gained, the screws could back out of the construct and put the esophagus at risk of potential injury. There was no intrinsic stability in the fi xation system. However, in the mid-eighties the stainless steel screws were replaced by titanium plasma-sprayed screws which allowed a firm bone-screw interface, and purchase in the posterior cortex was no longer necessary. The fixation stability was also enhanced by excentric drilling which added a dynamiccompression-plate (DCP) concept in order to put the bone graft under axial compression ( Fig 7.2.1-1e ). The application of basic principles of osteosynthesis lead to a faster healing and incorporation of the bone interfaces [9].

2.2

The CSLP screws do not need to penetrate the posterior wall of the vertebral body. This is a constrained system. A constrained system is one that has a rigid locking mechanism between the individual components. Maximum rigidity is achieved by the segmental fi xation of each vertebra to the constrained system. Unlike its predecessor, the angle of the screw cannot be varied. This may be a disadvantage in certain situations and can lead to occasional difficulties in screw placement, especially at the two ends of the plate. However, the advantages of this design outweigh the disadvantages—the system has been used worldwide and is still being used by many surgeons.

12°

CONSTRAINED OR STATIC PLATES

Cervical spine locking plate (CSLP)

It was not until 1986 that Erwin Morscher first applied the concept of a plate with fixed-angle screws that could be securely locked to the plate [7, 10]—a concept which originally had been introduced for maxillofacial surgery. This system— the CSLP ( Fig 7.2.1-2 ) —revolutionized the technique for cervical spinal stabilization.

a

b

c

Fig 7.2.1-2a–c The CSLP (small stature and regular) and its basic principles: screws do not penetrate the posterior cortex, screws have fixed angulation and are convergent, and they are locked to the plates as a result of the expansion mechanism of the screw heads. The proximal screws are inserted 12° superiorly, the distal screws are oriented perpendicularly to the plate. For the cervicothoracic junction, the plate is turned 180° to facilitate insertion.

256

7 7.2

Spinal instrumentation Cervical spine

Subsequent to the success of the CSLP and the CSLP small stature (a smaller version for small-statured patients), the CSLP variable angle ( Fig 7.2.1-3 ) was created. This system is based on the success of the design and biomechanics of the CSLP. The CSLP variable-angle system retains the load sharing design of the plate and the intrinsic stability of a locking design, but it allows up to 20° of variability in the plate-screw angle, and up to 40° of screw conversion. The main advantage is that the screws can be inserted in an optimal direction chosen by the surgeon—for instance, with such a plate system insertion of screws in T1 or even T2 is facilitated thanks to the 20° of possible angulation.

Anterior cervical locking plate (ACLP)

The ACLP ( Fig 7.2.1-4 ) offers yet another design enhancement while capitalizing on the successful features of the original CSLP. This version incorporates a new one-step locking mechanism. This one-step mechanism obviates the need for the small 1.8 mm expansion-head screw and shortens the overall procedure by combining screw insertion steps. The threaded conical screw head allows the screw to be locked to the plate in a fi xed angle. The best indication for the ACLP are cervical fractures to prevent subsidence.

12°

20° 20°



a a

b

c

Fig 7.2.1-3a–c The CSLP variable angle allows a 20° arc of screw insertion in each direction providing greater variability of the plate-screw angle.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES



b

Fig 7.2.1-4a–b The ACLP is shown with its one-step locking mechanism. However, screw angulation is fixed as with the original CSLP. The fixed screw angulation is 12° cranial and 6° caudal. For use of the plate in the cervicothoracic junction, the plate can be turned 180°.

257

7.2.1 Modularity and evolution of instrumentation for the cervical spine

2.3

DYNAMIC AND HYBRID PLATES

Dynamic plate with translation capabilities—Vectra-T plate

Vectra plate

The newest addition to the CSLP family is the Vectra plate ( Fig 7.2.1-5 ). The Vectra plate incorporates some of the best features of all the plates—such as load sharing, low profi le, intrinsic stability, and one-step locking—yet, it adds more options. This plate system offers both locked variable-angle and fi xed-angle screws. Due to this variability, the surgeon can create a constrained system (using all fi xed-angle screws), a semiconstrained system (using all variable-angle screws), or a hybrid system (using a combination of the two screws). The plate can be used, not just as a buttress plate, but also as a system which produces a dynamic compression plating effect. Technology has come a long way since the advent of anterior plating of the cervical spine giving surgeons more choices in treating their patients and their specific pathologies. fixedangle screws variableangle screws a

b

Another advance of dynamic and hybrid plates is the Vectra-T plate. The Vectra-T plate enables axial settling along the lordotic curve (ie, translation) while preventing adjacent disc impingement due to plate migration. The translation capability ensures that the plate never extends beyond its original placement. This can significantly contribute to the prevention of adjacent level ossification development [11]. The carriage spacers are removed from the plate at the end of the construct and allow a 2–3 mm translation depending on the level (caudal or cranial) ( Fig 7.2.1-6 ). This indication is for multilevel fi xation of degenerative conditions with or without corpectomy.

8° 0° a

28° 28°

c 14°

Fig 7.2.1-5a–c The Vectra plate has a one-step, clip-based locking mechanism. The clip is built into the plate and prevents backout of the screw. Different construct options are possible with a constrained, a semiconstrained, or a hybdrid option, depending on the use of fi xed-angle or variable-angle screws. In the semiconstrained option, the screws can be inserted at a variable angle with a 28° cranial/caudal range of direction. In the axial plane, screws can be inserted with a 14° mediolateral range of direction (c).

1

b

c

Fig 7.2.1-6a–c Once the Vectra-T plate is in place the carriage spacers (1) are removed to allow a translation of 3 mm cranially or 2 mm caudally.

258

7 7.2

Spinal instrumentation Cervical spine

3

2.4

SCREWS

The initial screws ( Fig 7.2.1-7a ), or the Morscher CLSP, require a four-step procedure for screw insertion: drilling, tapping, screw insertion, and locking of the screw. With the current Vectra system the screws are either fi xedangle screws or variable-angle screws. Each screw is either self-tapping with a blunt tip ( Fig 7.2.1-7b ), or self-drilling and self-tapping with a sharp tip ( Fig 7.2.1-7c ). Different sizes of screws can be used, either 4.0 mm or 4.5 mm in diameter. The screws are color-coded: the variable-angle screws ( Fig 7.2.1-7d ) are purple (4.0 mm) and blue (4.5 mm), and the fi xed-angle screws ( Fig 7.2.1-7e ) are brown (4.0 mm) and green (4.5 mm). The 4.5 mm screws are used as rescue screws when the initial 4.0 mm screw does not provide good enough purchase or when a new trajectory is necessary.

a

b

c

d

The original description of posterior cervical fixation with spinous process wiring was done by Hadra in 1891 [12]. This system was modified with figure-of-eight wiring in 1942 by Rogers [13], and then by Bohlman with the triple wiring technique [14]. In the 1980s, Roy-Camille introduced the lateral mass screw fixation technique [15]. Then Magerl modified the drilling trajectory of the lateral masses of the cervical spine with a divergent orientation [16]. More recently, Abumi reported on the use of a pedicle screw fi xation of the lower cervical spine [17]. The goal of posterior cervical fi xation is to act as a tension band. In order to use an implant as a posterior tension band, intact compressive load-bearing ability is required. The purpose of the tension band is to resist tensile forces and bending movements. The dynamic compression through the weightbearing column is the mechanism that encourages fusion.

e

Fig 7.2.1-7a–e a Standard screw tip (CSLP) requiring predrilling and tapping. b Self-tapping screw tip: these screws only require predrilling before insertion, their tip is blunt, and they can be used to maximize length and orientation. c Self-drilling and self-tapping screw tip: these screws have a one-step mechanism for insertion. Because of their sharp tip, they have disadvantages; the screw needs to be shorter than the self-tapping screw and controlling their direction may not be as accurate as the self-tapping screws. d Head of variable-angle screw. e Head of fixed-angle screw.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

POSTERIOR CERVICAL FIXATION

Fig 7.2.1-8 Interspinous wire application.

259

7.2.1 Modularity and evolution of instrumentation for the cervical spine

3.1

WIRES AND CABLES

Wires

Early tension bands were made of wire. They were used through the spinous processes (interspinous wiring) or passed sublaminarly. The interspinous wiring was the least dangerous method, but was prone to wire breakage and wire cutout through the spinous process ( Fig 7.2.1-8 ). This wiring technique was also poor with regard to stabilizing rotation and was not able to maintain lordosis. Sublaminar wires have a higher biomechanical stability, but are difficult to shape and pass under the lamina, and, therefore, put the spinal cord at risk for a potential injury ( Fig 7.2.1-9 ).

Cable system

In order to increase the safety of the existing wiring techniques and to facilitate their use, a cable system has been introduced ( Fig 7.2.1-10 ). The cable is designed for increased flexibility and control. The current design incorporates a retrieval loop at the end of the cable, which facilitates the passing of the cable under the lamina (chapter 7.2.2 Upper cervical spine). These sublaminar techniques, however, when compared to screw-rod or screw-plate systems, provide less torsional control and they poorly resist anterior translation. In many cases, when using cable techniques, external cervical immobilization is recommended until fusion occurs. Most surgeons use complementary cable fi xation as an additional measure.

b

a

Fig 7.2.1-9 Sublaminar wire for a posterior C1/2 fixation may put the spinal cord at risk during the passage of the wires.

c

Fig 7.2.1-10a–c a Magerl C1/2 fixation supplemented with titanium cables according to Gallie [18]. b Retrieval loop of double-lead cable. c Retrieval loop of single-lead cable.

260

7 7.2

Spinal instrumentation Cervical spine

3.2

HOOK PLATES

3.3

In order to further enhance fi xation to the cervical spine, hook plates were once a popular choice of fi xation [16]. A hook plate is a prestressed construct, which is inherently stable. Any resistance to compression is due to the intervertebral joints and the interspinous H-graft. These form a triangle in the horizontal plane. The compression force is provided when the screws are tightened ( Fig 7.2.1-11). The advantages of these hook plates are their primary stability, which is superior to that of wiring, and their reduced risk of injury to the spinal cord in comparison to sublaminar wiring. The overwhelming disadvantage, however, is that the procedure is technically more demanding than wiring. It requires exact contouring of the plate component to perfectly fit the anatomy of the posterior lamina. The incorrect placement of the hook could result in stenosis, which precludes its use with certain preexisting conditions ( Fig 7.2.1-11).

Using screws with plates or rods in posterior cervical fi xation allows for a segmental fixation with increased rigidity in torsion, lateral bending, flexion, and extension when compared to wire or cable techniques. With these systems, fi xation to the pedicles or lateral masses is possible. The early posterior plates were not initially designed for the spine, but were used in other orthopedic procedures. The titanium notched plate and the one-third tubular plate are examples of these early plates ( Fig 7.2.1-12 ). Their advantages were that they were biomechanically superior to wiring and could be used in the presence of a fracture of the lamina and spinous process, or for laminectomies. The disadvantages of such plates were that they required a very demanding technique because ideal screw placement was compromised by hole spacing, which increased the potential for neurovascular damage.

a a

b

Fig 7.2.1-11a–b Diagram representation of hook-screw fixation for the cervical spine according to Magerl plus application of an H-graft.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

PLATE SYSTEMS

b

Fig 7.2.1-12a–b Principles of plate fixation with titanium cervical plates. Screws are inserted in the lateral masses of the cervical spine in an outward direction. Bone graft is applied in the decorticated laminae.

261

7.2.1 Modularity and evolution of instrumentation for the cervical spine

3.4

ROD-SCREW SYSTEMS

The latest generation of posterior cervical implants focuses on stability and ever increasing ease of use and variability of screw placement. Cervifix and Starlock systems

Both systems are modular and offer a choice of clamps and hooks fi xed on a 3.5 mm rod by means of a set screw. Screws can be optimally positioned through the clamps in any desired motion segment and in any desired direction( Fig 7.2.1-13, Fig 7.2.1-14 ).

same features and advantages of the Cervifix system and is compatible with Cervifi x implants, but it allows the surgeon to load the clamps onto the construct sequentially. The Starlock implants create a rigid, locked variable axis construct. The clamps may be placed after screw insertion, so that screw positioning is optimized and accommodated. In turn, the rod may be attached to the clamps after clamp placement to allow options at every level of the construct. The other advantage of the Starlock system is that removal of one screw is possible without taking out the whole rod construct. Top-loading polyaxial screw system or Axon

The rods of the Cervifi x system may also be connected to other rod systems (ie, USS) to continue caudally, or they may be used with a rod that has one end shaped like a reconstruction plate for attachment to the occiput. The disadvantage of this system is that the clamps for all levels must be preassembled to the rod before the construct is put into the wound. This does not allow for many intraoperative options. It is for this reason that the Starlock system was designed. It shares many of the

Fig 7.2.1-13 The Cervifix clamp needs to be mounted onto the rod before screw insertion.

Fig 7.2.1-14 Starlock clamp—the screws are first inserted into the bone, then the clamps are placed onto the screws and the rod inserted into the clamp.

To further ease implantation, the Axon system ( Fig 7.2.1-15 ) was created. The system is interchangeable with the Cervifi x and Starlock components, yet offers top-loading variable axis screws. The top-loading ability provides even greater flexibility when needed to accommodate variations according to a patient’s anatomy. These screws offer up to a 30º angulation in all directions so that the local anatomy is able to dictate screw placement, not just the limitation of the system.

a

30°

30°

b

30°

30°

Fig 7.2.1-15a–b Polyaxial cancellous bone and cortex screws allow an arc of 60° in total.

262

7 7.2

Spinal instrumentation Cervical spine

4

Modularity of current posterior fixation systems

The 3.5 mm rod can be used with the Cervifi x, Starlock, or Axon system and all these systems are compatible with each other. Extensive modularity is best shown in the following pictures ( Fig 7.2.1-16 ).

CONCLUSION

The evolution of cervical systems has been geared to simplifying techniques in order to decrease their potential for neurovascular injury and to offer options to the surgeons so that the patient’s anatomy and pathology dictate the implant of choice and the screw trajectory. The anterior plate systems offer a versatile choice of dynamic, constrained, or hybrid constructs. Despite all of these technological advances, surgeons should not forget that the most critical part of the surgery consists of an adequate decompression, end-plate preparation, and bone grafting. The choice of the type of plate system is often of secondary importance except in rare situations, such as, multilevel corpectomies. The posterior systems have evolved in line with a perfect modularity, that allow the surgeon to operate from the occiput to the upper thoracic spine with the same set of instrumentation.

a

b

Fig 7.2.1-16a–b Examples of the modularity of the Axon system with two different means of fixation to the occiput. a The use of the Cervifix or Starlock clamps in order to get a midline purchase in the occiput. b The use of a rod-plate system. Lateral offset connectors, hooks, 3.5 mm/6.0 mm domino connectors, and tapered rods to cross the cervicothoracic junction are used in this extensive instrumentation.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

7.2.1 Modularity and evolution of instrumentation for the cervical spine

5

1.

2.

3.

4.

5.

BIBLIOGRAPHY

Cloward RB (1958) The anterior approach for removal of ruptured cervical discs. J Neurosurg; 15(6):602–617. Smith GW, Robinson RA (1958) 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. Bailey RW, Badgley CE (1960) Stabilization of the cervical spine by anterior fusion. J Bone Joint Surg Am; 42-A:565–594. Orozco-Delclos R, Llovet TR (1986) Osteosintesis en las lesions traumaticas y degeneratives de la columna vertebral. Revista Traumatol Cirurg; 57:702–707. Senegas J, Gauzere JM (1976) [In defense of anterior surgery in the treatment of serious injuries to the last 5 cervical vertebrae.]

Rev Chir Orthop Reparatrice Appar Mot; 62(2 Suppl):123–128. Caspar W, Barbier DD, Klara PM (1989) Anterior cervical fusion and Caspar plate stabilization for cervical trauma. Neurosurgery; 25(4):491–502. 7. Morscher E, Sutter F, Jenny H, et al (1986) [Anterior plating of the cervical spine with the hollow screw-plate system of titanium.] Chirurg; 57(11):702–707. 8. Brodke DS, Gollogly S, Alexander Mohr R, et al (2001) Dynamic cervical plates: biomechanical evaluation of load sharing and stiffness. Spine; 26(12):1324–1329. 9. Aebi M, Zuber K, Marchesi D (1991) Treatment of cervical spine injuries with anterior plating. Indications, techniques, and results. Spine; 16(3):38–45. 10. Morscher E, Moulin P, Stoll T (1992) [New aspects in anterior plate osteosynthesis of injuries of the cervical spine.] Chirurg; 63(11):875–883. 11. Park JB, Cho YS, Riew KD (2005) Development of adjacent level ossifi cation in patients with an anterior cervical plate. J Bone Joint Surg Am; 87(3):558–563. 6.

12. Hadra BE (1891) Wiring of the vertebrae as a means of immobilization in fracture and Potts disease. Med Times and Register; 22(May 23). 13. Rogers WA (1942) Treatment of fracture dislocations of the cervical spine. J Bone Joint Surg; 4:245. 14. Bohlman HH (1985) Surgical management of cervical spine fractures and dislocations. Instr Course Lect; 34:163–187. 15. Roy-Camille R, Saillant G, Mazel C (1989) Internal fi xation of the unstable cervical spine by a posterior osteosynthesis with plate and screw. Sherk HM (ed), The Cervical Spine. Philadelphia, PA: Lippincott, 390–403. 16. Grob D, Magerl F (1987) [Dorsal spondylodesis of the cervical spine using a hooked plate.] Orthopade; 16(1):55–61. 17. Abumi K, Kaneda K (1997) Pedicle screw fi xation for nontraumatic lesions of the cervical spine. Spine; 22(16):1853–1863. 18. Gallie WE (1939) Fracture and dislocations of the cervical spine. Am J Surg; 46:495–499.

263

264

7 SPINAL INSTRUMENTATION 7.2 CERVICAL SPINE 7.2.2 UPPER CERVICAL SPINE

1

Introduction ………………………………………………………………………………………………… 265

2 2.1 2.2 2.3

Anterior screw fixation of odontoid (dens) fractures ……………………………………………… Standard lag screw technique …………………………………………………………………………… Cannulated screw technique …………………………………………………………………………… Single screw odontoid fixation …………………………………………………………………………

3

Anterior C1/2 screw fixation …………………………………………………………………………… 271

4 4.1 4.2

Posterior wiring technique ……………………………………………………………………………… 273 Standard technique (according to Gallie) …………………………………………………………… 273 Posterior wiring technique using titanium cable (modified Brooks and Jenkins technique) … 275

5 5.1 5.2 5.3 5.4

Transarticular C1/2 screw fixation ……………………………………………………………………… Standard technique ……………………………………………………………………………………… Cannulated screw technique …………………………………………………………………………… Posterior C2 pedicle screw fixation—hangman’s fracture ………………………………………… Posterior C1/2 screw fixation with C2 pedicle screws and C1 lateral mass screws ……………

6

Bibliography ………………………………………………………………………………………………… 287

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

265 267 269 270

278 279 280 283 284

265

Vincent Arlet, Jason C Datta

7 7.2

SPINAL INSTRUMENTATION CERVICAL SPINE

7.2.2

1

UPPER CERVICAL SPINE

INTRODUCTION

For a long time stabilization of the upper cervical spine relied on posterior cerclage wiring using either the Gallie or the Brooks and Jenkins technique [1, 2]. Improved understanding of the anatomy and biomechanics of the C1/2 complex now allows for a much better fi xation. The anterior screw fi xation of the odontoid and the anterior C1/2 fixation are good examples of this improved fixation. Posteriorly, the transarticular C1/2 fixation according to Magerl and the C1/2 pedicle screw fi xation according to Harms have rendered the need for postoperative rigid immobilization in a minerva cast or brace unnecessary in most cases. All the techniques described require a careful review of the local anatomy. The MRI shows the trajectory of the vertebral artery and the neural elements. The CT scan with different reconstructions makes the bony anatomy and screw trajectory visible.

2

ANTERIOR SCREW FIXATION OF ODONTOID (DENS) FRACTURES

Principle

Anterior screw fi xation [3] provides a direct fi xation of a type II odontoid fracture by using the axial compression lag screw principle and avoiding a fusion of C1/2 ( Fig 7.2.2-1).

a

5° 5°

b

Fig 7.2.2-1a–b Osteosynthesis of an odontoid fracture using compression screws.

266

7 7.2

Spinal instrumentation Cervical spine

Indication

The ideal indication for an anterior screw fi xation is a transverse fracture of the neck of the odontoid process—type II and certain type III injuries (shallow type) according to the Anderson and D’Alonzo classification [4].

Fig 7.2.2-2 Anterior screw fixation is contraindicated in an oblique flexion fracture of the odontoid.

Contraindications

• Nonunion of the odontoid. • Type II fracture with osteoporotic bone in the elderly. • Type I and III odontoid fractures. Advantages

• This procedure preserves the C1/2 motion segment. • Postoperative care and immobilization are simple. • Anterior approach is less traumatic than posterior surgery. Disadvantages

• Anterior screw fi xation should not be used in oblique flexion fractures of the odontoid’s neck ( Fig 7.2.2-2 ). • Technically difficult or impossible in short-necked patients, in patients with limited motion of the cervical spine, and in patients with a pronounced kyphosis of the upper thoracic spine or a barrel chest. • Requires high-resolution two-plane imaging ( Fig 7.2.2-3 ). • Spinal stenosis is a contraindication because of the danger of cord injury associated with hyperextension of the neck. • Postoperative dysphagia especially in the elderly.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.2.2-3 Two image intensifiers are necessary to identify the odontoid process in the AP and lateral projections.

267

7.2.2 Upper cervical spine

Surgical approach

The patient is in a supine position. Two image intensifiers are necessary to identify the odontoid process in the AP and lateral projections ( Fig 7.2.2-3 ). Without this help, the technique cannot be carried out. The head is placed in the extended position to reduce the fracture and to facilitate the insertion of the screws. The use of a Mayfield headrest makes reduction of the fracture and biplane imaging of the fracture easier than the Gardner-Wells tongs and horseshoe headrest. In the case of a persistent anterior displacement (despite extension of the neck) further reduction of the odontoid can be achieved by pushing directly on the anteriorly displaced C1/2 segment through the mouth with the index fi nger. An anteromedial approach is used. The surgical incision should not be started until the fracture is reduced. The placement of the incision is determined by placing a long K-wire along the side of the neck in the intended direction of the screw and viewing on the image intensifier. The transverse incision can then be made in the neck where the K-wire is likely to exit the skin (in most cases at the C4/5 level).

2.1

STANDARD LAG SCREW TECHNIQUE

A long 2.5 mm drill bit is inserted into the anterior inferior edge of the body of C2. In the sagittal plane, the drill should be angled slightly posteriorly in order to exit at the posterior half of the odontoid’s tip ( Fig 7.2.2-4b–c ). Furthermore, in the frontal plane, the drill should be angled a few degrees toward the midline. A second drill is inserted in the same manner ( Fig 7.2.2-4b ). One drill bit is removed and the entire hole in the distal fragment is overdrilled with a 3.5 mm drill bit ( Fig 7.2.24d ). The depth of the hole to the tip of the odontoid is measured, tapped, and a 3.5 mm cortex screw of the appropriate length inserted ( Fig 7.2.2-4e ). The second screw is inserted applying the same technique. It is absolutely essential that tissue protectors are used when drilling and tapping to avoid damaging vital structures. The oscillating attachment should be used to avoid soft-tissue damage. In patients of small stature, 2.7 mm screws are used with the appropriate drill bit and tap.

The vertebral column is exposed anteriorly by blunt dissection and then exposed cranially until the inferior edge of the body of the second cervical vertebra is identified. Two Hohmann retractors or a specially curved radiolucent retractor are then inserted on either side of the odontoid (dens) to expose the body of the axis ( Fig 7.2.2-4a ).

a

Fig 7.2.2-4a–e a On either side of the odontoid, two Hohmann retractors are inserted in order to expose the body of the axis; a special large radiolucent carbon fiber retractor can also be used (as shown in figures b–e). The trajectory of the screw, as seen in the lateral image intensifi er view, with a K-wire placed on the side of the neck is represented by the dotted line.

268

7 7.2

b

Spinal instrumentation Cervical spine

c

Fig 7.2.2-4a–e b In the frontal plane the drill bit should be angled a few degrees toward the midline. A second drill bit is inserted in the same manner. c A long 2.5 mm drill bit is inserted into the anteroinferior edge of the C2 body. In the sagittal plane, the drill should be angled slightly posteriorly in order to exit at the posterior half of the tip of the odontoid.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

d

e

d One drill bit is removed and the entire hole in the distal fragment is overdrilled using a 3.5 mm drill bit. e After the first 3.5 mm cortex screw is inserted, the second 3.5 mm cortex screw is applied in the same way.

269

7.2.2 Upper cervical spine

2.2

CANNULATED SCREW TECHNIQUE

To facilitate the placement of screws and to decrease the risk of neurovascular damage, a cannulated screw technique has been developed, which is based on the odontoid cannulated screw set. Using lateral image intensifier control, a 20 cm long, 1.2 mm K-wire is inserted in a sagittal direction on both sides. The K-wire should be angled slightly posteriorly in order to exit at the posterior half of the tip of the odontoid. Furthermore, in the frontal plane, the K-wire should be angled a few degrees toward the midline ( Fig 7.2.2-5a–b ). The length of the

a

e

b

K-wire in the bone is measured with the special ruler ( Fig. 7.2.2-5c ), indicating the length of screw required. In order to allow the self-drilling screw to start entering the bone in the near cortex, the cortex is perforated with the special cannulated countersink ( Fig 7.2.2-5d ). The odontoid-type cannulated screw with the appropriate length is inserted using the special cannulated screwdriver ( Fig 7.2.2-5e ). During insertion of the cannulated screw, it is essential to observe this procedure on the lateral image intensifier to ensure that the K-wire does not advance superiorly.

c

d

Fig 7.2.2-5a–e a–b Insertion of a 1.2 mm K-wire with a special drill sleeve on either side of the odontoid. c The length of the screws to be inserted is measured using a specially designed ruler. d The near cortex is opened with the cannulated countersink. e The cannulated self-drilling odontoid-type screws are inserted with the special cannulated, angled screwdriver.

270

7 7.2

Spinal instrumentation Cervical spine

Postoperative care after application of standard lag screw and cannulated screw technique

2.3

Patients are immobilized in a fi rm collar for a period of 10–12 weeks, but are allowed to remove the collar for daily care. After 6 weeks, the collar can be discarded while resting.

Single screw odontoid fixation is also possible, depending on the surgeon’s preference. Biomechanically, there does not seem to be a major difference between using one or two screws, yet the two-screw construct controls rotational forces better during screw insertion [5]. For the single screw fi xation one 4.0 mm or 4.5 mm screw can be inserted in a strict sagittal position. Review of the CT scan and measurement of the odontoid diameter is required before choosing between one or two screws for the fi xation.

Case example

Treatment of an odontoid fracture type II applying the cannulated screw technique is well demonstrated in Fig 7.2.2-6 . In the case of osteoporosis, it is necessary to remove the anterior part of the C2/3 disc to gain a slightly posterior entry point and in order to avoid a cutout of the screw.

a

b

Fig 7.2.2-6a–c Type II odontoid fracture treated with two 3.5 mm cannulated screws.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c

SINGLE SCREW ODONTOID FIX ATION

271

7.2.2 Upper cervical spine

3

ANTERIOR C1/2 SCREW FIXATION

Principle

Surgical technique

Anterior C1/2 screw fi xation [6] provides anterior stability and is as stable as a posterior transarticular fixation in all clinically significant planes of motion [7].

The initial part of the approach and technique ( Fig 7.2.2-7a ) is identical to the odontoid screw fi xation. In the case of an odontoid fracture, stabilization with only one odontoid screw is necessary. The insertion technique of the anterior C1/2 screw fi xation is described in Fig 7.2.2-7b–f.

Indications

• Odontoid fracture type II in elderly who cannot be operated on in a prone position. • Failed anterior odontoid screw fi xation. • Instability of C1/2. • Unstable Jefferson fracture. Advantages

Postoperative care

Postoperative immobilization in a soft collar is sufficient. Case example

The anterior screw fixation of C1/2 of an unstable odontoid fracture type II in an elderly patient is shown in Fig 7.2.2-8 .

• This technique does not require prone positioning. • The same approach as for odontoid screw fi xation is used. Disadvantage

The setup for anterior C1/2 screw fi xation requires two C-arms.

a

Fig 7.2.2-7a–f a A special large radiolucent carbon fiber Hohmann retractor is used to retract the anterior C1/2 joint.

272

7 7.2

Spinal instrumentation Cervical spine

25°

5 mm

b

c

d

Fig 7.2.2-7a–f b The anterior portion of both atlantoaxial joints is then denuded of cartilage using a small angled curette under direct visualization and fluoroscopic guidance. c–d The landmarks for anterior C1/2 fixation are identified. The starting point for the C1/2 stabilization screws lies on the undersurface of the overhanging lip of the lateral mass of C2, 5 mm lateral to the base of the odontoid process or 8–10 mm from the midline. Under AP and lateral fluoroscopic control, the guide wire is advanced in a 25º lateral direction and engaged in the lateral mass of C1.

a

b

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

e

f

e–f The length of the K-wires are then measured, the screw holes are overdrilled, and 3.5 or 4.0 mm cannulated self-tapping titanium screws are inserted. Due to the saucer-like shape of the C1 lateral mass, the image intensifier picture overestimates the length of the transarticular screw. Therefore, it is useful to have a preoperative measurement on the CT scan of the appropriate length of the screws to be selected. The anterior transarticular C1/2 screws are 1.5–2.0 cm long.

c

Fig 7.2.2-8a–c a Preoperative CT reconstruction of an unstable odontoid fracture in an elderly patient. b–c Postoperative AP and lateral x-rays.

273

7.2.2 Upper cervical spine

4

POSTERIOR WIRING TECHNIQUE

4.1

STANDARD TECHNIQUE (ACCORDING TO GALLIE)

Principle

The standard posterior wiring technique according to Gallie [1] provides a stable posterior construct for C1/2 instability and is particularly resistant to flexion forces. Indications

• Fracture of the odontoid (dens) with anterior displacement. • Rupture of the transverse ligament of C1.

Surgical technique

The patient is placed in a prone position with Gardner-Wells traction or with a Mayfield clamp installed. Lateral image intensification is used to check the position of the head and reduction of the fracture. A midline incision extends from the occiput to C3. The soft tissues are cleared from the occiput, C1, and C2. Lateral dissection beyond a maximum of 2 cm from the midline, particularly at C1, is to be avoided in order to prevent an injury to the vertebral artery and venous plexus. The soft tissues are cleared circumferentially around the posterior arch of C1 in the midline to allow easy passage of the wire.

Advantages

• This technique is relatively easy to perform. • The graft is fi rmly fi xed between the two arches of C1 and C2. • Prevents flexion and extension. • Posterior wiring can be combined with a C1/2 transarticular screw fi xation or a C1/2 pedicle screw fi xation. Disadvantages

• It is a sublaminar wiring technique. • Does not control translation, lateral bending, and rotation. • Posterior wiring cannot be used with associated fractures of the C1 arch or spina bifida of C1. • Not suitable for the treatment of a posterior displacement of the odontoid. • Requires postoperative immobilization. • High rate of pseudarthrosis in C1/2 instability due to rheumatoid arthritis.

A 1.2 mm wire is fashioned into a loop with a hook configuration. The wire is passed from the inferior aspect of C1 cranially and looped over the superior aspect of C1 ( Fig 7.2.2-9a ). The loop is carefully pulled backward until it is distal enough to be looped over the spinous process of C2 ( Fig 7.2.2-9b ). Unless the vertebrae are sufficiently reduced prior to insertion of the wire, there is a risk of damage to the spinal cord during the procedure. The arch of C1 and the lamina of C2 are decorticated using a high-speed burr.

274

7 7.2

Spinal instrumentation Cervical spine

A rectangular corticocancellous bone graft measuring 3–4 cm is removed from the posterior iliac crest. The graft is fashioned into an H configuration and laterally notched ( Fig 7.2.2-9c ) to snugly fit around the spinous process and over the lamina of C2. The cancellous surface of the graft is shaped to conform with the slope of the arch of C1 and the lamina of C2 in order to provide maximum contact. The two free ends of the wire, which are laterally placed, are then brought across the midline after the graft has been applied to the posterior surfaces of C1 and C2 ( Fig 7.2.2-9d ). The notches provide better fi xation of

a

b

c

Fig 7.2.2-9a–e a The 1.2 mm wire is formed into a loop. It is passed from the inferior aspect of C1 in an anterior and cranial direction and looped over the superior surface of C1. b The loop is carefully pulled back distally to be looped over the spinous process of C2.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

the bone graft when the wires are tightened. During tightening of the wires the remaining reduction is achieved. Fragments of cancellous bone can then be packed around the bone graft between C1 and C2 ( Fig 7.2.2-9e ). Postoperative care

A fi rm collar preventing extension of the neck for a period of 6–10 weeks is recommended. It may be removed for daily care and, after 6 weeks, while resting.

d

e

c H-shaped and laterally notched corticocancellous bone graft of about 3–4 cm. d After placement of the bone graft to the posterior surface of C1 and C2, the ends of the wire are brought across the midline. e Fragments of cancellous bone are packed around the applied bone graft between C1 and C2.

275

7.2.2 Upper cervical spine

4.2

POSTERIOR WIRING TECHNIQUE USING TITANIUM CABLE (MODIFIED BROOKS AND JENKINS TECHNIQUE)

The wedge compression technique of Brooks and Jenkins has been applied for a long time using stainless steel wire. Such a fi xation can nowadays be done with 1.0 mm titanium cables, which have the advantage of being MRI compatible. Principle

The construct of posterior wiring with cable is similar to applying the standard wiring technique according to Gallie, but it provides more rotational and tensile strength. Indications

The indications are the same as those for the standard technique according to Gallie. Advantages

• The cable wiring is biomechanically superior to the Gallie technique. • The retrieval loop at the end of the cable facilitates a sublaminar passing of the cable. • MRI compatibility.

Surgical technique

The surgical approach is the same as for the standard technique. In addition, the soft tissues anterior to the arch of C2 must be cleared, leaving the atlantoaxial membrane intact. Applying this technique requires passing the leader of a double-lead cable under the laminae of C1 and C2 ( Fig 7.2.210a ) or, alternatively, the leader of two single-lead cables ( Fig 7.2.2-10b ). Following is a description of the use of the doublelead cable. The cable retriever catches the loop of the lead cable. The double-lead cable is cut and separated into a right and left cable. Then, the loop is gently pulled until the leader is completely exposed ( Fig 7.2.2-10c–d ). The same maneuver is repeated on the opposite side so that the two cables (one for each side) are passed under the lamina. One or two corticocancellous bone grafts measuring 1.5 × 3.5 cm (two grafts) or 1.5 × 4 cm (one graft) are taken from the posterior iliac crest. They are fashioned into a wedge with the cortical portion on the posterior side. The wedges of bone are placed between the two vertebral arches ( Fig 7.2.2-10e ) after the undersurface of C1 and the superior aspect of C2 have been decorticated with a high-speed burr. The leader of the cable is then cut and threaded through the crimp sleeve ( Fig 7.2.2-10f ). The slack in the cable is pulled up until the cable fits snugly over the lamina.

Disadvantages

• The posterior cable wiring technique cannot be used for the stabilization of associated fractures of the C1 arch. • It requires the passage of sublaminar wires at two levels and is therefore more dangerous for the spinal cord than the standard Gallie technique. • Tensioning of the cable requires careful attention to detail for a balanced tension on either side of the cable.

a

b

Fig 7.2.2-10a–j a Double-lead cable with its retrieval loop. b Single-lead cable with its retrieval loop.

276

7 7.2

Spinal instrumentation Cervical spine

The cable is passed in the pretensioner for provisional tightening ( Fig 7.2.2-10g ). The cable is automatically locked by depressing the button of the pretensioner. Each cable (right and left) are sequentially tightened by pulling gently for provisional tightening. The tensioner/crimper is then adjusted over the crimp sleeve ( Fig 7.2.2-10h ). A gentle squeeze of the handle is applied until a single audible click is heard. The torque-limiting handle is attached to the tensioner/crimper. Tension is applied by rotating the torque-limiting handle in a clockwise direction until it slips at the desired tension ( Fig

c

d

Fig 7.2.2-10a–j c Retrieval of the cable leader after it was passed under the lamina of C1 and C2. d Cutting of the cable leader with the cable cutter.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

7.2.2-10i ). Squeezing the handles of the tensioner/crimper

will defi nitively crimp the cable. Remove the tensioner and pretensioner and cut the cable. Repeat the same maneuver on the opposite side ( Fig 7.2.2-10j). Postoperative care

Postoperative care is the same as described for the standard Gallie technique: A rigid collar preventing extension of the neck for a period of 6–10 weeks is recommended. It may be removed for daily care and, after 6 weeks, while resting.

e

f

e Two wedges of bone (each 1.5 × 3.5 cm) are inserted on either side between the posterior arches of C1 and C2 f Passage of the cable in the crimp sleeve and adjustment of the cable over the lamina on both sides.

277

7.2.2 Upper cervical spine

g

h

i

Fig 7.2.2-10a–j g Attaching of the pretensioner on the cable. Note proper alignment of the pretensioner’s button. Pressing the button locks the cable. h Adjustment of the pretensioner/crimper over the crimp sleeve and final tightening of the cable with the torquelimiting handle that was preset at the desired tension. i Crimping of the cable by squeezing the handles of the tensioner/crimper. j Final representation of the Brooks and Jenkins fixation with titanium cable.

j

278

7 7.2

Spinal instrumentation Cervical spine

5

TRANSARTICULAR C1/2 SCREW FIXATION

Principle

Surgical approach

The original transarticular screw fi xation decribed by Magerl allows a rigid stabilization of the C1/2 segment that is stable in all directions [8].

The patient lies in the prone position and the reduction of C1/2 is checked using lateral image intensifier control. The reduction is facilitated by traction either with Gardner-Wells tongs/halo or a Mayfield clamp, the latter of which controls translation more easily. The neck is flexed as much as possible to facilitate insertion of the screws, and the image intensifier is used to exclude redislocation.

Indications

• Acute and chronic atlantoaxial instability. • Unstable Jefferson fracture. Advantages

• The transarticular C1/2 screw fi xation is biomechanically superior to the wiring technique. • Maintenance of reduction is possible. • Integrity of the posterior arch of C1 is not necessary. • Can easily be combined with a fusion according to Gallie. Disadvantages

• This technique is demanding. • There is a potential risk for bleeding related to the injury of the venous plexus located laterally. • Difficulties in patients with upper thoracic kyphosis. • Requires perfect reduction of C1/2 before screw insertion. • Erosion of the transverse foramen or the aberrant trajectory of the vertebral artery may render it impossible. • If bilateral transarticular screw fi xation is not possible, then unilateral transarticular screw fi xation and posterior C1/2 wiring is a viable option [9].

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

A midline incision is performed from the occiput to the tip of the spinous process of C3. The arch of C1, the spinous processes, lamina, and inferior articular processes of C2 are exposed subperiosteally. Persistent anterior dislocation of C1 or C2 may be reduced by pushing on the spinous process of C2 and/or by pulling gently on the posterior arch of C1, either with a towel clamp or with a sublaminar wire. Persistent posterior dislocation requires opposing forces. A small dissector is used to expose the cranial surface of the lamina and isthmus of C2 by careful subperiosteal dissection up to the posterior capsule of the atlantoaxial joint ( Fig 7.2.2-11a , Fig 7.2.2-12a ). Medial to the isthmus, the atlantoaxial membrane is visible. The laterally situated vertebral artery is not exposed.

279

7.2.2 Upper cervical spine

5.1

STANDARD TECHNIQUE

Using lateral image intensifier control, a long 2.5 mm drill bit is inserted in a strictly sagittal direction ( Fig 7.2.2-11b ). The oscillating attachment prevents soft tissues from being wrapped around the drill bit. The entry point of the drill bit is at the lower edge of the caudal articular process of C2 (see Fig 7.2.2-11a ). The drill bit goes through the isthmus near to its posterior and medial surface. It then enters the lateral mass of the atlas close to its posteroinferior edge. Anteriorly, the drill perforates the cortex of the lateral mass of C1. The screw length is measured and the direction of the screw canal is checked using the image intensifier ( Fig 7.2.2-11c ).

a

b

Fig 7.2.2-11a–f a Identification of the entry points. A small dissector or nerve hook is needed to expose the isthmus of C2. The entry point of the drill is the lower edge of the caudal articular process of C2, 3 mm above the C2/3 joint line. b The 2.5 mm drill bit is inserted in a strictly sagittal direction so that it passes 3 mm outside the medial border of the isthmus. It then enters the lateral mass of the atlas.

The 3.5 mm cortex screws are inserted after tapping with a 3.5 mm tap across the C1/2 joint—the anterior cortex of C1 must not be tapped ( Fig 7.2.2-11d–e ). Proper caudocephalad drilling may sometimes be difficult because the neck muscles and the upper torso prevent the correct placement of the drill bit. Gently pulling the spinous process of C2 cranially with a towel clamp facilitates drilling. It is often necessary to drill through a distal percutaneous stab wound in order to place the drill in the correct angle. Prepping and draping of the upper thoracic spine is necessary to be able to enter the drill bit through a stab incision.

c

d

c Measuring the screw length and tapping the screw hole with a 3.5 mm tap. d Insertion of two 3.5 mm cortex screws of appropriate length. A cerclage wire according to Gallie may be added to increase stability.

280

7 7.2

Spinal instrumentation Cervical spine

Drilling in a horizontal direction must be avoided because:

5.2

• At the level of C2 the vertebral artery runs upward anteriorly to the C1/2 joint and could easily be damaged. • The screw could exit C2 anteriorly and not enter the atlas.

Using lateral image intensifier control, a 1.2 mm K-wire is inserted with a surgical drill guide in a strictly sagittal direction into each hole ( Fig 7.2.2-12b ). The entry point of the K-wire is at the lower edge of the caudal articular process of C2 (see Fig 7.2.2-12a ). The length of the screw is established with a special ruler by measuring the protruding part of the guide wire ( Fig 7.2.2-12c ). Before inserting the cannulated screw over the K-wire, the entrance for the screw is prepared by a special 3.5 mm cannulated countersink ( Fig 7.2.2-12d ) in order to facilitate the starting purchase of the screw. The appropriate 3.5 mm self-drilling, cannulated cancellous bone screw (same screw as for odontoid fi xation, usually around 4.5 cm long) is inserted over the guide wire ( Fig 7.2.2-12e ). The progress of the screw must be observed under image intensifier control to ensure that the K-wire does not migrate proximally beyond the C1 arch.

Postoperative care

Patients are immobilized in a fi rm collar for a period of 6–8 weeks, but are allowed to remove the collar for daily care. After 6–8 weeks, the collar can be discarded when resting. If additional posterior wiring has been used, a soft collar can be worn instead of a fi rm collar.

CANNULATED SCREW TECHNIQUE

In severe degenerative diseases the sclerotic subcortical bone of the C2 joint may prevent insertion of the self-drilling screw. In this case a 2.7 mm cannulated drill bit is used to cross the joint and a cannulated fully threaded 3.5 mm screw can be inserted after regular tapping of the predrilled hole. e

f

Fig 7.2.2-11a–f e The inclination of the screw trajectory is assessed on lateral fluoroscopy images. f Final representation with a fusion according to Gallie.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Following bilateral screw fi xation, a posterior C1/2 fusion is performed. It is preferable to supplement the graft with a posterior cable as this increases the stability of the fi xation and hence the fusion rate ( Fig 7.2.2-12f–g ).

281

7.2.2 Upper cervical spine

a

e

b

c

f

g

Fig 7.2.2-12a–h a–g Schematic representation of the C1/2 transarticular screw technique using two cannulated screws. Care must be taken to avoid upward migration of the K-wire during drilling.

d

282

7 7.2

Spinal instrumentation Cervical spine

When there is a defect or fracture of the posterior arch of C1, a fusion of the atlantoaxial joint must be performed. For visualization of the atlantoaxial joint, K-wires are drilled into the posterior aspect of the lateral mass of the atlas. For this purpose, the greater occipital nerve is retracted cranially. The soft tissues containing the greater occipital nerve and its accompanying venous plexus can also be retracted. The atlantoaxial joints are exposed by opening the posterior capsule, thus, making the C1/2 joint visible ( Fig 7.2.2-12h ). The articular cartilage of the posterior half of the facet joint is removed with either a small chisel or a sharp curette, after which the joints are packed with cancellous bone and the screws are inserted.

Postoperative care

Patients are immobilized in a fi rm collar for a period of 6–8 weeks, but are allowed to remove the collar for daily care. After 6–8 weeks, the collar can be discarded when resting. If additional posterior wiring has been used, a soft collar can be worn instead of a fi rm collar. Case example

The transarticular screw fixation of C1/2 with Gallie-type titanium cable fi xation in a 55-year-old rheumatoid patient is shown in Fig 7.2.2-13.

h Fig 7.2.2-12a–h h In the case of a fracture or a defect of the posterior arch of C1, C1/2 fusion is achieved by removing the posterior half of the C1/2 cartilage. The greater occipital nerve (not shown) must be retracted cranially, in order to expose the C1/2 joint. With a small curette the cartilage can be removed and bone graft can be added to the C1/2 joint.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

a

b

Fig 7.2.2-13a–b a Preoperative lateral x-ray of a 55-year-old patient with C1/2 instability due to rheumatoid arthritis. b Lateral x-ray at the 4-year follow-up examination with C1/2 transarticular screw fixation and fusion according to Gallie.

283

7.2.2 Upper cervical spine

5.3

POSTERIOR C2 PEDICLE SCREW FIX ATION— HANGMAN’S FRACTURE (FRACTURE OF THE PEDICLE OF C2)

Principle

The technique originally described by Robert Judet in 1962 is based on direct screw fi xation of the pedicle of C2 in order to maintain motion.

only the exposure of the posterior arch of C2 and its isthmus is necessary. The landmarks for the C2 pedicle screw insertion must be identified ( Fig 7.2.2-14a–b ). The dissection is carried out over the superior surface of the C2 pars interarticularis, and with a small Penfield elevator or nerve hook the medial part of the C2 pars interarticularis is identified. Under C-arm control a 2.5 mm drill bit with an oscillating attachment and a protective sleeve is positioned halfway between the upper and lower articular surfaces of C2 in a vertical line bisecting

Indications

• Displaced hangman’s fracture type II (Effendi type II). • Type III fractures will require an additional fi xation of C3. Advantages

• Does not immobilize the C2/3 segment. • Does not require the use of a postoperative halo vest. Disadvantages

• Cannot be performed without prior reduction of the fracture. • The Judet technique is technically challenging. • Cannot be used if the C2/3 disc is totally disrupted. • There is a risk of injury to the vertebral artery; therefore, a review of the CT scan and the MRI is required to assess the trajectory of the artery and its patency.

Surgical technique using a 3.5 mm cortex screw

The fracture must be reduced prior to surgical stabilization. The patient is in a prone position with Gardner-Wells tongs or a Mayfield headclamp attached. The head is positioned in slight extension, and the reduction controlled with lateral fluoroscopy. The surgical approach is similar to the one used for C1/2 transarticular screw fi xation, yet more limited because

a

b

Fig 7.2.2-14a–c a Identification of the landmarks for the C2 pedicle screw. The entry point for the C2 pedicle c screw is halfway between the upper and lower articular surface of C2 at a vertical line bisecting the articular mass. b–c The screw holes are predrilled in a 25° upward orientation and a 15–25° medial direction. A 3.5 mm cortex screw is then inserted.

284

7 7.2

Spinal instrumentation Cervical spine

the C2 articular mass. The drill is then oriented 25° upward as seen on the C-arm ( Fig 7.2.2-14b ) and 15–25° medially as checked clinically by the palpation of the medial wall of the isthmus of C2 ( Fig 7.2.2-14c ). A depth gauge measures the screw length. A 3.5 mm cortex screw or a lag screw is inserted after tapping the cortex at the entry point. The usual length of the screw is 30–35 mm.

5.4

POSTERIOR C1/2 SCREW FIX ATION WITH C2 PEDICLE SCREWS AND C1 LATERAL MASS SCREWS

Principle

The technique originally described by Harms is based on a pedicle screw fi xation of C2 according to Robert Judet and a lateral mass screw fi xation of C1 using a polyaxial screw-rod system [10].

Postoperative care

For 6 weeks, the patient’s neck is immobilized in a fi rm collar that can be removed for daily care. Case example

The posterior fi xation of a displaced hangman’s fracture with a C2 pedicle screw is demonstrated in Fig 7.2.2-15.

Indications

• • • •

C1/2 instabilities after odontoid fracture. Transverse ligament rupture. Unstable Jefferson fracture. Inflammatory conditions.

Advantages

a

b

c Fig 7.2.2-15a–c a Example of a hangman’s fracture with marked displacement. b–c Lateral x-ray and CT scan after treatment with bilateral pedicle screws according to Judet.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

• In patients with a kyphotic spine, the transarticular screw fi xation according to Magerl may be difficult because of the orientation of the drill bit. • Erosion of the transverse foramen at C2 may make the Magerl screw fi xation technique dangerous. • C1/2 pedicle screws may be removed in specific indications after fracture healing to regain C1/2 range of motion. • Reduction of C1/2 is facilitated with the polyaxial C1/2 screw-rod system. Disadvantages

• Bleeding related to the injury of the venous plexus. • Injury or irritation of the C2 nerve by the screw. • Injury of the internal carotid artery in front of the C1 lateral mass. • This technique is technically difficult.

285

7.2.2 Upper cervical spine

Surgical technique using the Axon system

The surgical approach is similar to the C1/2 transarticular screw fi xation, except that the greater occipital nerve must be mobilized distally and not proximally to assess the entry point in the lateral masses of C1. Insertion of the pedicle screw in C2

The technique for C2 pedicle screw fi xation is identical to the previously described technique for a hangman’s fracture. A polyaxial regular Axon screw of appropriate length is inserted instead of a regular cortex screw (see Fig 7.2.2-14 ). Insertion of C1 lateral mass screw

The landmarks for the C1 screws are on the same vertical line as the C2 pedicle screws—just below the posterior lamina of C1 and above the C1/2 joint. The dorsal root ganglion of C2 is identified and reclined distally with a smooth elevator

a

b

c

10°

Fig 7.2.2-16a–e a Identification of the landmarks for the C1 screw insertion. The posterior C2 root ganglion is reclined distally. b–c The screw hole for the C1 polyaxial Axon screw is prepared. The drill bit is inserted parallel to the C1 lateral mass under image intensifier control.

( Fig 7.2.2-16a ). The venous plexus is coagulated with bipolar coagulation and soaked thrombin patties. The lateral mass of C1 is identified just above the C1/2 joint. The 2.5 mm drill bit with the oscillating attachment and its protective sleeve are used and the position confirmed with a control image. The drill is then oriented parallel to the C1 posterior arch toward the anterior aspect of C1, usually with a 10–20° ascending direction. In the axial plane the drill is oriented 10° toward the midline. Drilling is checked under fluoroscopy. The drilling must be bicortical and stop 3 mm before the anterior tubercle of C1 as seen in the lateral C-arm projection ( Fig 7.2.2-16b–c ). The length for the screw is measured with a small depth gauge, and the path for the C1 lateral mass screw is tapped, if necessary, and a special polyaxial 3.5 mm Axon shaft screw is inserted penetrating both cortices. The distal part of the screw is not threaded in order to avoid irritating the C2 nerve.

d

e

d–e Lateral and AP image of the final construct with the 3.5 mm rod connecting the two Axon screws (note the anterior tubercle of C1 that lies 3 mm in front of the tip of the screws).

286

7 7.2

Spinal instrumentation Cervical spine

Connection of the screws with the 3.5 mm rods

Case example

Two short 3.5 mm rods are used to connect the polyaxial screws on both sides. Further additional reduction maneuvers (translation, compression, distraction) can be carried out at this time ( Fig 7.2.2-16d–e ). For additional stability, a C1/2 cerclage wire may be added.

The C1 pedicle screw fixation is well demonstrated in this patient with a complex odontoid fracture ( Fig 7.2.2-17 ).

Postoperative care

Patients are immobilized in a fi rm collar for a period of 6–8 weeks, but are allowed to remove the collar for daily care. After 6–8 weeks, the collar can be discarded when resting. If additional posterior wiring has been used, a soft collar can be worn instead of a fi rm collar.

a Fig 7.2.2-17a–c a Preoperative lateral x-ray showing the complex odontoid fracture.

b b Postoperative lateral x-ray. The tip of the screw in C1 stops 3 mm before the anterior tubercle of C1.

c c Sagittal CT reconstruction shows that the C1 screw is inserted bicortically.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

PSpi_I_7_2_2_FS19.indd 286

19.1.2007 9:23:15 Uhr

7.2.2 Upper cervical spine

6

BIBLIOGRAPHY

Gallie WE (1939) Fracture and dislocations of the cervical spine. Am J Surg; 46:495–499. 2. Brooks AL, Jenkins EB (1978) Atlantoaxial arthrodesis by the wedge compression method. J Bone Joint Surg Am; 60(3):279–284. 3. Aebi M, Etter C, Coscia M (1989) Fracture of the odontoid process. Treatment with anterior screw fi xation. Spine; 14(10):1065–1070. 4. Anderson LC, D’Alonzo RT (1974) Fractures of the odontoid process of the axis. J Bone Joint Surg Am; 86(9):2081. 5. Sasso R, Doherty BJ, Crawford MJ, et al (1993) Biomechanics of odontoid fracture fi xation. Comparison of the one- and twoscrew technique. Spine; 18(14):1950–1953. 6. Reindl R, Sen M, Aebi M (2003) Anterior instrumentation for traumatic C1/2 instability. Spine; 28(17):E329–333. 7. Sen MK, Steffen T, Beckmann I, et al (2005) Atlantoaxial fusion using anterior transarticular screw fi xation of C1–C2: technical innovation and biomechanical study. Eur Spine J; 14(5):512–518. 8. Jeanneret B, Magerl F (1992) Primary posterior fusion C1/2 in odontoid fractures: indications, techniques, and results of transarticular screw fi xation. J Spinal Disord; 5(4):464–475. 9. Grob D, Jeanneret B, Aebi M, et al (1991) Atlanto-axial fusion with transarticular screw fi xation. J Bone Joint Surg Br; 73(6): 972–976. 10. Harms J, Melcher RP (2001) Posterior C1/2 fusion with polyaxial screw and rod fi xation. Spine; 26(22): 2467–2471. 1.

287

288

7 SPINAL INSTRUMENTATION 7.2 CERVICAL SPINE 7.2.3 MIDDLE AND LOWER CERVICAL SPINE

1

Introduction ………………………………………………………………………………………………… 289

2 2.1

Anterior cervical discectomy and fusion ……………………………………………………………… 289 Semiconstrained plate …………………………………………………………………………………… 290

3

Posterior wiring technique using cable ………………………………………………………………… 293

4

Posterior fixation with lateral mass screw—general principles …………………………………… 295

5

Posterior cervical fixation using titanium rods with polyaxial screw system …………………… 298

6

Bibliography ………………………………………………………………………………………………… 303

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

289

Jason C Datta, Michael E Janssen

7 7.2

SPINAL INSTRUMENTATION CERVICAL SPINE

7.2.3

1

MIDDLE AND LOWER CERVICAL SPINE

INTRODUCTION

The middle and lower cervical spine is by far the most commonly instrumented area in the cervical spine. Strict adherence to the basic principles of instrumentation with anterior plating and posterior lateral mass screw fi xation achieves a satisfactory outcome in the majority of cases. More challenging pathologies, such as osteoporosis, cervical myelopathy, tumors, or deformities, have led to the development of versatile and advanced instrumentation, the use of which allows most problems to be tackled successfully.

2

ANTERIOR CERVICAL DISCECTOMY AND FUSION

The concept of anterior plate fi xation with fi xed-angle screws was introduced by Erwin Morscher [1]. The cervical spine locking plate (CSLP)—the fi rst plate of this kind—is based on the AO principles and techniques for internal fi xation. The clinical performance of the CSLP has been documented in more than 20 peer-reviewed studies. The original concept of locking plate has been copied by many and several generations of the CSLP have been developed by AO. This chapter will discuss the latest CSLP version—the Vectra plate.

290

7 7.2

Spinal instrumentation Cervical spine

2.1

SEMICONSTRAINED PLATE

Principle

Anterior plate fi xation is used to increase the stability of the anterior column following grafting techniques. This semiconstrained plate—Vectra plate—functions as a tension band in extension and as a buttress plate in flexion. The Vectra plate allows for self-locking variable-angle screws or fi xedangle screws, depending on the surgical need (chapter 7.2.1 Modularity and evolution of instrumentation for the cervical spine).

Disadvantages

• Locking mechanism requires special instrumentation to remove the screws. • Learning curve due to the multiple constructs that the plate makes possible (locking, dynamization and various possible screws).

8° fixed-angle screws b

Indications

• To support the anterior column when instability persists, particularly when associated with a loss of height of the vertebral body following a severe wedge compression fracture. • Following partial or total vertebrectomy, for decompression of the spinal cord. • Anterior spinal reconstruction from C2 to T2. • Deformity correction. Advantages

• More versatility with the possibility of having fi xed-angle screws, variable-angle screws, self-tapping, or self-drilling screws ( Fig 7.2.3-1). • No need to have a bicortical fi xation because the screw heads lock into the plate. • Locking prevents the migration of the screw anteriorly. • Axial compression and dynamization are possible. • One-step locking plate. • CSLP offers a stronger fi xation than with simple nonlocking plates [2]. • Large graft visibility window.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

14° 8° 28°

c

variable-angle screws 8° offset a

28° d

Fig 7.2.3-1a–d a Vectra plate with the variable-angle and fixed-angle screws. b Possible mediolateral angulation of the variable-angle screw . c The screw is either self-tapping with a blunt tip (left) or self-drilling and self-tapping with a sharp tip (right). Different sizes of screws can be used, either 4.0 mm or 4.5 mm in diameter. The screws are color-coded: the variable-angle screws are purple (4.0 mm) and blue (4.5 mm), and the fixed-angle screws are brown (4.0 mm) and green (4.5 mm). The 4.5 mm screws are used as rescue screws when the initial 4.0 mm screw has not provided sufficient purchase or when a new trajectory is necessary. d Locking clip mechanism that prevents the screw from backing out.

291

7.2.3 Middle and lower cervical spine

Surgical technique

The patient is positioned supine on a horseshoe headrest, preferably using Gardner-Wells tongs with slight traction (5– 7.5 kg). Alternatively, a Mayfield headclamp can be used. The use of lateral fluoroscopy is recommended for the optimum placement of the screws and choice of graft height. The spine is approached anteriorly through a transverse incision at the appropriate level. Discectomy will be started with small pituitary rongeurs. A Caspar retractor is then used to distract the disc space and to facilitate the discectomy. Following approach and decompression, the graft site may be measured by the caliper or by inserting an interbody template (sizes available from 6–9 mm) ( Fig 7.2.3-2a). With lateral C-arm control the appropriate graft height can be chosen by comparing the disc space above and below and by measuring the segmental lordosis and the parallelism of the facets posteriorly. If necessary, the other disc levels can be treated according to the same principle.

After thorough preparation of the end plates, a tricortical bone graft from the iliac crest or spacers (allograft, cage) are inserted into the disc spaces that have been prepared. A tricortical bone graft from the iliac crest or a spacer (allograft or cage) is inserted into the disc space ( Fig 7.2.3-2b ). At this time, cervical traction is decreased to facilitate compression of the graft(s). A cervical plate template ( Fig 7.2.3-2c ) may be used to aid in estimating the correct plate length. Overriding of a normal disc by the plate must be avoided as it may lead to a fusion of the disc space. The plates are prelordosed, but the lordosis may be adjusted by using the bending pliers that come as part of the set ( Fig 7.2.3-2d ). When using standard self-tapping screws, the plate is positioned with the drill guide ( Fig 7.2.3-2e ) and the fi rst screw hole is drilled using a special drill that has an automatic stop, so that overdrilling is impossible ( Fig 7.2.3-2f–g ). (If desired, hold the plate in place with a temporary fi xation pin.) Color bands on the drill guides correspond to the colors of screws

d

a

b

c

Fig 7.2.3-2a–h Plate fixation for anterior cervical discectomy and fusion. a Insertion of the trial spacer. The Caspar distractor is released to check the intrinsic stability and the appropriate height is checked under lateral fl uoroscopy. b Insertion of the interbody graft. c Determining the plate length using a template. d Preshaping of the plate to the desired lordosis with the bending pliers.

292

7 7.2

Spinal instrumentation Cervical spine

Fig 7.2.3-2a–h e Positioning of the drill guide (either variable-angle or fixed-angle) into the plate. Introduce it into the plate at an angle as shown and then rotate forward until the clip of the plate is engaged. f–g Select the appropriate length drill bit with a stop-and-quick coupling handle to predrill the screw hole. Remove the drill bit and insert either a variable-angle screw or a fi xed-angle screw. Advance the head of the screw until it is seated in the plate. Do not fully tighten the screw. Insert the second screw diagonally to the first, and then insert all other screws. Tighten all the screws so that the screw heads are completely sunk into the plate. h Final construct for a two-level anterior cervical discectomy and fusion.

associated with each guide. A special compression drill guide allows an additional 0.5 mm of compression intraoperatively. When using the self-tapping self-drilling screws, an awl may be used at any trajectory to break the cortex before the screw is inserted. The awl also has an automatic stop and will center the screw within the screw hole. Once the hole is prepared, insert the screw using the cruciform screwdriver. The fi rst screw should not be fully tightened. The second screw should be placed diagonally to the fi rst and then all other screws are inserted in a similar manner. When the last screw is inserted, the screws should all be tightened so that the screw heads are completely sunk into the plate ( Fig 7.2.3-2h).

e

f

g

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

h

Video 7.2.3-1 Two-level anterior cervical discectomy and fusion using the Cervios cages and the CSLP.

293

7.2.3 Middle and lower cervical spine

3

Postoperative care

A soft collar is worn for 6 weeks, but may be removed for daily care. Case example

A 55-year-old male with severe pain and weakness of the left upper extremity (wrist extension and elbow extension). His reflexes of the lower extremities were a little brisk ( Fig 7.2.3-3a ). After a corpectomy of C6 a fibular allograft was inserted and a semiconstrained Vectra plate with variable-angle screws was applied from C5 to C7 ( Fig 7.2.3-3b–c ).

POSTERIOR WIRING TECHNIQUE USING CABLE

Titanium cables have advantageously replaced stainless steel wires in the cervical spine. There are many wiring techniques for posterior fi xation of the cervical spine (chapter 7.2.2 Upper cervical spine). The most simple and least dangerous is the interspinous wiring technique. Principle

The posterior wiring technique applies the tension band principle. Indications

• Injuries of the posterior complex involving predominantly soft tissue with insignificant damage to the vertebral body. • Enhancement of other posterior fusion techniques. Advantages

a

b

c

Fig 7.2.3-3a–b A case example of an anterior cervical discectomy and fusion with the Vectra plate in a patient with cervical myelopathy.

• The posterior wiring technique is easy to perform and safe. • Easy complementary fi xation (anterior or posterior). • Large surface area for fusion. • Short segment stabilization. • Cables are MRI compatible as opposed to stainless steel wires. • The current design incorporates a retrieval loop at the end of the cable, which facilitates the passing of the cable. A double-lead cable allows for the passage of two cables simultaneously. Disadvantages

• Cannot be used in fractures of the vertebral arch, including the spinous processes. • Offers poor biomechanical fi xation—especially in rotation. • Failure to maintain lordosis. • Cannot be used as the sole form of fi xation.

294

7 7.2

Spinal instrumentation Cervical spine

Surgical technique

Postoperative care

The patient is placed prone and a midline posterior approach is used. It is essential to identify radiographically the levels that are to be fused. A hole is drilled on each side of the base of the spinous process of the upper vertebra of the injured segment ( Fig 7.2.3-4a ). The entry point corresponds to the junction of the base of the spinous process and the lamina. A towel clip is placed in the holes, and with gentle rocking movements the holes are connected ( Fig 7.2.3-4b ). The leader of a 1.0 mm titanium cable is passed through the hole and then pushed under the base of the inferior spinous process, leaving the interspinous soft tissues intact ( Fig 7.2.3-4c ). The leader of the wire is passed in the crimping sleeve and tightened with a special tensioner. The laminae are decorticated with a high-speed burr and a cancellous bone graft is applied ( Fig 7.2.3-4d ).

Postoperative immobilization depends on the associated fi xation as the posterior wiring technique cannot be used as a stand-alone fi xation.

a

b

c

d

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.2.3-4a–d Interspinous wiring with cable. a Using a drill bit, a hole is made on each side of the base of the spinous process of the upper vertebra of the injured segment. b The two tips of a towel clamp are placed in the holes, and with a gentle rocking movement the holes are connected. c The leader of a 1.0 mm titanium cable is passed through the hole and then under the base of the inferior spinous process, leaving the interspinous soft tissues intact. The cable is then tensioned and crimped. d The laminae are decorticated, a cancellous bone graft applied, the cable tightened using the torquelimiting handle, and excess cable is cut.

295

7.2.3 Middle and lower cervical spine

4

POSTERIOR FIXATION WITH LATERAL MASS SCREW—GENERAL PRINCIPLES

Principles

Surgical principles

Posterior fi xation of the cervical spine can be achieved with cervical reconstruction plates or a rod-screw system, such as Cervifi x and Starlock, and lately with the Axon system. The techniques are based on the insertion of screws into the lateral masses. The principles of lateral mass screw insertion are reviewed and demonstrated here using the most advanced system—the Axon system.

Following exposure, the boundaries of the lateral mass are identified. The medial border is the valley at the junction of the lamina and lateral mass. The lateral boundary is the far edge of the lateral mass. The superior and inferior borders are the respective cranial and caudal facet joints. According to Magerl, the starting point for screw insertion is 2 mm medial to the center of the lateral mass ( Fig 7.2.3-5a ). The screw orientation is about 20–25º outward ( Fig 7.2.3-5b ) and 30–40º cranially. The surgeon should attempt to make the cranial angulation parallel the facet joints ( Fig 7.2.3-5c ). Roy-Camille (1980) has described an alternative technique in which screws are placed perpendicular to the posterior cortex in the sagittal direction ( Fig 7.2.3-5e–f ) and placed straight forward in a slightly divergent (0–10°) direction in the axial plane. The entry point is the junction of the upper and middle third of the lateral mass in the midline ( Fig 7.2.3-5d ). However, the authors do not recommend it as a standard technique because the screw length is shorter (usually 10–12 mm) and the purchase in the bone is less satisfactory. The authors feel that there is more likelihood of damage to neurovascular structures.

Understanding the anatomy is very important. Viewed posteriorly, a valley exists at the junction of the lamina and lateral mass. The vertebral artery and the most posterior aspect of the exiting nerve root lie anteriorly. Screws placed into the lateral mass must start laterally in relation to the valley and can be directed outward to avoid neurovascular damage. Screw placement from C3 to C6 is in the lateral mass and not transpedicular (except in rare cases where cervical pedicle screws may be justified—chapter 7.2.5 Cervicothoracic junction). Indication

Posterior stabilization of the cervical spine from C3 to C7. Advantages

• Segmental fi xation at each level providing intrinsic stability. • Biomechanically stronger than posterior wiring techniques and anterior plating [3, 4]. • Relatively easy and not as demanding as the technique for cervical pedicle screw fi xation. • Can be performed before or after a laminectomy. Disadvantage

Risks of injury to the vertebral artery and segmental nerve. A careful review of the CT scan is necessary to identify any aberrant trajectory of the vertebral artery.

The Roy-Camille method can represent a rescue procedure when purchase of the screws inserted according to the Magerl technique is lost. Following predrilling of the screw hole, a lateral mass screw of appropriate length can be inserted. If a cervical reconstruction plate is used, K-wires are inserted in each predrilled hole and the plate is positioned with the K-wires in place; the K-wires are then sequentially removed and replaced by screws (see Fig 7.2.3.6e–f). If the Cervifi x system is used, the screws have to be inserted through the clamps that are attached to the rod. With the Starlock system the side-opening clamps are secured to the screws and then attached to the rod (chapter 7.2.1 Modularity and evolution of instrumentation for the cervical spine).

296

7 7.2

Spinal instrumentation Cervical spine

Magerl technique

a

b

c

20–25°

Roy-Camille technique

1/3 2/3

d

e

f

0–10° Fig 7.2.3-5a–f Lateral mass screw insertion. a In the Magerl technique the entry point for the screws lies 2 mm medially and cranially to the center of the articular mass. b Each screw diverges by 20–25º anterolaterally and runs parallel to the surface of the intervertebral joints. Note the relationship to the neurovascular structure. c The inclination of the surface can be determined by inserting a fine dissector into the joint. d–f Alternative technique according to Roy-Camille. The screw direction is slightly divergent in the axial plane and perpendicular to the spine in the sagittal plane. The screw is shorter.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

297

7.2.3 Middle and lower cervical spine

Once the starting point has been identified, a 2.5 mm drill bit is used. An adjustable 2.5 mm drill guide is set to allow an initial maximal penetration of 14 mm ( Fig 7.2.3-6a–b ). The length of the adjustable drill guide is increased by 2 mm increments until the drill penetrates the far cortex. The hole is checked for penetration with a depth gauge ( Fig 7.2.3-6c ) and the depth is measured. Only the proximal cortex is tapped with a 3.5 mm tap ( Fig 7.2.3-6d ).

a

e

b

f

c

d

Fig 7.2.3-6a–f a–b A 2.5 mm drill bit with an adjustable drill guide is used. The drill guide is primarily set at 14 mm. The adjustable drill guide is increased by 2 mm increments until the drill penetrates the far cortex. c The hole is checked for penetration with the depth gauge. d Only the proximal cortex is tapped with the 3.5 mm tap for cancellous bone screws. e–f Once drilling of the holes in the lateral masses is done, K-wires are inserted. A cervical reconstruction plate is applied over the temporarily inserted K-wires, which are then replaced by lateral mass screws.

298

7 7.2

Spinal instrumentation Cervical spine

5

POSTERIOR CERVICAL FIXATION USING TITANIUM RODS WITH POLYAXIAL SCREW SYSTEM

Principles

This is a modular tension band system for posterior fi xation of not only the middle and lower cervical spine, but also the occipitocervicothoracic spine, the upper cervical spine, and the upper thoracic spine. The Axon system is compatible with the Cervifi x and the Starlock systems [5, 6]. The screws ( Fig 7.2.3-7 ) may be assembled to the 3.5 mm titanium rod ( Fig 7.2.3-8 ). They are polyaxial, have a 30º arch of motion in each direction, and come as either 3.5 mm cortex screws or 4.0 mm cancellous bone screws. Special lengths for the transarticular screw fi xation according to Magerl are also available. Furthermore, shaft screws for the C1 pedicle fi xation are available (chapter 7.2.2 Upper cervical spine).

a

30°

b

30°

30°

30°

c

30°

Indications

• Instabilities in the middle and lower cervical spine (traumatic instabilities, tumors, iatrogenic instabilities following laminectomy, etc). • Degenerative and painful posttraumatic conditions in the lower cervical spine. • Anterior fusions that require additional posterior stabilization.

d

e 15°

15°

Fig 7.2.3-7a–e Different Axon screws—cortex screw, cancellous bone screw, transarticular screw, shaft screw for C1, and the locking mechanism.

a b Fig 7.2.3-8a–b The 3.5 mm rod and the tapered rod used to cross the cervicothoracic junction.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

30°

299

7.2.3 Middle and lower cervical spine

Advantages

• The system allows for optimal screw insertion at all levels. • Allows for optimal bone grafting. • Postoperative magnetic resonance imaging (MRI) is possible. • Can be connected to other rod systems, such as the USS. • Increased stability compared to the reconstruction plate [7]. • Rigid stability afforded by the rod technique eliminates the use of a postoperative halo jacket. • Compression or distraction possible.

Defi ne the landmarks for screw insertion as described above (see 4 Posterior fi xation with lateral mass screw—general principles). In order to prevent displacement of the drill bit during insertion, a pilot hole should be created using the awl or a 2 mm high-speed burr. With the drill sleeve set at 14 mm, the screw holes are drilled in the lateral masses using a 2.4 mm drill bit either by hand or with the oscillating drill attachment according to Magerl ( Fig 7.2.3-9a ). The drill sleeve depth is increased 2 mm by 2 mm until both cortices are predrilled ( Fig 7.2.3-9b ). Ensure the correct path with a ball tip probe and, with a depth gauge, measure the exact length of the screws to be inserted bicortically.

Disadvantages

• Top-loading system that requires removal of the whole rod to modify or remove a screw. • Destruction of the vertebral bodies with loss of stability in compression cannot be treated by a posterior approach alone. Such instabilities require a reconstruction of the anterior column. • Higher profi le and more expensive than a simple reconstruction plate. • Not as strong in the fatigue testing as the reconstruction plate.

Surgical technique—Axon system

For application of the Axon system from C4 to C6 the patient is placed in the prone position and the lateral masses of the cervical spine are exposed to their lateral margins. If a laminectomy for decompression is planned, the authors recommend performing it after the instrumentation has been inserted in order to minimize the risks of neurological injuries. (The only exception would be the need for an urgent decompression, such as the loss of an intraoperative monitoring signal when operating on a patient with cervical myelopathy.)

A self-tapping Axon screw of appropriate diameter (3.5 mm cortex screw or 4.0 mm cancellous bone screw) is inserted in the lateral mass ( Fig 7.2.3-9c ). Further screws are then inserted into each lateral mass ( Fig 7.2.3-9d ). In case of sclerotic and brittle bone the fi rst 3 mm of the outer cortex may require tapping. A trial rod is contoured to the desired anatomy and then the rod is bent accordingly using the bending pliers ( Fig 7.2.3-9e–f ). A rod of appropriate length is then inserted into the proximal Axon screw and first secured proximally, then inserted and secured into the other variable axis screws heads ( Fig 7.2.3-9g ). A persuader may be of help in pushing the rod into the Axon screw ( Fig 7.2.3-9h ). The set screws are introduced through the persuaders if necessary. Judicious compression/ distraction completes the fi xation ( Fig 7.2.3-9i–j). A cross-link may be added for further stability ( Fig 7.2.3-9k ). The spine is then decorticated with a high-speed burr and the bone graft applied.

300

7 7.2

a

e

Spinal instrumentation Cervical spine

b

f

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c

d

Fig 7.2.3-9a–k Instrumentation with the Axon system. a Drilling the lateral masses with the drill guide preset at 14 mm. (Increase the drill guide length 2 mm by 2 mm until the trajectory is bicortical.) b Use of the depth gauge to confirm the screw hole length and make sure the hole is bicortical in the lateral mass. c Insert the selected preassembled self-tapping 3.5 mm (or 4.0 mm) Axon screw using a 2.0 mm hexagonal screwdriver. d Six Axon screws in place from C4 to C6. e Contour the rod template for a 3.5 mm rod to the desired anatomy. f Shape the 3.5 mm rod according to the prebent trial rod to fit the desired anatomy.

301

7.2.3 Middle and lower cervical spine

g

h

k

i

j

Fig 7.2.3-9a–k g The rod is being introduced into the heads of the variable axis screws proximally into C4 and distally into C6. h With the persuader the rod is inserted into the screw head in C5 by pushing it into the groove of the Axon screw. i–j Final tightening of the screw after having achieved compression or distraction if necessary. k A transverse bar/transconnector may be added for further stability.

302

7 7.2

Spinal instrumentation Cervical spine

Postoperative care

A Philadelphia collar is worn for a period of 6–12 weeks depending on the underlying pathology. It may be removed for daily care and while resting. Case example

Video 7.2.3-2 Instrumentation with the Axon system from C3 to C7 after laminectomy.

A clinical example of a posterior instrumentation is shown with the Axon system and a titanium cable from C2 to C4 for posterior C2–4 instability ( Fig 7.2.3-10 ). In C2, only a uni lateral pedicle screw fi xation was done because the contralateral pedicle was very thin. In C3 and C4 lateral mass screws according to Magerl were inserted.

a

b

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c

Fig 7.2.3-10a–c Posterior reconstruction with C2 pedicle screws and C3 and C4 lateral mass screws in a posttraumatic cervical instability. A posterior interspinous wiring was added to increase stability.

7.2.3 Middle and lower cervical spine

6

1.

2.

3.

4.

5.

6.

7.

BIBLIOGRAPHY

Morscher E, Sutter F, Jenny H, et al (1986) [Anterior plating of the cervical spine with the hollow screw-plate system of titanium.] Chirurg; 57(11):702–707. Smith SA, Lindsey RW, Doherty BJ, et al (1995) An in-vitro biomechanical comparison of the Orosco and AO locking plates for anterior cervical spine fi xation. J Spinal Disord; 8(3):220–223. Gill K, Paschal S, Corin J, et al (1988) Posterior plating of the cervical spine. A biomechanical comparison of different posterior fusion techniques. Spine; 13(7):813–816. Singh K, Vaccaro AR, Kim J, et al (2003) Biomechanical comparison of cervical spine reconstructive techniques after a multilevel corpectomy of the cervical spine. Spine; 28(20):2352–2358. Jeanneret B (1996) Posterior rod system of the cervical spine: a new implant allowing optimal screw insertion. Eur Spine J; 5(5):350–356. Borm W, Konig RW, Hubner F, et al (2003) First clinical experiences with a new cervical fi xation device—technical report. Zentralbl Neurochir; 64(3):123–127. Grubb MR, Currier BL, Stone J, et al (1997) Biomechanical evaluation of posterior cervical stabilization after a wide laminectomy. Spine; 22(17):1948–1954.

303

304

7 SPINAL INSTRUMENTATION 7.2 CERVICAL SPINE 7.2.4 CRANIOCERVICAL JUNCTION

1 1.1

Introduction ………………………………………………………………………………………………… 305 Anatomical basics for screw insertion in the occiput ……………………………………………… 305

2

Occipitocervical plate fixation with a titanium occipital plate/rod system ……………………… 306

3

Occipitocervicothoracic fixation …………………………………………………………………………

312

4

Bibliography …………………………………………………………………………………………………

313

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

305

Michael E Janssen, Jason C Datta

7 7.2

SPINAL INSTRUMENTATION CERVICAL SPINE

7.2.4

1

CRANIOCERVICAL JUNCTION

INTRODUCTION

Instrumentation of the craniocervical junction requires a thorough understanding of the bony architecture of the occiput and the upper cervical spine. For a long time, craniocervical fi xation relied on cerclage wires passed through the occiput and fi xed to a rod system. These methods of fi xation were not very stable and therefore required a halo vest for postoperative immobilization [1]. With current fi xation methods postoperative immobilization can be limited to a minimum, for example, by using a neck collar for only a short period of time [2, 3]. By using modular instrumentation, such as the Axon, Cervifix, or Starlock system, a broad combination of fi xations are possible and allow for instrumentation from the occiput to the cervical and the thoracic spine to be used (if necessary) while applying the same instrumentation principles.

1.1

ANATOMICAL BASICS FOR SCREW INSERTION IN THE OCCIPUT

Screws can safely be placed in the occiput, but this requires a careful understanding of the occipital anatomy [4]. Screws should not be inserted above the inion to avoid damage to the intracranial sinus. The bone present at the midline ( Fig 7.2.4-1) allows strong screw purchase. The occipital cortex becomes thin 2–3 mm from the midline. Care must be taken while drilling and placing screws in order to avoid injury to the cerebellum. Dural laceration with a CSF leak (cerebrospinal fluid leak) is not uncommon and is dealt with by screw insertion into the predrilled screw hole.

306

7 7.2

Spinal instrumentation Cervical spine

2

In the midline, a 2.5 mm drill bit is used with an adjustable drill guide, initially set at 8 mm and increased by 2 mm increments until the far cortex is penetrated. The oscillating drill attachment is used to prevent soft tissue from wrapping around the drill. The thickness of the cortex lateral to the midline is only 3–7 mm and care must be taken when predrilling the screw hole. If the entry point of the screw is lateral to the midline, the adjustable drill guide should be set at 4 mm initially, and the screw trajectory should be convergent toward the midline (where the bone is thick) in order to maximize the length of the trajectory. After each drilling procedure the hole is checked for penetration with a depth gauge, and when penetration of the distal cortex has occurred the depth is measured. With a tap for the 3.5 mm cortex screws, the screw holes are also tapped bicortically. The screws are then inserted.

SNL INL

OCCIPITOCERVICAL PLATE FIXATION WITH A TITANIUM OCCIPITAL PLATE/ROD SYSTEM

Principles

The function of the Axon’s plate on the occiput is mainly as a buttress and partially as a tension band. Indications

• Cranial settling or basilar invagination. • Pseudoarthrosis of atlantoaxial arthrodesis. • Occipitocervical and upper cervical spinal instabilities (rheumatoid arthritis, C1/2 anomalies, posttraumatic conditions, tumors, and infections). Advantages

• The Axon implants are versatile ( Fig 7.2.4-2 ). • The occipitocervical plate fi xation allows for optimal screw insertion at all levels to be instrumented. • Optimal bone grafting is possible. • Postoperative magnetic resonance imaging (MRI) can easily be done. • The Axon system can be connected to the adult 6 mm rod or the pediatric 5 mm USS rod thanks to side connectors. • Rigid stability afforded by this technique, eliminates the use of a postoperative halo jacket. Disadvantages

SNL = superior nuchal line INL = inion nuchal line

Fig 7.2.4-1 Areas of thin and thick bone in the occiput—optimal location for screw placement is 2–3 mm to both sides from the midline (marked area). Screws should be inserted laterally to the midline and should be aimed medially.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

• This technique is technically demanding and there is a potential for neurovascular injuries. • It requires a perfect analysis of the bony anatomy with CT scans and reconstructions to evaluate each screw trajectory and the location of the transverse foramen at the C2 level for safe insertion of transarticular C2/1 or Judet pedicle screws. • Poor purchase of occipital screws due to the thin cortex, if the plate is applied in a lateral position.

307

7.2.4 Craniocervical junction

Surgical technique—fusion from the occiput to C2

a

b

30°

30°

30° 30°

c Fig 7.2.4-2a–c a Occipital plate/rod. The 3.5 mm rod can be connected to either the pediatric or the adult USS. b Cortex bone screws (3.5 mm or 4.0 mm) and cancellous bone screws (3.5 mm or 4.0 mm). Long Axon screws can be used for transarticular C2/1 screw fixation or pedicle screw fixation. c Preassembled transconnector to link both rods in order to increase stability.

The patient is placed in a prone position. The head rests on a horseshoe headrest with Gardner-Wells traction or with a Mayfield head clamp. The table is slightly tilted in a reverse Trendelenbourg position—it is important that the face is looking horizontally when positioning the patient at the beginning of the operation ( Fig 7.2.4-3 ). In most cases a lateral x-ray or fluoroscopic image of the cervical spine is recommended to assess the position of the head and the reduction of C1 on C2. If C1/2 is unstable a closed reduction is necessary. A posterior midline incision is performed. The occiput, posterior ring of the atlas, posterior elements of C2, spinous processes, vertebral arches, and articular masses of the lower vertebrae of the cervical spine, which are all to be included in the fusion, are exposed subperiosteally. In order to perform a transarticular screw fi xation at C1/2, the isthmus of C2 must be exposed on both sides.

Fig 7.2.4-3 The patient’s face is positioned looking horizontally in relation to the table.

308

7 7.2

Spinal instrumentation Cervical spine

If a cerclage of the posterior arch of C1 is to be performed, it must be done first. The C1 cerclage may be used to pull the posterior arch of C1 backward in case of an incomplete reduction ( Fig 7.2.4-4 ).

Fig 7.2.4-4 A sublaminar wire can be placed beneath either side of the C1 arch in order to facilitate reduction.

A template ( Fig 7.2.4-5a ) is contoured in such a way that its cranial end lies adjacent to the midline and is situated just caudally to the occipital protuberance. The rod passes over the lateral rims of the articular processes of those levels that are to be incorporated in the fusion. The occipital plate/rod is then bent and cut according to the template. The bend of the plate/ rod must start at the level of the plate part. Then the holes for the transarticular screws in C2/1 are drilled (chapter 7.2.2 Upper cervical spine). The holes may be started using an awl or a 2 mm high-speed burr, which helps to prevent displacement of the drill after initial insertion. Drill the screw hole to the desired depth using a 2.0 mm oscillating drill bit attachment and the drill sleeve ( Fig 7.2.4-5b ). Use the depth gauge to confi rm the depth of the screw hole and choose selftapping Axon screws of the appropriate length. In order to provisionally stabilize C1/2, the drill bit is left in place on one side while the opposite side is instrumented ( Fig 7.2.4-5c–d ).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

The occipital plate/rod is then applied in the polyaxial heads of the Axon screws ( Fig 7.2.4-5e ). The set screws are left loose so that the rod portion of the plate/rod can be mobilized against the occiput. Fine-tuning of the plate/rod by additional bending is often necessary. The plate must be bent medially in order to be as central as possible to where the screws will be inserted in the occiput. The rod portion is left loose in the Axon screw heads. The occipital screw holes are drilled through the plate toward the midline and 3.5 mm cortex screws (Cervifi x screws) are inserted. Care must be taken to place the cranial end of the plate as close as possible to the midline of the skull, in order to provide the best possible purchase. The occipital screws are then tightened ( Fig 7.2.4-5f ). After final adjustment of the construct, fi rmly tighten all locking screws with the locking screwdriver shaft and the torque-limiting handle, which is preset to a torque of 2 Nm ( Fig 7.2.4-5g ). A cross-link/transconnector is then applied ( Fig 7.2.4-5h ). A cancellous bone graft is applied over the decorticated laminae and articular mass. Between the occiput and the spinous process of C2, a corticocancellous bone graft is inserted acting as a buttress.

a

Fig 7.2.4-5a–h a Contouring and prebending of a rod template.

309

7.2.4 Craniocervical junction

b

e

h

c

d

f

g

Fig 7.2.4-5a–h b Predrilling of the first screw hole for the C2/1 transarticular screw fixation. c Drilling of the second C2/1 transarticular screw hole while leaving the drill bit in the first screw hole for provisional stabilization. d Insertion of the transarticular C2/1 screws on both sides (fully or partially threaded screws). e Application of the plate/rod into the polyaxial Axon screw head. The set screw is not tightened to allow positioning of the plate on the occiput. f Insertion of the cortex screws into the occiput after predrilling and tapping of the screw holes. g Final tightening of the Axon locking screws using the torque-limiting screwdriver. h Final construct after application of the transconnector and tightening of the sublaminar cables over the plate.

310

7 7.2

Spinal instrumentation Cervical spine

Case example

A basilar invagination with C1/2 rotatory luxation in a 17year-old boy was treated by traction and fusion from C0 to C2 ( Fig 7.2.4-6 ).

b

a

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.2.4-6a–b a Preoperative x-ray, MRI, and CT scans. b Occipitocervical fixation with a plate/rod system and C2/1 transarticular screws (screws and clamps from the Starlock system were used).

311

7.2.4 Craniocervical junction

Alternatives to the plate/rod system

Choices of screw fixation

A rod can be used instead of a plate/rod system ( Fig 7.2.4-7 ). Thanks to the modularity of the system, it is possible to use a 3.5 mm rod and Cervifi x clamps, which are used to insert the screws in the occiput (see Fig 7.2.4-8 ). This has the advantage of an easier contouring of the instrumentation and insertion of the screws in the thick midline bone of the occiput.

Other alternatives are the use of C1 pedicle shaft screws and C2 pedicle screws (chapter 7.2.2 Upper cervical spine). If a C1 posterior cerclage wire or cable is used, it can be attached to the rod to increase stability (see Fig 7.2.4-6b ).

A titanium reconstruction plate can also achieve, in a more cost-effective manner, an occipitocervical fi xation, but it is technically more demanding [5]. Such plate systems have been shown to be biomechanically as stable as the new rod systems [6].

a

Fig 7.2.4-7a–b a Example of an occipitocervical fixation using a 3.5 mm rod system and the Cervifix clamps and screws for fixation to the occiput. In this illustration, C1 lateral mass screws (shaft screws) and C2 pedicle screws are used instead of applying a Magerl fixation technique.

Postoperative care

A Philadelphia collar should be worn for a period of 6–12 weeks depending on the underlying pathology. It may be removed for daily care and while resting.

b b Drilling of screw holes in the occiput should start with the most cephalad screw. The direction is toward the midline to maximize the length of the screw. The predrilled screw hole is then tapped and a Cervifix screw is inserted bicortically through the clamps. The rest of the screws are then inserted in the same way from proximal to distal.

312

7 7.2

Spinal instrumentation Cervical spine

3

OCCIPITOCERVICOTHORACIC FIXATION

In rare cases, a fi xation from the occiput to the thoracic spine is necessary (upper cervical and subaxial involvement of the spine as observed in a severe form of rheumatoid arthritis). Principles of such a fi xation require a perfect understanding of the different areas of the cervical spine to be instrumented.

it is necessary to obtain long x-rays in order to judge the head position. The choice of implants is either a plate/rod system with a cross-link to the USS for the upper thoracic spine, or a tapered USS rod and the use of a Cervifi x clamp in order to fi x the rod to the occiput.

Technically, it is essential that the head’s position be perfect before fi xation because no anatomical adaption of the spine will be possible. When positioning the head, great attention must be given that it is parrallel to the floor, when the table is flat, and parallel to the trunk. For the preoperative planning,

Clinical example

a

b

Fig 7.2.4-8a–f a–c Preoperative diagnostic pictures.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

This female patient presented with long-lasting inflammatory arthropathy, cranial settling, and involvement of the subaxial spine ( Fig 7.2.4-8a–c ). Tretament of choice was occipitocervicothoracic fusion ( Fig 7.2.4-8d–f ).

c

313

7.2.4 Craniocervical junction

4

1.

2.

3.

4.

5.

6.

*

d

e

f

Fig 7.2.4-8a–f d–f Intra- and postoperative photographs after occipitocervicothoracic fusion using Cervifix clamps in the occiput with cortex screws, Axon screws in the cervical spine, and USS II screws for the upper thoracic spine. A USS tapered rod is inserted from the upper thoracic spine to the occiput. A unicortical bone graft (asterisk) from the posterior iliac crest is added to achieve fusion between the occiput and C3.

BIBLIOGRAPHY

Hurlbert RJ, Crawford NR, Choi WG, et al (1999) A biomechanical evaluation of occipitocervical instrumentation: screw compared with wire fi xation. J Neurosurg; 90(1 Suppl):84–90. Deutsch H, Haid RW Jr, Rodts GE Jr, et al (2005) Occipitocervical fi xation: long-term results. Spine; 30(5):530–535. Vender JR, Rekito AJ, Harrison SJ, et al (2004) The evolution of posterior cervical and occipitocervical fusion and instrumentation. Neurosurg Focus; 16(1):E9. Hertel G, Hirschfelder H (1999) In vivo and in vitro CT analysis of the occiput. Eur Spine J; 8(1):27–33. Lieberman IH, Webb JK (1998) Occipito-cervical fusion using posterior titanium plates. Eur Spine J; 7(4):308–312. Puttlitz CM, Melcher RP, Kleinstueck FS, et al (2004) Stability analysis of craniovertebral junction fi xation techniques. J Bone Joint Surg Am; 86-A(3):561–568.

314

7 SPINAL INSTRUMENTATION 7.2 CERVICAL SPINE 7.2.5 CERVICOTHORACIC JUNCTION

1

Introduction ………………………………………………………………………………………………… 315

2

Posterior stabilization techniques ………………………………………………………………………

3 3.1 3.2

Anterior stabilization technique (cages) after cervicothoracic corpectomy …………………… 322 Expandable cervical cage ………………………………………………………………………………… 323 Mesh cage ………………………………………………………………………………………………… 326

4

Combined anterior and posterior techniques ………………………………………………………… 327

5

Bibliography ………………………………………………………………………………………………… 327

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

315

315

Brian JC Freeman, Frank Kandziora

7 7.2

SPINAL INSTRUMENTATION CERVICAL SPINE

7.2.5

1

CERVICOTHORACIC JUNCTION

INTRODUCTION

The cervicothoracic junction is difficult to image and to access surgically. The use of the swimmer’s view, computed tomography scanning with reconstruction, and magnetic resonance imaging assists in the visualizing and planning of the surgery. The transition from the mobile cervical to the rigid thoracic spine makes the cervicothoracic junction particularly unstable. In most cases, stabilization of the cervicothoracic junction is done posteriorly, however, anterior decompression, reconstruction, and stabilization may be necessary in selected cases.

2

POSTERIOR STABILIZATION TECHNIQUES

Principles

The Axon system is used for the cervical spine and the USS II for the upper thoracic spine. The cervicothoracic junction is a potentially unstable site due to the transition from the highly mobile cervical spine to the rigid thoracic spine. As a result, posterior instrumentation must ensure excellent stability with the use of pedicle screws and/or lateral mass screws at all the levels. The advent of the dual diameter transition rod system has allowed greater versatility in the instrumentation of the cervicothoracic junction [1]. Indications

• Cervicothoracic instability (posttraumatic instability, infections, tumors). • Complementary fi xation after anterior surgery. • Correction of cervicothoracic kyphosis (see AOSpine Manual—Clinical Applications, 4.2 Ankylosing spondylitis).

316

7 7.2

Spinal instrumentation Cervical spine

Advantages

• The use of the same modular system makes the instrumentation versatile. • The use of Axon screws can be extended from the cervical spine to the upper thoracic spine, and down to T3. • The use of USS II screws or hooks can be extended in the cephalad direction up to T1. • The tapered titanium rod from the USS II is used for instrumentation in the transition from the cervical to the thoracic spine.

because lateral fluoroscopy may not reveal the pedicles hidden by the shoulders and one may have to rely mostly on the AP fluoroscopy for the placement of the screws in large patients. The operating table should be placed in the head-up position to reduce bleeding and the shoulders should be taped to allow perioperative visualization of the cervicothoracic junction under fluoroscopic control. The bony processes are palpable posteriorly with a large spinous process felt at C7 and T1. In the case of an occipitocervicothoracic fusion, assessment of the head’s position is particularly important.

Disadvantages

Insertion of the screws

• The bulkiness of instrumentation for the cervicothoracic junction when using USS II screws or hooks at these levels. • The difficulty in fitting the 3.5 mm rods if lateral mass screws are used in C6 and pedicle screws in C7, due to the mismatch of the lateral mass/pedicle screw insertion. • One segmental level between C7 and T2 must usually be “skipped” when using the tapered rod.

In the lower cervical spine Axon screws are inserted down to C7/T1 and USS II screws from T1/T2 downward. In some cases where the anatomy is not favorable, it may be necessary to skip a level in order to accommodate for the offset between lateral mass screws and pedicle screws, or due to the flute of the tapered rod.

Surgical technique Preparation

CT scans are needed to evaluate the direction and the width of the pedicles of the cervical spine and of the upper thoracic vertebrae. Additional fiber-optic endoscopic intubation may be necessary in cases of limited extension, stiffness, or instability of the spine. The patient is put in a prone position and may be supported with a halo, Mayfield clamp, or Gardner-Wells pins and a horseshoe headrest to avoid undue pressure sores. Intraoperative monitoring (SSEP and MEP) is essential. The use of a radiolucent table, such as the Jackson table, is advisable

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Insertion of screws into C6 and above

In most cases, the surgeon chooses lateral mass screws at C6 and the upper levels. This technique is described in chapter 7.2.3 Middle and lower cervical spine. However, in selected and in rare indications (osteoporosis) it may be advantageous to insert pedicle screws in C6 and above. The authors recommend the keyhole foraminotomy technique. A 2 mm diamond burr is used to create a starting hole just lateral to the midpoint of the lateral mass and 2 mm below the inferior facet margin of the cephalad vertebrae. A power drill (with the oscillation attachment) is used, with the stop set at 18 mm, to drill through the pedicle into the vertebral body. The orientation of the screw should be parallel to the superior end plate of C6 and should aim medially by 25–45° depending on

317

7.2.5 Cervicothoracic junction

the pedicle’s position ( Fig 7.2.5-1). A 20 mm long 3.5 mm Axon screw is then placed into the C6 pedicle. The screw tip should reach the midpoint of the C6 vertebral body. During these maneuvers the palpation of the pedicle’s medial wall with a nerve hook is recommended. Because the vertebral artery enters the transverse foramen at C6, the risk of an injury to the artery exists at this level. Careful review of the CT scan, in order to measure the direction of the pedicle, its width, and the length of the screws to be inserted, is mandatory before such techniques are carried out. Other techniques of cervical pedicle screw placement involve opening the cortex of the pedicle and drilling inside with a small high-speed burr or with a curette [2]. The use of computer-assisted surgery or a special guide may increase the accuracy of pedicle screw placement. The insertion of the pedicle screws through a complementary stab incision in the skin may also be safer and more accurate, because the soft tissues do not prevent the desired medial angulation. [3]

Insertion of the pedicle screws into C7

The surgeon needs to review the preoperative CT scans in order to analyze the angulation of the C7 pedicle and to check for any aberrant vertebral anatomy at this level ( Fig 7.2.5-2 ). Due to the fact that the pedicle of C7 is wider than the pedicle of C6, and because there is usually no vertebral artery at this level, the technique for placing C7 pedicle screws can be done under C-arm control. Alternatively, a keyhole foraminotomy at the C6/7 interval can be done as described above [4]. The C7 pedicle screw fixation exhibits superior fixation when compared to the lateral mass screw technique at C7. Depending on the size of the pedicle, a 3.5 or 4.0 mm cancellous bone Axon screw is inserted into the pedicle of C7.

*

25–45° a

b

Fig 7.2.5-1a–b Axial CT scan at C6. The vertebral arteries are just entering the transverse foramen (asterisk), and the direction of the pedicle is 45°. If a pedicle screw fixation is contemplated at this level, a keyhole foraminotomy is recommended in order to feel the medial border of the pedicle.

Fig 7.2.5-2 Axial CT scan of C7 showing the convergence of the C7 pedicle.

318

7 7.2

Spinal instrumentation Cervical spine

Pedicle screw fixation in the upper thoracic spine T1, T2, T3

The pedicles of T1 and T2, and to a lesser degree of T3, tend to be more widespread than the pedicles lower in the thoracic spine. This is very important to know, as they do not lie in line with the inferior pedicles ( Fig 7.2.5-3 ). Their transverse angulation ranges from 30° at the level of T1, to 15° at the level of T3. Then, from T4 downward the transverse angulation is almost sagittal (chapters 6.3 Cervicothoracic junction, 6.4 Thoracic spine). The upper thoracic screws are placed in the pedicles and not in the articular masses. The entry point lies just below the rim of the upper facet joint, 3 mm lateral to the center of the joint near the superior border of the transverse process. The screws

should be angled between 30° and 10° toward the midline, depending on the level ( Fig 7.2.5-4 ). Pedicle preparation may be performed using the small 4 mm pedicle probe. The small straight ball tip probe is used to confirm by palpation the accurate placement within the pedicle. If a medial breech of the pedicle is detected, the hole should simply be packed with bone wax and not used. The small pedicle markers, with either spherical bulges or long bulges, may be used during radiographic imaging to confirm position and orientation. The screws should not pass more than 80% through the depth of the vertebral body. The dimensions of the pedicles of C7, T1, and T2 allow relatively small margins for error. In most cases a 4.2 mm USS II screw (or 5.2 mm) is inserted. If difficulties are encountered with insertion of pedicle screws, pedicle hooks from the USS II set may be used as an alternative. Claws using lamina hooks, specialized pedicle hooks, or transverse hooks are other possibilities at these levels.

a Fig 7.2.5-3 Landmarks for pedicle screw insertion at T1, T2, and T3. Note the spread of the T1 and T2 pedicles compared to T3.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

Fig 7.2.5-4a–b a CT scan at the level of T1. The pedicles are widespread and their orientation is 30° toward the midline. b CT scan at the level of T3 showing the pedicle orientation. Note that at this level the medial angulation is only 15–20°. The spread of the pedicle is less than in T1.

319

7.2.5 Cervicothoracic junction

Rod preparation

Insertion of the rod

A flexible template is inserted into the polyaxial heads of the Axon screws and in the side-opening grooves of the USS II screws. The template is adjusted to the local sagittal anatomy. In cases where reduction is needed, the template is contoured to the desired sagittal profile. The tapered rod is then bent according to the template ( Fig 7.2.5-5 ).

The 3.5 mm part of the rod is inserted in the cephalad Axon screws, then the rod is attached to the upper thoracic screws. The match between the lateral mass screws and the pedicle screws of C7 is assured by inclining the polyaxial head of the Axon screws in the opposite direction (inclining the polyaxial head laterally for the lateral mass screws and medially for the pedicle screws in C7) ( Fig 7.2.5-6 ). At times, lateral offset connectors or the contouring of the rod in the coronal plane may be required. The flute of the tapered rod should be pushed cranially as far as possible to the bottom of the Axon screw.

b

a

Fig 7.2.5-5a–b A rod template is used to get the exact curve of the spine. The tapered rod is then shaped accordingly using the rod bender.

Fig 7.2.5-6 Alignment of the screws from C5 to T4. Lateral mass Axon screws are inserted into C5 and C6; Axon pedicle screws into C7. USS II screws are inserted in the pedicles of T1, T2, T3, and T4. Note that the pedicles of T1, T2, T3, and T4 are not in a straight line.

320

7 7.2

Spinal instrumentation Cervical spine

Compression-distraction and cross-linking of the two rods

Once the construct is in place, compression and/or distraction at the appropriate levels is achieved if necessary. An Axon transconnector is applied for the 3.5 mm part of the rod, and a USS II cross-link is applied for the 6 mm part of the rod ( Fig 7.2.5-7 ). Variation according to the possible constructs—combination of Axon and USS II or Cervifix/Starlock and USS II without a tapered rod

In extending thoracic instrumentation to the lower cervical spine or vice versa, the easiest combination possible for crossing

Fig 7.2.5-7 The rods are connected with the 3.5 mm rod cross-link cranially and with the USS II cross-link distally.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

the cervicothoracic junction is achieved by using a number of parallel connectors to provide a lateral offset. The USS II also allows a possible offset of the rod depending on whether the rod is placed laterally or medially in the pedicle screws. Depending on whether one uses polyaxial Axon, Cervifix, or Starlock screws in the cervical spine, the cervical 3.5 mm rod may be positioned either on the medial or lateral side of the thoracic 6 mm rod. Parallel 3.5/6.0 mm rod connectors perfectly accommodate for such an offset. In addition, a 3.5/5.0 mm parallel connector and a 3.5/3.5 mm parallel connector ( Fig 7.2.5-8 ) are available for linking rods of all sizes (small or large stature USS).

Fig 7.2.5-8 Connection of the 3.5 mm cervical spine rod system to the 6.0 mm USS rod system using the parallel rod-to-rod connectors.

321

7.2.5 Cervicothoracic junction

Clinical examples

Conclusion

Fig 7.2.5-9a shows the destruction of the vertebral body T1

Due to the transition from the mobile cervical to the rigid thoracic spine, surgeons would be strongly advised to perform a posterior fusion in conjunction with stabilization in this region. If left unfused, it is likely that the tapered rods will fracture between 9 and 12 months after insertion. Similarly, if there is extensive loss of anterior support such as that seen in certain tumors, fractures, and infections, additional anterior stabilization/interbody fusion should be considered.

in a patient with lung cancer. Treatment was a posterior decompression and stabilization with Cervifi x and USS linked with side-to-side connectors. Two clinical examples of other possible cervicothoracic instrumentations are shown in Fig 7.2.5-9b–c . They exemplify the modularity of the instrumentation.

a

b

Fig 7.2.5-9a–c a Example of a cervicothoracic instrumentation with the Cervifix and USS. b Example of a cervicothoracic instrumentation with a tapered rod: the Starlock system for the cervical spine, and the USS II for the thoracic spine. c Example of a cervicothoracic instrumentation using the Axon and the connection with lateral connectors (6 mm/3.5mm) to the USS II.

c

322

7 7.2

Spinal instrumentation Cervical spine

3

ANTERIOR STABILIZATION TECHNIQUE (CAGES) AFTER CERVICOTHORACIC CORPECTOMY

Principles

In the case of a tumor, cervical instability, or cervical disc herniation, the cervicothoracic junction may require anterior stabilization. Such stabilization relies on anterior column support and anterior plate fi xation. The anterior support can be achieved with a tricortical iliac bone graft, PMMA cement, an expandable cervical cage (expandable corpectomy device) ( Fig 7.2.5-10 ), or the Synmesh cage ( Fig 7.2.5-11) [5, 6]. The expandable corpectomy device is a continuously expandable vertebral body replacement for the cervical (C3–7) and the high-thoracic (T1/2) region. It is used to stabilize the anterior column of the spine. The expandable corpectomy device should always be used with an additional internal fi xation (anterior and/or posterior) system to bear tensile forces as well as torsion, flexion, and extension moments. There are different implant sizes available with varying heights and end-plate inclinations. This device is available in PEEK (radiolucent) and titanium alloy. Synmesh is a vertebral body replacement with titanium imprints available in various footprints and heights to fill a range of vertebral defects. Parallel and angled end rings are designed to restore normal spinal alignment and resist subsidence. The open architecture of the mesh promotes bony ingrowth. Synmesh cages may be used in all areas of the spine (chapter 7.3.4 Instrumentation for the degenerative thoracolumbar spine), ie, the cervical, thoracic, and lumbar spine (C3–L5) to replace collapsed, damaged, or unstable vertebral bodies due to tumor or trauma. The round meshes for the cervical spine are available in diameters of 10 mm, 12 mm, and 15 mm and in different heights from 4 mm up to 88 mm. Supplementary fi xation is required with a cervical spine locking plate (CSLP) or a posterior fixation. Round press-fit end rings are available with 0° of angulation or 2.5° of angulation. Locking screws can be used to lock the end rings to the Synmesh.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.2.5-10a–b The expandable vertebral body replacement (PEEK) for the cervical spine is available in different sizes. A locking clip secures the implant after final distraction.

a

b

1.5 mm

Synmesh height

1.5 mm

Fig 7.2.5-11 Synmesh cage: vertebral body replacement for the cervical spine. The addition of the end rings adds 1.5 mm of height on each side. The end rings are available in 0º or 2.5º of angulation.

323

7.2.5 Cervicothoracic junction

Indications

• Corpectomy for tumor or cervical disc herniation between C7 and T2. • Cervicothoracic instability and fractures. • Cervicothoracic deformity. Advantages

In most cases, anterior surgery addresses the pathology directly and allows optimal decompression of the spinal cord. Disadvantages

• The access may be difficult in patients with short necks and may require splitting of the manubrium in a reverse T-fashion or splitting of the whole sternum. • Proximity of the large vessels and the risk of injury to the innominate vein.

3.1

Fig 7.2.5-12 Positioning of the patient.

EXPANDABLE CERVICAL CAGE

Surgical technique

The surgical technique for an anterior reconstruction after corpectomy of T1 and T2 is described in the following. The patient is positioned supine with slight traction on the head while resting on a horseshoe headrest ( Fig 7.2.5-12 ). Prepping of the entire neck and chest is done for proper access and possible extension with sternotomy. Preoperative MRI or CT scans at the level of the manubrium are needed to plan for a possible sternotomy [7]. The cervical Synframe ( Fig 7.2.5-13 ) is positioned across the cervicothoracic junction, halfway between the xyphoid and the chin to allow maximum exposure. With the use of Hohmann retractors and blades

Fig 7.2.5-13 Application of the cervical Synframe (ring retractor) and the use of Hohmann blades or adjustable blades for proper exposure of the cevicothoracic junction.

324

7 7.2

Spinal instrumentation Cervical spine

the cervicothoracic junction is approached on the left side with or without a sternotomy from C5 to T2 according to the classic approaches [8]. The appropriate implant height is determined before surgery by using the preoperative planning template. The partial or complete corpectomy is performed as required ( Fig 7.2.5-14 ). Disc material and superficial layers of the cartilaginous parts of the adjacent end plates are excised. However, care should be taken not to weaken the end plate in order to maintain mechanical strength. A calliper is used to determine the size of the resulting spinal defect and an appropriately sized implant is chosen. The implant must be inserted with the cranial convex end plate at the top. The optimal position for an expandable cervical cage is in the center of the vertebral end plate ( Fig 7.2.5-15 ). Space should be left around the implant to allow bony fusion to take place.

The final position of the expandable corpectomy device is confi rmed intraoperatively using an image intensifier. Three x-ray markers serve to control the position of the PEEK implants. Once in the correct position, the expandable corpectomy device is expanded in situ using the holding and distraction instrument ( Fig 7.2.5-16 ). Once the maximum expansion position has been reached, the fi nal position is again checked in the frontal and sagittal plane intraoperatively using an image intensifier. A locking clip is then inserted to secure the fi nal position (see Fig 7.2.5-10b, Fig 7.2.5-10b ). Additional fixation with an anterior plate or a posterior fi xation is mandatory ( Fig 7.2.5-17 ). The implant is then packed, in and around the cage, with bone chips ( Fig 7.2.5-18 ) or ß-tricalcium phosphate granules (Chronos).

+45° C7

C7

T3

Fig 7.2.5-14 The vertebral bodies of T1 and T2 are removed.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

T3

Fig 7.2.5-15 The expandable corpectomy device is inserted with the implant holder/ distractor, which allows implant insertion at 45° for easy reconstruction of the cervicothoracic junction. The cage is placed on the superior end plate of T3.

325

7.2.5 Cervicothoracic junction

C7

b

T3 a

Fig 7.2.5-16a–b Once the cage is placed properly and firmly, the appropriate length is achieved by distracting the expandable corpectomy device using the implant holder/distractor. A locking clip (b) is applied after distraction of the PEEK cage at the level of the cogwheel. (The locking clip can, if necessary, be removed in the same manner.)

Fig 7.2.5-17 After final distraction and insertion of the locking clip, a complementary fixation with a Vectra plate and variable-angle screws adds to the stability of the final construct. Patients with a long neck may not require sternotomy for corpectomy of T1 and T2.

Fig 7.2.5-18 The expandable cervical cage is then covered with autogenous bone chips or ß-tricalcium phosphate granules.

326

7 7.2

Spinal instrumentation Cervical spine

Case example

3.2

A 58-year-old patient shoes a destruction of C5 as a result of a breast metastasis ( Fig 7.2.5-19a ). The destructed vertebra was treated with an expandable corpectomy device and anterior plate fi xation ( Fig 7.2.5-19b–c ).

MESH CAGE

Surgical technique—similar to thoracolumbar technique (chapter 7.3.4 Instrumentation for the degenerative thoracolumbar spine)

A corpectomy is performed and with the use of a parallel distractor, the height of the defect is measured. The addition of two round end rings adds a total of 3 mm to the total height of the construct (see Fig 7.2.5-11). The mesh may be cut if required. One end ring is attached to the Synmesh and held in place with a 2 mm locking screw. The Synmesh is fi lled with bone graft or cement as appropriate. The second end ring is attached and locked in place with a locking screw. The space is distracted and the implant is inserted. The distractor is removed. Supplemental fi xation is required either with a CSLP or a posterior fi xation depending on the stability of the anterior reconstruction. a

b

c

Fig 7.2.5-19a–c Anterior reconstruction from C4 to C6 with a titanium expandable cervical cage and an anterior CSLP (as tumor treatment).

Fig 7.2.5-20a–c Anterior reconstruction of the cervicothoracic junction from C5 to T2 with a Synmesh and an anterior CSLP (as tumor treatment). The reconstruction did not require a sternotomy.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Case example

The instrumentation is shown in a clinical example ( Fig 7.2.520 ). A 38-year-old patient with spasticity in his lower extremities due to destruction of the vertebral body C7 (from a metastasis of a bladder carcinoma).

a

b

c

327

7.2.5 Cervicothoracic junction

4

COMBINED ANTERIOR AND POSTERIOR TECHNIQUES

With the type of pathology encountered at the cervicothoracic junction a long posterior instrumentation for stabilization combined with a posterior fusion to prevent implant failure might need to be performed. With significant anterior column insufficiency, consideration should be given to anterior column reconstruction, either directly after the posterior stabilization and fusion or as a staged procedure. The anterior approach to the cervicothoracic junction is difficult, and the decision to operate on the anterior column will clearly depend on the patient, the pathology, and the prognosis—particularly in the case of a metastatic disease.

5

1.

2.

3.

4.

5.

6.

7.

8.

BIBLIOGRAPHY

Le H, Balabhandra R, Park J, et al (2003) Surgical treatment of tumors involving the cervicothoracic junction. Neurosurg Focus; 15(5):E3. Review. Abumi K, Kaneda K (1997) Pedicle screw fi xation for nontraumatic lesions of the cervical spine. Spine; 22(16):1853–1863. Richter M, Cakir B, Schmidt R (2005) Cervical pedicle screws: conventional versus computer-assisted placement of cannulated screws. Spine; 30(20):2280–2287. Albert TJ, Klein GR, Joffe D, et al (1998) Use of cervicothoracic junction pedicle screws for reconstruction of complex cervical spine pathology. Spine; 23(14):1596–1599. Gokaslan ZL, York JE, Walsh GL, et al (1998) Transthoracic vertebrectomy for metastatic spinal tumors. J Neurosurg; 89(4):599–609. Kandziora F, Pflugmacher R, Schaefer J, et al (2003) Biomechanical comparison of expandable cages for vertebral body replacement in the cervical spine. J Neurosurg; 99(1 Suppl):91–97. Fraser JF, Diwan AD, Peterson M, et al (2002) Preoperative magnetic resonance imaging screening for a surgical decision regarding the approach for anterior spine fusion at the cervicothoracic junction. Spine; 27(7):675–681. Cauchoix J, Binet JP (1957) Anterior surgical approaches to the spine. Ann R Coll Surg Engl; 21(4):234–243.

328

7 SPINAL INSTRUMENTATION 7.2 CERVICAL SPINE 7.2.6 CERVICAL LAMINOPLASTY

1

Introduction ………………………………………………………………………………………………… 329

2

General principles ………………………………………………………………………………………… 329

3

Surgical technique ………………………………………………………………………………………… 330

4

Bibliography ………………………………………………………………………………………………… 335

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

329

Christopher I Shaffrey

7 7.2

SPINAL INSTRUMENTATION CERVICAL SPINE

7.2.6

1

CERVICAL LAMINOPLASTY

INTRODUCTION

The history of laminoplasty started with the description of the technique by Hirabayashi in 1997 [1]. In 1983, he reported a technique known as the “expansive open-door laminoplasty”, which was stabilized by simple sutures on the hinge side. Since that time many modifications of the technique have been proposed. To name a few, the French door technique (or midline technique) and the Z-plasty technique represent the two basic variations of laminoplasty. Finally, an expansive open-door laminoplasty using minititanium plates and allograft spacers to keep the open door has also been described [2]. Results of cervical laminoplasty have been encouraging and Hirabayashi gives a recovery rate, after expansive open-door laminoplasty, of 62% after a 5-year follow-up [3]. The defi nitive advantage of cervical laminoplasty is its ability to maintain lordosis and the range of motion of the cervical spine [4]. Expansive open-door laminoplasty should be viewed as an alternative technique to anterior surgery or laminectomy in specific indications for cervical myelopathy [5].

2

GENERAL PRINCIPLES

Numerous techniques for cervical laminoplasty have been described, but the open-door technique has gained the most widespread acceptance. In this chapter the technique of cervical open-door laminoplasty with segmental instrumentation and fusion using the Arch fi xation system is described. The system offers a range of allograft spacers of different sizes (from 4 mm to 12 mm) and shapes (parallel or angled) in order to optimize the fit of the spacer. The titanium miniplates used for fi xation have a very low profi le, are prebent (single bend or double bend), and are available in five sizes corresponding to the Arch open-door laminoplasty spacers. Indications

Indications include, cervical myelopathy due to multilevel congenital spinal stenosis or developmental stenosis caused by spondylosis or ossification of the posterior longitudinal ligament with a lordotic or neutral sagittal cervical alignment.

330

7 7.2

Spinal instrumentation Cervical spine

3

Contraindications

• • • •

Single- or two-level stenosis. Kyphotic sagittal alignment. Severe anterior focal compression of the spinal cord. Predominance of radicular symptoms.

Advantages

• Laminoplasty is biomechanically superior to laminectomy. • It reduces postoperative kyphosis compared to laminectomy alone. • The incidence of perioperative complications is reduced compared to multilevel anterior decompression and fusion. Disadvantage

Cervical laminoplasty is technically demanding.

a

SURGICAL TECHNIQUE

A typical laminoplasty from C3 to C7 is described here in detail, but the procedure can be performed from C2 to the upper thoracic spine. The opening of the laminoplasty can be performed on either side of the spine. Factors, such as the side of predominant symptoms, maximum anatomical compression, or the presence of coexisting radiculopathy due to foraminal stenosis ( Fig 7.2.6-1a ), can influence which side is chosen for the opening. A Mayfield clamp is applied and the patient is placed in the prone position with the cervical spine in a neutral or mildly lordotic position. The shoulders can be taped, or the arms positioned for intraoperative longitudinal traction in order to facilitate radiographic exposure during the procedure. Lateral image intensifier control can be used for the localization of the level and for the placement of the instrumentation. A midline incision is performed from the spinous process of C2 to the tip of the spinous process of T1. A subperiosteal dissection is

b

Fig 7.2.6-1a–c a Axial view demonstrating the cervical spinal stenosis. b With a thin cut the lamina is separated on the side with the predominant symptoms. On the opposite side, a hinge is created in the lamina. c Posterior view of the vertebral levels showing the laminar cuts and the hinges.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c

331

7.2.6 Cervical laminoplasty

made of the spinous processes and laminae of C3 to C7 with care taken to preserve the intraspinous ligaments and the musculotendinous attachments to the spinous process of C2. The exposure is extended to the lateral aspect of the lateral mass on the open-door side with an attempt made to maintain the integrity of the facet joint capsules. On the contralateral side, the dissection is continued to the medial aspect of the lateral masses at each level. If a foraminotomy is needed, it is to be performed on the open-door side prior to performing the laminoplasty—take care to not resect excessive bone in order to avoid destabilization and to maintain an adequate site of attachment for the allograft spacer and the miniplate. A high-speed drill is used to transect the lamina at the junction of the lamina with the lateral mass. An M8 or equivalent small drill bit is used to thin the lamina to the yellow ligament while avoiding contact with the dura. A 1 mm Kerrison punch can complete the cut. On the contralateral side, the junction of the lamina with the lateral mass is decorticated to the deep

Fig 7.2.6-2 Posterolateral view showing the elevation of the lamina.

(anterior) cortex of the lamina ( Fig 7.2.6-1b–c ). The yellow ligament is released, and the epidural vessels are divided on the side of the opening. The opening of the lamina is performed either with gentle manipulation of the spinous process toward the contralateral side or with the use of a specially designed laminar elevator ( Fig 7.2.6-2 ). If there is resistance, with the gentle manipulation of the spinous process the surgeon should check to assure the opening had been completed and that the scoring on the contralateral side is sufficiently deep. Excessive force should not be applied. Once the lamina can be manipulated into an expanded position, a trial spacer is used to determine the proper size of the allograft spacer by inserting it into the newly created laminar gap ( Fig 7.2.6-3 ). The chosen Arch open-door laminoplasty allografts are inserted by slightly overdistracting the lamina and inserting it so the notched ends in the graft fit into the cut edges of the lamina and lateral mass. A specialized graft holder is available to facilitate graft placement. It is also an option to

Fig 7.2.6-3 Use a trial spacer and with the lamina in expanded position, determine the appropriate size and shape of the allograft.

332

7 7.2

Spinal instrumentation Cervical spine

attach the miniplate to the graft prior to insertion or to secure the miniplates after the grafts are in place ( Fig 7.2.6-4 ). A range of self-drilling and self-tapping screws exist, from which the most appropriate can be chosen to secure the miniplate to the lamina and lateral mass. The miniplate is centered in the middle of the lamina and lateral mass for ideal placement. A drill bit with a built-in stop is used for the preparation of the screw site for insertion of the self-tapping screws ( Fig 7.2.6-5a ). Alternatively, a drill guide can be utilized for the preparation of the screw hole. The screw length is selected to avoid excessive penetration of the lamina or violation of the facet joint during screw placement in the lateral mass area. Preoperative planning is essential to determine the thickness of the lamina and the lateral mass. Often two screws can be placed in both the lamina and lateral mass. After all screws are inserted ( Fig 7.2.6-5b–d , Fig 7.2.6-6 , Fig 7.2.6-7 ), a multilayer closure is performed. The use of a drain at the surgical site is left to the discretion of the surgeon.

a

b

c

d Fig 7.2.6-4 According to the shape and size of the allograft, the appropriate miniplate is chosen and secured on the allograft with a screw. The construct is now ready to be placed.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.2.6-5a–d a Preparation of screw hole. b–c Axial and posterolateral view demonstrating placement of the first self-tapping cortex screw. d Insertion of the last screw.

333

7.2.6 Cervical laminoplasty

Postoperative care

Patients are immobilized in a fi rm collar for a period of 6–8 weeks. They are allowed to remove the collar for daily care. After 6–8 weeks, the collar is gradually weaned and the patient is encouraged to begin using an active range of motion of the neck.

Fig 7.2.6-6 Open-door laminoplasty from C3 to C7 using the Arch system.

Fig 7.2.6-7 Intraoperative view showing completed laminoplasty. The interspinous ligament and capsules of the facet joints were left intact.

334

7 7.2

Spinal instrumentation Cervical spine

Case example

A 45-year-old male patient presented with a 3-year history of hand numbness and a 6-month history of weakness and spasticity in the upper and lower extremities, which progressed to a near inability to ambulate ( Fig 7.2.6-8 ). Follwing surgery, he regained his ability to walk but his gait remained spastic.

a

c

b

d

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.2.6-8a–d a Preoperative mid-sagittal T2-weighted MRI of a patient with cervical myelopathy. The sagittal alignment of the spine is lordotic. Spinal stenosis is present at three levels. b Preoperative flexion and extension views. c T2-weighted MRI 1 year postoperatively shows the spinal cord nicely decompressed after laminoplasty. d Lateral flexion and extension x-rays 3 years postoperatively.

7.2.6 Cervical laminoplasty

4

1.

2.

3.

4.

5.

BIBLIOGRAPHY

Hirabayashi K, Watanabe K, Wakano K, et al (1983) Expansive open-door laminplasty for cervical spinal stenotic myelopathy. Spine ; 8(7):693–699. Shaffrey CI, Wiggins GC, Piccirilli CB, et al (1999) Modifi ed open-door laminoplasty for treatment of neurological defi cits in younger patients with congenital spinal stenosis: analysis of clinical and radiographic data. J Neurosurg ; 90(2 Suppl):170–177. Satomi K, Ogawa J, Ishii Y, et al (2001) Short-term complications and long-term results of expansive open-door laminoplasty for cervical stenotic myelopathy. Spine J ; 1(1):26–30. Maeda T, Arizono T, Saito T, et al (2002) Cervical alignment, range of motion, and instability after cervical laminoplasty. Clin Orthop Relat Res ; (401):132–138. Truumees E, Herkowitz HN (2000) Cervical spondylotic myelopathy and radiculopathy. Instr Course Lect ; 49:339–360. Review.

335

336

7 SPINAL INSTRUMENTATION 7.3 THORACOLUMBAR AND SACROPELVIC SPINE 7.3.1 MODULARITY OF THE UNIVERSAL SPINE SYSTEM

1

Basic concepts ……………………………………………………………………………………………… 337

2 2.1 2.2 2.3 2.4 2.4.1 2.4.2 2.4.3 2.4.4 2.4.5 2.4.6 2.4.7

The system ………………………………………………………………………………………………… Fracture module …………………………………………………………………………………………… Low back surgery module ……………………………………………………………………………… Scoliosis and deformity module ………………………………………………………………………… Specific implants and instruments ……………………………………………………………………… USS II side-opening pedicle screws …………………………………………………………………… USS hooks—lamina hooks ……………………………………………………………………………… USS hooks—specialized pedicle hook ………………………………………………………………… USS hooks—transverse process hooks ………………………………………………………………… USS rod introduction into side-opening implants …………………………………………………… Complex reduction forceps—persuader ……………………………………………………………… Rod connectors ……………………………………………………………………………………………

3

Bibliography ………………………………………………………………………………………………… 355

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

339 341 342 343 345 345 347 348 350 350 351 354

337

Christopher Cain

7 7.3

SPINAL INSTRUMENTATION THORACOLUMBAR AND SACROPELVIC SPINE

7.3.1

1

MODULARITY OF THE UNIVERSAL SPINE SYSTEM

BASIC CONCEPTS

The universal spine system (USS) was developed with several basic concepts in mind. • Independent anchorage of hooks and/or screws to the spine. • Independent rod contouring and positioning. • Simple and uniform connection between the anchorage points (hooks or screws) and the rod. • Standardization and simplification of implants and instrumentation. • True “segmental” correction and stabilization. • Universal instrumentation and implants applicable for use in treatment of a range of spinal pathologies and versatile in their application. • Top-loading system.

These basic concepts allow a completely free choice with regard to anchorage points of the screws or hooks to the spine, utilizing the best possible anatomical locations. The connection between the rod and screws/hooks is uniform and can accommodate variable distances between the anchorage site, the hook or screw, and the rod. In addition, the rod can be contoured to the desired curvature of the spine, and the spine is reduced to the rod segmentally using unique, purpose-built reduction forceps. The anatomy of the deformed or degenerated spine does not always allow the placement of all the screws or hooks to be in a straight line. These implants should be placed in the appropriate anatomical position, without having to consider as a primary concern the position of the rod. It is also desirable to limit the number of implants required to facilitate reduction and

338

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

stabilization of the spine and to simplify their use, regardless of the pathology. Therefore, a concept was developed that allows a simple connection between the rods and the anchorage points (hooks and/or screws) to accommodate this anatomical variability.

the end vertebrae, cranially as well as caudally [3, 4]. These implants are then attached to the two vertical rods and the intervening vertebrae are then reduced individually to, and rotated around, the rods, which completes the structure of the frame.

In addition, the universal spine system was designed so that identical instrumentation can be used to couple the hooks and screws to the rod, and to enable the use of the screws and rods as an anterior system for deformity correction [1].

Experience with the Schanz screw used in the internal fi xator made it clear that for the manipulation of a vertebra the strongest lever arm is a pedicle screw with a long attached handle, which transforms it into something like a Schanz screw. In order to employ the power of this type of implant, instrumentation was designed that could convert the hook or screw, once attached securely to the spine, into a Schanz screw temporarily in order to facilitate the derotation maneuver. For this to be effective it is imperative that the anchorage of the implant to the vertebra is very strong. This can be readily achieved with a pedicule screw and, uniquely in the case of the USS, also with a specialized pedicle hook ( Fig 7.3.1-1) [5]. Over the last 15 years, clinical experience with the USS has shown its effectiveness in treating different pathologies [6–12].

Experience with the Cotrel-Dubousset instrumentation has shown that the concept of derotation in scoliotic spines through the rotation of the rod does not, in reality, always involve significant derotation of the affected vertebrae. When it does, this “global” derotation may be associated with the transmission of forces above or below the instrumented segments leading to postoperative imbalance or persistent deformity [2]. The Cotrel-Dubousset principle is based on the assumption that by adapting the rod to the curve in the frontal plane and attaching it along the concavity of the curve, the scoliosis in the thoracic spine, which is in reality a lordosis, can be transformed into a kyphosis by rotating the bent rod from the frontal plane into the sagittal plane. The spinal area which is fi xed with the instrumentation is pulled along with the rotated rod into the desired kyphotic form. This global rotation has a significant torque effect on the rest of the spine. The USS concept focuses on true segmental derotation, stable fi xation of implants to each vertebra, and true segmental reduction and derotation of the individual vertebrae. Clinical experience of scoliosis with the internal fi xator led to the development of the concept of a frame technique, whereby the frame is constructed by the anchorage of screws or hooks to

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.3.1-1 The specialized USS pedicle hook incorporates a fixation screw through the base of the pedicle, which provides stable fixation to the pedicle.

339

7.3.1 Modularity of the universal spine system

2

THE SYSTEM

The universal spine system (USS) was built on a modular concept. The three major areas of spine surgery are: • Trauma and other pathological entities causing mechanical insufficiency of the spine. • Deformity treatment for both children and adults. • Degenerative disease of the spine. The management of these areas presents unique clinical challenges and demands individual techniques and specific instrumentation solutions. The USS utilizes a common basic instrumentation set and additional modular instruments and implants to address each of the individual problems specifically. Since the introduction of the USS in the early 1990s several modifications have been incorporated into the system. This initially occurred through the introduction of the small stature USS-asset of implants and instruments using a 5 mm rod and lower profi le dual-opening implants, to which the rod can be coupled on either side of the implant ( Fig 7.3.1-2 ) reducing the inventory of implants required, and improving the ease of use and versatility of the system.

Further rationalization has occurred with the introduction of USS II, which effectively represents an amalgamation or merger of the USS and small stature USS sets. This enables the use of one versatile set of instruments and implants for adult and pediatric deformity indications. Implants and instruments

The fi rst common implant is the rod. Rods are smooth and are available with a 5 mm or 6 mm diameter; 6 mm hard (TiCP) and extra hard (TAN), and 5 mm extra hard (TAN) are available in lengths from 50 to 500 mm. Tapered rods are also available, 5–6 mm and 3.5–5.0 mm or 3.5–6.0 mm, which enable the transition of instrumentation constructs from the cervical through to the thoracic and into the lumbosacral region if required. Side-to-side and end-to-end connectors are also available in five different configurations, enabling the linkage of all of the various rod-to-rod combinations ( Fig 7.3.1-3 ). These devices have set screws that accept the hexagonal screwdriver from the small fracture fragment set.

a a b Fig 7.3.1-2a–b a Pedicle screw with dual opening and dual core. b Pedicle hooks.

b

Fig 7.3.1-3a–b a Extension connector. b Parallel connector. c Parallel connector for a 6.0/3.5 mm rod.

c

340

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Screws come in diameters from 4.2 mm to 9.0 mm and lengths from 25 mm to 80 mm in 5 mm increments. The same screw will accept either a 5 mm or 6 mm rod by selection of a different collar or sleeve ( Fig 7.3.1-4 ). A different sleeve with grooves is used to connect a screw or hook to a transverse connector ( Fig 7.3.1-5 ). The USS II also has a module for anterior surgery utilizing the same dual-core screw principle. The addition of a washer, single screw staple, and double screw staple (for anterior tworod system) add to the stability of the construct ( Fig 7.3.1-6 ).

In order to handle these implants properly, special instruments have been designed. There are standard instruments to hold and bend the rods, ie, a rod holder and rod-bending irons, with straight or angled notches for in situ contouring of the rod. The USS tubular rod bender enables the fabrication of rods that need more pronounced bends with a greater degree of control. USS bending pliers and a 5 mm or 6 mm bending template are also available. Distraction and compression forceps are included in the set. Cross-links were created to link the longitudinal bars together. Fixed to adjustable length cross-links are available in different sizes (Fig 7.3.1-7) and types (Fig 7.3.1-8). For the adjustable length cross-link, one of the clamps can be rotated or swiveled along its longitudinal axis to accommodate the local instrumentation anatomy.

Fig 7.3.1-5 Transverse connector or transverse bar.

Fig 7.3.1-4 Sleeve or collar for 6 mm rod.

a

b

c

d 45°

Fig 7.3.1-6a–c a Screw shaft with a dual-core design and a specific thread for cancellous and cortical bone. b Washer. c Staple. d Double screw staple.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

45° Fig 7.3.1-7 USS cross-link system with an angulation of ± 45°.

341

7.3.1 Modularity of the universal spine system

An easy-to-use, new low-profi le, cross-link clamp has also been introduced for use in fractures, deformities, tumors, and degenerative conditions ( Fig 7.3.1-8 ). Right and left couplings are supplied for 5 mm and 6 mm rods. One of each ( Fig 7.3.1-9a ) or two identical clamps can be used depending on the space available ( Fig 7.3.1-9b ). If the distance between the two rods to be connected is less than 30 mm, one of the two crosslink clamps must be replaced by a cross-link clamp with rod ( Fig 7.3.1-9c ). Couplings allow 15° angulation of the 3.5 mm diameter cross-link rod in either a cephalic or caudal direction relative to the vertical 5 mm or 6 mm rod ( Fig 7.3.1-9d ).

Fig 7.3.1-8 Low-profile cross-link connector.

+15° a 0° –15° b

c

2.1

FRACTURE MODULE

The fracture set contains specific implants with instrumentation similar to the basic USS set. The main component of this instrumentation has been adapted for use with transpedicular Schanz screws of 5.2, 6.2, or 7.0 mm diameter and five thread lengths from 35 to 55 mm. The Schanz screws have a bulletshaped tip for safer insertion, a double thread for easier insertion, and a dual core to optimize hold in the body of the vertebra and pedicle ( Fig 7.3.1-10 ). The Schanz screws provide a significant lever arm to facilitate reduction of the fracture and restoration of the sagittal contour of the spine. The second-generation internal fi xator clamp with posterior nuts ( Fig 7.3.1-11a )—a significant advance from the original, laterally tightened clamp—enables easy access to secure all components of the implants following completion of the reduction maneuver. A recent addition is the cranial clamp for use in patients with smaller anatomy ( Fig 7.3.1-11b ). Both clamps have a 36° arch of movement in the sagittal plane and a single set screw tightens the clamp onto the rod. If a larger degree of sagittal correction is needed, the rod can be contoured in situ to increase the amount of lordosis achieved. It is however, important to remember that it is not possible to slide the fracture clamp over a contoured rod.

d Fig 7.3.1-9a–d a Right and left cross-link clamps and rod. b Two left cross-link clamps and rod. c One cross-link clamp with rod and a right cross-link clamp. d Possible angulation of the cross-link rod.

Fig 7.3.1-10 Schanz screw with dual core and double thread.

342

7 7.3

a

Spinal instrumentation Thoracolumbar and sacropelvic spine

resist screw pullout in cases of osteoporosis and infi ltrative disease. Pedicle screws can also be used in conjunction with a fracture coupling and double threaded, specialized Schanz screws to facilitate reduction of a spondylolisthesis in a controlled and precise manner (chapter 7.4.5 Spondylolysis, spondylolisthesis—reduction and stabilization).

b

Fig 7.3.1-11a–b a Fracture clamp mounted on rod. b Cranial clamp with set screws (arrows).

Video 7.3.1-1 Implants for the instrumentation of the USS for thoracolumbar fracture treatment.

2.2

The side-opening screws have the same head configuration as the hooks, and can be coupled to either a 5 mm or 6 mm rod, as described above, by selection of the appropriate sleeve. The side opening allows the insertion of additional screws, if needed, once a rod is in place. Likewise, the side-opening screws or hooks can be removed at any time without having to take down the whole instrumentation. They are, therefore, easier to replace without removing the complete construct, if a revision is necessary. The fact that the rod can be located and secured in either side of the screw provides up to 4 mm offset from the center of one screw to the center of a 6 mm rod (Fig 7.3.1-12, see Fig 7.3.1-23), which can be utilized to accommodate minor fixed deformity or

LOW BACK SURGERY MODULE

The universal spine system can be used to manage a full range of degenerative low back pathologies, but the true value of the system becomes evident when the nature of the disease process necessitates instrumentation over longer segments, particularly in the presence of deformity. The lower profi le of the USS II reduces the chance of proximal implants impinging on the adjacent facet joint, and the double thread and dual core enhance fi xation in the pedicle and vertebral body. The versatility of the system also allows for the simple addition of supplementary implants (hooks) to improve the hold and

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

4 mm

4 mm

Fig 7.3.1-12 Extent of the lateral offset when using the two opposite side openings of the implants.

a

b

Fig 7.3.1-13a–b a Frontal-opening pedicle hook. b Lamina or transverse process hook.

343

7.3.1 Modularity of the universal spine system

variation in alignment of consecutive implants. This allows the rod to be more easily accommodated to various positions of the screws, reducing the need to contour or bend the rod. For even more versatility, rod connectors (see Fig 7.3.1-5 ) allow the complete adjustment of an independently inserted pedicle screw to the rod. The rod connector also allows indirect connection of the side-opening screw to the rod. In addition, the same connectors can be used with the frontal-opening pedicle hook ( Fig 7.3.1-13a) and the lamina hook ( Fig 7.3.1-13b), or the transverse process hook. Rod connectors can be connected to the rod once the rod and implants aligned with the rod have been secured.

2.3

True segmental derotation and translation can only be achieved when the implants are fi rmly anchored to the spine. Optimal fi xation to the spine is accomplished via a pedicle screw, but the use of this type of fi xation carries with it the additional risk of pedicle and canal penetration and spinal cord injury. In cases of pediatric deformity the concave pedicles are often hypoplastic, increasing the difficulty of pedicle screw placement. In order to optimize fi xation in this situation, specialized pedicle hooks have been designed to allow fi xation to the adjacent pedicle and the vertebra via a 3.2 mm self-tapping screw. The hook is positioned and held, while a guide is used to drill through the base of the inferior aspect of the pedicle and through the superior end plate of the vertebra (see Fig 7.3.1-1). These special hooks allow, for the fi rst time, translation and rotational

SCOLIOSIS AND DEFORMIT Y MODULE

The scoliosis instrumentation is the most innovative and was developed around a new concept of scoliosis treatment. At its introduction, the USS offered for the fi rst time true segmental instrumentation, correction, and realignment of the spine to the sagittally placed rods. This includes reducing the spine to the rod and enabling true rotation of the vertebral segment around the rod through the Schanz pin extensions, which are applied to implants securely fi xed to individual vertebrae. Fixation of the individual implants, pedicle hooks, and screws to each vertebra avoids the need to apply distraction in order to keep hooks in place. This reduces tension on the paravertebral soft tissues and optimizes the ability of the surgeon to realign the spine. The modifications introduced with the release of the USS II enabled application of the rod to either side of the hook or screw, and the use of common implants and instruments when using either a 5 mm or 6 mm rod. This reduces the inventory of implants required and further increases the versatility of the system.

Fig 7.3.1-14 Hook/screw holder (mounted on the handle for hook/screw holder) acts as an extension to the hook or screw to facilitate placement of the whole implant (screws, hooks, sleeves, or collars).

344

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

forces to be applied directly to the vertebra without the hooks loosening or becoming dislodged, and they do not require distraction to keep them in place. The mechanism locking the hooks to the rod is exactly the same as for screws. The hooks and screws are inserted and manipulated with the screw holders, which are often called “sticks” and are fi xed within the hook or screw head, extending the lever arm imitating a Schanz screw ( Fig 7.3.1-14 ) and facilitating the application of the sleeves and locking nuts. Since all the hooks open either laterally or frontally, additional hooks or screws can be added to supplement fi xation if needed at any time during the insertion or fi xation of the implants and rods.

The side openings in the hooks or screws are especially useful in the treatment of kyphosis, where fi xation of the kyphosis from a cranial to a caudal direction allows progressive and gradual correction of the deformity. The addition of supplementary lamina hooks at the end of the construct to “protect” the distal pedicle screws and prevent their pullout is another feature of the system. The open USS implants simplify pre- and intraoperative planning, and allow easy connection to either the 5 mm or 6 mm rods as indicated by the clinical circumstances. Manipulation of the implanted vertebrae is made possible by use of the Schanz pin extensions and the complex reduction

Pedicle hooks ( Fig 7.3.1-15a ) and lamina hooks ( Fig 7.3.1-15b ) come in different sizes to accommodate the variations of the local anatomy in the thoracic and lumbar spine. Pedicle hooks should only be inserted between T1 an T9. Lamina hooks can be inserted from T1 to L5. There are also right and left angled lamina or transverse process hooks ( Fig 7.3.1-15c ). All three hook configurations are also available with a frontal opening for use with a rod connector (see Fig 7.3.1-13 ).

a

b

c

Fig 7.3.1-15a–c Dual-opening design of the hooks allows rod attachment from either side of the hook. a Pedicle hook. b Lamina hook. c Transverse process hook.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.3.1-16 The vertebra is translated posteriorly and laterally to the fixed, sagitally oriented rod with the complex reduction forceps or persuader.

345

7.3.1 Modularity of the universal spine system

forceps (rod introduction pliers), often referred to as the persuader. This instrument allows the simultaneous elevation and translation of the spine and facilitates the coupling of the implanted hook or screw to the rod in one maneuver ( Fig 7.3.1-16 ). This underlines the concept of a modular system with a high degree of versatility. 2.4

SPECIFIC IMPLANTS AND INSTRUMENTS

2.4.1 USS II SIDE-OPENING PEDICLE SCREWS Insertion technique

Side-opening screws are inserted while attached to an implant holder (stick) held in a universal handle (Fig 7.3.1-17a). The stick is connected to the screw by rotating the coupling extension, which allows the thread of the stick to engage the top of the screw ( Fig 7.3.1-17b ); alternatively, the implant holder can be tightened to the screw by hand. The screw can now be inserted into the bone ( Fig 7.3.1-17c ). The screw holder is disconnected from the handle by pressing the top of the handle ( Fig 7.3.1-17d ). A 5 mm or 6 mm rod can now be inserted into the screw. The rod is then held in place with the appropriate sleeve, for either the 5 mm or the 6 mm rod (see Fig 7.3.1-4 ), and secured with a locking nut ( Fig 7.3.1-18 ). The same locking nut is used to secure all collars or sleeves and all implants. The sleeve has an eccentric opening for the rod and is introduced over the stick onto the hook/screw and rod. The nut and sleeve can be inserted together onto the hook/screw-rod complex using the holding sleeve. The nut can be tightened with the handle. Final tightening of the nut is performed using the 11 mm L-handle socket wrench and the 6 mm socket wrench, which are used to engage the top of the stick in order to control the screw to prevent torque of the implant when the nut is tightened ( Fig 7.3.1-19 ).

a

Fig 7.3.1-17a–d a Attach the handle to the stick by pressing the knurled button on the upper end of the handle and simultaneously attaching the hook and the screw holder to the handle. b Pick up the implant by connecting the dual-opening implant to the stick and rotating the release button to the handle.

b

346

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Fig 7.3.1-17a–d c Insert the pedicle screw into the prepared pedicle until the screw head is well seated. d Disconnect the stick from the handle by pressing the release button.

c

d

1

2

3

Fig 7.3.1-18 Different collars and nuts used to secure the construct. 1 collar for 6 mm rod 2 collar for 5 mm rod 3 collar with grooves for transverse connector

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.3.1-19 Final tightening of the locking nut using the 11 mm L-handle socket wrench. The 6 mm T-shape socket wrench is used for applying countertorque.

347

7.3.1 Modularity of the universal spine system

2.4.2 USS HOOKS—LAMINA HOOKS Insertion technique

The lamina hook (see Fig 7.3.1-15b ) can be placed around either the superior or inferior portion of the lamina. The yellow ligament is carefully removed with a rongeur. A small portion of the lamina is then removed with a Kerrison punch to square off the lamina in order to ensure the hook fits snugly ( Fig 7.3.1-20a ). The lamina feeler is carefully put around the lamina to ensure a close fit for the hook ( Fig 7.3.1-20b ). The appropriate-size hook is mounted onto a stick and attached to a universal handle. The lamina hook is gently eased into place around the lamina. Additional control during this maneuver

is provided by using the hook positioner ( Fig 7.3.1-20c ). It is important to ensure that a hook of the appropriate size is selected to prevent the foot of the hook encroaching unduly on the dura and the spinal cord. The inferior part of the hook must fit closely to the anterior surface of the lamina ( Fig 7.3.1-20d ). The hook is attached to the implant holder in the same manner as the screw. All of the various hooks are attached to either a 5 mm or a 6 mm rod using the same technique as with a screw. The lamina hooks should be removed from the spine until the surgeon is ready to attach them to the rod in order to avoid putting the spinal cord under pressure.

d

a

b

Fig 7.3.1-20a–d Insertion of a supralaminar hook. a A Kerrison punch squares off a small notch at the tip of the lamina. b The lamina feeler prepares the final placement of the hook.

c

c–d The lamina hook is correctly placed with the help of the hook positioner.

348

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

2.4.3 USS HOOKS—SPECIALIZED PEDICLE HOOK

The special pedicle hook (see Fig 7.3.1-1 ) is attached to the screw holder and universal handle. They can be inserted between T1 and T9. Bone preparation

The level in the thoracic spine for the insertion of the specialized pedicle hook is identified. The pedicle feeler is placed between the inferior and superior facet joints. Ensure that the feeler is placed into the articular space and not into the bone of the inferior facet. To facilitate insertion of the pedicle hook, a small portion of the inferior facet may need to be removed. Ensure the pedicle feeler is around the pedicle by exerting a cephalad and lateral force, making sure the tongues of the pedicle feeler are engaged around the inferior aspect of the pedicle ( Fig 7.3.1-21a ). Do not push medially. Once the feeler is securely seated around the pedicle, it may be necessary to remove further bone from the inferior facet to allow proper seating of the pedicle hook around the pedicle and inferior aspect of the lamina. The pedicle feeler has six lines on the blade; when the last line is reached, sufficient bone has been removed to accommodate the hook around the pedicle. The osteotome provided is used to remove this bone, and should only be applied in combination with the pedicle feeler ( Fig 7.3.1-21b ). The osteotome is directed down onto the pedicle feeler in order to prevent accidental penetration of the spinal canal.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Insertion technique

The specialized pedicle hook, which is attached to a stick and universal handle, is now inserted into the prepared site. The hook positioner is placed into the small hole in the back of the pedicle ( Fig 7.3.1-21c ). The hook is then eased around the inferior aspect of the pedicle. Once in place, check the hook is properly seated around the pedicle by exerting a cephalad and lateral load. No displacement of the hook should occur. If the hook does not move it is around the pedicle. The hook positioner is then gently tapped with a hammer to fi nally seat the hook in place. The pedicle hook drill guide is inserted into the hole in the back of the pedicle hook ( Fig 7.3.1-21d ). A 2.0 mm drill bit is placed into the drill guide and, using the oscillating drill attachment, the tip is advanced until it passes through the inferior aspect of the pedicle to the superior end plate of the vertebra. The stick should be perpendicular to the posterior aspect of the spine prior to commencing, and it is important to ensure there is bone at the tip of the drill before activating the power. The drill bit should be advanced slowly. The drill guide is removed and the depth to the end plate is measured with a small depth gauge (this is usually 25–30 mm, Fig 7.3.1-21e ). A self-tapping 3.2 mm cortex screw is inserted ( Fig 7.3.1-21f ). The specialized pedicle hook is now fi rmly fi xed to the pedicle and end plate.

349

7.3.1 Modularity of the universal spine system

3–4 mm

a

d

b

e

Fig 7.3.1-21a–f a Insertion of the pedicle feeler until it fits the inferior edge of the pedicle. b Osteotomy of the medial part of the inferior facet until the 6th line of the pedicle feeler is exposed. c Insertion of the pedicle hook with the hook positioner.

c

f

d Drilling with a 2 mm oscillating drill bit. The depth distance is set at 20, then 25, and 30 mm until the path is felt to be bicortical. e Depth of the screw path is measured with a depth gauge of the small fragment set. f Insertion of the 3.2 mm self-tapping screw.

350

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

2.4.4 USS HOOKS—TRANSVERSE PROCESS HOOKS

2.4.5 USS ROD INTRODUCTION INTO SIDE-OPENING IMPLANTS

Technique

The chosen transverse process is cleared of soft tissue. The lamina feeler is placed around the transverse process elevating the soft-tissue attachments from its anterior part ( Fig 7.3.1-22a ). An appropriate-size lamina hook is mounted onto a stick with handle, and the hook is eased around the transverse process using the hook positioner ( Fig 7.3.1-22b ).

a

Technique

Placement of the 5 mm or 6 mm rod may be achieved with the rod holding forceps, in which case the collar and nut are applied with the holding sleeve as described above. The bilateral opening of the screw/hook heads results in a potential offset, of 7.8 mm when a 5 mm rod is used and 8.8 mm if a 6 mm rod is used, from the center of one implant relative to an adjacent implant with the rod placed on the opposite side ( Fig 7.3.1-23 ). This enables implants to be placed with greater flexibility both before and after placement of the rod. Such an offset is also of great advantage when creating claws, such as the laminopedicle-hook claw, or the lamina-hook/pedicle-screw claw.

b

Fig 7.3.1-22a–b a The lamina elevator is placed above the transverse process and cuts the superior costovertebral ligament to allow for the placement of the hook. b Insertion of the hook above the transverse process with the help of the hook positioner.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

7.8 mm for 5 mm rod 8.8 mm for 6 mm rod

Fig 7.3.1-23 Introduction of the rod into the side opening of the lamina-hook/pedicle-hook claw construct.

351

7.3.1 Modularity of the universal spine system

2.4.6 COMPLEX REDUCTION FORCEPS—PERSUADER Principle

At times, it is not possible to maintain the position of the rod within the opening of the screw or hook to enable placement of the collar and nut. The distance between the rod and implant may also prevent this. The complex reduction forceps or so-called persuader can be used to lift and translate the screw/vertebra to the rod so that the side opening of the screw engages into the rod; this relationship must be maintained during the application of the collar. Technique

A holding sleeve is placed over the shaft of the persuader ( Fig 7.3.1-24a ). A collar is loaded from the holding tray ( Fig 7.3.1-24b ) with the small arm of the collar facing in the direction of the rod. The collar clips into place. The persuader is then introduced over the stick, which is attached to the implant that is to be reduced to the rod ( Fig 7.3.1-24c ). The free limb of the reduction forceps is placed on the rod. The stick clip is then placed on the top of the stick ( Fig 7.3.1-24d–e ). The forked opening of the stick clip must face upward to engage the stick correctly. Controlling the stick clip will prevent rotation of the hook or screw. If translation alone is required to facilitate reduction of the implant to the rod, then the forceps are gently closed. The persuader is a powerful instrument and care must be taken to ensure the implant is not dislodged by excessive force being applied to a rigid deformity.

If elevation and translation are required, then, while controlling the forceps and ensuring the limb of the forceps remains on the rod, a spreader is placed between the stick clip and the top of the cylinder of the persuader. The spreader is then slowly opened, elevating the hook or screw and the attached vertebra upward toward the rod ( Fig 7.3.1-24f ). When the side opening of the implant is opposite the rod, the persuader is closed, allowing the implant to engage the rod ( Fig 7.3.1-24g ). The sleeve is pushed down the cylinder and placed over the rod and implant. If the sleeve will not ease onto the implant, place the hook positioner into the hollow on the sleeve of the persuader and gently tap the sleeve into place ( Fig 7.3.1-24g ). The stick clip and the persuader are removed and a nut dropped over the stick and loosely attached to the hook or screw.

Video 7.3.1-2 Deformity correction demonstrated on a wooden model using the USS II.

352

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

d

a

b

c e

Fig 7.3.1-24a–g a The holding sleeve fits in the shaft of the persuader. b The persuader picks a collar from the holding tray. c Introduction of the persuader over the stick. d–e Distraction clip is placed over the stick.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

353

7.3.1 Modularity of the universal spine system

*

f

Fig 7.3.1-24a–g f The distraction forceps (asterisk) is applied against the distracting clip while squeezing the rod introduction pliers until the rod is seated into the hook or screw. g The hook positioner is used to tap down the collar retaining sleeve until it is seated on the rod.

g

354

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

2.4.7 ROD CONNECTORS Principle

There are times when the nature or rigidity of a spinal deformity or just the spinal anatomy will make it impossible to reduce the implant to the rod using the persuader. It is important to note that if too much force is exerted on the implant by the persuader, the implant will cut out of the bone. When this situation becomes evident it is necessary to use the rod connectors to complete the construct. When a rod connector is required, it is necessary to use the frontal-opening hooks, and screws must be turned 90° to enable coupling of the implant to the connecting rod.

the connector clamp as this will prevent its reduction over the rod. The ribbed part of the connecting rod is then placed in the screw or frontal-opening hook. The sleeve and nut are then dropped onto the implant using the holding sleeve and the nut is secured ( Fig 7.3.1-25b ). If required, the screw or hook can be further approximated to the rod with the compression forceps if required. The nut is fi nally tightened on the implant using the socket wrench with the L-handle, and rotation of the implant as it is tightened is controlled by the small 6 mm socket wrench applied to the top of the stick. A derotation force can be applied to the stick while it is attached to the implant prior to fi nal tightening of the small screw securing the connecting rod to the rod.

Technique

The appropriate-length rod connector (range 15–25 mm) is placed on top of the rod ( Fig 7.3.1-25a ). The position of the rod and implant dictates this position. The small hexagonal screwdriver can be used to help manipulate the connecting rod while it is being positioned. Ensure the screw is not protruding inside

Fig 7.3.1-25a–c a Positioning of the transverse bar to the rod. b The transverse bar is fitted into the frontal-opening hook, and the locking mechanism (collar and nut) is dropped onto the hook. c The screw is turned 90° to make it a frontal-opening screw.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

a

c

355

7.3.1 Modularity of the universal spine system

3

BIBLIOGRAPHY

Laxer E (1994) A further development in spinal instrumentation. Technical Commission for spinal surgery of the ASIF. Eur Spine ; 3(6):347–352. 2. Moore MR, Baynham GC, Brown CW, et al (1991) Analysis of factors related to truncal decompensation following Cotrel-Dubousset instrumentation. J Spinal Disord ; 4(2):188–192. 3. Aebi M, Etter C, Kehl T, et al (1988) The internal skeletal fi xation system. A new treatment of thoracolumbar fractures and other spinal disorders. Clin Orthop Relat Res ; (227):30–43. 4. Aebi M (1988) Correction of degenerative scoliosis of the lumbar spine. A preliminary report. Clin Orthop Relat Res ; (232):80–86. 5. Arlet V, Papin P, Marchesi D, et al (1999) Adolescent idiopathic thoracic scoliosis: apical correction with specialized pedicle hooks. Eur Spine J ; 8(4):266–271. 6. Remes V, Helenius I, Schlenzka D, et al (2004) Cotrel-Debousset (CD) or Universal Spine System (USS) instrumentation in adolescent idiopathic scoliosis (AIS): comparison of midterm clinical, functional, and radiologic outcomes. Spine ; 29(18):2024–2030. 7. Burwell RG, Aujla RK, Cole AA, et al (2002) Anterior universal spine system for adolescent idiopathic scoliosis: a follow-up study using scoliometer, realtime ultrasound and radiographs. Stud Health Technol Inform ; 91:473–476. 8. Macchiarola A, Di Carlo FP, Di Pietro FP, et al (2000) USS internal fi xator in lumbar and thoracolumbar vertebral fractures. Chir Organi Mov ; 85(2):177–184. 9. Lindsey RW, Dick W (1991) The fixateur interne in the reduction and stabilization of thoracolumbar spine fractures in patients with neurologic deficit. Spine; 16(3 Suppl):S140–145. 10. Marre B (2005) Management of posttraumatic kyphosis: surgical technique to facilitate a combined approach. Injury ; 36 (Suppl 2):B73–81. 1.

11. Jia QZ, Gao JC, Zhang CM (2003) [Surgical treatment of kyphosis with universal spine system.] Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi ; 17(4):276–278. 12. Berlet GC, Boubez G, Gurr KR, et al (1999) The USS pedicle hook system: a morphometric analysis of its safety in the thoracic spine. Universal Spine System. J Spinal Disord ; 12(3):234–239.

356

7 SPINAL INSTRUMENTATION 7.3 THORACOLUMBAR AND SACROPELVIC SPINE 7.3.2 STABILIZATION AND RECONSTRUCTION TECHNIQUES FOR THE THORACOLUMBAR SPINE (FRACTURES, TUMORS, DEGENERATIVE)

1

Introduction ………………………………………………………………………………………………… 357

2 2.1 2.1.1 2.1.2 2.1.3 2.1.4 2.1.5

Posterior instrumentation techniques ………………………………………………………………… Universal spine system (USS) —fracture module …………………………………………………… Reduction and fixation of fractures with intact posterior wall (type A1 and A2) ……………… Reduction and fixation of fractures with fractured posterior wall (type A3) …………………… Reduction and fixation of transverse bicolumn fracture (B2.1) …………………………………… Reduction and monosegmental fixation of predominantly ligamentous injuries (B1.1) ……… Reduction and fixation of type C injuries ………………………………………………………………

357 359 362 364 367 368 369

3 3.1 3.2 3.2.1 3.2.2 3.2.3 3.2.4 3.2.5 3.3 3.3.1 3.3.2

Anterior instrumentation techniques ………………………………………………………………… Thoracolumbar spine locking plate (TSLP) …………………………………………………………… Anterior titanium rod systems …………………………………………………………………………… Double-rod clamp ………………………………………………………………………………………… Fracture clamp ……………………………………………………………………………………………… Single-clamp-double-rod configuration ……………………………………………………………… Single-rod configuration ………………………………………………………………………………… Telescoping fixation system (Telefix) ………………………………………………………………… Anterior application of the USS for thoracolumbar stabilization ………………………………… Surgical technique—anterior construct ………………………………………………………………… Surgical technique—anterior vertebral body construct ………………………………………………

371 371 373 374 377 377 378 379 380 380 382

4 4.1 4.2

Anterior reconstruction with cage technology ……………………………………………………… 384 E xpandable cage …………………………………………………………………………………………… 384 Mesh cage ………………………………………………………………………………………………… 387

5

Bibliography ………………………………………………………………………………………………… 389

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

357

Dante G Marchesi

7 7.3

SPINAL INSTRUMENTATION THORACOLUMBAR AND SACROPELVIC SPINE

7.3.2

STABILIZATION AND RECONSTRUCTION TECHNIQUES FOR THE THORACOLUMBAR SPINE (FRACTURES, TUMORS, DEGENERATIVE)

1

INTRODUCTION

The stabilization of fractures in the thoracolumbar spine can be obtained through an anterior or posterior fi xation technique with or without the association of an anterior column recon struction. The choice will be dictated by biomechanical factors, such as the amount of spinal instability, physical constraints, patient’s anatomy and age, and other considerations.

2

POSTERIOR INSTRUMENTATION TECHNIQUES

Modern posterior instrumentations of the thoracolumbar spine are nowadays based on pedicle screws. These techniques require perfect knowledge of the anatomy and the size of the pedicles (chapters 6.4 Thoracic spine, 6.6 Lumbar spine and lumbosacral junction). The diameter of the pedicles varies from patient to patient and depends on the thoracolumbar level in question. The exact assessment of the pedicles is an essential preoperative requirement and can be done on plain x-rays or preferably on CT scans. Anatomical landmarks are used to identify the entry points of the screws. In the thoracic spine the entry points are located just below the inferior facet at the superior aspect of the root of the transverse process at its junction with the superior facet ( Fig 7.3.2-1). In the lumbar spine these correspond to the intersection of a horizontal line bisecting the transverse process with a vertical line tangential to the lateral border of the facet joint ( Fig 7.3.2-2 ). Entry points are marked using a pedicle awl to perforate the posterior cortex. Proper screw placement in the sacrum is more difficult due to its variable anatomy. The entry point is located on a

358

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

vertical line tangential to the lateral border of the facet joint, just underneath the superior facet. Partial osteotomy of the inferior facet of L5 may help proper identification for the S1 pedicle screw placement. The most solid biomechanical purchase will be obtained by aiming and passing a screw through the anterior corner of the promontorium with a 15–20° convergent direction toward the midline ( Fig 7.3.2-3 ).

20° 0°

a

b

c 7–10°

7–10°

5–10°

a

b

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

5–10°

L1

15–20°

c

15–20°

L5

Fig 7.3.2-1a–c Determining the position of the transpedicular screw in the thoracic spine. a The entry point is just below the rim of the upper facet joint. b–c The screw should be angled 7–10° toward the midline and 0–20° caudally depending on wether a monoaxial screw (0°) or a polyaxial screw (0–20°) is used. Fig 7.3.2-2a–c Determining the position of the transpedicular screw in the lumbar spine. a The entry point for the pedicle is at the intersection of a vertical line tangential to the lateral border of the superior articular process and a horizontal line bisecting the transverse process. b–c The screws should converge by 5–10° in the upper lumbar spine increasing to 15–20° at L5.

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

15–20°

15–20°

2.1

UNIVERSAL SPINE SYSTEM (USS) —FRACTURE MODULE

Principles

a

b

c

Fig 7.3.2-3a–c a The entry point in the S1 pedicle is located at the intersection of a vertical line tangential to the lateral border of the S1 facet and a horizontal line tangential to its inferior border. b–c The screws converge toward the midline by 15–20° and aim toward the anterior corner of the promotorium. At this level the screw can be inserted bicortically.

The internal fi xator—ie, the USS fracture module—allows stabilization and reduction of the spine using the USS transpedicular Schanz screws. The original concept was described by Dick in 1985 [1]. The USS transpedicular Schanz screws are provided with a unique screw thread with a dual core for a stable anchorage. They are available in three diameters, 5.2 mm, 6.2 mm, 7.0 mm, and five thread lengths, 35–55 mm. They are advanced into the pedicle with a T-handle and are connected to a 6.0 mm (diameter) hard rod using the specially designed USS clamps, which are fully adjustable posterioropening clamps. Each clamp allows a 36° range of motion of the screws in the sagittal plane ( Fig 7.3.2-4 ). The implants act as a tension band, a buttress, and a neutralization system. They allow lordosis, distraction, compression, as well as fi xation in a neutral position. The fulcrum of corrective forces can be adjusted thanks to the USS half rings. All posterior instrumentation, if used by itself without interbody fusion, will need to be complemented by a posterolateral intertransverse fusion. After fracture reduction and stabilization, anterior surgery may be required for biomechanical purposes in case of significant vertebral body comminution, osteoporosis, or incomplete clearance of the spinal canal with persistant neurological deficit. Indications

Indications include fractures in the low thoracic and lumbar spine or alternatively, stabilization of the degenerative spine or tumors of the spine.

359

360

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Advantages

• • • • • •

The USS is a simple versatile system. It can be used for all types of fractures. It offers short segmental fi xation (bi- or monosegmental). The use of Schanz screws allows easy reduction. Clearance of the canal is assured by ligamentotaxis. Half-ring clamps can move the fulcrum of the corrective forces away from the posterior wall.

• Indirect decompression of the spinal canal should not be done in the presence of a reverse cortical sign (AOSpine Manual—Clinical Applications, 1.3.2 Thoracolumbar and lumbar spine). • A bursitis over the end of the Schanz screws can develop in these patients.

Surgical technique Disadvantages

• The USS fracture module cannot be used in the upper thoracic spine because the pedicles are too small and the instrumentation may be too prominent at these levels. At these levels the dual-opening USS should be used instead. • Additional anterior column reconstruction (cage or bone graft) may be necessary when there is a persistent anterior column defect.

a

b

Fig 7.3.2-4a–b a USS fracture clamps allow the screws a range of motion of ±18° in the sagittal plane. b The single set screw (arrow) tightens the clamp onto the rod, which is inserted into the posterior nut.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

After a posterior midline approach and subperiosteal dissection to the tip of the transverse processes, the entry points of the Schanz screws are determined according to the pedicle landmarks. The entry point is started with the pedicle awl ( Fig 7.3.2-5a , Fig 7.3.2-6a ) or a 2 mm high-speed burr that perforates the posterior cortex for approximately 5 mm. The entry point is deepened with the straight and blunt pedicle probe for 3 cm ( Fig 7.3.2-5b, Fig 7.3.2-6b ). The direction of the pedicle probe may be slightly ascending toward the superior end plate of the cephalad vertebra, and slightly descending toward the rostral vertebra for a biomechanical advantage (see case example). Using a 2 mm blunt tip probe or a depth gauge, the medial pedicle wall is palpated to ensure that the path of the pedicle probe has remained intraosseous and does not breach the anterior cortex ( Fig 7.3.2-6c ). Blunt K-wires can be inserted in the pedicle

a

b Fig 7.3.2-5a–b Instruments for the preparation of the screw canal: USS pedicle awl (a) and blunt USS pedicle probe with a 3.8 mm diameter (b).

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

probe path and checked with AP and lateral fluoroscopy. The 6.2 or 7.2 mm Schanz screws are then inserted with a simple handle ensuring they do not perforate the anterior cortex ( Fig 7.3.2-6d ). This is checked under lateral fluoroscopic control and fi nally with an AP control, thereby making sure that the tips of the Schanz screws do not cross the midline. The subsequent reduction technique depends on the type of fracture.

When all four Schanz screws have been inserted, the rods of the fracture module are applied to the Schanz screws using fracture clamps with the rods lying medially to the Schanz screws. The clamps are left loose ( Fig 7.3.2-6e–f ).

3 cm

a

e

b

c

f

d

Fig 7.3.2-6a–f a–b The entry point for the Schanz screws is started with the pedicle awl and deepened with the pedicle probe. c Insertion of temporary K-wires in order to check proper positioning and length with fluoroscopy. d K-wires are replaced by USS Schanz screws, which are advanced into the pedicle using a T-handle. e 6.0 mm hard rods are connected to the Schanz screws using the USS fracture clamp (arrow). f The small posterior set screws (arrow) are initially left loose (in case of a fracture of the posterior wall, half rings have to be inserted).

361

362

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

2.1.1 REDUCTION AND FIX ATION OF FRACTURES WITH INTACT POSTERIOR WALL (T YPE A1 AND A2)

The posterior ends of the Schanz screws are manually approximated until the desired correction of the kyphosis has been attained ( Fig 7.3.2-7a–b ). The set screws on the clamps must remain loose so that the clamps can slide freely toward each other during the reduction maneuver. The center of rotation then lies at the posterior edge of the vertebral body. By creating the lordosis, the vertebral body will be distracted anteriorly and disc space and disc height will be restored by ligamentotaxis. Place the 11 mm cannulated socket wrenches over the caudal Schanz screw and tilt them cranially to create lordosis in the spine ( Fig 7.3.2-7c ). The posterior nuts are locked. The same procedure is performed on the cranial Schanz screw in

a

order to reestablish the correct sagittal plane ( Fig 7.3.2-7d ). The appropriate posterior nuts are tightened to fi x the angle between the Schanz screws and the rods. At this stage, it is necessary to distract the Schanz screws to reestablish the normal height of the injured segment (disc). A half ring is placed and locked in the center of each 6 mm rod between the clamps (Fig 7.3.2-7e). Distract the spreader forceps (Fig 7.3.2-7f–g) and check the procedure with the image intensifier. When the desired distraction is obtained, tighten the set screws ( Fig 7.3.2-7h ) and remove the half rings. Alternatively, the 6 mm rod holders can be used instead of the half rings ( Fig 7.3.2-7g ). This distraction force reconstitutes the height of the anterior portion of the vertebral body, relieves the pressure from the intervertebral disc, and completes the reduction.

b

Fig 7.3.2-7a–j a–b Manual reduction maneuver for a type A fracture with intact posterior wall (arrow). Correction of the kyphosis by manual approximation of the posterior ends of the Schanz screws.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

2

1 2

c

d

e

f

* g

Fig 7.3.2-7a–j c–d Manipulation of the Schanz screws into their final position using the 11 mm socket wrench (1). Once the appropriate lordosis is obtained, the nuts are tightened (2) to lock the Schanz screws in their final position (caudal and cranial). e–g A half ring is placed in the center of each 6 mm rod and distraction is applied with the spreader forceps. Alternatively, the 6 mm rod holder (arrow) can be used instead of the half ring (asterisk).

363

364

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

2.1.2 REDUCTION AND FIX ATION OF FRACTURES WITH FRACTURED POSTERIOR WALL (T YPE A3)

In type A3 injuries ( Fig 7.3.2-8a ), there is a theoretical danger that the posterior wall fragments might displace posteriorly into the spinal canal during the correction of the kyphosis by compressing the posterior-opening clamps of the Schanz screws. It is important to protect the posterior wall against compression. Distraction is used to reconstitute the height of the vertebral body.

h

i

j

Fig 7.3.2-7a–j h–i Final reduction with restored height and spinal alignment— the set screws are tightened (arrow). j The protruding ends of the USS Schanz screws are cut using the appropriate bolt cutter.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Two 5 mm half rings are placed on each of the 6 mm rods prior to reduction with the Schanz screws; a distance of 5 mm between the half rings or the rod holder and the clamp is allowed for every 10° of attempted kyphosis correction ( Fig 7.3.2-8b ). When approximating the ends of the Schanz screws, the clamps will soon touch the half rings, and the center of rotation is transferred posteriorly to the level of the rods instead of the posterior wall as described previously in this chapter (2.1.1 Reduction and fi xation of fractures with intact posterior wall). The force now required to correct the kyphosis is much greater. The lordosis is checked with a lateral image intensifier view. The posterior-opening nuts are tightened to secure the correction and the set screws on the clamps are fi xed ( Fig 7.3.2-8c ). This procedure is repeated for the other Schanz screws ( Fig 7.3.2-8d ). The half rings are removed. Distraction, as described for type A1 and A2 fractures, is now exerted to restore the original height of the vertebra ( Fig 7.3.2-8e–f ).

365

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

10° 5 mm = 10°

* * a

5 mm = 10°

10°

b

1 2

2

1

c

d

e

f

Fig 7.3.2-8a–f a Type A3 burst fracture (with fractured posterior wall). b Half rings (asterisks) are mounted on the rods. For each 10° of kyphosis correction, a 5 mm distance between the half ring and the clamp is needed.

c–d Using the cannulated socket wrench (1), the caudal and the cranial nuts are tightened (2) to secure the correction. e–f Distraction is applied with the spreader forceps and final correction is achieved.

366

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Video 7.3.2-1 Application of the USS for thoracolumbar fractures.

b

a

Case example

A 26-year-old female suffered an A3 burst fracture of L2 without any neurological deficit after a motor vehicle accident ( Fig 7.3.2-9a ). The injury was reduced and fi xed posteriorly with the fracture module of the universal spine system ( Fig 7.3.2-9b ). The x-rays taken at the 3-year follow-up examination show a good result ( Fig 7.3.2-9c ). The vertebral body height has been restored and the segmental lordosis maintained without any loss of correction.

c

Fig 7.3.2-9a–c a A3 burst fracture of L2. b Postoperative AP and lateral x-rays. c Follow-up x-rays 3 years after surgical treatment with the USS fracture module. (Observe the slightly divergent orientation of the Schanz screws for better biomechanics).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

367

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

2.1.3 REDUCTION AND FIX ATION OF TRANSVERSE BICOLUMN FRACTURE (B2.1)

In these pure bony Chance fractures, the implant acts as a pure tension band. Therefore, after reduction in the manner described above with the clamp elements sliding freely, no distraction is carried out, but slight compression is applied ( Fig 7.3.2-10a–c ). To achieve compression, the half rings or the 6 mm rod holders are placed between the posterior- opening clamps and the compression forceps are applied across the posterior clamps and the half rings or 6 mm rod holder ( Fig 7.3.2-10d ). The set screws on the clamps are tightened. The protruding ends of the Schanz screws are cut with the bolt cutter ( Fig 7.3.2-10e ). Once the fracture has healed after several months, the instrumentation can be removed to regain range of motion.

1

b

c

a

d

e

Fig 7.3.2-10a–e a Reduction maneuver of a type B2.1 injury (transverse bicolumn fracture). b–c Tightening of the distal and proximal nuts using the cannulated socket wrench (1) after having restored lordosis. d Distally, the rod is locked by tightening the small set screws. Slight compression can now be used to achieve final reduction. The proximal set screws are tightened. e Using the bolt cutter with handles, the protruding ends of the Schanz screws are cut off.

368

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

2.1.4 REDUCTION AND MONOSEGMENTAL FIX ATION OF PREDOMINANTLY LIGAMENTOUS INJURIES (B1.1)

The USS fracture module acts as a tension band. As a type B1.1 injury is predominantly ligamentous, a monosegmental fi xation can be achieved. If the pedicles and the inferior half of the vertebral bodies are intact, the Schanz screws can even be inserted into the fractured vertebral body in judicious orientation. Before surgery, an MRI is necessary to rule out any posterior disc herniation that could cause a neurological injury during the posterior reduction maneuver which is based on compression ( Fig 7.3.2-11).

a

Case example

This case example shows the reduction and monosegmental fixation in a patient with a predominantly ligamentous posterior disruption injury type B1.1 ( Fig 7.3.2-12 ).

b a

c Fig 7.3.2-11a–c a Reduction maneuver of a posterior disruption injury, predominantly ligamentous (type B1.1 injury). b Insertion of the Schanz screw into the intact portion of the fractured vertebral body. c Compression of the Schanz screws and final view of the monosegmental fixation.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

c

Fig 7.3.2-12a–c a Preoperative images of a posterior disruption injury, predominatly ligamentous, type B1.1. b Intraoperative view showing the facet dislocation. c Lateral x-ray at the 2-year follow-up.

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

2.1.5 REDUCTION AND FIX ATION OF T YPE C INJURIES

The USS fracture module acts as a neutralization device. The reduction technique is similar to that described for fractures with disruption of the posterior wall. However, as type C injuries are rotation injuries, it is essential to add a cross-link between the two 6 mm rods ( Fig 7.3.2-13 ). The technique for the application of such a cross-link is described in detail in chapter 7.3.1 Modularity of the universal spine system. In case of severe dislocation in the thoracic spine it is advisable to extend the instrumentation two levels above and two levels below the dislocation to facilitate the reduction.

a

b

Fig 7.3.2-13a–b Final construct after application of a cross-link between the two USS rods. In type C fractures one or two cross-links are necessary to neutralize the rotational forces.

Additional measures

1. In the majority of cases, reduction results in a large defect of bone stock in the vertebral body. It is an integral part of the fi xator instrumentation to fi ll the anterior defect with autogenous bone graft. This is possible using the same posterior approach by the transpedicular bone-grafting procedure as described by Daniaux [2]: a channel of 6 mm diameter is made into the pedicle of the fractured vertebra, and the pedicle entered as previously described. The hole is made initially with the pedicle probe ( Fig 7.3.2-14a ) and the position is controlled with the image intensifier. A depth gauge is placed down the pedicle to ensure bone surrounds the hole. The hole is then enlarged using the oscillating drill attachment with the drill bit directed slightly cranially toward the defect ( Fig 7.3.2-14b ). In order to not only fi ll the fractured vertebral body with cancellous bone but to enhance an anterior fusion between the vertebral body above and the fractured vertebra, disc material can be removed through the 6 mm channel in the pedicle with a pituitary rongeur, and the lower end plate of the vertebra above can be cleared ( Fig 7.3.2-14c ). The entry point can be reached without impairment because it is situated lateral to the longitudinal rod and lies in line with the already inserted Schanz screws. The special long funnel can be inserted into the drilled hole until the stop touches the posterior cortex of the lamina; its tip then automatically reaches into the center of the vertebral body and protects the vertebral canal from

369

370

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

unintended intrusion of bone graft particles through a fracture gap in the pedicle. The bone graft is pushed anteriorly into the defect zone ( Fig 7.3.2-14d ). 2. If the patency of the spinal canal has not been restored and posterior wall fragments continue to protrude, a hemilaminectomy and resection of one pedicle of the fractured vertebra is performed to reduce by impaction, or to extract the protruding fragments. Alternatively, an anterior approach can be performed to decompress the spinal cord.

a

3. Posterolateral fusion is recommended. 4. In case of a significant defect with mechanical impairment of the anterior column, the incidence of fatigue fractures of the Schanz screws increases; anterior bone grafting is therefore necessary.

b

Fig 7.3.2-14a–d a For transpedicular bone graft insertion, the hole is initially made with the pedicle probe. b The prepared hole is enlarged using the 6 mm oscillating drill inserted in a slightly cranial direction. c Disc material can be removed and the lower end plates are cleared with a pituitary rongeur inserted through the pedicle. d Bone graft is pushed anteriorly into the defect using the special long funnel with a stop rim and the bone pusher.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c

d

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

3

ANTERIOR INSTRUMENTATION TECHNIQUES

Anterior fi xations of the lumbar spine are indicated to support the anterior column when instability exists, particularly in association with loss of height of the vertebral body following a severe wedge compression or burst fracture, or following a partial or total vertebrectomy. Anterior spinal stabilization can be done with plates—preferably anterior locking plates [3]—or rod systems, such as the Ventrofi x. Alternatively, the USS can also be used anteriorly. Anterior column reconstruction can be done with a tricortical autograft from the iliac crest, a femoral allograft, mesh cages, or expandable cages [4, 5].

3.1

THORACOLUMBAR SPINE LOCKING PLATE (TSLP)

Principles

The thoracolumbar spine locking plate (TSLP) is designed to secure and stabilize the anterior column of the spine and to act as a buttress plate. It is an evolution of the fi rst anterior thoracolumbar locking plate [3]. The TSLP provides stable fi xation of the thoracolumbar spine. Used in combination with an anterior cage system it allows a less invasive anterior approach to the spine. This low-profi le system (4.5 mm) contains plates with lengths from 40 mm to 109 mm in 3 mm increments, 5.5 mm cancellous bone locking screws, and instrumentation designed to simplify implant placement. Indications

• Support of the anterior column along with anterior spinal reconstruction using a cage or bone graft when there is loss of vertebral body height following a severe wedge compression fracture. • Following partial or complete vertebrectomy for decompression of the spinal cord in case of tumors or thoracic disc herniation.

Advantages

• The TSLP offers a low profi le (4.5 mm). • Short segment fi xation is possible. • A less extensive approach compared to other systems is possible. • Locking screws are used. • Multiple screw fi xation (up to four screws per vertebra). • The implants are MRI compatible. Disadvantages

• The position of the screws in the vertebral bodies is fi xed by the plate. • Correction of the deformity cannot be achieved with the plate and may require an expandable cage. • Compression across the graft is difficult to achieve. • The TSLP cannot be used above T5 and rarely for L5. • The TSLP cannot be used in scoliosis. Surgical technique

The anterior and anterolateral aspect of the lumbar spine is fi rst exposed using a standard approach with the patient in a lateral decubitus position. After corpectomy ( Fig 7.3.2-15a ) and decompression of the canal, the coronal width across the vertebral bodies superior and inferior to the corpectomy site is measured using the depth gauge. The length of the self-tapping screws to be inserted monocortically should be approximately 5.0 mm less than the coronal width of the vertebral body. If the bone is osteoporotic, the far cortex should be engaged with the screw. The anterior column defect is distracted using a vertebral body spreader or the expansion of an expandable cage in order to obtain the desired lordosis ( Fig 7.3.2-15b ). A plate with the appropriate length is chosen after measuring the distance between the vertebrae with measuring forceps. The plate with threaded drill guides attached to it is positioned

371

372

7 7.3

a

Spinal instrumentation Thoracolumbar and sacropelvic spine

b

Fig 7.3.2-15a–b a Removal of the vertebral body. b Expansion of the corpectomy site using the expandable cage in order to restore sagittal alignment and vertebral height.

a

b

on the lateral aspect of the vertebral body. Temporary fi xation pins are inserted into two of the threaded drill guides and the plate is pushed to the bone so it fits snugly without any gap ( Fig 7.3.2-16a). Before plate application, in some cases it may be necessary to smoothen out the overhang of the lateral end plate with a rongeur, because these irregularities would prevent a good application to the lateral aspect of the vertebral body. An awl is then used directly through the threaded drill guide to create a 22 mm hole ( Fig 7.3.2-16b ). Further depth can be achieved with a standard USS pedicle probe. In osteoporotic bone a bicortical purchase is preferred. The awl and the threaded drill guide are removed. With a depth gauge the appropriate screw length is measured ( Fig 7.3.2-16c ). A self-locking 5.5 mm titanium screw is inserted with the self-retaining 3.5 mm hexagonal screwdriver ( Fig 7.3.2-16d ). The thread of the screw head engages the threaded screw hole of the plate. These different steps are repeated for the insertion of four to eight screws ( Fig 7.3.2-16e ).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c Fig 7.3.2-16a–e a Insertion of two temporary fixation pins. b Preparation of the screw hole by application of the threaded drill guide applicator (arrow) into a free hole and then insertion of the awl through the drill guide applicator. c The screw length is measured with a depth gauge.

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

d

e

Fig 7.3.2-16a–e d Screws of appropriate length are selected, inserted, and locked into the plate. e Final view after application of the TSLP.

Case example

A 50-year-old woman suffered from severe thoracic myelopathy with a calcified disc ( Fig 7.3.2-17a ). Treatment included anterior corpectomy and reconstruction with a Synmesh and a TSLP ( Fig 7.3.2-17b ).

3.2

ANTERIOR TITANIUM ROD SYSTEMS

Principle

Anterior rod systems that perform similarly to the thoracolumbar spine plate system, but allow compression and distraction, are the Ventrofi x, the Telefi x, and the USS in its anterior application. Their surgical techniques are described separately together with the different applications and particularities. Indications—Ventrofix

• Support of the anterior column along with anterior spinal reconstruction using a cage or bone graft when there is loss of vertebral body height following a severe wedge compression fracture. • Following partial or complete vertebrectomy for decompression of the spinal cord in case of tumors. Advantages—Ventrofix

• The rod system allows compression and distraction. • Correction of the sagittal deformity is possible through the system. • Application of the anterior graft or cage after insertion of the Ventrofi x is possible. Disadvantages—Ventrofix

a

b

Fig 7.2.3-17a–b The severe thoracic myelopathy with a calcified disc was treated with a corpectomy and reconstructed with a mesh cage and TSLP.

• An anterior titanium rod system requires a more extensive exposure than the thoracolumbar spine locking plate. • The rod system has a higher profi le than the TSLP. • Direction of the screw is as critical as it is with the TSLP. • It cannot be used in scoliosis.

373

374

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

3.2.1 DOUBLE-ROD CLAMP

Assembly of the montage

Approach

The anterior and anterolateral aspect of the spine is fi rst exposed using a standard approach with the patient in a lateral decubitus position. After the corpectomy and the decompression of the canal, the coronal width across the vertebral bodies superior and inferior to the corpectomy is measured using the depth gauge. The length of the unicortical self-tapping screws to be inserted should be approximately 5.0 mm less than the coronal width of the vertebral body. If the bone is osteoporotic, the far cortex should be engaged with the screw. The anterior column defect is distracted using a vertebral body spreader to obtain the desired lordosis and reconstructed with either a tricortical graft or a cage ( 7.3.2-18a ). The length of the construct is measured with a rod template ( 7.3.2-18b ).

a

b

Fig 7.3.2-18a–b a The anterior column defect is spread with the vertebral body spreader. b The intervertebral spacer (iliac crest graft, allograft, or cage) is inserted and the length of the construct is measured using a rod template.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

The fi xation system is fi rst assembled outside the operating field. Two 6 mm hard rods of appropriate length are selected using the rod template and placed into two clamps ( Fig 7.3.2-19). The set screws over the incomplete holes are tightened and the length protruding is adjusted only to allow compression or distraction. The threaded drill guides with their fi xation pins are inserted into the posterior screw holes of each clamp of the construct, which is now ready to be implanted ( Fig 7.3.2-20a ). Positioning of the Ventrofix

The assembled construct montage is placed on the lateral part of the vertebral body ( Fig 7.3.2-20b ). The superior and inferior end plates are trimmed to ensure the clamps lie in direct contact with the vertebral body. The self-tapping 2.5 mm

a

b

Fig 7.3.2-19a–b Assembly of the construct montage. a 6 mm USS rods of the appropriate length are placed into the clamps and the set screws in the blind holes are tightened. b The protruding ends of the rods are adjusted.

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

fi xation pins are inserted in the posterior hole of the clamp and tightened to ensure close approximation of the implant against the vertebral body. Make sure that the direction of the drill holes does not penetrate the spinal canal. At this stage it is necessary to ensure the height of the anterior column is correct and the sagittal plane has been restored. A lateral x-ray is absolutely essential. The construct is temporarily fi xed to the spine by inserting the fi xation pins into the posterior holes ( Fig 7.3.2-20c ).

Preparation of the screw hole

A threaded drill guide with an awl is attached to the anterior hole of one of the clamps. The awl is advanced to prepare the hole in the vertebral body. Since purchase in the opposite cortex is not necessary, the awl has a stop at 25 mm to prevent overdrilling ( Fig 7.3.2-21). If the bone is osteoporotic, the purchase of the screw in the opposite cortex may be necessary. In this case the length of the screw is determined by measuring the path of the screw hole with the depth gauge.

a

b

c

Fig 7.3.2-20a–c Positioning of the Ventrofix. a The threaded drill guides/fixation pins are attached to the clamps. b The Ventrofix is placed on the vertebral column. c The fixation pins are inserted into the posterior holes, thereby, temporarily fixing the clamps to the spine.

a

b

Fig 7.3.2-21a–b Preparation of the screw holes. a The drill guide/awl is attached to the anterior hole of one clamp (arrow). The awl perforates the cortex and is advanced to 25 mm (or to the far cortex in case of osteoporosis). b After removal of the drill guide/awl measure the screw length with the depth gauge.

375

376

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Insertion of the screws

A 7.5 mm titanium self-tapping, self-locking screw of the correct length is then inserted ( Fig 7.3.2-22a ). The same maneuver is performed for the other clamp ( Fig 7.3.2-22b ). Two additional screws are inserted in the remaining posterior holes after removal of the temporary fi xation pins. Compression and locking of the instrumentation

Compression can be applied by loosening the set screws on the open hole of each clamp so that the rods can slide through the clamps. The desired compression is applied using the compression forceps before tightening the set screws ( Fig 7.3.2-23a ). If the clamps are spanned over too great a distance for the compression, then the compression forceps can be applied between one of the Ventrofi x clamps and a half ring attached to the posterior rod to maintain the lordosis ( Fig 7.3.2-23b ). The Ventrofi x clamps are designed for a triangular

orientation of the screws increasing the resistance of the construct to pull out. The screws incorporate a machine thread section close to the screw head which locks the screw to the plate to minimize the risk of screw backout and allows the screw to be countersunk into the clamps providing a lowprofi le implant. This design provides a secure angle-stable construct, eliminating the need for bicortical screw purchase in nonosteoporotic bone. A single clamp may be added to the double clamp configuration if purchase in the bone graft is necessary ( Fig 7.3.2-23c ).

a

a

b

Fig 7.3.2-22a–b Insertion of the screws. a Insert the appropriate 7.5 mm titanium self-tapping, self-locking screws. b View after insertion of the screws and axial view of the screw directions.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c

b Fig 7.3.2-23a–c Compression and locking of the instrumentation. a The Ventrofix compression forceps is used to lock the graft (or cage), the set screws can then be tightened. b View of the construct in its final position. c A single clamp may be added anteriorly if purchase in the bone graft is needed.

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

3.2.2 FRACTURE CLAMP

3.2.3 SINGLE-CLAMP-DOUBLE-ROD CONFIGURATION

The advantage of this application lies in the preservation of the inferior disc space, if the distal part of the vertebral body is intact; for example, in the case of a monosegmental fi xation.

The advantage of this application is the independent positioning of the screws on the vertebral body. The implant consists of four single clamps, one parallel connector, and two 6 mm rods of appropriate length ( Fig 7.3.2-25 ). The implant is assembled, the set screws of the clamps are lightly tightened, and the screws of the parallel connector are fi rmly tightened.

The application of this instrumentation is the same as for the double-rod clamp configuration. The only difference is that the holes in the distal clamp are transverse and not offset in an oblique manner ( Fig 7.3.2-24 ).

Fig 7.3.2-24 Double-rod configuration with fracture clamp to spare the inferior disc.

Fig 7.3.2-25 Single-clamp-double-rod configuration.

The surgical technique is similar to the technique used for the double clamp (see Fig 7.3.2-20 , Fig 7.3.2-26a ). The length of the rods is determined with a template. The two anterior clamps are held with drill sleeves that have an integrated fi xation pin. The construct is placed on the vertebral column after reconstruction (see above), the clamps are adjusted to the desired position and the set screws of the posterior clamps are fi rmly tightened. Both fi xation pins are exposed and inserted to hold the implant on the vertebral bodies. Make sure the direction of the drill holes is such that they do not penetrate the spinal canal. At this stage, it is necessary to ensure the height of the construct is correct and the sagittal plane is preserved—a control x-ray is absolutely essential. The anterior clamps are adjusted to the desired position and the set screws fi rmly tightened. The drill sleeve with integrated awl is attached to one anterior clamp. Introduce the drill through the anterior hole using the awl and insert the appropriate 7.5 mm locking screw. The hole of the second anterior clamp is drilled in a similar way, and the second anterior screw inserted. After removal of the fi xation pins and penetration of the awl through the posterior hole of each clamp, the posterior locking screws are inserted. To apply compression, loosen the two distal set screws and compress between the parallel connector and distal clamps with the Ventrofi x compression forceps and tighten the set screws.

377

378

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

3.2.4 SINGLE-ROD CONFIGURATION 12

These constructs should not be used by themselves in the lumbar spine and thoracolumbar junction, because they have poor rotational stability. However, it may be used in the thoracic spine or to supplement fi xation of an anterior reconstruction in the treatment of a metastatic disease. In the thoracic spine the rod may need to be bent for correct placement (see Fig 7.3.2-23a ). At the thoracolumbar or lumbar levels such a configuration can be used to supplement an anterior or posterior reconstruction ( Fig 7.3.2-26b–c ).

Fig 7.3.2-27a–b 67-year-old patient with an A3 burst fracture at level L1 was treated with anterior decompression, anterior fusion, and a Ventrofix.

40% 20°

2 3

a

Case example

This case example shows a type A3 burst fracture of L1 in a 67-year-old patient ( Fig 7.3.2-27a ) treated with anterior decompression, anterior fusion with rib grafts, and a Ventrofi x (double-rod-clamp configuration, Fig 7.3.2-27b ).



b

a

b

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

c

Fig 7.3.2-26a–c a Single-clamp-single-rod configuration. b Double-clamp-single-rod configuration. c Single-clamp-single-rod configuration in combination with a posterior fixation.

379

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

3.2.5 TELESCOPING FIX ATION SYSTEM (TELEFIX)

The telescoping fi xation system (monosegmental or bisegmental version) combines rods with a plate. The main advantage of the Telefi x over the Ventrofi x are the posterior polyaxial screws which permit an angulation of 15° to tailor screw insertion to the individual anatomical situation. Directing the posterior screws away from the spinal canal is rendered easy. The usage of self-locking screws enables a stable anchorage ( Fig 7.3.2-28 ). Minimally invasive application of this anterior telescoping fi xation system is possible (thoracoscopic insertion), but extensive fluoroscopic exposure is necessary. Instrumentation of scoliosis is not possible.

a

b

c

Video 7.3.2-2 Telefix application for anterior stabilization of a type A fracture of L1.

Fig 7.3.2-28a–c a Monosegmental version: self-locking screws to be used on the anterior side, special polyaxial screws (± 15° angulation) can be inserted posteriorly. b Bisegmental version. c Instrumentation with a bisegmental Telefix and an expandable cage after corpectomy of L1.

380

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

3.3

ANTERIOR APPLICATION OF THE USS FOR THORACOLUMBAR STABILIZATION

The USS regular or small stature can be used to reconstruct or stabilize the anterior column (also chapter 7.3.3 Universal spinal instrumentation for deformity). The same indications as for other anterior constructs described in this chapter are valid. Advantages

• • • • • • •

Ease of application. Best for correction of sagittal malalignment of the spine. Allows compression across the bone graft. Placement of screws is not dictated by the plate. Versatile instrumentation. Can be used in case of a scoliotic deformity. Bicortical purchase is easy to achieve.

Disadvantages

• Higher profi le than the plate system and/or Ventrofi x. • Requires conventional open or miniopen surgery as opposed to the Telefi x. • Requires the screws to be inserted parallel to the end plate as opposed to the posterior polyaxial screws of the Telefi x.

3.3.1 SURGICAL TECHNIQUE—ANTERIOR CONSTRUCT

The spine is exposed and after treatment of the pathology, the sagittal plane is restored with a bone graft or a spacer. The standard 6 mm USS screws are used. First a staple is inserted with a special staple holder ( Fig 7.3.2-29a ). The staple is placed

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

toward the posterior aspect of the vertebral body near the base of the pedicle. The staple provides a fi xed angle between itself and the screw preventing pullout or cutout of the screw. The cortex is opened through the staple with a short sharp awl, and then a long blunt awl (pedicle feeler) is inserted into the vertebral body. It is useful to aim the tip of the awl at the tip of the index fi nger which has been placed at the planned exit point ( Fig 7.3.2-29b ). Direction of the probe should be parallel to the posterior wall or slightly away from it. With a depth gauge the appropriate screw length is measured, and a 5.0 or 6.0 mm screw of appropriate length is inserted. When penetrating the far cortex, a washer can be used in order to better stabilize the screw in a porotic vertebral body ( Fig 7.3.2-29c ). A similar procedure is performed in the caudal vertebra. The length of the 6 mm hard rods is determined with the template ( Fig 7.3.2-29d ). The two rods attached to the parallel connector are now fi xed to the posterior screw and held with the sleeves and nut. The anterior rod is slid distally, the opening for the 6 mm USS screw is prepared as described above, and the screw is inserted so that the opening of the screw aligns with the rod ( Fig 7.3.2-29e ). The rod is then slid proximally and the distal screw is inserted ( Fig 7.3.2-29f ). Compression is now applied by locking the parallel connector to the rods. The proximal screw should be locked to the rods. The compression clamp is used to apply compression to the graft ( Fig 7.3.2-29g ). The distal rods are then locked to the pedicle screws. Alternatively a double screw staple construct can be used (chapter 7.3.1 Modularity of the universal spine system).

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

a

b

d

e

c

f

g

Fig 7.3.2-29a–g a The tip of the staple impactor is driven through the cortical bone in order to start the entry hole for the screw and to anchor the staple in its position. b The pedicle feeler is inserted until it breaches the far cortex. c The 6 mm titanium USS screw is inserted until it penetrates the far cortex and engages into the washer (if a washer is needed). d A template rod is used to assess the required length of the 6 mm hard rod. e The posterior rod with the attached parallel connector is fixed to the posterior screws. The anterior rod is then slid distally and the cranial screw can now be inserted; the anterior rod can then be moved proximally into the screw. f The anterior rod is moved further proximally and the distal screw is inserted. The anterior rod is now attached to both screws. g Compression is applied to the graft and the distal screws are tightened.

381

382

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

3.3.2 SURGICAL TECHNIQUE—ANTERIOR VERTEBRAL BODY CONSTRUCT

This efficient and cost-effective technique is only used to reconstruct vertebral bodies that have been destroyed by malignant tumors. A multilevel disease that requires more than two vertebral excisions is a contraindication.

a

b

Fig 7.3.2-30a–h a Staples are placed into the end plates using the staple holder. b The length of the required 6 mm hard rod is assessed with the rod template.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

After removal of the posterior vertebrae the titanium staples are embedded into each normal end plate ( Fig 7.3.2-30a ). A rod template is used to assess the length of the 6 mm hard rod, which should protrude into a third of each vertebral body ( Fig 7.3.2-30b ). The 6 mm rod holder is attached to the rod and pushed into the proximal end plate; if difficulty is encountered, start the entry hole with the sharp awl or a right

c

d

c The 6 mm rod holder is attached to the rod and pushed into the proximal end plate. d After the rod is pushed into the proximal vertebra, it is eased into the distal vertebra.

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

The distal half-ring clamp is now locked ( Fig 7.3.2-30f ). This construct is reinforced with a 6 mm hard rod and two 6 mm pedicle screws inserted laterally into the vertebral body ( Fig 7.3.2-30g ). The defect is now packed with methylmethacrylate. However, it is important that before applying the cement, the dura is protected with a sheath of collagen substitute. A blunt elevator is used to ensure the cement does not extrude into the spinal canal ( Fig 7.3.2-30h ).

angle high-speed burr. The rod is pushed into the proximal vertebral body ( Fig 7.3.2-30c ) and then eased into the distal vertebral body ( Fig 7.3.2-30d ). Half-ring clamps are placed loosely onto the rod so that they abut against the proximal and distal staple. The proximal half-ring clamp is locked ( Fig 7.3.230e ). The USS spreader is placed onto the rod and between the half-ring clamp and the height of the space is restored.

e

f

g

h

Fig 7.3.2-30a–h e The proximal half ring is locked against the proximal staple and tightened. f The USS spreader is placed on the rod and the height of the space is restored. The distal half ring is tightened. g The construct is reinforced with a 6 mm hard rod and two 6 mm screws. h Cement is used to fill the defect—the dura must be protected.

383

384

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

4

ANTERIOR RECONSTRUCTION WITH CAGE TECHNOLOGY

The implants for this technique (cages or spacers) are vertebral body replacement devices intended for use in the thoracic and lumbar spine (T1–L5) to replace a collapsed, damaged, or unstable vertebral body resulting from a tumor or a fracture. These implants can be packed with bone and should be used with supplemental anterior internal fi xation systems, such as Ventrofi x, Telefi x, USS, TSLP, or posterior fi xation systems. Alternatively, femoral shaft allograft packed with autograft inside the medullary canal can be used for anterior column reconstruction.

4.1

EXPANDABLE CAGE

Indications

• Anterior column reconstruction from S1 to T1. • Fractures and tumors. Advantages

• Allows a less invasive approach to the spine limited to the level of the vertebrectomy. • Helps restoration of the sagittal alignment due to its expansion mechanism. • May be inserted without any fusion in certain indications (ie, tumors). • Can be inserted through a posterior extracavitary approach as recently reported [6, 7].

Principles

Disadvantages

Expandable cages allow restoration of the anterior column, with restoration of the vertebral body height thanks to its expansion mechanism. The Synex implant is an expandable vertebral body replacement device. It is available in two types, one for the mid and upper thoracic spine with small end plates, the other one for the low thoracic and lumbar spine ( Fig 7.3.2-31). Each type is available in different heights with various end plates. The parallel and angled end plates (0°, 10°, 20°, – 6°) allow for restoration of the lordosis or kyphosis of the spine. The self-locking ratchet expansion mechanism with a 2.5 mm distraction in each ratchet interval provides an efficient means of restoring proper spinal alignment using rapid and controlled in situ expansion. The convex end plates are anatomically designed to minimize subsidence while optimizing bone graft placement following expansion.

• Little room is left over for the bone graft. • Subsidence in the end plates with overdistraction and insufficient anterior or posterior stabilization. • Long-term rate of fusion unknown.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

20°

20° a

b

Fig 7.3.2-31a–b a Expandable cage, Synex, is shown with potential expansion capacity. b Different end-plate configurations are available to match the local anatomy with a range from – 6° to 20°.

385

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

Surgical technique

repositioning, if necessary ( Fig 7.3.2-32b ). The Synex implant is then expanded in situ using the parallel distractor forceps until the desired amount of spinal correction is achieved ( Fig 7.3.2-32c ). Each step of the ratchet mechanism produces 2.5 mm of distraction. A stop will engage if the implant reaches its maximum height, thereby preventing the implant from separating. When ideal alignment and height are achieved, the closure of the locking ring must be checked to make sure it is secure ( Fig 7.3.2-32d ). Bone graft is then packed in and around the cage, especially around the anterior aspect of the corpectomy site ( Fig 7.3.2-32e )—in case there is not sufficient bone graft available, it is not necessary to pack the center of the cage. The graft around the cage is more important.

A lateral or anterolateral approach is used depending on the spinal level involved. A partial or complete corpectomy is performed with special attention paid to removing the superficial layers of the entire cartilaginous end plates and exposing bleeding bone. Excessive removal of subchondral bone may weaken the vertebral end plate and cause implant subsidence and loss of segmental stability. After selection of the ideal cage size to be implanted, bone graft is placed into the bone cups of the implant’s end plates ( Fig 7.3.2-32a ). Using the implant holder, the cage is then positioned in the corpectomy site while taking care that the release opening is facing the implant holder to allow for

1

a

b

Fig 7.3.2-32a–f a The end plates of the Synex cage are filled with bone chips. b Use the implant holder to guide and position the cage. c The positioned cage is expanded in situ using the parallel distractor.

c

2

d d Check if the locking ring is secure by inspecting the gap. If the gap is 1 mm, the system is locked (2); if the gap is more than 1 mm, the system is not locked (1).

386

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Once the cage has been applied and expanded a supplementary anterior and/or posterior fi xation is mandatory ( Fig 7.3.2-32f ). If it is necessary to reposition the implant, the Synex cage can still be collapsed to its neutral height. The disconnecting instrument is introduced into the slot between the two ends of the locking ring and turned 1/4 of a revolution. The collapsed implant can be removed using the implant holder.

Case example

Multiple bony metastases ( Fig 7.3.2-33a ) from a renal cell carcinoma in a 60-year-old woman with marked osteoporosis, intractable pain and weakness in her lower extremities. The patient was treated with insertion of an expandable cage and supplementary anterior and posterior instrumentation ( Fig 7.3.2-33b ).

a

e

f

Fig 7.3.2-32a–f e Pack the cage with autogenous bone graft; it is most important to add bone graft on the anterior side of the cage. f Complementary fixation with, for example, the Ventrofix.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b Fig 7.3.2-33a–b a Preoperative AP and lateral x-rays and CT scan. b Corpectomy reconstruction with expandable cage and complementary anterior and posterior fixation with the USS.

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

4.2

MESH CAGE

Principles

Fig 7.3.2-34 Synmesh spacer with end rings to be fitted inside the mesh.

Mesh cages are designed with an open architecture in order to optimize bony ingrowth. The application of such a mesh cage is described using the Synmesh spacer. The Synmesh implant is available in different heights and with six different footprints. End rings fit into the ends of the round or oblong mesh, providing an improved bone-to-implant interface ( Fig 7.3.2-34 ). The end rings have teeth on the top surface to minimize migration and are available in a round form with 0° and 2.5° angles and in an oblong form with 0° and 5° angles. Indications

• Anterior column reconstruction from S1 to C2. • Fractures and tumors. Advantages

• Optimization of the bony ingrowth and fusion. • Superior and inferior end rings prevent subsidence of the spacer. • Different sizes and potential to customize the spacer. Disadvantages

• Requires distraction to restore vertebral body height. • Compression across the spacer is necessary for stability.

Surgical technique

The approach is similar to the one used for treatment with the Synex cage, but may require a more extensive exposure of the spine. The approach may preserve the opposite lateral wall of the vertebral body and the anterior wall in order to promote fusion. Disc and end plates are removed and a corpectomy is carried out. The parallel distractor is inserted and used to restore the lordosis ( Fig 7.3.2-35a ). The height of the defect is measured with a special caliper, and the appropriate Synmesh spacer is chosen.

387

388

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

When choosing the spacer, the surgeon needs to take into account that the height of the end rings will add between 3 and 7 mm to the mesh height depending on the spacer size. With a special cutter, the mesh is then trimmed to the appropriate height taking care to cut perpendicular to the long axis of the mesh. The bottom end ring is attached to the mesh ( Fig 7.3.2-35b ). The cage is then fi lled with bone from the vertebrectomy site ( Fig 7.3.2-35c ). The bone is impacted into

* a

f

b

g

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

the cage, and the second end ring is fitted into the cage and secured. The Synmesh spacer is now ready for insertion. Under distraction, the Synmesh spacer is inserted using the appropriate implant holder ( Fig 7.3.2-35d ). Once the implant is in place, the distractor is gently removed. A supplementary fi xation (eg, Ventrofi x or TSLP) is added and additional bone is applied around the cage ( Fig 7.3.2-35e–g ).

* c

d

e

Fig 7.3.2-35a–g a Parallel distractor restores lordosis after complete corpectomy. b The mesh cage is cut with a special cage cutter either on diagonal or horizontal (asterisks). c The end ring is fit into the mesh and locked with a screw. d The cage is filled with cancellous bone graft and the end ring is fit on the top “closing the cage”. A screw locks the top end ring. e The cage is inserted with the parallel distractor left in place. The blade of the distractor allows the sliding of the end rings of the cage (arrow). f–g In order to maintain compression and ensure stability, a supplementary fixation is applied and additional bone graft is placed around the cage.

7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine (fractures, tumors, degenerative)

5

Case example

1.

A 17-year-old female sustained a type C burst fracture and paraplegia ( Fig 7.3.2-36a ). Spinal reconstruction with posterior USS and anterior decompression reconstruction with Synmesh and anterior USS ( Fig 7.3.2-36b ). Complete neurological recuperation at the 2-year follow-up.

2.

3.

4.

5.

a

6.

7.

b Fig 7.3.2-36a–b a Preoperative x-rays and CT scan. b Postoperative x-rays 2 years after surgery.

a

BIBLIOGRAPHY

Dick W, Kluger P, Magerl F, et al (1985) A new device for internal fi xation of thoracolumbar and lumbar spine fractures: the “fi xateur interne”. Paraplegia; 23(4):225–232. Daniaux H (1986) [Transpedicular repositioning and spongioplasty in fractures of the vertebral bodies of the lower thoracic and lumbar spine.] Unfallchirurg; 89(5):197–213. Thalgott JS, Kabins MB, Timlin M, et al (1997) Four years experience with the AO anterior thoracolumbar locking plate. Spinal Cord; 35(5):286–291. Dvorak MF, Kwon BK, Fisher CG, et al (2003) Effectiveness of titanium mesh cylindrical cages in anterior column reconstruction after thoracic and lumbar vertebral body resection. Spine; 28(9):902–908. Knop C, Lange U, Reinhold M, et al (2005) [Vertebral body replacement with Synex in combined posteroanterior surgery for treatment of thoracolumbar injuries.] Oper Orthop Traumatol; 17(3):249–280. Snell BE, Naser FF, Wolfla CE (2006) Singlestage vertebra thoracolumbar vertebrectomy with circumferential reconstruction and arthrodesis: surgical technique and results in 15 patients. Neurosurgery; 58(4):ONS-263–269. Hunt T, Shen FH, Arlet V (2006) Expandable cage placement via a posterolateral approach in lumbar spine reconstructions. Technical note. J Neurosirg Spine; 5(3):271–274.

389

390

7 SPINAL INSTRUMENTATION 7.3 THORACOLUMBAR AND SACROPELVIC SPINE 7.3.3 UNIVERSAL SPINAL INSTRUMENTATION FOR DEFORMITY

1

Introduction ………………………………………………………………………………………………… 391

2

Basic principles …………………………………………………………………………………………… 392

3 3.1 3.2

Scoliosis—posterior correction and stabilization …………………………………………………… 394 Right thoracic curve ……………………………………………………………………………………… 394 Double curve ……………………………………………………………………………………………… 401

4

Scoliosis—anterior correction and stabilization ……………………………………………………… 405

5

Kyphosis—posterior correction and stabilization …………………………………………………… 412

6

Bibliography ………………………………………………………………………………………………… 420

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

391

Christopher Cain

7 7.3

SPINAL INSTRUMENTATION THORACOLUMBAR AND SACROPELVIC SPINE

7.3.3

1

UNIVERSAL SPINAL INSTRUMENTATION FOR DEFORMITY

INTRODUCTION

The universal spine system (USS) was designed around a new concept of scoliosis correction. No distraction or compression is used in the reduction of the deformity; the spine is brought to its corrected position and it is allowed to fi nd its own length. Since its introduction, more than 10,000 spinal deformities have been treated using such instrumentation. The application of the universal spine system has been discussed in numerous peer-reviewed articles [1–11]. The governing principles of the USS incorporate:

• Stable instrumentation of the apical vertebrae, via unique pedicle hooks or pedicle screws. • The instrumented spine is then reduced to the desired position by segmental manipulation. • Emphasis is also placed on the application of the convex rod, which allows translation of the apex toward the midline. • In rigid deformities the concave implants are reduced to the rod after placement of the convex rod. • The stable attachment of the implants to the instrumented vertebral bodies and the Schanz pin extension allow true segmental derotation of each individual vertebra.

392

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

2

BASIC PRINCIPLES

• Soft-tissue release, posterior and/or anterior as indicated by the clinical circumstance. • Stable anchorage of implants to individual vertebrae, particularly the apical vertebra. • Contouring of the rods to the desired sagittal plane. • Reduction of the spine to the 5 or 6 mm rod, facilitated by the so-called persuader. • Segmental derotation of the individual vertebra, facilitated by the Schanz screw extensions for the implants. • End vertebrae are used as the cornerstones of the construct onto which the spine is reduced. • Inclusion of stable low-profi le cross-links increases the stability of the construct. • No distraction or compression is required to maintain stability of the implants.

It is important to note that although this is the way the system was intended to be used, the versatility of the instrumentation and implants enables surgeons to modify their surgical plan as required to address the pathology of each individual case. In more flexible curves, implants can be placed with greater separation between the end vertebrae of the construct without compromising the corrective power of the system. Where more significant deformity or greater rigidity is evident, instrumentation of each vertebra may be required to achieve the desired correction.

T4

Frame concept

The neutral cranial and caudal vertebrae are used as the cornerstones of a virtual frame onto which the deformed spine is reduced (Fig 7.3.3-1) [12]. The stable pedicular instrumentation in the caudal end vertebra becomes a base on which the construct is built. Implants are inserted; the concave side of the cranial end is instrumented with pedicle screws when the anatomy allows or unique specialized pedicle hooks, and on the convex side a claw construct with lamina hook and pedicle screw or specialized pedicle hook is recommended for the vertebra below. With this construct the upper convex hook must be a lamina hook and not a transverse process hook because a transverse process hook would pull off the transverse process as the convex rod is used to translate the spine to the midline.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

L1

Fig 7.3.3-1 The scoliosis frame. The spine will be reduced onto the frame that is built between the bottom foundation and the top anchorage.

393

7.3.3 Universal spinal instrumentation for deformity

Basic principle of instrumentation

The usual procedure is to fi rst instrument the concave side of the curve and then instrument the convex side of the curve. First establish the fi xation points at the caudal ends of the construct with pedicle screws, and then at the top of the construct with specialized pedicle hooks and/or pedicle screws. Then the rest of the spine is instrumented with either specialized pedicle hooks, screws and/or lamina hooks depending on the local anatomy and the surgeon’s choice ( Fig 7.3.3-2 ). The bilateral

claw

or

opening of the USS II implants avoids the need to consider which side the implant should open on in order to facilitate the reduction of the spine to the rod. The rigidity of the deformity, and to some extent the placement of implants on the concave side, will determine the placement of the implants on the convex side of the curve. As a general rule it is desirable to instrument the apical vertebrae on both sides. With greater acceptance and utilization of thoracic pedicle screws [13], and as the manipulative power of the system is enhanced by the

claw or

or

or

or

or

optional

a

b

Fig 7.3.3-2a–b Possible instrumentation with pedicle screws and/or pedicle or lamina hooks according to the surgeon’s choice.

394

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

3

additional fi xation provided by these implants, the use of pedicle screws is recommended where the spinal anatomy permits their safe application. At the curve’s apex on the concave side a small laminectomy may be required to control the medial border of the pedicle and therefore allow safer insertion of pedicle screws [11]. When the experience of the surgeon is limited, when the preoperative imaging suggests that the use of pedicle screws is inappropriate, dangerous, or when intraoperative difficulties are encountered, then specialized pedicle hooks should be used instead of pedicle screws. However, these specialized pedicle hooks can only be used from T1 to T9 (and in some cases T10). The principles for implant insertion and correction of the curve are best demonstrated in the correction of a simple right thoracic curve (see next paragraph for a step-by-step description of the procedure).

SCOLIOSIS—POSTERIOR CORRECTION AND STABILIZATION

To illustrate the principles of correction of a right-sided scoliosis, the example of the instrumentation of a right thoracic curve instrumented from T4 to L1 is given.

3.1

RIGHT THORACIC CURVE

Preoperative plan

The proximal and distal end vertebrae are identified on AP standing and lateral bending, fulcrum bending, or traction x-rays. In the classic situation instrumentation extends from T4 to L1. Surgical technique

The spine is exposed with subperiosteal dissection and the levels identified by palpating the 12th rib and the fi rst transverse process of L1. Intraoperative x-rays with a marker on the pedicle of T12 or of L1 confi rm the appropriate level. The implants can then be inserted by starting from the concave side. Concave side

The foundation of the construct is established with a pedicle screw placed at T12 and Ll. The cranial vertebra, T4, is also instrumented with a pedicle screw or specialized pedicle hook ( Fig 7.3.3-3a ). The apical vertebra is usually T8 or T9 and is also instrumented with either a pedicle screw or specialized pedicle hook ( Fig 7.3.3-3c–d ). Additional instrumentation is usually placed at alternating levels, however, more frequent use of implants may be needed in larger and stiffer curves. Caudally, the T11 vertebra is instrumented with a pedicle screw, or a downward pointing lamina hook ( Fig 7.3.3-3b ).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

395

7.3.3 Universal spinal instrumentation for deformity

or

T4

or pedicle hook

T7 or

T9

or lamina hook

T11

L1 a

b

c

d

Fig 7.3.3-3a–d a–b Instrumentation of the concave side with a hybrid construct or whole pedicle screw construct. c–d Lateral view of instrumentation in the concave side with either pedicle screws or specialized pedicle hooks.

396

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Convex side

A proximal claw is applied at T4 and T5. A downward lamina hook is placed over the T4 lamina and either a pedicle screw or a specialized pedicle hook is placed in or around the T5 pedicle ( Fig 7.3.3-4a ). Once fitted, the lamina hook is removed to prevent its inadvertent displacement onto the spinal cord while the remainder of the implants are secured. The lamina hook can easily be replaced around the rod once it has been coupled to the implant inserted at T5. The spacing of implants on the convex side will be determined by the magnitude and

or

claw or

pedicle hook

claw

or

or

pedicle hook

or

or

or lamina hook

rigidity of the deformity. The general principle would be to instrument the apical vertebra with either a pedicle screw or specialized pedicle hook. The intervening vertebrae cephalad and caudad to the apex are instrumented as needed with either pedicle screws or specialized pedicle hooks. A pedicle screw is inserted at Ll and the use of a pedicle screw or a downgoing lamina hook can be added at T12 to protect the distal screw from pullout if it is a stiff curve or if there is a significant kyphotic deformity ( Fig 7.3.3-4b ).

or

or

or

lamina hook optional

a

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

Fig 7.3.3-4a–b Final implant configuration after the concave and the convex side have been instrumented with either a hybrid or pedicle screw configuration.

397

7.3.3 Universal spinal instrumentation for deformity

Reduction

Once all the implants are attached to the vertebra, a 5 mm or 6 mm rod template is placed in the desired sagittal plane on the concave side between the T4 hook and Ll screw. This template is used for the calculation of the fi nal rod length bearing in mind that the spine will automatically and passively elongate during the correction maneuvers. The rule of thumb is that for a 50° curve the automatic elongation of the spine may reach 1.5 cm and for a 70° curve 2.5 cm. The rod template is then removed and the appropriate size hard or extra hard rod is contoured to the desired sagittal plane and cut to length ( Fig 7.3.3-5a–b ). When the rod is prepared and ready for implantation, it is important to decorticate the posterior element and excise the facet joints that are likely to be inaccessible after placement of the rods. Resection of the facet joints will also facilitate the mobilization of a stiff curve and help the correction maneuvers. The rod can then be inserted into the T4 and L1 implants, collars and nuts are applied, but the nuts are only tightened at L1 while maintaining the sagittal orientation of the rod ( Fig 7.3.3-5c–d ). If a T12 screw has also been used this can be coupled to the rod and secured at this stage. A rod holder is used to maintain the sagittal position of the rod while the distal implants are being tightened. It is important to remember that the proximal end of the rod is left loose (nut at T4 is not tightened). In the case of a supple curve, the implants on the concave side are then reduced to the 5 or 6 mm rod. Using the complex reduction forceps, the intervening implants, in this case at T7 and T11 are brought to the rod ( Fig 7.3.3-6a ). It is important not to apply force beyond that which the bone can withstand. The apical screw or hook on the concave side is then brought to the rod with the persuader. If the curve is stiff, do not proceed and direct your attention to the convex side. The convex contoured rod is now applied. The rod is inserted in the proximal claw made of the supralaminar hook at T4 and

the specialized pedicle hook or pedicle screw at T5. The claw is then compressed and the nuts tightened to ensure stability of the implants while the rod is maintained in a strict sagittal position ( Fig 7.3.3-6b ) The rod is then pushed toward the midline using the rod holder to laterally engage the specialized pedicle hooks or screws sequentially at T8, at T10, at T12, and then at L1. It is important not to tighten these nuts at this stage ( Fig 7.3.3-6c ) because this would prevent further correction on the concave side. The apical screw or hook on the concave side will now have been translated toward the concave rod. The complex reduction forceps are then used to further bring the apical vertebra and its specialized pedicle hook or screw to the concave rod (see Fig 7.3.1-24c , Fig 7.3.1-24f–g , chapter 7.3.1 Modularity of universal spinal instrumentation). If this cannot be achieved due to the rigidity and/or the magnitude of the deformity, connecting rods can be used as described previously (see Fig 7.3.1-25, chapter 7.3.1 Modularity of universal spinal instrumentation).

Video 7.3.3-1 Posterior correction of a right thoracic scoliosis with USS II.

398

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

T4 loose

T7

T9

T11

a

b

c

d

L1

Fig 7.3.3-5a–d a–b The preshaped rod template is used to precontour the rod to the desired sagittal profile using the “French” rod bender. c–d The precontoured rod is then inserted in a strictly sagittal plane into the cranial implant and the distal pedicle screw.

Only the distal nut is tightened. The rod is left loose in the proximal implant and not cut to allow the spine to elongate during the correction maneuvers.

Having completed the coupling of implants to the rod on both sides of the spine, in this way the spine will have passively found its own length ( Fig 7.3.3-6d ). The individual instrumented vertebrae are then sequentially derotated as they are secured to the rod starting from each end. This is achieved by placing a derotation force through the sticks attached to each of the implants as the L-handle and 6 mm socket wrench

is used to tighten the nuts ( Fig 7.3.3-6e ). It is important to hold the end vertebrae in their normal, neutral position prior to commencing this process in order to avoid transference of torque from the instrumented to the uninstrumented spine. Cross-links are secured ( Fig 7.3.3-6f ) and decortication completed where necessary. Bone graft is then applied and the wound is closed.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

399

7.3.3 Universal spinal instrumentation for deformity

loose

tightened

loose loose

loose

loose loose loose

loose loose

a

b

c

Fig 7.3.3-6a–f a The implants on either side of the apex are brought to the rod using the complex reduction forceps (not shown here). b The rod on the convex side is inserted into the proximal claw that is compressed and tightened. Then the rod is pushed across into the lateral side opening of the T8 implant. The T8 nut is not tightened. c The convex rod is further engaged in the lateral opening of the distal implants at T10, T12, and L1. All the nuts (except the ones of the proximal convex claw are left loose). The rod maneuver on the convex side has translated the spine, and the concave apical screw or specialized pedicle hook is now ready to be engaged to the concave rod.

d

e d All the implants have been engaged to the rods on the concave and convex side. e Derotate the spine around the rod and tighten the nut.

400

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Case examples

Example of a patient with a right thoracic curve treated exactly according to the same principles as the technique illustrated with the hybrid instrumentation ( Fig 7.3.3-7 ). A second example shows a patient with a right thoracic curve treated with a pedicle screw instrumentation ( Fig 7.3.3-8 ).

f Fig 7.3.3-6a–f f Cross-links are inserted.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

a

b

Fig 7.3.3-7a–b a Preoperative x-rays showing a right thoracic curve. b Postoperative x-rays after insertion of a hybrid instrumentation with lumbar pedicle screws and specialized pedicle hooks.

401

7.3.3 Universal spinal instrumentation for deformity

3.2

DOUBLE CURVE

Principle

The same basic concept of reduction used in a spine with a single curve is applicable to double curves. However, in this situation the foundation of the frame onto which the spine is reduced is created on the neutral vertebra between the curves rather than on the distal end vertebra ( Fig 7.3.3-9 ). Each curve is considered separately and the central neutral vertebra becomes the end vertebra for each of the two curves, one extending distally, the other proximally. To clearly demonstrate these principles the instrumentation in a double curve (right thoracic and left lumbar) from T4 to L4 is explained. a

b

Fig 7.3.3-8a–b a AP and lateral preoperative x-rays showing a right thoracic curve. b Postoperative x-rays following correction of the right thoracic scoliosis with pedicle screw instrumentation.

T12 central neutral vertebra

Fig 7.3.3-9 A double curve is reduced using the neutral vertebra between the curves as a central point for the instrumentation.

402

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Pedicle screws are inserted bilaterally into the neutral vertebra—here at T12 ( Fig 7.3.3-10a ). The upper curve is instrumented as previously described in the single right thoracic curve ( Fig 7.3.3-10b ). The distal curve is instrumented in a similar way, and the use of pedicle screws is recommended ( Fig 7.3.3-10c ). The rods are contoured to create a normal sagittal profi le in the thoracic and lumbar spine. The posterior facets are decorticated and the rod inserted into the upper concave side of the thoracic curve. This rod is then locked to the neutral vertebra, in this case at T12, while holding the rod with a rod holder in the correct sagittal alignment. The concave side of the thoracic curve is reduced as described in the text above relating to a single curve ( Fig 7.3.3-10d ). The convex lumbar vertebrae (L2, L3, and L4) are then sequentially brought to the sagittally fi xed rod starting with the vertebrae that are closest to the rod. The complex reduction forceps (persuader) will be used to bring each vertebra across to the fi xed sagittal rod. In some circumstances, when the lumbar curve is very stiff, it may be necessary to apply the second rod on the concave side of the lumbar curve in order to reduce the distal spine to the rod ( Fig 7.3.3-10e–f ). This will facilitate reduction of the convex lumbar spine to the rod. The only nuts that should be tightened at this stage are those at T12. Once all the implants have been reduced to the rod, judicious compression/distraction of the distal lumbar vertebrae will be applied to horizontalize L4 ( Fig 7.3.3-10g ). An intraoperative AP x-ray or fluoroscopic view may be necessary to judge the horizontal plane (horizontalization) of L4. The nuts are then locked, cross-links are attached, the rest of the spine decorticated, and the bone graft is applied ( Fig 7.3.3-10h ).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

T4 T5 T6 T7 T8 T9 T10 T11 T12 L1 L2 L3 L4

a Fig 7.3.3-10a–h a Instrumentation of the neutral vertebra bilaterally (here T12) with pedicle screws.

403

7.3.3 Universal spinal instrumentation for deformity

loose T4 T5

T6 T7 T8 T9 T10 tightened

T11 T12 L1 L2 L3 L4

b

c

d

Fig 7.3.3-10a–h b First, insertion of implants into the vertebrae of the upper curve. c Second, insertion of pedicle screws into the vertebrae of the lower curve. d The thoracic curve is reduced according to the same principles as described for the simple right thoracic curve.

404

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

T4 T5 T6 T7 T8 T9 T10 T11 T12 L1 L2 L3 L4

e

f

Fig 7.3.3-10a–h e–f The thoracic curve is now instrumented and reduced. The convex lumbar vertebrae are brought to the rod with the persuader. In a stiff curve working on the concave side of the lumbar curve, it may be necessary to translate the spine.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

g

h

g Judicious compression/distraction of the distal part of the lumbar curve may be necessary to horizontalize L4. h Two cross-links are applied to increase the rigidity of the system.

405

7.3.3 Universal spinal instrumentation for deformity

4

Case example

A 12-year-old girl with a double curve was treated with posterior spinal fusion and hybrid instrumentation from T2 to L3 ( Fig 7.3.3-11).

SCOLIOSIS—ANTERIOR CORRECTION AND STABILIZATION

The USS II makes it possible to address a scoliotic deformity from an anterior approach using the same basic set of instruments and modified screws with an expanded thread for cancellous bone. The coupling of the implants to the rod is undertaken in the same way, using identical collars and nuts, as other implants contained in the system. Concept

Modified threaded pedicle screws for cancellous bone of 6.2 mm and 7.0 mm diameters are used for anterior anchorage into the vertebral bodies. The standard 5 mm or 6 mm rods are used. The concept of correction is the same as in posterior surgery. The vertebrae to be instrumented are determined during preoperative planning and are instrumented with a single transverse screw. The prebent rod is inserted and coupled to the end vertebrae. The adjacent instrumented vertebrae are then gradually reduced, translated, and derotated onto the rod using the persuader, resulting in a 3-D correction of the deformity. The anterior approach to the spine at any level of the thoracic and lumbar spine is considered to be more demanding for the surgeon and the patient, but the correction of the deformity is likely to be superior to posterior surgery. a

b

Fig 7.3.3-11a–b a Preoperative AP and lateral x-rays show the double curve. b Postoperative AP and lateral x-rays following hybrid instrumentation with the universal spine system from T2 to L3.

A combination of anterior surgery, to release and mobilize the spine, followed by posterior correction and instrumentation may be indicated in severe and rigid curves. In this situation an anterior release is undertaken without anterior instrumentation. With the introduction of more powerful segmental instrumentation systems such as the USS II, and the use of preoperative traction x-rays with the patient anesthetized prior to commencing corrective surgery, anterior release alone is indicated in only the most significant and rigid curves. Where this approach is warranted the release may be done thoracoscopically or via a miniopen approach reducing the operative morbidity of this surgery.

406

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Principles

Indications

All the vertebrae involved in the deformity from the end vertebra above to the end vertebra below are usually instrumented with either 6.2 mm or 7.0 mm screws. The screws are inserted into the vertebral body from the lateral convex side of the vertebral body, usually slightly above the central portion of the vertebra near the origin of the pedicle, but leaving enough space to enable passage of the screw and application of a staple without endangering the content of the spinal canal. Screws are inserted through staples at the upper and lower end vertebrae, and a washer of the appropriate size is used over the screw in the intervening levels. The rod is prebent from a template contoured in the desired sagittal plane, ie, with a lordosis in the lumbar and kyphosis in the thoracic spine. The thoracolumbar region (T12/L1) usually remains fairly straight. The prebent rod is attached and held in the desired sagittal plane to the upper and lower screws. The nut is tightened at only one end in order to maintain the orientation of the rod. The other end is kept loose so the rod can glide during the corrective maneuver and the spine will usually elongate during this process. As the rod is to be placed on the convexity, with the disc spaces open on this side, sequential compression is required to facilitate correction and realignment of the spine. Once coupled to the rod, the individual vertebra can be derotated around the rod prior to fi nal tightening of the nut. This is done while controlling the position of adjacent vertebrae via the stick extensions on the screws inserted in these vertebrae.

• Lumbar scoliosis. • Thoracolumbar scoliosis. • Thoracic scoliosis.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Advantages

• Derotation is achieved more powerfully and directly via anterior instrumentation. • Instrumentation can be shorter than with a posterior procedure, preserving motion segments. • Stable anterior fusion prevents a crankshaft phenomenon in skeletally immature patients. Disadvantages

• The surgical approach is considered to be more demanding, and there may be a need to enlist the assistance of an abdominal and/or thoracic surgeon. • The patient may need to stay in an intensive care unit postoperatively. • It is more difficult to control and balance a more rigid curve via anterior surgery alone.

407

7.3.3 Universal spinal instrumentation for deformity

Surgical technique

Depending on the curve, a thoracotomy, a retroperitoneal approach, or a combination of both with reflection of the diaphragm is performed on the convex side of the curve in order to obtain good access. The exposure of the spine in preparation for the instrumentation is well described in standard textbooks. The parietal pleura and/or the parietal peritoneum are mobilized around and into the concavity of the curve, starting from the lateral aspect of the vertebral body. This is important in understanding the orientation of the vertebral body, and it may also be necessary in order to gain access with the fi nger to the contralateral side of the vertebral bodies (concave side of the curve) to assess penetration of the awl or screw. All the discs between the end vertebrae are removed; the cartilage is removed from the end plates in order to have bare, bleeding bone at the end plates to facilitate fusion. It is also vital to clear the convex posterolateral corner of the disc and end plate in order to optimize correction. The preferred entry point of the vertebral screw is in the posterosuperior quadrant of the vertebral body, just above the waist of the vertebra and anterior to the pedicular origin, leaving

sufficient space superiorly and posteriorly to accommodate the screw and staple or washer without encroaching on the spinal canal ( Fig 7.3.3-12a ). The awl is used to make a starting hole in the near cortex ( Fig 7.3.3-12b ). The pedicle probe is then used to create the path for the screw across the vertebra ( Fig 7.3.3-12c ), and the depth gauge is used to assess the length of screw required to reach the far cortex ( Fig 7.3.3-12c–d ). The staple must then be positioned and secured prior to insertion of the screw. This is achieved using the staple inserter, which is directed along the tract made by the pedicle awl across the vertebral body ( Fig 7.3.3-12e–f ). Once the staple is fully impacted, a screw of the appropriate length is selected and directed through the staple perpendicular to the vertebra along the path made by the pedicle awl to the contralateral side. The tip of the screw should just penetrate the far cortex ( Fig 7.3.3-12g ). Radiological imaging or a fi nger placed in the concavity of the vertebral body will indicate when this has been achieved. Staples are placed at both ends of the construct and maintain a fi xed angle with the screw, reducing the chance of both angulation and pullout of the screw. The intervening vertebrae are instrumented following similar steps with the starting awl, pedicle probe, and depth gauge, but instead of staples, washers are used with their convex side resting on the concavity of the vertebral body ( Fig 7.3.3-12h–i ).

408

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

a

b

c

e es a

e

f

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

g

d

Fig 7.3.3-12a–i a The entry points for the USS II screws are anterior to the origin of the pedicle and at the level of the waist of the vertebral body. b The cortex is opened up using the awl. c–d The blunt pedicle probe is pushed through the contralateral cortex and the penetrating awl can be felt with the index finger on the opposite side of the vertebral body. The screw hole is measured with a depth gauge. e–g Insertion of staples in the proximal and distal vertebrae using the staple inserter that is positioned in the tracts of the pedicle probes. The staple is impacted and an anterior screw with its holding stick is inserted in the vertebral body.

409

7.3.3 Universal spinal instrumentation for deformity

h

i

Fig 7.3.3-12a–i h–i The intermediate screws with their holding sticks are inserted over washers.

The length of the rod and its contour is chosen with the help of the template ( Fig 7.3.3-13a–b ). The appropriate size and length of a hard rod is matched to the template using the rod bender or the bending irons. The prepared rod is introduced into the most cranial and caudal screws. The most cranial screw is tightened with the rod in the position of the planned correction. The most caudal screw is left loose so that the rod can glide during the correction maneuver ( Fig 7.3.3-13c ). The remaining screws are now reduced to the rod using the complex reduction forceps, beginning from the cranial and caudal ends, working toward

the apex. The screw in the apical vertebra is thus gradually brought closer to the rod, and is then itself reduced and coupled to the rod using the persuader ( Fig 7.3.3-13d–e ). Morselized bone graft, bone blocks, or in some cases intervertebral cages are inserted into the disc spaces (see case example Fig 7.3.3-14 ). The graft or the appropriately sized and configured cages are placed anteriorly toward the concavity to assist in the correction/ restoration of spinal alignment and to maintain lordosis.

410

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

The fi nal step is to compress each interspace of the intervertebral graft or cage ( Fig 7.3.3-13f ). Compression should be undertaken sequentially from the end of the construct where the nut has already been tightened to fi x the orientation of the rod relative to the spine. Subsequent implants are then compressed against the adjacent fi xed implant, and a derotation force is applied as the nut is tightened using the L-handle and 6 mm socket wrench ( Fig 7.3.3-13g ). The rotation of the already fi xed adjacent vertebra should be controlled during this process by applying a counter force on the stick, which is attached to the screw implanted in this vertebra and should be left in place until this maneuver has been completed. It is important to not allow the screw to angulate during compression. The L-handle and 6 mm socket wrench must be kept parallel with the adjacent fi xed vertebra during this process. This technique restores the normal sagittal plane of the spine ( Fig 7.3.3-13h ). Alternatively, in a stiff lumbar curve with a kyphosis it is possible to use an anterior double-rod construct with the use of double screw washers.

Video 7.3.3-2 Anterior scoliosis correction with USS II.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

T11

T12

L1

L2

L3

a

b

Fig 7.3.3-13a–h a–b The rod template is used to measure the final rod length and to precontour the rod.

411

7.3.3 Universal spinal instrumentation for deformity

tightened

tightened

loose

c

g

d

e

h

f

Fig 7.3.3-13a–h c–d In the sagittal plane the rod is fixed and locked in the proximal screw, but the distal nut is left loose, to allow autoelongation of the spine during the correction maneuver. The complex reduction forceps are used to bring each intermediate implant sequentially to the rod. The nuts are not yet tightened. e–f Bone graft (red) is inserted into the disc spaces, and all disc spaces are compressed sequentially starting from the top. g–h During the compression maneuvers each vertebra is ”derotated” while the nut is tightened with the L-handle and the 6 mm socket wrench.

412

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

5

KYPHOSIS—POSTERIOR CORRECTION AND STABILIZATION

Case example

Principles

This patient with a lumbar curve and a compensatory thoracic curve was treated with an anterior selective lumbar fusion using the USS I and a Synmesh cage as anterior column support was inserted ( Fig 7.3.3-14 ).

The construct that is applied posteriorly acts like a tension band. Compression causes shortening of the posterior column of the spine—distraction is not recommended in the correction of kyphotic deformities. Not only will posterior distraction tend to accentuate a kyphotic deformity, it also leads to overall elongation of the spinal column and potentially its contained neural elements, which may affect cord perfusion and have neurological consequences.

b

a

Fig 7.3.3-14a–c a Preoprative AP and lateral x-rays show the lumbar and c compensatory thoracic curve. b Intraoperative photograph after insertion of Synmesh cages for anterior column support. c AP and lateral x-rays at the 2-year follow-up.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

The side-opening implants of the USS II sets make the assembly and correction of a kyphotic deformity relatively simple. Preoperative planning is of course a vital step and should include an assessment of the placement and nature of the implants to be used. The standard construct for a posterior correction of a relatively flexible thoracic Scheuermann kyphosis would involve instrumentation between T2 and Ll. In the management of this condition and in other forms of fi xed kyphotic deformity, it is imperative that instrumentation extends up to at least T3 but in most cases T2 should also be included. This is due to the presence of significant anterior restrictive forces resulting from tension in the anterior longitudinal ligament and probably more importantly the rib cage, sternum, and abdominal musculature, which will usually lead to the development of a junctional kyphosis if instrumentation is terminated at T4 or T5.

413

7.3.3 Universal spinal instrumentation for deformity

There may be some variations in the distal extent of the instrumentation on the basis of a lateral standing view of the whole spine and a forced supine extension view over a bolster under the region of maximum deformity. Preoperative plan

The area to be instrumented is dictated by the length of the curve and its flexibility. An anterior release via an open or endoscopic transthoracic approach may be indicated in patients with larger and more rigid kyphotic deformity. In cases of more localized and fi xed kyphotic deformity in older patients, the performance of a posterior closing decancellation osteotomy may be considered by experienced hands. It is important to note the potential risks of this approach and that this type of surgery should only be attempted following appropriate training and experience. For standard thoracic kyphotic deformity, instrumentation is usually indicated from T2 and Ll. Cranial implants are inserted in a claw configuration on each side. Claws can be staggered, on one side at T2 and T3, the other at T3 and T4 if

there is concern regarding the encroachment within the canal of two lamina hooks at the same level ( Fig 7.3.3-15a ). If the capacity of the spinal canal is such that this is not a concern, both claws can be placed at T2 and T3 ( Fig 7.3.3-15b ). Pedicle screws or specialized pedicle hooks are inserted in T3; or in T3 on one side and T4 on the other side as the case may be, and downward facing lamina hooks of an appropriate size bilaterally in T2; or staggered, one in T2 and the other in T3. The spacing of more distal implants will be determined by the rigidity of the deformity. Pedicle screws or specialized pedicle hooks should be used so that sequential compression of the more distal implants toward the proximal claw constructs can be undertaken, gradually correcting the deformity, working toward and beyond the apex of the curve ( Fig 7.3.3-15c–d ). The preference at the caudal end is to utilize pedicle screws supplemented by lamina hooks to resist the pullout force that will be exerted on the distal implants by the tension created in anterior structures ( Fig 7.3.3-15e–f ). Axial compression of the implants along the rods toward the upper claw increases the fi nal correction achieved prior to fi nal tightening of the implants onto the rods. Cross-links are then applied.

414

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

T2

or

or

or T3 T4

or T5

T5

T6

T6

or

or

T7 T7

T8

T8

T9 T10 T11 T12

L1

a

b

c

Fig 7.3.3-15a–f a Staggering insertions of the top two claws on either side. b Alternative symmetrical configuration of the top claws (increased risk of encroachment in the spinal canal). c Proximal fixation of the kyphosis with a staggered insertion pattern.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

415

7.3.3 Universal spinal instrumentation for deformity

T2

T3 T4 T5

or

T6

or

T7 T8 T9 T10

T10

T11

T11

T12

T12

L1

L1

d

e

Fig 7.3.3-15a–f d Alternative symmetrical configuration for the proximal kyphosis. e Distal fixation with pedicle screws and protective infralaminar hooks at T11 and T12. f Alternative distal fixation with pedicle screws and distal infralaminar hooks at L1.

f

416

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Surgical technique

The spine is exposed with subperiosteal dissection between T2 and L1. Pedicle screws or specialized pedicle hooks are inserted as per the preoperative plan. The lamina hooks are inserted either bilaterally in T2 or in T2 on one side and T3 on the other side. The yellow ligament is carefully and minimally removed with a Kerrison rongeur and the superior margin of the lamina is “squared” off by the removal of a small amount of bone at its lateral margin to allow a good, close fit of the lamina hook to the lamina itself. The lamina hooks can be removed at this stage to prevent their inadvertent displacement into the canal while the remainder of the instrumentation is being inserted. The more distal implants are then inserted as per the preoperative plan down to Ll. A 6 mm rod template is placed between T2 and L1 to identify the length of the rod that will be required ( Fig 7.3.3-16a ). Using the rod benders the rod is precontoured to the desired sagittal plane kyphosis. The magnitude of the kyphosis can be assessed by holding the rod in two rod holders, one at each side, and assessing the angle projected between the handles. Some give will be evident in the rods used and for this reason the degree of kyphosis created in the rods must be less than that desired as an outcome of the correction. The T12/L1 section of the rods are also straightened to allow a smooth transition from the thoracic kyphosis to the lumbar lordosis ( Fig 7.3.3-16b ). The 6 mm rods are now inserted into the upper pedicle screws or specialized pedicle hooks, and the sleeves and nuts dropped onto the hooks. These nuts can be secured ensuring the rods are orientated in the correct sagittal plane. If the proximal lamina hooks have been removed to prevent displacement into the canal, ensure enough of the rods remain proximal to the upper pedicle screw or specialized pedicle hook to enable coupling of the lamina hooks to the rod ( Fig 7.3.3-16c ). The lamina hooks are then inserted and attached to the rod and

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

the sleeves and nuts are dropped into place ( Fig 7.3.3-16d ). The hook is controlled with the hook/screw holder extension, or compression is applied between the lamina hook and the adjacent pedicle screw or specialized pedicle hook, and the nut is tightened using the L-handle and 6 mm socket wrench. The claw is now compressed. This is repeated on the opposite side. The rods are then reduced to the next distal implants by applying uniform force on each rod with a rod pusher or rod holder. If more control is required the persuader is placed over the sticks attached to these implants on each side, and using the distraction clip the rod is reduced down onto the implant and the collar applied. The persuader can then be removed and the nut applied while maintaining the correction. Compression is then applied to the implant toward the proximal claw construct by placing a rod holder distal to the most recently coupled implant and applying distraction between the rod holder and pedicle screw or special pedicle hook ( Fig 7.3.3-16e ). The nut can then be tightened to maintain the correction achieved before moving on to the next distal implants and repeating the procedure ( Fig 7.3.3-16f–g ). Prior to fi nal reduction of the rods to the distal implants at T12 and L1 an assessment of the rod length should again be undertaken and the rod shortened using bolt cutters at this stage if indicated. Two rod holders are then applied to the rods, which are gently pushed down together onto the pedicle screws at Tl2 and Ll ( Fig 7.3.3-16h ). The use of two persuaders and distraction clips may again be required at this stage in the management of more rigid curves. Once the rods are reduced onto the implants, the collars and nuts are applied ( Fig 7.3.3-16i ) and compression is applied against a more proximally placed rod holder, or an adjacent fi xed implant, and the nuts are tightened. The lamina hooks are then applied as per the operative plan to protect the distal implants against pullout if indicated ( Fig 7.3.3-16j–k ). Finally two cross-links are applied ( Fig 7.3.3-16l ).

417

7.3.3 Universal spinal instrumentation for deformity

T2

T2

T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 T12

L1

L1 L1

a

b

c

Fig 7.3.3-16a–l a The flexible rod template is placed between T2 and L1 to identify the length of the rod. b–d The rod is inserted into the proximal claw and then sequentially into the other implants in the proximal portion of the kyphosis.

d

418

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

T2

T2

T2

T2 T3 T4 T5

L1

e

f

g

Fig 7.3.3-16a–l e–g Compression is applied toward the proximal claw, starting with the implant at T5 and then at T7. h–i Cantilevering the rod in order to insert it into the pedicle screws in T12 and L1.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

L1

L1

L1

h

419

7.3.3 Universal spinal instrumentation for deformity

T2

T2

T11

L1

L1

i

j

k

Fig 7.3.3-16a–l j A lamina hook is added at T11 to protect the distal screws from pulling out. k Alternative use of lamina hooks under the lamina of L1. l Final construct with two cross-links in place.

l

420

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

6

1.

Case example

A 14-year-old girl with a 70° Scheuermann kyphosis was treated with a posterior fusion from T2 to L1 and a staggered claw construct in the proximal part and pedicle screws protected with lamina hooks in the distal part of the kyphosis ( Fig 7.3.3-17 ).

2.

3. 4.

5.

6.

7.

8.

9.

10.

a

b

c

Fig 7.3.3-17a–c Preoperative and postoperative x-rays of a 14-year-old girl with a 70° Scheuermann kyphosis.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

11.

12. 13.

BIBLIOGRAPHY

Muschik M, Schlenzka D, Robinson PN (1999) Dorsal instrumentation for idiopathic adolescent thoracic scoliosis: rod rotation versus translation. Eur Spine J ; 8(2):93–99. Jia QZ, Gao JC, Zhang CM (2003) [Surgical treatment of kyphosis with universal spine system.] Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi ; 17(4):276–278. Laxer E (1994) A further development in spinal instrumentation. Technical Commission for Spinal Surgery of the ASIF. Eur Spine J ; 3(6):347–352. Pratt RK, Burwell RG, Cole AA, et al (2002) Patient and parental perception of adolescent idiopathic scoliosis before and after surgery in comparison with surface and radiographic measurements. Spine ; 27(14):1543–1552. Muschik MT, Kimmich H, Demmel T (2006) Comparison of anterior and posterior double-rod instrumentation for thoracic idiopathic scoliosis: results of 141 patients. Eur Spine J ; 15(7):1128–1138. Burwell RG, Aujla KK, Cole AA, et al (2002) Anterior universal spine system for adolescent idiopathic scoliosis: a follow-up study using scoliometer, real-time ultrasound and radiographs. Stud Health Technol Inform ; 91:473–476. Remes V, Helenius I, Schlenzka D, et al (2004) Cotrel-Dubousset (CD) or Universal Spine System (USS) instrumentation in adolescent idiopathic scoliosis (AIS): comparison of midterm clinical, functional, and radiologic outcomes. Spine ; 29(18):2024–2030. Cohen-Gadol AA, Dekutoski MB, Kim CW, et al (2003) Safety of supplemental endplate screws in thoracic pedicle hook fi xation. J Neurosurg ; 98(1 Suppl):31–35. Berlet GC, Boubez G, Gurr KR, et al (1999) The USS pedicle hook system: a morphometric analysis of its safety in the thoracic spine. Universal Spine System. J Spinal Disord ; 12(3):234–239. Arlet V, Papin P, Marchesi D, et al (1999) Adolescent idiopathic thoracic scoliosis: apical correction with specialized pedicle hooks. Eur Spine J ; 8(4):266–271. Arlet V, Jiang L, Ouellet J (2004) Is there a need for anterior release for 70–90 degrees thoracic curves in adolescent scoliosis? Eur Spine J ; 13(8):740–745. Aebi M (1988) Correction of degenerative scoliosis of the lumbar spine. A preliminary report. Clin Orthop Relat Res ; (232):80–86. Kuklo TR, Lenke LG, O‘Brien MF, et al (2005) Accuracy and effi cacy of thoracic pedicle screws in curves more than 90 degrees. Spine ; 30(2):222–226.

421

422

7 SPINAL INSTRUMENTATION 7.3 THORACOLUMBAR AND SACROPELVIC SPINE 7.3.4 INSTRUMENTATION FOR THE DEGENERATIVE THORACOLUMBAR SPINE (SPONDYLOSIS)

1

Introduction ………………………………………………………………………………………………… 423

2 2.1 2.2 2.3 2.4

Posterior instrumentation techniques ………………………………………………………………… Universal spine system (USS)—degenerative module ……………………………………………… USS variable axis screw …………………………………………………………………………………… Polyaxial top-loading system …………………………………………………………………………… Translaminar screw fixation ………………………………………………………………………………

3 3.1 3.2

Posterior techniques with interbody spacers (PLIF/TLIF) ………………………………………… 431 Posterior lumbar interbody fusion (PLIF) ……………………………………………………………… 431 Transforaminal lumbar interbody fusion (TLIF) ……………………………………………………… 436

4 4.1 4.2 4.3 4.4 4.5

Anterior lumbar interbody fusion devices (ALIF) …………………………………………………… Anterior titanium interbody spacer …………………………………………………………………… Machined femoral ring allograft spacer system ……………………………………………………… Polyetheretherketone (PEEK) spacers ………………………………………………………………… Anterior tension band system for the lumbosacral spine ………………………………………… Stand-alone anterior lumbar interbody device ………………………………………………………

5

Bibliography ………………………………………………………………………………………………… 454

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

424 424 424 425 429

441 441 445 446 447 449

423

Dante G Marchesi

7 7.3

SPINAL INSTRUMENTATION THORACOLUMBAR AND SACROPELVIC SPINE

7.3.4

INSTRUMENTATION FOR THE DEGENERATIVE THORACOLUMBAR SPINE (SPONDYLOSIS)

1

INTRODUCTION

Surgical treatment of the degenerative thoracolumbar spine is by far the most frequently performed instrumentation in the spine. The procedures described represent only part of the surgery that involves decompression of the neural elements and bony fusion. Lumbosacral fusion can still be achieved with very simple and cost-effective instrumentation, such as translaminar screw fi xation and posterolateral fusion [1, 2] or, for example, anterior screw fi xation across L5/S1 with tricortical bone graft from the iliac crest. Recent developments in lumbosacral fi xation for the degenerative spine have led to very sophisticated but expensive technology consisting of polyaxial pedicle screw systems that render the posterior fi xation easier. However, in most instances such fi xation can be achieved with simpler and more cost-effective monoaxial screw fi xations such as the universal spine system (USS). Posterior interbody fusion, either with PLIF or TLIF, is progressively replacing posterolateral fusion with reported clinical success [3, 4]. Likewise, the

cage technology along with bone substitutes and minimally invasive bone harvesting tools have decreased the morbidity of iliac crest bone harvesting [5]. For a long time, stand-alone anterior interbody fusion cages were “banned” from the AO’s spinal community due to the high rate of failures observed. New stand-alone devices with built-in locking screw fi xation (Synfi x) have proved their biomechanical effectiveness for clinical use [6]. However, it is not clear today if all these expensive and complex technological advances have changed the long-term outcome of the pathology treated [7]. This chapter provides a review of the different instrumentation possible for the degenerative lumbar spine. The motion-preserving technology is addressed in chapter 7.5.2 Arthroplasty in lumbar spine surgery.

424

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

2

POSTERIOR INSTRUMENTATION TECHNIQUES

2.1

UNIVERSAL SPINE SYSTEM (USS) —DEGENERATIVE MODULE

The surgical technique is described in chapter 7.3.3 Universal spinal instrumentation for deformity.

Principles

The modularity and general principles of the USS degenerative module and the dual side-opening USS pedicle screws are described in chapter 7.3.1 Modularity of the universal spine system. The USS II can be used for all degenerative conditions. Advantages

• The dual opening allows for adjustment of any possible malalignment of the pedicle screw fi xation due to local anatomy changes (such as scoliosis or imperfection in the pedicular screw fi xation technique). • The removal or addition of a screw while the rod is in place is possible thanks to the side-opening configuration. • Can be used in combination with other systems (eg, VAS, Click’X, Pangea). • More cost-effective than polyaxial screws. • Same system for all posterior instrumentation. • The use of a 6 mm soft rod for the degenerative spine may decrease adjacent segment disease, compared to stiff rods.

2.2

USS VARIABLE A XIS SCREW

Principles

The USS variable axis screw (VAS) complements the USS system as a special feature screw. This screw has the same thread dimensions as the standard USS side-opening pedicle screws and is available in 6.2 mm, 7.2 mm, and 8.0 mm diameters, and in 30–60 mm lengths. The USS variable axis screw consists of the screw itself with a spherical head, the rod-screw connector, the locking ring, the collar, and the 11 mm nut ( Fig 7.3.4-1). Indications

• Degenerative lumbar spine. • Delta fi xation of spondylolisthesis (chapter 7.3.5 Fixation of the sacrum and pelvis. • Degenerative scoliosis (preferably not at the ends of the construct). Advantages

Disadvantages

• Fixed-angle screws may make the bone-screw interface fragile, particularly when forcing the screw into the rod in osteoporotic bone. • Requires perfect placement of pedicle screws, which need to be perpendicular to the rod. • Perfect rod contouring to match the lordosis is necessary for insertion of the rod into the fi xed-angle screws. • The use of lateral connectors connecting the rod to the screws may be needed, but they increase the bulkiness of the instrumentation.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

• A ±36° freedom of movement facilitates the rod-to-screw connection and minimizes the need for contouring the rod. • The use of a VAS instead of a USS screw may obviate the need for a lateral connector. • Can be added or removed without taking down the whole instrumentation. • Reduced risk of bone-screw-interface failure in osteoporotic bone.

425

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

2.3

Disadvantages

• It is a bulky and high-profi le implant. • When used concomitantly with USS II screws, there is a vertical offset that needs to be accomodated by leaving the USS II screws raised so that the 6 mm rod can fit both VAS and USS II screws. • It can only be used in the lumbar spine. • Risk of stripping the screw head in hard bone. • The locking ring may disengage during forceful reduction of the screw to the rod. The surgical technique for the side-opening locking system is similar to the one for the USS using a collar and nut. 1

POLYA XIAL TOP-LOADING SYSTEM

Principles

This polyaxial top-loading pedicle screw system (Click’X) consists of pedicle screws with a special spherical head, 3-D heads that are clicked on the pedicle screws, and locking caps ( Fig 7.3.4-2 ). Indications

• Degenerative spinal instabilities (uni- or multisegmental), for example, degenerative spondylolisthesis. • Iatrogenic instability after decompressive surgery. • Posterior stabilization after posterior lumbar interbody fusion (PLIF) or transforaminal lumbar interbody fusion (TLIF).

2 3 4

5

a

b

Fig 7.3.4-1a–b a Variable axis screw assembly. 1 nut 2 collar 3 locking ring 4 rod-screw connector 5 screw

32°

32°

b 32° angulation around the axis of the screw.

a

b

25°

25°

Fig 7.3.4-2a–b a 3-D pedicle screw. b Pedicle screw and its locking cap mounted onto the rod; allows ± 25° of movement in all directions.

426

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Contraindications

• Any polyaxial system is contraindicated when treating spinal fractures. • Spinal deformities. • Reduction of a high-grade spondylolisthesis. • Tumors. Advantages

• • • •

It is a top-loading system that is easy to use. It allows for ± 25° movement in all directions. There is no need for complex rod bending. It is forgiving when dealing with a malaligned screw placement. • Better for the bone-screw interface in osteoporotic bone. Disadvantages

• As in any top-loading system it is impossible to remove or add a component without taking out the whole implant. • This implant is bulkier than the USS screw. • Can only be used in the lumbar spine. • It is more difficult to have a convergent screw trajectory than with a monoaxial USS screw with its extension stick. • The removal of a polyaxial screw during revision surgery may be far more complex than the removal of a USS screw.

align the top of the screws in one plane. Pick up the 3-D head directly from the screw rack using the self-holding positioning holder. Click the 3-D heads on each inserted pedicle screw ( Fig 7.3.4-3b ). Use the rod template to predetermine the correct length and contour of the 6 mm USS soft rod ( Fig 7.3.4-3c ). Note that the ± 25° freedom of movement compensates a certain lateral screw offset and reduces the need for precise contouring of the rod. Titanium rods should not be bent more than 45°; care must be taken that the rods are not bent forward and backward (danger of breakage). Insert the rod into the aligned 3-D head using the USS rod holder ( Fig 7.3.4-3d ). With the rod pusher, push the rod into the 3-D head ( Fig 7.3.4-3e ). For instrumentations with more than two pedicle screws on each side, start with the most central 3-D head. Pick up the locking cap from the screw rack using the selfholding screwdriver, slide it through the already placed rod pusher onto the 3-D head and tighten it ( Fig 7.3.4-3f ). The rod is now secured in the 3-D head while maintaining its polyaxial freedom. Repeat this for all remaining 3-D heads before tightening the set screws with the hexagonal screwdriver. The rod pusher is used to counteract the tightening movement ( Fig 7.3.4-3g ).

Surgical technique

The pedicle is opened as described in chapter 7.3.2 Stabilization and reconstruction techniques for the thoracolumbar spine, and the depth is measured. The appropriate pedicle screw is picked up directly from the screw rack using the self-holding screwdriver and inserted into the pedicle ( Fig 7.3.4-3a ). Make sure there is enough space around the ball top of the pedicle screw to be able to click on the 3-D head. Take care to

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Compression/distraction with the mobile 3-D head

Compression or distraction with a mobile 3-D head can be carried out as long as the set screw of the pedicle screw has not been tightened using the compression/distraction forceps ( Fig 7.3.4-3h ). After compression or distraction the set screws can then be tightened.

427

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

Remobilization

Video 7.3.4-1 Lumbosacral fusion with Click’X.

The 3-D heads can be remobilized at any time using the extraction forceps ( Fig 7.3.4-3i ). After having removed the rod, insert the working end of the forceps into the 3-D head and press the handle. However, remobilization can become a challenge during revision surgery, and the rod might have to be cut on either side of the screw in order to remove the screw. While the screw head is still locked to the rod, it is then possible to remove the screw with a rod holder (placed in the small portion of the rod) that is untwisted.

a

b

Fig 7.3.4-3a–i a Insertion of screw with the self-holding screwdriver. b 3-D head is in the self-holding positioner holder to be inserted into the pedicle screw. c Using the rod template, the correct length and curve of the rod is determined.

c

428

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

d

f

e

g

h

Fig 7.3.4-3a–i d Insert the correct rod with the 6 mm USS rod holder. e Use the rod pusher to push it into the 3-D head. f With the self-holding screwdriver, insert the locking cap through the rod pusher. g The screws are tightened using the hexagonal screwdriver. h Compression (or distraction) can be applied as long as the set screws are loose. i The 3-D head can be remobilized using the extraction forceps after having removed the rod.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

i

429

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

2.4

TRANSLAMINAR SCREW FIX ATION

This traditional fi xation technique is biomechanically less stable than instrumentation with pedicle screws, but it is regaining popularity when combined with an interbody spacer (PLIF or ALIF) [1, 2]. Principles

Stabilization of a posterior fusion over one or two motion segments is achieved by transfi xing the facet joints. Indications

• Supplementary fi xation of a degenerative lumbar segment treated with ALIF, PLIF, or TLIF. • Augmentation of posterolateral lumbar fusion.

Surgical technique

Using a 3.2 mm oscillating drill attachment and the corresponding drill guide, the screw hole is started at the base of the spinous process. The hole is then drilled through the lamina and traverses the facet joint, so that the drill exits near the base of the transverse process ( Fig 7.3.4-4a ). The length of each hole is measured and the beginning of the screw canal is tapped with a 4.5 mm tap. The tap only needs to cross the facet joints if the joints are very sclerotic. The appropriate 4.5 mm cortex screw is then inserted ( Fig 7.3.4-4b ). When preparing the fi rst screw hole on the spinous process, it is important to realize that two screws need to be inserted ( Fig 7.3.4-5 ). The fi rst screw should therefore enter the spinous process and lamina slightly off of the ideal position, either cranially or caudally, in order to leave enough space for the second screw.

Contraindications

• Multilevel (more than two levels) stabilization of the spine. • Spondylolithesis. • Deformities and fractures. • Osteoporosis. Advantages

• Translaminar screw fixation can be performed with minimal access to the spine (cannulated screw technique). • It is a fast and cost-effective technique. Disadvantages

• This fi xation is biomechanically less stable than a fi xation with pedicle screws. • It is not applicable with laminectomy and difficult with extended laminotomy. • Lordosis cannot be restored as well as with pedicle screws.

a

b

Fig 7.3.4-4a–b a Positioning and direction of a special long 3.2 mm drill bit, protected with a drill sleeve. With the oscillating drill attachment the surgeon drills through a stab wound of the skin. b Insertion of the 4.5 mm cortex screw of appropriate length.

430

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Given their diverging direction, translaminar screws cannot be used as lag screws. They function as threaded bolts, which prevent motion in the respective segment, but do not exert compression in the facet joint. To avoid a long incision and extensive stripping of the soft tissues, especially when this technique is used in combination with an anterior interbody spacer, drilling and screw placement can be performed percutaneously after having exposed just the spinous process and its base. A posterolateral fusion is then carried out. At L5/S1, because the lamina of L5 is thinner, translaminar screw fi xation may be more challenging. Such fi xation can therefore be replaced by a transfacet screw fi xation or Boucher technique. Another alternative is a translaminar screw on one side and a “Boucher screw” on the opposite side at L5/S1.

Video 7.3.4-2 Translaminar screw fixation in the lumbar spine.

Case example

A 36-year-old woman with a degenerative disc disease at level L5/S1 was treated with an anterior titanium cage and posterior translaminar screw fi xation ( Fig 7.3.4-6 ).

b

a

c

Fig 7.3.4-5a–c a Posterior view of the positioning of two translaminar screws. b Screws seen in the axial plane. c View from the side of the two inserted screws.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

a

b

Fig 7.3.4-6a–b Postoperative AP and lateral x-rays.

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

3

POSTERIOR TECHNIQUES WITH INTERBODY SPACERS (PLIF/TLIF)

Spacers or cages can be positioned through a posterior approach using PLIF or TLIF devices (posterior lumbar interbody fusion or transforaminal lumbar interbody fusion), or vertebral body reconstruction devices. It is not recommended to use spacers as stand-alone implants, but to supplement them with instrumentation in compression.

3.1

POSTERIOR LUMBAR INTERBODY FUSION (PLIF)

Principles

Through a posterior approach anterior column support is provided and the disc height is restored in order to open the neural foramen. Different devices are available, including allograft spacers or cages (such as the contact fusion cage). Indications

• • • •

Degenerative lumbar segments (from L3 to the sacrum). Degenerative spondylolisthesis. Discogenic low back pain. Pseudarthrosis of a posterolateral fusion.

Contraindications

• • • • •

Severe osteoporosis. Infection. More than three affected levels. Grade III spondylolisthesis. Severe epidural scarring.

Advantages

• PLIF provides anterior column support through a posterior approach. • This technique keeps the posterior facets intact for stability, fusion, and translaminar screw fi xation.

Disadvantages

• It requires nerve retraction. • Epidural bleeding may render the surgery difficult. Surgical technique

A midline incision over the levels to be instrumented is performed. The muscles should not be stripped more laterally than the lateral aspect of the facet joints unless a posterolateral fusion between the transverse processes is planned. In the case of intact posterior elements the lamina of the cephalad vertebra can be used for the application of translaminar screws. Pedicle screws for additional posterior instrumentation can be inserted at this time or after implantation of the cage. The rod, however, is mounted on the screws only after the cages have been inserted. The epidural space is then exposed: The spinous processes of the vertebrae to be fused are freed from all soft tissue. They are shortened in order to gain wider access to the interlaminar space ( Fig 7.3.4-7a ). The removed bone is stored in a container under a moistened gauze, to serve as graft material. In order to allow as free an access as possible to the dural sac and roots, a partial inferior laminotomy (1/3) of the upper adjacent vertebra is performed. The medial half of the facet joints is removed with a partial resection of the overlying inferior facet and lateral part of the laminar edge ( Fig 7.3.4-7b ). At the L5/S1 level, the distal half of the lamina of L5 usually needs to be removed in order to ensure instrument access to the disc space. The underlying superior facet of S1 is then nibbled away to the level of the medial aspect of the pedicle. (It is essential to make sufficient room laterally to avoid excessive retraction of the neural tissue, but great care should be taken to protect the nerve root inside the foramen.)

431

432

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

The next step is to evacuate the disc space by a bilateral discectomy. The dural sac is then carefully pushed toward the midline to expose the disc space. The posterior annulus is opened and resected to allow the introduction of the appropriate cage probes. The nucleus has to be cleaned out as completely as possible and the end plates need to be freed from the cartilage without perforating the bone. A good instrument to prepare the end plates with is the ring curette in different sizes ( Fig 7.3.4-7c–d ). During these maneuvers the nerve roots and the dura are protected with the appropriate root retractor.

a

b

The selection of the defi nitive size of the device is obtained through fluoroscopic and tactile judgment confi rming the good fit of the trial spacer. Once the disc space is prepared, it is opened with a PLIF distractor—trial spacers of different sizes can be inserted and rotated at 90° to open up the disc space. Starting on one side, while leaving the distraction on the other side, the fi rst cage is introduced into its defi nitive vertical position in the disc space ( Fig 7.3.4-7h ). Extra graft material can now be packed in the anterior and medial aspect of the disc space before placing the second cage on the other side ( Fig 7.3.4-7i–j ).

c

Fig 7.3.4-7a–j a–b Exposure of the epidural space, partial resection of the spinous processes, and laminotomy with special attention paid to preserving the facets. c Cleaning of the intradiscal space and removal of the cartilage of the end plates with the ring curette.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

d

g

e

f

h

Fig 7.3.4-7a–j d The end plates are prepared using the intervertebral disc shaver or a ring curette. e Distraction of the disc space with the help of the PLIF distractor. f Distraction of the disc space can alternatively be done with the PLIF trial spacer that can be inserted horizontally and rotated at 90° to open up the disc space. g Sizing of the implant is done by leaving the PLIF distractor on one side and inserting the implant trial on the other side. h Insert the PLIF allograft spacer on the opposite side to the PLIF distractor (alternatively, controlateral to the implant trial left on the other side or the detachable spacer), in order to avoid simultaneous bilateral dural retraction.

433

434

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Different PLIF implants can be used:

i

j Fig 7.3.4-7a–j i After the first PLIF implant has been inserted, pack autogenous bone graft medially and anteriorly in the disc space. The second spacer is then inserted. j Final representation with bilateral pedicle screw instrumentation (for a one-level fixation translaminar screws can be used instead, if enough laminae and facets have been respected—see PLIF case example).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

• PLIF allograft—surgical technique described above. • The contact fusion cage is a titanium cage that has a rectangular cross-section. Viewed from the side, it has a lenticular shape that prevents it from backing out posteriorly into the spinal canal. Contact fusion cages are available in different sizes (up to 17 mm in height) and are designed according to the anatomical shape of the disc space and the end plates. The cage is open at the top and bottom where the bone graft is packed, so that the graft comes in contact with the adjacent end plates. They are introduced on the flat side and rotated 90° in order to spread the disc space and to bring the cage into its fi nal position with the bone graft in contact with the end plates ( Fig 7.3.4-8 ). • Plivios: These PEEK (polyetheretherketone) spacers have the same external shape as the contact fusion cages. They have a scaffold-like structure, which reduces the overall amount of synthetic material needed. A central hole allows the use of graft. The vertebral spacer contains two radiopaque pins for proper placement under fluoroscopy and recognition during follow-up evaluation. The selection of the spacers and their placement in the intervertebral disc space is the same as for the contact fusion cage. The graft material is placed prior to the introduction of the second implant in the anterior and medial aspect of the disc space in order to create a lining between the two spacers ( Fig 7.3.4-9 ).

435

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

a

90°

b

c

Fig 7.3.4-8a–c The contact fusion cage is an alternative to the allograft PLIF spacer. Its insertion requires rotation of the implant 90° to increase the disc space height as well as for the bone graft inside the cage to be in contact with the end plate.

Video 7.3.4-3 L4/5 PLIF with contact cages and Click’X pedicle screws.

Fig 7.3.4-9 Plivios is a PLIF cage made of PEEK. The implant has a scaffold-like structure, which reduces the overall amount of synthetic material needed. A central hole is used to pack the bone graft inside the cage. The vertebral spacer PR contains two radiopaque pins for proper placement under fluoroscopy and recognition in follow-up evaluations.

436

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Case example

3.2

A 50-year-old man suffered from a degenerative disc disease at the level L3/4. He was treated with a circumferential fusion with contact cages and translaminar screw fi xation. The facets were well preserved which allowed the translaminar screw fi xation ( Fig 7.3.4-10 ).

TRANSFORAMINAL LUMBAR INTERBODY FUSION (TLIF)

Principles

TLIF is a unilateral alternative to the PLIF approach that restores the height of the disc space and the lumbar lordosis. Like the PLIF, it relies on distraction of the height of the disc space to decompress the neural elements. Spacers are available as allograft spacers in different sizes (7–17 mm) or in PEEK. Indications

• • • • •

Degenerative lumbar segments (from L3 to the sacrum). Degenerative and isthmic spondylolisthesis. Discogenic low back pain. Pseudarthrosis of a posterolateral fusion. Degenerative conditions that require complete facetectomies. • Anterior column support for the lumbosacral junction for degenerative scoliosis. a

b

Fig 7.3.4-10a–b Postoperative AP and lateral x-rays.

Contraindications

• Infections. • Destruction of the end plates. • Major scarring of the epidural space. Advantages

• Less retraction of the neural elements is needed, and theoretically it is a more expeditious procedure because it is unilateral. • Can be done with a minimally invasive approach or a muscle-splitting approach (Wiltse).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

437

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

Disadvantages

• Possible injury to the exiting nerve root at the level of the dorsal root ganglion. • Decompression of the opposite side is indirect. • Epidural bleeding may make surgery difficult. • Incomplete disc excision.

Surgical technique

Preoperative assessment of the lateral standing lumbar x-rays is essential for planning and sizing of the spacer. The height of the disc space of the adjacent segments needs to be assessed. The goal is to restore the height of the disc space in order to achieve maximum stability. Assessment of the shape of the end plates (flat or concave may influence the selection/sizing of the spacer). The side of the TLIF approach is chosen according to which side the patient’s leg symptoms are on. Careful assessment of the opposite foramen to the TLIF approach is done to decide on the need for restoration of foraminal height or direct decompression. Through a standard midline incision, or a minimally invasive muscle-splitting approach, the landmarks for pedicle screw insertion are identified (alternatively, regular USS screws or polyaxial screws from the Click’X, VAS, or Pangea system can be used). Positioning of the pedicle screws is checked under AP and lateral fluoroscopy. Distraction of the interlaminar space is achieved with either a lamina spreader (preferred technique) ( Fig 7.3.4-11a ) or with the lateral distractors. The lamina spreader should be progressively distracted as the facetectomy and discectomy are carried out. Too vigorous a retraction can lead to fractures of the spinous processes. The lateral distractor may be too cumbersome to perform the discectomy and the preparation

of the end plates ( Fig 7.3.4-11b ). Too much of an opening of the lateral distractor encourages a loosening of the pedicle screws in patients with osteoporotic bone. Then the foraminal window needs to be created ( Fig 7.3.4-11c ). With the 12 mm TLIF osteotome, fi rst remove the inferior articular process, then the superior articular facet using an ostetome or a 3 mm Kerrison rongeur. Identification of the epidural space medial to the facet may be necessary before removing the superior facet. The disc space is then identified and the traversing and exiting nerves are identified. The exiting nerve root with its dorsal root ganglion (DRG) is visible in the superolateral aspect of the disc. It is mobilized in a cephalad direction and protected with a thrombin-soaked pattie. In spondylolithesis, the exiting nerve root is in an even closer relationship to the disc space and more at risk of injury. The traversing nerve root is identified and pushed distally and protected with a pattie. Epidural bleeders are coagulated with bipolar coagulation. The working channel of the TLIF spacer has now been created ( Fig 7.3.4-11c ). It is limited proximally by the DRG of the exiting nerve, medially by the dura that can be gently retracted with a dura retractor,

Video 7.3.4-4 Transforaminal lumbar interbody fusion with a Travios cage and Click’X dual core.

438

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

inferiorly by the pedicle, and laterally by the descending DRG and its nerve. With a 15 mm blade the discectomy is started and with the help of pituitary rongeurs, a straight and an angled curette, rasps, and a narrow Cobb the disc is separated from the end plates ( Fig 7.3.4-11d ). Once the disc has been removed the end plates are prepared with a right angled rasp. During all this preparation it is essential that the end plates are kept intact to prevent subsidence of the spacers. A TLIF trial spacer of appropriate height is then inserted to determine the correct implant height ( Fig 7.3.4-11e ). Stability of the trial spacer is tested after partially releasing the lamina spreader; lateral fluoroscopy is helpful in judging the appropriateness of the spacer height. The trial is then removed.

a

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

The prepared disc space is then fi lled with bone graft using an autogenous bone graft from the iliac crest, any bone substitute, or an osteoinductor-soaked collagen sponge. The use of the bone graft funnel facilitates its insertion ( Fig 7.3.4-11f ). Graft is packed anteriorly and on the sides of the disc space. An implant of correct size is then inserted with its implant holder ( Fig 7.3.4-11g ). Lateral C-arm control images may help to determine the correct position of the implant. Its fi nal postioning is achieved using special implant impactors ( Fig 7.3.4-11h–j ). Distraction is released, and the stability of the TLIF spacer is checked. Additional bone graft is inserted behind the spacer, and the rods are inserted into the pedicle screws.

Fig 7.3.4-11a–j a Place the lamina spreader at the base of the spinous processes and apply distraction. This maneuver opens the posterior disc space and promotes increased exposure for decompression. b Alternatively, the lateral distractor can be used. Distraction can be applied between the heads of the inserted screws.

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

c

d

e

Fig 7.3.4-11a–j c Create the transforaminal window with the osteotome and/or Kerrison punch. View with the inferior articular facet removed, and transforaminal window created by unilateral facetectomy. d Using the straight, the reverse angled, and the rectangular curette, and the hockey stick rasp to remove disc material and prepare the end plates. e Introduction of the TLIF trial spacer.

439

440

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

*

f

j

g

h

i

Fig 7.3.4-11a–j f Insertion of cancellous bone graft into the cleaned disc space using a special funnel (see h). g Insert the TLIF spacer with the implant holder. h Axial view of implant insertion (asterisk = cancellous bone graft filled in before insertion of the cage). i Final insertion if the implant with special impactors. j Final construct with bilateral pedicle screw fixation. Fig 7.3.4-12 Alternatively, the TLIF side can be instrumented with pedicle screws, and the opposite side with a translaminar screw in order to minimize the approach (mini-TLIF).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

4

Postoperative care

The patient is mobilized the next day, and activities are progressively resumed as the fusion becomes solid. Variations in the TLIF technique—mini-TLIF

In the case of a unilateral approach, the opposite side can be instrumented with a translaminar screw according to Magerl (except if spondylolisthesis is present) or with percutaneous pedicle screw fi xation ( Fig 7.3.4-12 ).

ANTERIOR LUMBAR INTERBODY FUSION DEVICES (ALIF)

Different lumbar interbody devices can be used for anterior fusion, such as a femoral ring allograft or a Syncage. These implants should not be used as stand-alone devices. They require either an additional anterior fi xation (anterior tension band plate) or a posterior fi xation, such as a translaminar screw fi xation or a pedicular screw system. 4.1

ANTERIOR TITANIUM INTERBODY SPACER

Principles

Case example

A 30-year-old man with axial pain and pain in his left leg was treated with a transforaminal interbody fusion and USS pedicle screw fi xation ( Fig 7.3.4-13 ). Two allograft TLIF spacers were used in this specific case as opposed to the described technique with only one spacer. The symptoms completely resolved after surgery.

This titanium anterior lumbar intervertebral spacer (Syncage) has been developed to restore lordosis and disc height. It has a wedge-shaped design in order to match the anatomical endplate curvature. The denticulated surface of the cage increases its initial stability. The large implant surface reduces the risk of subsidence and the open implant structure facilitates bone ingrowth through the cage [8, 9]. Indications

• Anterior lumbar interbody fusion for degenerative lumbar disease in the middle and lower lumbar spine. • Spondylolisthesis. • Failed laminectomy syndrome, pseudarthrosis of posterior fusion. Contraindications

• • • • a

b

Fig 7.3.4-13a–b a Preoperative MRI. b Postoperative lateral and AP x-rays.

Major loss of disc space height with osteoporosis. Spondylolisthesis (grade III and IV). Infections. Fractures.

441

442

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Advantages

• ALIF with a Syncage recreates lumbar lordosis. • It provides immediate secure anterior support in the lumbar spine. • The cage can be fi lled with cancellous bone from the adjacent vertebral body (chapter 7.4.2 Bone harvesting tools).

a secure fit between the end plates to maximize segmental stability. The selected Syncage is then connected to the cage holder, packed with graft material, and defi nitively introduced into the disc space always using the distractor blades as a guide ( Fig 7.3.4-14d–f ) [9]. If necessary, fi nal seating of the implant is done with gentle blows of the appropriate impactor.

Disadvantages

• Difficulty in assessing the bony fusion. • It requires intact end plates to prevent subsidence. • An additional anterior or posterior fi xation is necessary.

*

Surgical technique

The Syncage is preferably inserted through a retroperitoneal approach to the disc space. At the level L5/S1 its insertion is best in a sagittal direction; at the levels L4/5 and higher the insertion can be either sagittal, oblique, or even lateral to the disc space. The discectomy is fi rst carried out through an anterior midline window. The lateral walls of the annulus must be preserved to provide adequate support to the cage. The superficial layers of the cartilaginous end plates are removed to expose bleeding bone. Adequate preparation of the end plates is important to facilitate vascular supply into the graft material. However, excessive cleaning may weaken the end plates and cause cage subsidence. Intervertebral disc distraction is performed by using the specially designed distractor and placing its two blades on the midline ( Fig 7.3.4-14a ). Once the desired level of distraction is achieved, the size of the implant is determined using trial implants corresponding to the four Syncage sizes (large, medium, small, extra-small). The trials are slid through the distractor blades, which guide their introduction ( Fig 7.3.4-14b–c ). The trial must be fi rmly seated with

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

a

b

Fig 7.3.4-14a–f a Insertion of the distractor. b Attachment of the implant holder to the trial implant (asterisk = track which fits exactly the distractor blade guiding the implant to be seated in the intervertebral space).

443

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

c

f

d

e

Fig 7.3.4-14a–f c Distraction applied with insertion of the trial implant. d The appropriate cage is attached to the implant holder and packed with cancellous bone. e Insertion of the appropriate sized cage (guided by the distractor blades). f View of the inserted cage.

444

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

The Syncage can also be inserted through a lateral or an anterolateral approach to the disc. The implantation is essentially the same as for an anterior approach. The trials and the cages are laterally connected to their holders ( Fig 7.3.4-15a ). The lateral distractor is inserted, the disc space distracted, and the appropriate size cage is inserted ( Fig 7.3.4-15b–c ). Posterior fi xation with translaminar screw fi xation or pedicle screws is mandatory.

Case example

A 43-year-old female with isthmic spondylolisthesis and low back pain, but without any leg pain, was treated with and anterior interbody titanium cage followed by posterior pedicle screw fi xation ( Fig 7. 3.4-16 ). Her lordosis was completely restored and her back pain disappeared.

a

a

b

c

Fig 7.3.4-15a–c a Attachment of the trial implant holder in an oblique and lateral position. b Anterolateral approach (30° offset) and distraction of the disc space. c Insertion of the trial implant before final cage insertion.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b Fig 7.3.4-16a–b a Preoperative x-ray and MRI. b Postoperative lateral and AP x-rays.

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

4.2

MACHINED FEMORAL RING ALLOGRAF T SPACER SYSTEM

Principles

The principles of the machined femoral ring allograft (FRA) are identical to the Syncage. The ring allograft spacers are designed with a curved superior and inferior surface with an anatomical footprint to interface with the end plates. They have a slot on the top and bottom of the spacer to facilitate the use of the distractor. The center portion can be fi lled with autograft ( Fig 7.3.4-17 ).

Fig 7.3.4-17 Machined femoral ring allograft (FRA spacer).

Advantages

• It is similar to allograft. • The middle portion can be fi lled with autograft. • It is less likely to migrate than a classic FRA because of its denticulated surface. Disadvantages

• The allograft spacer may break during insertion. • There is a risk of disease transmission. • This spacer is more likely to migrate or subside than a metallic or PEEK spacer, because of the lack of superior and inferior surface. • Is far more expensive than a classic femoral ring allograft (FRA) that is carved out of a femoral shaft using an oscillating saw. • An additional anterior stabilization to prevent migration (anterior screw with or without washer or buttress plate) might be necessary before the posterior fi xation.

Indications

Surgical technique

• Can be used instead of the Syncage. • The FRA spacer is useful for multilevel interbody support (ie, adult scoliosis).

The surgical technique to be used is identical to the one used for the Synmesh cage. When inserting the FRA spacer, great care must be taken not to break the FRA cage during its impaction in the disc space.

Contraindications

Same as for Syncage.

445

446

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Case example

4.3

A 60-year-old patient suffered from painful adult scoliosis. Through a subdiaphragmatic approach, an anterior release and anterior column support from L1 to the sacrum was done. FRA spacers were inserted at each discectomy level ( Fig 7.3.4-18 ). In order to prevent displacement of the spacers before the posterior surgery, 30 mm long, 6.5 mm cancellous bone screws were inserted.

a

c

b

Fig 7.3.4-18a–c a Preoperative AP and lateral x-rays. b Postoperative AP and lateral x-rays before definitive posterior fusion. c AP and lateral x-rays after posterior spinal fusion.

POLYETHERETHERKETONE (PEEK) SPACERS

Visios spacers have the same concept as the Syncage, but they are made of pure PEEK (polyetheretherketone), a linear and semi-cristalline (35%) thermoplastic polymer. Visios is a radiolucent anterior lumbar intervertebral spacer ( Fig 7.3.4-19 ) and can be inserted through an anterior, anterolateral, or lateral approach. Their posterior curvature enables optimal anatomical adaptation. A large central apperture allows for good bone ingrowth. The wedge-shaped convex design permits restoration of lordosis and ensures conformity with the geometry of the end plate. Its denticulate surface provides immediate stability. The Visios spacer is available in five different heights for adaptation to the natural disc height. This radiolucent spacer is provided with three x-ray markers to facilitate location control. This allows easy assessment of the cage position and fusion progression using conventional imaging methods. An additional posterior fi xation is recommended. Syncage-LR intervertebral spacers ( Fig 7.3.4-20 ) with a slightly different design exist in two footprints (regular and large), two profi les (regular and lumbosacral), and each one in five different heights. The implants are also radiolucent and provided with three x-ray markers. The surgical technique is similar to the Syncage insertion and requires additional fi xation.

Fig 7.3.4-19 Visios spacer made of PEEK.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.3.4-20 Syncage-LR made of PEEK.

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

4.4

ANTERIOR TENSION BAND SYSTEM FOR THE LUMBOSACRAL SPINE

Principle

The anterior tension band system (ATB) is a low-profi le anterior plating system that can be used in the lumbar spine specifically at the levels L5/S1 and L4/L5. Indications

• Complementary fi xation of the lumbar spine after anterior column support with femoral ring allograft or cages. • Anterior stabilization of the lumbosacral junction after anterior corpectomy for tumor treatment. Advantages

• Its low profi le. • Obviates the need for posterior complementary fi xation. • One-step locking cancellous bone screw system. Disadvantages

• Despite its low profi le it may represent a risk for the anterior vascular structures if it is implanted so that it comes into contact with them. In some situations, where the vascular anatomy is not favorable, these plates should not be inserted and a complementary posterior fi xation or an alternative (stand-alone cage with built-in anterior locking plates) should be used. • As in any anterior lumbosacral surgery there is a risk of retrograde ejaculation. • Potentially dangerous for the same reasons in case there is a need to revise and remove the plate.

Contraindications

• • • •

Marked osteoporosis. Spondylolisthesis. Scoliosis. Lack of adequate interbody support.

Surgical technique

After anterior interbody fusion with a femoral ring allograft or other interbody spacers, the plate is positioned over the disc space ( Fig 7.3.4-21a ). Anterior osteophytes or prominent anterior lips of the vertebral bodies may have to be smoothed out before instrumentation. At L5/S1 the small step on the plate allows perfect fitting of the plate over the superior sacral end plate ( Fig 7.3.4-21b ). At L5/S1 the plate is positioned anteriorly, at L4/L5 the plate is positioned anterolaterally to avoid contact with the left common iliac vein. Small threaded drill guides are inserted into the plate. Temporary fi xation pins, which are diagonally opposed, are used to secure the plate to the bone ( Fig 7.3.4-21c ). One should make sure that the temporary fi xation pins lag the plate snugly to the bone. With an awl or a drill bit the screw paths are prepared. Self-locking screws of appropriate length are then inserted into the plate ( Fig 7.3.4-21d–f ). A special compressor system can be used if one wishes to enhance compression across the disc space. Postoperative care

A soft lumbar brace can be used for the fi rst 6 weeks for patient comfort. Isometric exercises for the back and abdominal muscles are encouraged afterward. At 3 months there are no restrictions, including sports.

447

448

7 7.3

a

Spinal instrumentation Thoracolumbar and sacropelvic spine

b

d

c

e

f

Fig 7.3.4-21a–f a–b Positioning of the plate over the L5/S1 disc space with the step in the plate resting over the sacral promontorium. c Insert diagonally opposed fixation pins that hold the plate to the bone, thereby causing a lag effect. d–f Insertion of the self-locking screws and final aspect of the construct.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

449

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

Case example

4.5

A 35-year-old patient with discogenic low back pain was treated with an anterior spinal fusion with femoral ring allograft packed with cancellous autograft and an anterior tension band system ( Fig 7.3.4-22 ).

STAND-ALONE ANTERIOR LUMBAR INTERBODY DEVICE

Principles

Synfi x is an anterior stand-alone interbody cage with a built-in anterior locking plate that obviates the need for complementary fi xation. Its design is a further enhancement of the Syncage and the Syncage-LR. It has an integrated fi xation plate and with the use of locking head screws the implant acts as an anterior tension band ( Fig 7.3.4-23 ).

Fig 7.3.4-23 Synfix-LR is a cage with an integrated plate using locking head screws.

a

b

Fig 7.3.4-22a–b Postoperative AP and lateral x-rays after treatment with an anterior tension band system and femoral ring allograft.

Indications

The indications are the same as for the other anterior interbody devices. Contraindications

• • • • •

Spondylolisthesis. Severe segmental instability. Infections. Fractures. Osteoporosis.

450

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Advantages

• The use of a stand-alone anterior interbody device provides immediate stability. • There is no need for a complementary anterior or posterior fi xation. Disadvantages

• A strictly anterior approach is required. • Screw placement into L5 for an L5/S1 insertion is difficult. • Potentially difficult for revision surgery.

The aiming device holder is reattached to the aiming device, the aiming device is then loosened without disengaging it completely from the plate, so that it can then be rotated 180° ( Fig 7.3.4-24f ). The aiming device is then relocked to the plate. The third and fourth self-tapping locking head screws are inserted following the same principles as described for the fi rst and second screw ( Fig 7.3.4-24g ). The aiming device is then removed. The implant position ( Fig 7.3.4-24h ) is checked with AP and lateral fluoroscopic control images.

Surgical technique

Planning with preoperative x-rays and using the Synfi x x-ray template is necessary in order to determine the approximate size of the implant. Preparation of the disc and the end plates and insertion of the implant into the disc space is a very similar procedure to the one described for Syncage with the use of the anterior distractor ( Fig 7.3.4-24a ). With the implant placed correctly, carefully remove the implant holder and the distraction device. Once the implant is in place, an appropriate sized aiming device is mounted and screwed in the fi xation plate ( Fig 7.3.4-24b ). The aiming device holder is removed for improved viewing, leaving the aiming device mounted to the plate. The awl is inserted into the aiming device with the help of the tweezers ( Fig 7.3.4-24c ) in order to prepare the cortical rim of the vertebral body for screw insertion. The awl is then removed with the help of the tweezers and the fi rst self-tapping locking head screw (up to 30 mm of length) is inserted and tightened ( Fig 7.3.4-24d ). The second self-tapping locking head screw is then inserted, tightened, and locked into the plate ( Fig 7.3.4-24e ).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

a

Fig 7.3.4-24a–h a Insertion of the Synfix, which is mounted on the implant holder and then slid between the distractor blades into the prepared disc space.

451

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

c

b

d

Fig 7.3.4-24a–h b Mounting the aiming device on the Synfi x cage. Once the mounting device has been inserted it is recommended that the aiming device holder be removed to improve the view. c Insertion of the awl with the help of the tweezers. The tweezers help to keep control during insertion. d Insertion of fi rst locking head screw which is tightened when seated properly. e Insertion and tightening of the second locking head screw (after preparation using the awl).

e

452

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

g 180°

f

h

Fig 7.3.4-24a–h f Rotate the aiming device. This requires reattachment of the aiming device holder before rotation. g Preparation and insertion of the third and fourth self-tapping locking head screws. h Remove the instruments when the fixation plate is secured and check the position of the implant.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

453

7.3.4 Instrumentation for the degenerative thoracolumbar spine (spondylosis)

Postoperative care

A soft lumbar brace can be used for the fi rst 6 weeks. Return to full activities including sports can be done at 3 months. Case example

A 30-year-old man suffered from debilitating low back pain. He had a positive discogram at L5/S1 and a negative at L4/5 ( Fig 7.3.4-25a ). This male patient was treated with a Synfi x ( Fig 7.3.4-25b–d ).

a

b

Fig 7.3.4-25a–d a Preoperative MRI. b Clinical photopgraph after insertion of a Synfix cage. c–d Postoperative AP and lateral x-rays. (Courtesy of Dr Frank Kandziora, Charité, Berlin, Germany.)

c

d

454

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

5

BIBLIOGRAPHY

1.

2.

3.

4.

5.

6.

7.

8.

9.

Benini A, Magerl F (1993) Selective decompression and translaminar articular facet screw fi xation for lumbar canal stenosis and disc protrusion. Br J Neurosurg ; 7(4):413–418. Best NM, Sasso RC (2006) Effi cacy of translaminar facet screw fi xation in circumferential interbody fusions as compared to pedicle screw fi xation. J Spinal Disord Tech ; 19(2):98–103. Freeman BJ, Licina P, Mehdian SH (2000) Posterior lumbar interbody fusion combined with instrumented postero-lateral fusion: 5-year results in 60 patients. Eur Spine J ; 9(1):42–46. Hackenberg L, Halm H, Bullmann V, et al (2005) Transforaminal lumbar interbody fusion: a safe technique with satisfactory three to fi ve year results. Eur Spine J ; 14(6):551–558. Pavlov PW, Meijers H, van Limbeek J, et al (2004) Good outcome and restoration of lordosis after anterior lumbar interbody fusion with additional posterior fi xation. Spine ; 29(7):1893–1899. Cain CM, Schleicher P, Gerlach R, et al (2005) A new stand-alone anterior lumbar interbody fusion device: biomechanical comparison with established fi xation techniques. Spine ; 30(23):2631–2636. Soegaard R, Christensen FB, Christiansen T, et al (2006) Costs and effects in lumbar spinal fusion. A follow-up study in 136 consecutive patients with chronic low back pain. Eur Spine J ; in press. Steffen T, Tsantrizos A, Aebi M (2000) Effect of implant design and endplate preparation on the compressive strength of interbody fusion constructs. Spine ; 25(9):1077–1084. Steffen T, Tsantrizos A, Fruth I, et al (2000) Cages: designs and concepts. Eur Spine J ; 9(Suppl 1):S89–94.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

455

456

7 SPINAL INSTRUMENTATION 7.3 THORACOLUMBAR AND SACROPELVIC SPINE 7.3.5 FIXATION OF THE SACRUM AND PELVIS

1

Introduction ………………………………………………………………………………………………… 457

2

S1 screws and iliac screws ……………………………………………………………………………… 457

3

Maximum-width fixation of the sacropelvis

4

Lumbopelvic reconstruction with four-rod technique ……………………………………………… 464

5

Bibliography ………………………………………………………………………………………………… 465

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

………………………………………………………… 462

457

Travis Hunt, Vincent Arlet

7 7.3

SPINAL INSTRUMENTATION THORACOLUMBAR AND SACROPELVIC SPINE

7.3.5

1

FIXATION OF THE SACRUM AND PELVIS

INTRODUCTION

The fi rst instrumentation for spinal deformities of the pelvis was introduced with the work of Luque using segmental sublaminar instrumentation in the lumbar spine and rods inserted at right angles in the posterior iliac spine [1]. An improvement of the technique was represented by the LuqueGalveston technique which achieved a better fi xation with intrailiac insertion of bent rods [2]. This method of fi xation remained the gold standard for almost 15 years. Different modifications of the technique were published in order to either simplify the complex bending of the rods, or to increase the fi xation in the pelvis. Marchesi et al published a modified technique where S1 pedicle screws are inserted to increase the foundation of the sacropelvic construct [3]. However, due to the need to correct ever more challenging curves and to prevent pseudarthrosis in the lumbosacral junction, stronger fi xations to the sacropelvis are necessary [4]. Currently, the most advanced method of fi xation to the sacropelvic spine is represented by the S1 pedicle screws and iliac wing screws and the maximum-width fi xation[5, 6]. Lumbopelvic reconstruction after total sacrectomies may require a four-rod

2

S1 SCREWS AND ILIAC SCREWS

Principles

Complex spine surgery often requires instrumentation to the sacrum and pelvis, and an appropriate technique can provide considerable strength to posterior constructs. This fi xation is often required to establish a stable base for extended constructs in an adult deformity and can be crucial for correcting pelvic obliquity in neuromuscular scoliosis. Fracture fi xation and reduction of severe spondylolisthesis can involve the use of these techniques. Indications

• Degenerative scoliosis, revision of spinal deformities with extension to the sacrum. • Osteoporotic sacrum. • Pelvic obliquity in nonambulatory neuromuscular scoliosis. • High-grade spondylolisthesis. • Fracture fi xation.

458

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

Surgical technique Insertion of the sacral screws

Advantages

• Increased strength for a stable base in extended constructs. • Powerful reduction tools for the treatment of fractures and deformities.

After careful exposure of the posterior elements and location of the L5/S1 facet, the starting position for the S1 pedicle screw can be identified at the inferior lateral portion of this facet. A partial facetectomy of the inferior facet of L5 can facilitate a proper entry point for the screw insertion ( Fig 7.3.5-1). A highspeed burr or an awl is necessary to breach the cortex. With the C-arm in a lateral position, a pedicle fi nder is passed down the pedicle toward the sacral promontory with 30–45° of medial inclination. The starting point for the S1 screw from the USS II set should be as lateral as possible and as convergent as possible to allow matching of the iliac screw inferiorly ( Fig 7.3.5-2 ). The overhang of the iliac crest may hinder this lateral and convergent placement of the screw. In such cases a rongeur or a high-

Disadvantages

• It is technically demanding. • Increased surgery also means increased potential for complications.

a Fig 7.3.5-1 Osteotomy of the inferior facet of L5 and entry point into the S1 pedicle.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

Fig 7.3.5-2a–b Direction for pedicle screw insertion in the AP and lateral view.

459

7.3.5 Fixation of the sacrum and pelvis

speed burr can “thin out” the medial aspect of the iliac wing to allow such convergence of the screw. The anterior cortex of S1 should be breached at the level of the sacral promontory. It is safe here because the iliac vessels bifurcate above this level. Long screws can then be inserted bicortically after probing the pathway (between 40–45 mm long) ( Fig 7.3.5-3 ). Insertion of L5 pedicle screws is done according to the standard technique, taking care to maintain the alignment between the L5 and S1 screws ( Fig 7.3.5-4a ). Insertion of the iliac screws a

b

Fig 7.3.5-3a–b Insertion of the S1 pedicle screw toward the sacral promontory. The second screw is then inserted in the same way into the contralateral side of S1.

a

b

If the patient is young and the iliac crest is thin (neuromuscular scoliosis), the authors recommend the exposure of the lateral table of the iliac wing to control the direction and pathway of the screw. If the patient is an adult, the thickness of the iliac crest is large enough that it is not necessary to expose the iliac wing. The posterior cortex of the posterior iliac spine is removed to facilitate proper insertion ( Fig 7.3.5-4b ). The notch should be made flush with the plane of the sacrum. The

Fig 7.3.5-4a–b Insertion of L5 pedicle screws according to standard technique. With a rongeur the cortex of the posterior superior iliac spine is removed to prepare for the placement of the iliac screw.

460

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

starting position of the iliac screw is 5 mm lateral to the sacroiliac joint and the trajectory is approximately 20–40° outward between the two cortices of the iliac wing. In the sagittal plane the screw is directed toward the anteroinferior iliac spine. The trajectory is 1.5–2.5 cm above the iliac notch and superior to the hip joint as visualized on lateral fluoroscopy. A pedicle fi nder (5 or 6 mm in diameter) is used to create the trajectory and a ball-tip probe is used to ensure that the trajectory has remained in bone ( Fig 7.3.5-5 ). The insertion of two screws in each iliac wing might be necessary for an even stronger fi xation. Long (100 mm) 8.0 mm screws can provide incredible purchase. Correct positioning of the screws can be verified on modified Judet views using C-arm fluoroscopy. On these views the screw must project in the radiological teardrop and above the hip and the iliac notch.

Insertion of the rod

a

a

b

Fig 7.3.5-5a–b Insertion of an iliac screw 1.5–2.5 cm above the notch and the acetabulum toward the anteroinferior iliac spine.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

If the S1 screw has been inserted with a lateral starting point and is convergent enough, and if the iliac screw starting point is 5 mm outside the sacroiliac joint, then the side-openings of the USS screws will match perfectly, so connectors are not required to insert the 6 mm rods. Sagittally precontoured rods, or custom-contoured rods, can now be inserted distally in the iliac screw fi rst, and then into the S1 screw ( Fig 7.3.5-6 ). In the case of an oversized offset between the screws, a side connector is inserted. Postoperative care

The patient is mobilized 24–48 hours after surgery. A brace is not necessary.

b

Fig 7.3.5-6a–b The iliac screws and the S1 screws are connected with 6 mm rods on both sides. No connectors are necessary, if the S1 entry point is lateral enough, and if the iliac screw entry point is only 5 mm outside the sacroiliac joint.

461

7.3.5 Fixation of the sacrum and pelvis

Case example

Degenerative scoliosis in a 70-year-old female with marked osteoporosis and multiple previous orthopedic operations ( Fig 7.3.5-7a ). Because of her sagittal malalignment she was treated with an anterior release, an anterior column support, and posterior S1 pedicle and iliac screws ( F i g 7. 3 . 5 -7 b ) . No connectors were attached for the sacropelvic fi xation.

a

b

Fig 7.3.5-7a–b a Preoperative AP and lateral x-rays. b X-rays at the 4-month follow-up after staged anterior and posterior reconstruction with S1 pedicle screws and iliac screws. Note the absence of connectors in order to achieve the sacropelvic reconstruction.

462

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

3

MAXIMUM-WIDTH FIXATION OF THE SACROPELVIS

Principles

Surgical technique

The maximum-width (MW) sacropelvic construct is a combination of iliosacral screws and iliac screws.

This extremely rigid system can be used to correct a deformity including pelvic obliquity. First, the iliac wing needs to be exposed from the lateral side. A pedicle finder is used to perforate the iliac crest—usually 2 cm below the crest. Alternatively, a K-wire is used to prepare the trajectory for the cannulated screw. Two 7.0 mm fully threaded cannulated cancellous bone screws with washers are inserted from the iliac wings at a 45° trajectory into the fi rst sacral pedicles. A small laminotomy for better visibility of the S1 nerve is required and allows for safe screw insertion. Iliac screws are inserted 1.5–2 cm below the iliosacral screws using the technique described above. Side-opening lamina hooks from the USS II are positioned either upward or downward on the rod. The iliosacral screws are seated into the hooks and deformity correction is achieved through cantilever and compression or distraction of this very rigid assembly ( Fig 7.3.5-8 ).

Indication

Pelvic obliquity in neuromuscular scoliosis (AOSpine Manual—Clinical Applications; 5.1.3 Neuromuscular scoliosis). Advantages

• This is a very strong fi xation of the sacropelvis. • MW fi xation allows cantilevering of the pelvis as well as compression and distraction. • Powerful reduction tool for fractures and deformities. Disadvantages

• It is a technically demanding fi xation technique. • Increased surgery, which means that there is an increased potential for complications.

Postoperative care

The patient is mobilized 24–48 hours after surgery. A brace is not necessary.

2

1

1

2 a

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

Fig 7.3.5-8a–b Top and anterior views of the MW fixation in the sacropelvis show the iliosacral screws (1), and the iliac screws ( 2 ). Side-opening hooks are seated over or under the iliosacral screws. Each rod is connected to the side-opening hook and iliac screw. Compression and/or distraction can then be applied to correct the pelvic obliquity.

463

7.3.5 Fixation of the sacrum and pelvis

Case example

This 15-year-old patient suffered from cerebral palsy and severe pelvic obliquity ( Fig 7.3.5-9a ). The scoliosis measured 120° and the pelvic obliquity 45°. The surgical treatment consisted of an anterior release of the stiff thoracolumbar curve and a posterior segmental sacropelvic fi xation ( Fig 7.3.5-9b ).

a

b

Fig 7.3.5-9a–b a Cerebral palsy with fixed, severe pelvic obliquity. b AP and lateral postoperative x-rays after treatment with anterior release and posterior MW fixation.

464

7 7.3

Spinal instrumentation Thoracolumbar and sacropelvic spine

4

LUMBOPELVIC RECONSTRUCTION WITH FOUR-ROD TECHNIQUE

Principles

Case example

Restore the continuity of the lumbopelvic junction after total sacrectomy using four rods.

A 55-year-old male patient with a sacral chordoma. Following anterior and posterior surgery with an en bloc resection of the whole sacrum, the spine is reconstructed using a four-rod technique and the insertion of two titanium mesh cages ( Fig 7.3.5-10 ).

Indication

Reconstruction after total sacrectomy for primary bone tumors (chordoma). Advantages

• Most solid reconstruction. • Immediate mobilization without brace. Disadvantage

Technically demanding reconstruction technique. Surgical technique

After a complete sacrectomy has been done (in most cases following a front and back approach to the spine) lumbar USS pedicle screws are inserted from L2 to L5 in an alternate fashion using the Roy-Camille method of insertion and the Magerl method of fi xation. Two iliac screws are inserted on each side as described previously in this chapter (S1 screws and iliac screws). The side-opening of the USS screws are oriented to create maximum offset between the Roy-Camille screws and the Magerl screws leaving room for the insertion of two rods on each side. Appropriate cross-linking of the four longitudinal rods increases the overall stability. Anterior column reconstruction is best achieved with curved titanium mesh cages.

a

b Fig 7.3.5-10a–c a Preoperative CT scans. b Schematic representation of the four-rod technique applied on a bone model.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

465

7.3.5 Fixation of the sacrum and pelvis

5

1

2

c

Fig 7.3.5-10a–c c Four-rod technique for spinopelvic reconstruction after total sacrectomy. The lumbar monoaxial pedicle screws are inserted in an alternate Roy-Camille (1) and Magerl (2) method. Two iliac wing screws are inserted on each side and the anterior column is reconstructed with two curved mesh cages.

BIBLIOGRAPHY

1. Luque ER (1982) Segmental spinal instrumentation for correction of scoliosis. Clin Orthop Relat Res ; (163):192–198. 2. Allen BL Jr, Ferguson RL (1982) The Galveston technique for L rod instrumentation of the scoliotic spine. Spine ; 7(3):276–284. 3. Marchesi D, Arlet V, Stricker U, et al (1997) Modifi cation of the original Luque technique in the treatment of Duchenne‘s neuromuscular scoliosis. J Pediatr Orthop ; 17(6):743–749. 4. Islam NC, Wood KB, Tranfeldt EE, et al (2001) Extension of fusions to the pelvis in idiopathic scoliosis. Spine ; 26(2):166–173. 5. Arlet V, Marchesi D, Papin P, et al (1999) The MW sacropelvic construct: an enhanced fi xation of the lumbosacral junction in neuromuscular pelvic obliquity. Eur Spine J ; 8(3):229–231. 6. Carroll EA, Shilt JS, Jacks L, et al (2006) MW construct in fusion for neuromuscular scoliosis. Eur Spine J ; in press. 7. Shen FH, Harper M, Foster WC, et al (2006) A novel “four-rod technique” for lumbo-pelvic reconstruction: theory and technical considerations. Spine ; 31(12):1395–1401.

466

7 SPINAL INSTRUMENTATION 7.4 SPECIAL TECHNIQUES AND INSTRUMENTATION 7.4.1 CONCEPTS OF MISS/LISS

1 1.1 1.2

Introduction ………………………………………………………………………………………………… 467 Conventional spine surgery vs MISS/LISS …………………………………………………………… 467 Introduction to MISS/LISS ……………………………………………………………………………… 469

2

Concept of the Synframe ………………………………………………………………………………… 471

3 3.1 3.2 3.3 3.4 3.5 3.6

Indications ………………………………………………………………………………………………… Transperitoneal minilaparotomy for interbody fusion at the lumbosacral junction …………… Retroperitoneal L1/2–L5/S1 miniapproach …………………………………………………………… Thoracolumbar junction ………………………………………………………………………………… Minithoracotomy …………………………………………………………………………………………… Posterior surgery …………………………………………………………………………………………… Other applications …………………………………………………………………………………………

4

Bibliography ………………………………………………………………………………………………… 483

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

475 475 476 480 481 482 483

467

Max Aebi

7 7.4

SPINAL INSTRUMENTATION SPECIAL TECHNIQUES AND INSTRUMENTATION

7.4.1

CONCEPTS OF MISS/LISS

1

INTRODUCTION

1.1

CONVENTIONAL SPINE SURGERY VS MISS/LISS

It is argued by the proponents of minimally or less invasive spine surgery (MISS or LISS) that conventional spine surgery requires a long incision with a large scar formation and a lengthy recovery. Minimally invasive surgery uses “keyhole” surgery performed either by endoscopic procedures or by specific retractor systems using small incisions with small scars and reduced recovery time. However, the complication rate during this kind of surgery may be increased and the result is not as predictable when the surgery is performed by surgeons who have not become familiar with these new techniques. Today, minimally invasive spine surgery is not applied for all spine surgery, but only for selected procedures such as anterior and posterior lumbar fusion procedures in degenerative or traumatic disorders of the spine, in selected anterior deformity corrections, and in thoracic spine as well as in disc surgery.

Minimally invasive spine surgery has the same goals and objectives as conventional open spinal surgery. Just the approach is to a certain degree different. Most spine surgery patients require decompression and/or reduction of the deformity and/ or stabilization. MISS decompression, deformity reduction, and stabilization are typically performed following the same principles used with open surgery, but through small incisions with minimal or less muscle stripping and with minimal softtissue retraction. As in laparoscopic cholecystectomy or in knee arthroscopy, the surgery at the target structures does not change, but by reducing the surgical trauma related to the approach, the results may be better with regard to patient outcome, shorter hospitalization time, less blood loss, shorter recovery time, etc. These are assumptions that make sense, but no real randomized prospective study has ever proved them. There are concerns as

468

7 7.4

Spinal instrumentation Special techniques and instrumentation

to whether or not this surgery is cost-effective and related to a higher degree of complications, and indeed whether a better outcome in the mid to long term is to be expected. There are confl icting opinions as to what exactly constitutes minimally invasive surgery and what still constitutes conventional surgery. “Miniaturized” conventional surgery may be better described by the term less invasive spine surgery (LISS) than with minimally invasive spine surgery (MISS). The latter may be reserved for endoscopically supported surgery, for which the indication is defi nitely limited, since by defi nition surgery by means of an endoscope can only be performed properly in a preformed cavity. Therefore, the best application for endoscopic or minimally invasive surgery is transthoracic surgery for the thoracic spine and laparoscopic anterior surgery at the lumbosacral junction and possibly at L4/5. The major part of the lumbar spine is approached anteriorly through the retroperitoneal space, which is not a preformed cavity and cannot be used for endoscopic surgery without additional tools. Disc surgery may be performed through an endoscope, however, the indications are limited. All the rest of spine surgery that runs under “MISS” is more precisely “LISS”, as it basically imitates conventional surgery through smaller accesses, be it through a “tube”-surgery or a special blade-retractor system. Therefore, surgeons need to keep in mind the traditional or conventional principles of spine surgery because a proper decompression and/or fusion and not just a “minimal” one is still needed. Conventional surgery may also differ in the manner in which the fusion is performed. For example, in MISS or LISS iliac crest bone harvesting may be avoided because of related morbidity, and the classic posterior lateral fusion may be avoided equally due to the muscle stripping

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

necessary to achieve it. Increasingly, in LISS and MISS the fusion is concentrated on the accessible disc, be it in anterior surgery (ALIF) or posterior surgery (PLIF or TLIF), by using substances which either stimulate bone growth by bone induction or augment fusion by osteoconduction. All these procedures can be achieved through smaller accesses, however, as soon as pedicle screws need to be placed percutaneously through a posterior approach, by means of tubes and special retractor systems, this becomes extremely difficult in a deformed and degenerated spine where the anatomical landmarks are difficult to identify. The hybrid technology of MISS/LISS combined with computer navigation may be useful in the future. It is inevitable, however, that patient-driven needs will force surgeons to apply MISS or LISS techniques when feasible and more promising for the patient’s outcome. This again, however, necessitates rigorous training of these newer surgical skills and technologies.

469

7.4.1 Concepts of MISS/LISS

1.2

INTRODUCTION TO MISS/LISS

The concept of MISS—minimally invasive spine surgery—or probably more appropriately termed LISS—less invasive spine surgery—has evolved significantly in the last 5–10 years. There are three major elements to this surgical concept, which are crucial for its further development and its survival in practical surgery:

In this chapter the author focuses on a specific access technology consisting of a unique retractor system and bone graft or bone substitute technology. There is a separate chapter about fusion alternatives (7.4.2 Bone harvesting tools).

1. Access technology 2. Implant technology 3. Bone graft or substitute technology ( Fig 7.4.1-1)

Access technology is primarily necessary in anterior spine surgery, in particular lumbar and thoracic spine surgery, because the spine is located deep in the operating field and a major part of the surgery is spent creating a space that allows sufficient access to the spine and enough room for the necessary manipulations needed for a stabilizing or fusion as well as decompressive procedure.

The access technologies include endoscopic procedures [1–3] necessitating optical tools, image-guided navigation or computer-assisted spine surgery [4], minimally open procedures with specific retractor systems [5, 6], robot-aided surgery, and combinations of these technologies (hybrid technologies).

When retroperitoneal access to the lumbar spine is chosen there is no preformed cavity, which would allow the necessary space for the aforementioned procedures, such as endoscopic surgery, and it must therefore be created artificially. While the peritoneal and thoracic cavity are natural spaces where laparoscopic and thoracoscopic accesses offer major advantages

Minimally open approach

Access technolog y

• • • • •

Optical systems Computer- guided surger y Special retrac tor systems Hybrids Robotic s

New generation of implant s suitable for minimal access

• Anterior surger y • Posterior surger y

New solutions for fusion

• • • •

C ages Bone har vesting tools Bone substitutes Alternatives

Fig 7.4.1-1 Minimally invasive spine surgery.

470

7 7.4

Spinal instrumentation Special techniques and instrumentation

[7], the lumbar and thoracolumbar areas of the spine are literally hidden behind these spaces and hence are not ideal for a laparoscopic or thoracoscopic approach. Gas insuffl ation techniques are not helpful in spine surgery and the balloonassisted endoscopic retroperitoneal gasless (BERG) approach has not really added a practical procedure [8]. That is why other methods needed to be developed to obtain proper access to this area of the spine [9]. A retractor system is crucial for access to the lumbar and thoracolumbar spine, be it applied by additional ports in an endoscopic procedure or by a retractor system that maintains a small surgical opening.

the retractor system balanced in place. While these systems may be sufficient for specific procedures, they do not allow a universal approach and they do not provide the stability of the operating field that is necessary for standardized minimally open procedures on the spine. It is for this reason that the author was looking for a different concept [5], one that enables spine surgeons to apply surgical techniques with which they are familiar. The retractor system allows for direct vision and does not need indirect recognition through an endoscope [2] and, nevertheless, offers optimal illumination and high versatility in its surgical application (Fig 7.4.1-3).

Retractor systems that can maintain a laparatomy or thoracotomy open have long been available, and these kinds of retractors have also been adjusted to fit the requirements of anterior lumbar interbody fusion (ALIF) surgery ( Fig 7.4.1-2 ) [5]. These systems, however, are not fi xed to the table. A rectangular or square frame is laid over the incision, and the maintaining of the surgical opening relies on the retractor blades, which are fi xed on the frame, pulling in opposite directions, thus, keeping

Fig 7.4.1-2 Example of an access tool: retractor used for ALIF surgery (with relevant disadvantages compared to a ring retractor system).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.4.1-3 The original prototype ring mounted on the table and on the patient with blade retractors demonstrating the size of the exposure in relation to the ring.

471

7.4.1 Concepts of MISS/LISS

2

CONCEPT OF THE SYNFRAME

This retractor system is built on a stable ring and fi xed onto the operating table by two adjustable arms. The ring is laid concentrically over the surgical exposure and allows access from any point on the 360° ring. There are, therefore, no predetermined locations on the ring where, for example, the retractors need to be fi xed, and there is complete freedom regarding where to position a retractor blade or other tools needed for the surgery. The free assembly of the ring through adjustable clamps and arms tolerates a high degree of versatility and adaptation of the ring position over the patient, not only for the anterior approach to the lumbar spine, but also for the approach to the anterior thoracic and cervical spine as well as for all posterior approaches ( Fig 7.4.1-4 ).

a

c

The ring acts as a carrier for surgical tools and instruments. The basic instruments are the retractor blades or the modified Hohmann levers, which can be clicked onto the ring (Fig 7.4.1-4 ). With these instruments, either alone or in combination, the surgical exposure can be maintained minimally open and a standardized surgical procedure can be carried out. The connection of the blades and Hohmann levers to the ring is guaranteed by a clamp, which allows the adjustment of the retractor elements in all six degrees of freedom ( Fig 7.4.1-5 ). Each blade can be adjusted individually and tilted in the sagittal plane in order to hold back the tissue, especially in the depth of the exposure, and can be exchanged according to the depth of the exposure ( Fig 7.4.1-6 ).

b

Fig 7.4.1-4a–c a Stable operating field and surgical exposure. b Improved visibility (retractor blades). c Optimal retraction and customized exposure (Hohmann levers).

Fig 7.4.1-5 Special clamp, which is clicked onto the ring and holds the retractor arms or the fiberoptic light source. The clamps can be moved along the ring and rotated on the ring tube, which allows rotation of the retractor arms in the horizontal plane, thus, tolerating movements in all three planes.

472

7 7.4

Spinal instrumentation Special techniques and instrumentation

The ring can also be used to carry a fiberoptic adjustable light source with which the depth of the operating field can be perfectly illuminated ( Fig 7.4.1-7 ). In the same way, endoscopes of different angulations and sizes can be mounted on the ring, allowing a proper view of the procedure in the narrow space in the depth of the operating field ( Fig 7.4.1-8 ). This allows those people not directly involved in the surgery to follow the surgery on the video screen, therefore, supporting an important teaching mission by making visible the details of the procedure. The stable, precise illumination and the use of loupes by the surgeon make this procedure as good as microscopic spine surgery, but most probably faster and more straightforward; however, it is possible to use the microscope as an alternative.

a

b Fig 7.4.1-6a–b Different sizes and types of blades. Blade arm can be tilted in the sagittal plane by a hexagonal screwdriver. a Soft-tissue retractors with 25 mm blades, length 60–160 mm and hexagonal screwdriver. b Synframe Hohmann bone levers (width 8 and 18 mm, length 100/130/160 mm and length 130/150/170/190 mm).

a

Fig 7.4.1-7 Anterior approach to the lumbar spine with mounted ring, inserted retractors, and the light source (arrow).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

b

c

Fig 7.4.1-8a–c Endoscope (in a model) mounted onto the ring (b) with the intraoperative view of the L4/5 disc (c). In the left upper corner is the illuminated corridor to the disc (a).

473

7.4.1 Concepts of MISS/LISS

Fig 7.4.1-9a–f

Application of the Synframe, adapted to the cervical spine, in a combined odontoid and articular C1/2 fracture.

a

b

c a Application of the Synframe adapted to the cervical spine. b Complex odontoid fracture combined with an atlantoaxial joint fracture on the left side. c Exposure of the anterior cervical spine at the level of C5 with the Synframe, reduction of C2 through the mouth. d Anterior transarticular C2/1 screw fixation together with an odontoid screw fixation (chapter 7.2.2 Upper cervical spine).

d

e

f

e MISS retractor used in anterior cervical spinal tumor surgery. Synframe disassembled in the two half-ring components.

f Breast cancer at T1/2, Synframe retractor enables a surgery without sternotomy.

474

7 7.4

Spinal instrumentation Special techniques and instrumentation

The original concept of the ring was to use it as a carrier of retracting, optical imaging, and illumination systems and was fi rst applied for anterior surgery of the lumbar, thoracolumbar, and thoracic spine. It quickly became obvious that there were also applications in anterior cervical spine surgery ( Fig 7.4.1-9 ), cervical plexus surgery, and, more importantly, in minimally invasive as well as conventional posterior spine surgery ( Fig 7.4.1-10 ). Microdiscectomies with the Synframe carrying a light source, a miniendoscope for transmission to the screen, and, for example, a modified root retractor, has become very standardized and a simple one-surgeon disc operation ( Fig 7.4.1-11). Even the root retractor can become a part of a standard set, since it can be stably fi xed to the ring. The great potential of the Synframe as an instrument and tool carrier has been recognized, and it is also being used in orthopedic procedures other than spine surgery.

*

Fig 7.4.1-10 Synframe application for posterior surgery (PLIF, TLIF, ALIF with posterior instrumentation).

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Fig 7.4.1-11 The Synframe in a simple configuration for posterior disc surgery with a miniendoscope and a root retractor which is also fixed to the ring (asterisk).

475

7.4.1 Concepts of MISS/LISS

3

INDICATIONS

3.1

TRANSPERITONEAL MINILAPAROTOMY FOR INTERBODY FUSION AT THE LUMBOSACRAL JUNCTION

The patient, having had his bowels prepared for abdominal surgery, is positioned (as for conventional surgery) on a radiolucent kidney table, which allows hyperextension of the lumbar spine and the lumbosacral junction. The projection of the segment L5/S1 and/or L4/5 on the skin of the abdominal wall at the midline and on the flank on one side is marked with the image intensifier. A mini-Pfannenstiel incision is made (4–6 cm, depending on the size of the fat layer), followed by a longitudinal midline incision through the fascial sheets between the two rectus abdominis muscles ( Fig 7.4.1-12 ). The visceral peritoneum is opened between two stay sutures; this will facilitate closure after the intervention. At this point two to four blades are introduced into the handles, which are then positioned onto the Synframe and into the wound, thereby using their adjustment capabilities in all three planes of the space. This allows for sufficient opening of the abdominal

head

head

symphysis

a

b

symphysis

Fig 7.4.1-12a–b a Projection of the promontorium on the lateral body surface and the abdominal wall established by image intensifier. b Skin incision (mini-Pfannenstiel) to the promontorium by a transperitoneal approach.

cavity to identify, by blunt dissection using either the surgeon’s fi ngers or a mounted sponge, the promontorium of the sacrum by mobilizing as best as possible the emptied bowel slings cranially out of the way. The bowels may be retracted with an abdominal pad held by one of the retractors positioned at 12 o’clock and one each at 9 and 3 o’clock of the Synframe. A long retractor blade is mounted on the handle at 6 o’clock, retracting (over a sponge) the bladder out of the way. The parietal peritoneum is incised sagittally over the promontorium and possibly the bifurcation of the iliac vessels. By repositioning the retractors, with longer blunt blades that are angulated at the tip, underneath the borders of the parietal peritoneum the lumbosacral disc can now be exposed. Depending on the type of surgery to be performed, some small posteriorly oriented vessels of the iliac artery, and in particular of the vein and the median sacral vessels, may need ligation or clipping in order to mobilize the bifurcation. Depending on the requirements, the common iliac artery and vein on both sides or only the external or internal one may be slung temporarily on a vessel loop to allow for optimal exposure of the L5/S1 and/or the L4/5 vertebrae. This surgery does not differ essentially from a conventional approach, except that it is performed through a smaller, though completely stable, incision thanks to a standardized retractor system and optimal illumination. In order to increase the precision and stability of the operating field, instead of blades optimized Hohmann retractors, which are linked with flexible handles to the Synframe, are used (see Fig 7.1.4-6b ). After satisfactory exposure and again confi rming the level with an image intensifier, the surgery related directly to the spine can be performed (see corresponding chapter), ie, interbody fusion.

476

7 7.4

Spinal instrumentation Special techniques and instrumentation

3.2

RETROPERITONEAL L1/2–L5/S1 MINIAPPROACH

Anatomical considerations

There are different possibilities of entering the retroperitoneum on the left side. The projection of the targeted segment(s) is marked on the skin. A line is drawn laterally on the left side of the patient from posterior to anterior in the line of the proposed disc surgery. The direction of this line is identified with the image intensifier ( Fig 7.4.1-13a ). A vertical line is drawn in the midline, and then a horizontal line is drawn at the level of the projection of the disc through the anterior abdominal wall which has already been marked laterally ( Fig 7.4.1-13b). A horizontal incision is prepared in line with the horizontal marker line on the anterior abdominal wall. The incision is one third over the midline and two thirds over the medial half of the left rectus sheath. The left rectus muscle is retracted laterally (innervation from lateral), and the linea arcuata is identified distally. Here the retroperitoneal space can be entered, and the dissection extended toward the left psoas muscle. midline

di sc lin e

a

b

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Magnification

In the author’s experience a microscope is not needed for the anterior part of the surgery. One main advantage of the microscope is undoubtly the focused illumination, however, this issue has been resolved with the integrated illumination within the Synframe retractor system. The use of regular operating loupes is sufficient for the amount of magnification that may be needed. The anterior part of the surgery is basically a macroscopic surgery, where there is no need to work in a microscopic dimension. As long as the surgery is a pure stabilization and fusion procedure, a microscope is not needed. However, in case of an anterior or posterior precise decompressive surgery the microscope is helpful. Patient positioning

The bowel-prepared patient is positioned supine with the side of the approach elevated and with sandbags under the pelvis and rib cage, allowing a slight tilting of the patient to the opposite side. During insertion of a cage or disc arthroplasty the table can be tilted back, so the patient lays in a horizontal plane in order to facilitate the application of the device in the proper direction and plane. The soft-tissue preparation and the access is easier, when the tissue is not under tension. However, once the anterior surface of the spine becomes visible, a hyperextension of the targeted segments is achieved by kinking the flat table, with the top of the kink projecting to the appropriate level. This defi nitely opens the disc space, which is then more easily accessible. Fig 7.4.1-13a–b Disc level projected onto the body surface. a Direction of the L5/S1 disc. b Transverse line corresponding to the projection on the abdominal wall of the L5/S1 disc.

477

7.4.1 Concepts of MISS/LISS

Surgical technique

There are two routes to be taken for the skin incision: • A 4–6 cm long horizontal incision is made (length depending on the size of the patient) located over the lateral border of the left rectus muscle at the level of the targeted lumbar segment ( Fig 7.4.1-14 ). The left rectus muscle is exposed by opening the anterior sheath of the muscle. The muscle is retracted laterally to avoid damaging the lateral innervation of the muscle. The surgeon’s fi nger then glides down the inner sheath of the rectus to the arcuate line at the level of the middle to the lower third of the rectus sheath, where the thin transversal fascia is laying over the peritoneum ( Fig 7.4.1-15 ). From the arcuate line a blunt dissection between the transversalis fascia and the peritoneal sheath is done, preferentially in a lateral direction in order to identify the psoas muscle. There are, however, different options as to how to incize the rectus sheath, as outlined in Fig 7.4.1-16.

a

b

c

1

L2/3

• Alternatively, the horizonal 4–6 cm long incision is made in a more lateral direction in the external iliac fossa, where the fascia of the musculus obliquus externus is present. This muscle is split in the direction of the fibers, the same is done with the underlying internal oblique muscle and the transverse muscle. Finally, after splitting carefully the transversalis fascia one enters the retroperitoneal space fi lled with fat tissue. One reaches the psoas muscle through blunt dissection of this retroperitoneal fat.

L2/3 L3/4 L4/5

Fig 7.4.1-14 Abdominal skin incision for different levels (L2–5) in relationship to the underlying rectus sheath and muscle on the left side.

2 3

Fig 7.4.1-15 Exposure of the opened rectus sheath (1) and the transition to the peritoneum and transversalis fascia at the linea arcuata (2), and where to enter the retroperitoneal space (3).

d

Fig 7.4.1-16a–d Different access options. a Midline incision. b Incision through the anterior rectus sheath, and the rectus is retracted laterally. c Approach through a muscle split in the abdominal wall and the rectus is retracted medially. d Linea arcuata with incized posterior rectus sheath with the rectus muscle retracted laterally.

478

7 7.4

Spinal instrumentation Special techniques and instrumentation

It is easier to get in the retroperitoneal space with a more lateral direction. More medially, there is a higher chance that the peritoneum is adherent to the transverse abdominal muscle, and the peritoneum will be torn. For both approaches, once the psoas muscle is identified, blunt dissection of the retroperitoneal space toward the midline of the spine is done. Once the spine is undoubtly identified, a Hohmann lever is inserted into the vertebral body and linked to a clamp of the Synframe on the right side. This will open the corridor to the spine and move the incision over the midline ( Fig 7.4.1-17 ). From the anchorage of the Hohmann retractor the blunt dissection can be developed cranially, caudally, and laterally on both sides to prepare the operating field. The operating field consists of a window including the targeted disc space, with a rim of bone from the adjacent vertebral bodies.

1

disc

2

5

4 6

3

1 2 3 4 5 6

Hohmann retractor parietal peritoneum retractor blade internal iliac artery common iliac artery external iliac artery

Fig 7.4.1-17 Exposure of the L4/5 level by a minimally invasive anterior approach and the use of Hohmann retractors to keep the small wound open. The Hohmann retractors are inserted in L4 and L5, keeping the L4/5 disc stable, open, and easily accessible.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Usually, the segmental vessels do not need to be dissected, except later for the bone harvesting from the center of the vertebral body, either below or above the targeted disc space. In order to have an ideal surgical corridor, usually four Hohmann retractors are used by fi xing them above the end plate, to the right/left of the superior vertebra and below the end plate to the right/left of the inferior vertebra ( Fig 7.4.1-18 ). These Hohmann retractors are pulled with the connecting arms toward the clamp fi xation on the Synframe creating a concentric operating field within the ring. The identification of the spine in order to place the first Hoh mann retractor may not always be obvious and easy. Sometimes, it may become necessary to identify the common iliac artery and vein, and to mark them with vessel loops to pull them either laterally to the right or to the left. Sometimes it may be necessary to specifically ligate posteriorly exiting venous branches to avoid tearing the posterior wall of either the cava before the bifurcation, or the common iliac vein after the bifurcation. With dissectors followed by Hohmann retractors or bent retractor blades, the anterior surface of the spine is bluntly exposed, and by levering the incision toward the midline a direct anterior approach is achieved. This normally facilitates the symmetrical evacuation of the intervertebral disc space and the central placement of the interbody device, be it a bone graft or a cage with bone graft. This same access has major significance and implications also for the lumbar disc replacement in placing a disc prosthesis. Ideally, the ureter is not dissected but moved together with the retroperitoneal fat and tissue to the midline. Aiming toward L4/5, the right common iliac artery is identified and gently moved to the right side. A long blunt hook blade fi xed to the ring is used to hold the right-sided vessels toward the right side. The front side of the vertebral body L4 or the disc space L4/5

479

7.4.1 Concepts of MISS/LISS

is now stepwise exposed by continuous repositioning of the retractor blades, or more ideally of the Hohmann levers. With them a gentle traction to the right side can be applied, and the targeted operating field is gradually exposed ( Fig 7.4.1-18a ). The right common iliac vein may be identified and also pulled to the right side. If the retraction of the vessels is difficult, the left iliolumbar vessels may need to be ligated. Occasionally, branches need to be clipped or coagulated, and the vein itself taken on a vessel loop sling. If the incision is further up, then the vena cava and aorta are retracted to the right side. Generally speaking, the disc space only needs to be exposed in its anterior circumference. Segmental vessels usually do not need to be ligated. The disc space is exposed in such a way that a window can be incised in the annulus allowing easy delivery of an intervertebral device and an easy evacuation of the disc

*

space, removal of the cartilaginous end plates and cleaning of the bony end plates will be possible ( Fig 7.4.1-18b). The best and most precise exposure is reached when a Hohmann retractor is placed in the subchondral bone of the end plates above and below on the left and the right side ( Fig 7.4.1-18a ). In case of a retroperitoneal appraoch to the disc L5/S1, usually the left artery and vein may only be slightly and bluntly mobilized to the left and the right vessels to the right. The accompanying sympathetic chain may be damaged independently, whether we use a microscope or not; however, the incidence seems to be less when the microscope is used. An injury may lead to differences in the temperature of the legs in a high percentage of patients, with an equally high potential of recovery within no later than 1 year. If the integrity of the parietal peritoneal sac is kept intact, usually the fibers of the autonomous nerve system are more out of the way with the mobilized peritoneum.

* disc

* aa

* b

Fig 7.4.1-18a–d Exposure of the L4/5 disc by direct view or by transmission from the laparoscope to the screen. a Four Hohmann retractors (asterisks) exposing the targeted disc. b Incision of the disc at the upper and lower interface of the end plates. Introduction of the special round and sharp Cobb to

c

d

separate the cartilaginous from the bony end-plate layer. This usually permits the removal of the majority of the disc with the adherent cartilage in one piece. c Empty disc space and bleeding subchondral bone. d Measuring the disc space with a trial implant.

480

7 7.4

Spinal instrumentation Special techniques and instrumentation

Before the disc, which is to be replaced is excised, the table is kinked with the kink under the exposed disc to open it optimally. After incising the annulus vertically and horizontally, a Cobb elevator is used to seperate the whole disc from the end plate with the adherent hyaline cartilage ( Fig 7.4.1-18b ). If one succeeds, a substantial part of the disc can be removed in one piece, and the nude bony end plate shows small points of bleeding ( Fig 7.4.1-18c ). The excavation of the disc is then completed. The author does not use any burr to clean the end plate in order to not weaken it, but uses ring curettes and angulated small regular curettes. The disc space is spread with a special spreader— depending on the device to be applied (Fig 7.4.1-19); the spreader is positioned exactly in the center of the disc space. With the complete fit of the spreader blade to the surface geometry of the superior and inferior end plate of the Syncage, the cage acts as a spreader, which distributes the forces on the whole circumference of the vertebral end plates. This allows a perfect fit of the appropriate cage.

*

Fig 7.4.1-19 Intervertebral disc spreader (asterisk) allows insertion of a corresponding cage with a central rim, which uses rail tracks on the blades into the disc space.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Once the cage is delivered through the application of the spreader, the spreader is removed and then the table is brought back to the straight position. The described steps may alter in case of disc arthroplasty surgery. Obviously, this procedure can easily be applied for each individual or several levels at the same time from L2/3 to L5/ S1. The entry point for the skin incision may change in relation to the level of the umbilicus and according to the preoperative marking with the image intensifier (see Fig 7.4.1-14 ). Today, the bone graft should no longer be harvested from the iliac crest, because this procedure is a major source of persistent postoperative pain. Instead, the bone graft should be harvested from the adjacent vertebra using a special bone harvesting trephine system (chapter 7.4.2 Bone harvesting tool).

3.3

THORACOLUMBAR JUNCTION

The thoracolumbar junction can be partially reached by a small incision through the retroperitoneal space at the level of L2, pushing the skin to the edge by the leverage power of the Hohmann retractors, which can be anchored in L1. At the same time, the left crus of the diaphragm can be dissected from the spine to obtain access to L1, and through a minithoracotomy and by splitting along the fibers of the diaphragm, the T11–12–L1 transitional zone can be visualized and the two accesses can be joined. This is a very small double access instead of using a conventional wide thoraco-phrenico-lumbotomy. In this procedure, the use of very precisely placed Hohmann retractors clicked onto the Synframe is an extremely helpful tool. Usually, the placement of a maximum of four Hohmann retractors on the ring is sufficient and the Hohmann retractors can be exchanged with the retractor blades.

481

7.4.1 Concepts of MISS/LISS

3.4

MINITHORACOTOMY

The thoracic spine has been identified as relatively easy for a gas-free endoscopic approach, specifically for an anterior release in scoliosis surgery. However, it is also relatively easy for very localized thoracic disc or vertebral body surgery, most of the time using four ports and unilateral lung collapse. However, this area of the spine can probably be approached just as easily using the Synframe technology through a small incision.

An incision of 4–6 cm is made at the level of the upper rib belonging to its vertebral segment or to the one above, from the left or right side, between the anterior and posterior axillary line, with the patient in lateral position. The thoracotomy is made through a window in the rib, which is created by a double osteotomy of the rib (4–6 cm), which allows a relatively wide spreading of this small incision, again by specially designed thoracic retractor blades or by the Hohmann retractors. The light source, the endoscope, or other instruments can be placed on the Synframe to enhance the view ( Fig 7.4.1-20 ). The minithoracotomy is as straightforward as a pure endoscopic surgery for a thoracic disc herniation ( Fig 7.4.1-21). Fig 7.4.1-20a–b a Minithoracotomy with the Synframe and the use of an endoscope for better visualization. b Different techniques of a minithoracotomy (1–4).

a

1

2

3

4

b 1

intercostal space entry

2

rib distraction (window) by osteotomy

3

rib resection (flap) by osteotomy

4

rib displacement by osteotomy

482

7 7.4

Spinal instrumentation Special techniques and instrumentation

3.5

POSTERIOR SURGERY

The Synframe is equally suitable for posterior surgery, because the retractor blades or customized short Hohmann retractors allow for a very focused and particularly stable exposure of the posterior elements of the spine in conventional posterior spine surgery. 2

1

a 1 Head

b

2

c

Fig 7.4.1-21a–c a Thoracic disc T9/10, Brown-Séquard syndrome. 1 disc herniation 2 area of interest for the anterolateral decompression b Preparation for minithoracotomy with skin markers according to the x-ray projections. c Miniopen thoracic cavity. 1 lung 2 spinal cord

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

Microdiscectomy can be handled very elegantly by the Synframe application, through a 2–3 cm skin incision, a one-side fascial incision, and subperiosteal exposure of the unilateral lamina and the interlaminar space. A customized fi ne Hohmann retractor is placed on the outside of the facet joint (along the outside of the inferior facet of the superior vertebra). This surgical maneuver leads to a sufficient exposure, allowing the resection of the yellow ligament, limited laminotomy, and exposure of the dural sac at the level of the supposed disc. The soft-tissue trauma around the access opening is minimal (see Fig 7.4.1-10 ), the illumination and the miniendoscope, together with the loupes of the surgeon, create an excellent view of inside the canal and the exposure. The retraction of a specific nerve root with a root retractor, which can be mounted on the ring for stability, adds to this concept and allows this kind of surgery to be easily performed without any other medical assistance (see Fig 7.4.1-11). In the process of gaining experience with this minidiscectomy, it became obvious that the same very limited approach could be chosen for a PLIF (posterior lumbar interbody fusion) or TLIF (transforaminal lumbar interbody fusion) procedure, even when combined with a semipercutaneous translaminar screw fi xation or a transmuscular pedicle fi xation. Both sides of the posterior elements can be sequentially approached via a very limited access, making a combined posterior instrumented PLIF/TLIF procedure a small operation with regard to the patient’s recovery (see Fig 7.4.1-10 ).

483

7.4.1 Concepts of MISS/LISS

4

3.6

OTHER APPLICATIONS

This retractor system can also be very useful in cervical spine surgery and in shoulder and pelvic surgery for trauma, and has met with a lot of enthusiasm from nonspine surgeons. It is obvious that all the above procedures can be executed with the same technology, equipment and tools, making this system extremely versatile and cost effective in comparison with endoscopic technology, with all its disposables and the expensive special instruments required for endoscopic surgery in the retroperitoneum. The ring, which is fi xed to the table, is also stable in relation to the patient; it may also open a perspective for the integration of different new technologies, like computer-navigated surgery, where the ring could act as a stable reference platform for a computer-guided surgical procedure.

1.

2.

3.

4.

5.

6.

7. 8.

9.

BIBLIOGRAPHY

Aebi M (2005) Minimally invasive 360° lumbar fusion. Mayer HM (ed), Minimally Invasive Spine Surgery. A surgical manual . Berlin-Heidelberg-New York: Springer-Verlag, 435–449. Mayer HM (1997) A new microsurgical technique for minimally invasive anterior lumbar interbody fusion. Spine ; 22(6):691–700. Nolte LP, Zamorano LJ, Jiang Z, et al (1995) Image-guided insertion of transpedicular screws. A laboratory set-up. Spine ; 20(4):497–500. McAfee PC, Regan JR, Zdeblick T, et al (1995) The incidence of complications in endoscopic anterior thoracolumbar spinal reconstructive surgery. A prospective multicenter study comprising the fi rst 100 consecutive cases. Spine ; 20(14):1623–1632. Aebi M, Steffen T (2000) Synframe: a preliminary report. Eur Spine J ; 9(Suppl 1):S44–50. Nibu K, Panjabi MM, Oxland T, et al (1998) Intervertebral disc distraction with a laparoscopic anterior spinal fusion system. Eur Spine J ; 7(2):142–147. O’Dowd JK (2000) Laparoscopic lumbar spine surgery. Eur Spine J ; 9(Suppl 1):S3–7. Thalgott JS, Chin AK, Ameriks JA, et al (2000) Minimally invasive 360° instrumented lumbar fusion. Eur Spine J ; 9(Suppl 1):S51–56. Onimus M (1998) Minimal invasive anterior lower lumbar spine fusion. Presented at the Fifth Annual Meeting of the Society for the Study of Surgical Technique for Spine and Spinal Nerves; September 5/6, Tokyo.

484

7 SPINAL INSTRUMENTATION 7.4 SPECIAL TECHNIQUES AND INSTRUMENTATION 7.4.2 BONE HARVESTING TOOL

1

General principles ………………………………………………………………………………………… 485

2

Conclusion ………………………………………………………………………………………………… 487

3

Bibliography ………………………………………………………………………………………………… 488

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

485

Max Aebi

7 7.4

SPINAL INSTRUMENTATION SPECIAL TECHNIQUES AND INSTRUMENTATION

7.4.2

1

BONE HARVESTING TOOL

GENERAL PRINCIPLES

An important element of the so-called minimally invasive spine surgery is the new bone graft and substitute technology. The vertebral body bone harvesting with specialized instrumentation is one of these technologies. To harvest autogenous bone in order to fi ll a cage, the author uses the vertebral body adjacent to the disc that is to be fused [1–4]. Gentle soft-tissue preparation is done in either a cranial direction or a caudal direction. Therefore, it may be necessary to ligate or clip the pair of segmental vessels overlying the targeted vertebral body. Usually it is sufficient to position an additional Hohmann retractor in the cranial third of the vertebral body above or in the caudal third of the vertebral body below to lever all the soft tissue out of the way. A Steinmann pin with a stop at 2.5 cm penetration is inserted in the center of the vertebral body ( Fig 7.4.2-1a ) aligned as parallel as possible to the adjacent end plates in order to avoid a perforation of the end plate above or below the vertebra used for bone harvesting.

A large size trephine (diameter 12–18 mm) is inserted over the K-wire, which is used as a guide ( Fig 7.4.2-1b ). The size of the cylinder depth to be removed is prefi xed in the trephine instrument. The sharp trephine is inserted by hand through rotating movements and by hammer. Once the determined depth is reached, the trephine is removed and replaced by a trephine bone extractor. The properly positioned extractor is rotated 90° and will break the cylinder of bone out of the vertebral body. The cylindrical hole created in the center of the vertebral body is fi lled with a pressfit -tricalcium phosphate plug (Chronos) of the appropriate size ( Fig 7.4.2-1c–d ). The application of these cylindrical plugs usually leads instantly to an appropriate hemostasis. This cylindrical plug also restores the integrity of the vertebral body, so iatrogenic fractures of the vertebral body do not occur and the bony defect will be fi nally replaced by bone in a long-term remodeling process. If the bone obtained from the vertebral body is not sufficient, additional bone can be removed with a currette through the

486

7 7.4

Spinal instrumentation Special techniques and instrumentation

cylindrical canal, basically taking the bone from the lateral cylindrical wall, but not from the cranial or caudal parts. The bone cylinder and any additional bone is morselized and fi lled together with -tricalcium phosphate granula into the cage with the help of a special fi lling device ( Fig 7.4.2-2 ). Some

spare bone is put at the end, in front of the cage, in order to help the fusion. It is obvious that one or two simple tricortical bone grafts from the iliac crest or a femoral allograft fi lled with autogenous bone can be used, as well as a technical variant instead of cages.

1

*

2 a

b

Fig 7.4.2-1a–d Vertebral bone harvesting trephine. a Positioning of a K-wire in the center of the vertebral body with a stop at a depth of 2.5 cm in the middle of the L4 vertebral body (1). Trial cage in the disc space (2). b Insertion of the sharp trephine (hollow cylinder with a diameter of 12–18 mm) over the K-wire.

c

d

c Chronos cylinder ( -tricalcium phosphate) in a delivery instrument. The hole created where the bony cylinder has been harvested is visible (asterisk). d The bony cylinder in a pressfi t position in the vertebral body.

Fig 7.4.2-2 Device used to fill the cage with bone and Chronos.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

487

7.4.2 Bone harvesting tool

2

CONCLUSION

This minimally invasive procedure has defi nitively reduced the average number of hospitalization days; the surgical time is not any longer than in wide open surgery. A concise evaluation of the economic benefit of this surgery (technology vs use of hospital services and infrastructure, such as occupation of the operating room or hospitalization time) is still outstanding. The presented technique can be used from at least L2/3 to L5/S1. In fact, even L1/T12 and L1/2 can be approached by splitting the diaphragma from below. Although the risk of vessel damage had been considered high when introducing this technique, this has not been substantiated. In fact, there are very low vascular hazards in this surgery [3].

This technique is reproducible, has little postoperative and perioperative morbidity, and the patient can be mobilized early with a soft brace. It is not necessary to learn this technique from scratch, because these surgical techniques are very similar to what surgeons usually do in a macroscopic approach. This technique does not require a microscope, works well with loupes only, and has optimal illumination with the fiberoptic light.

The major advantages of this surgery are: • Access to the whole lumbar spine. • Assistants are no longer necessary—in fact this surgery can be done alone or with maximum one assistant. • Minimal exposure, less surgical pain, reduction of hospitalization and rehabilitation time. One other important advantage compared to other ALIF (anterior lumbar interbody fusion) procedures is that there is no longer any bone harvesting from the iliac crest, and as a result the iliac crest as a pain source is excluded. Therefore, generally speaking donor site complications are almost excluded. The removal of the cylindrical plug from the vertebral body adjacent to the involved disc space has not shown any relevant complications, as long as the rules of the direction of the bone harvesting tools are respected. The -tricalcium phosphate ( -TCP) or Chronos plug becomes totally incorporated ( Fig 7.4.2-3 ).

*

a

b

Fig 7.4.2-3a–b X-rays after an anterior and posterior L5/S1 fusion for axial low back pain. a The -TCP plug is clearly visible on the postoperative x-ray (asterisk). b X-ray control at the 2-year follow-up. The -TCP plug is no longer visible.

488

7 7.4

Spinal instrumentation Special techniques and instrumentation

3

BIBLIOGRAPHY

1. Steffen T, Downer P, Steiner B, et al (2000) Minimally invasive bone harvesting tools. Eur Spine J ; 9(Suppl 1):S114–118. 2. Kumar N, Wild A, Webb JK, et al (2000) Hybrid computer-guided and minimally open surgery: anterior lumbar interbody fusion and translaminar screw fi xation. Eur Spine J ; 9(Suppl. 1):S71–77. 3. Arlet V, Jiang L, Steffen T, et al (2006) Harvesting local cylinder autograft from adjacent vertebral body for anterior lumbar interbody fusion: surgical technique, operative feasibility and preliminary clinical results. Eur Spine J ; 15(9):1352–1359. 4. Khanna G, Lewonowski K, Wood KB (2006) Initial results of anterior interbody fusion achieved with a less invasive bone harvesting technique. Spine ; 31(1):111–114.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

489

490

7 SPINAL INSTRUMENTATION 7.4 SPECIAL TECHNIQUES AND INSTRUMENTATION 7.4.3 VERTEBROPLASTY

1

Historical review …………………………………………………………………………………………… 491

2

Today’s clinical experience ……………………………………………………………………………… 492

3 3.1 3.2 3.3

Indications ………………………………………………………………………………………………… Osteoporosis ……………………………………………………………………………………………… Tumors ……………………………………………………………………………………………………… Contraindications …………………………………………………………………………………………

4

Preoperative workup ……………………………………………………………………………………… 493

5

Strategy for reinforcement ……………………………………………………………………………… 494

6 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8

Surgical technique ………………………………………………………………………………………… Suggested surgical technique …………………………………………………………………………… Positioning ………………………………………………………………………………………………… Monitoring/anesthesia …………………………………………………………………………………… Material needed to perform the procedure ………………………………………………………… Visualization/imaging …………………………………………………………………………………… Insertion of a guide wire/cannula ……………………………………………………………………… Cement preparation ……………………………………………………………………………………… Cement injection …………………………………………………………………………………………

7

Postoperative procedure ………………………………………………………………………………… 502

8

Pitfalls ……………………………………………………………………………………………………… 503

9

Bibliography ………………………………………………………………………………………………… 504

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

492 492 492 493

494 494 495 495 496 497 498 500 501

491

Paul Heini

7 7.4

SPINAL INSTRUMENTATION SPECIAL TECHNIQUES AND INSTRUMENTATION

7.4.3

1

VERTEBROPLASTY

HISTORICAL REVIEW

Vertebroplasty means the (percutaneous) reinforcement of vertebrae that are weakened from osteoporosis or tumors. This procedure was fi rst performed in 1984 by Galibert and Deramond (French neurosurgeons/interventional radiologists from Amiens) in the case of a C2 hemangioma when they injected PMMA using an open technique; they reported their experience of seven patients with vertebral hemangiomas of the cervical and lumbar spine in 1987. They showed that the destruction of an angioma and consolidation of the vertebral column can be obtained by a percutaneous intrasomatic injection of acrylic cement [1]. Another group from Lyon (France) reported their experience on the treatment of vertebral metastases in 20 patients [2]. In 1990 Galibert reported an extension of the use of percutaneous cement application for osteoporotic fractures and myelomas [3]. In 1994, Gangi from Strasbourg presented the use of the combined technique of CT guidance and fluoroscopy in the treatment of ten patients [4]. Weil et al from Paris presented their results in the treatment of 37 patients with different pathologies [5]. Among French

radiologists and neurosurgeons, percutaneous and open reinforcement of vertebrae became popular especially for the treatment of hemangiomas [6–9] and for vertebral metastases [5, 10, 11]. The fi rst reports on the treatment of a larger series of patients with osteoporotic fractures were published in 1997 by Jensen [12]. Biomechanical studies on cadaver vertebrae were performed in order to demonstrate the effect of cement injection. Different fi lling materials as well as different fi lling patterns were analyzed [13–21]. In all studies the cement fi lling was from a mechanical perspective highly effective. PMMA shows a more pronounced effect in comparison to CaP cements [17, 18]. A mono or bilateral approach is equally efficient. With lower injection volumes the height loss of vertebral bodies is more pronounced [13, 18, 22]. When the whole motion segment is analyzed, there seems to be an increased risk of adjacent vertebrae failure [23].

492

7 7.4

Spinal instrumentation Special techniques and instrumentation

2

TODAY’S CLINICAL EXPERIENCE

In the last years, numerous papers on the clinical experience of vertebroplasty have been published [24–38]. The consensus of opinion in these papers is that vertebroplasty is an efficient means of treatment for osteoporotic fractures. The clinical results for the treatment of osteoporotic fractures is superior to that of metastatic lesions [5]. Controversies remain regarding the efficiency of treatment with regard to age of the lesion and the expected effect of reinforcement, and in particular, of the failure of adjacent vertebrae. Maynard found an improvement of symptoms in 93% of 27 patients having 35 fractures with positive bone scans. Kaufmann did not see any significant difference between the age of fracture (1 week to 2 years) and the outcome, but reported that the amount of analgesic consumption before treatment is of importance in the outcome [30, 35]. A higher incidence of new fractures after reinforcement is described by Grados and Uppin [27, 39]. The natural history of osteoporosis on the other hand shows an increasing risk for further fractures with increasing number of vertebral fractures already present [40, 41].

3

INDICATIONS

3.1

OSTEOPOROSIS

The main indication for cement reinforcement is the osteoporotic spine. • Osteoporotic fractures with concordant ongoing pain for 3–8 weeks after an initial incident respond satisfactorily in a high number of cases. • Patients with severe pain who remain bedridden for longer than 1 week can profit from an augmentation. • Progressive compression fractures of one or multiple vertebrae with subsequent development of increasing kyphosis; in such situations, multilevel injections in repeated sessions may be necessary. • Patients with persisting instability after a fracture (ie, Kummel-Verneuil disease) can be treated with closed reduction and cement injection. • Combined procedures with internal fi xation in severe osteoporosis.

3.2

TUMORS

Osteolysis due to myeloma is probably the most important indication for vertebroplasty. The indication for treatment is based on the amount of destruction of the vertebra and, providing the pedicles and facet joints are not affected, the injection of PMMA can be helpful. Osteolysis by itself does not need to be reinforced, unless there is a significant destruction of the vertebral body with a clinical correlation of mechanical back pain. A CT scan is recommended to assess the bony destruction and the mechanical situation, which cannot be achieved with an MRI scan. Solid metastatic lesions, such as are found breast and prostate cancer, can be reinforced. During the

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

493

7.4.3 Vertebroplasty

4

injection of cement the solid part of the tumor cannot migrate into the circulation, but the increased pressure induced by the cement injection can lead to cement or tumor migration into the spinal canal or foramina. The reinforcement of the vertebrae after radiation therapy is less problematic as the tumor has shrunk and the cement volume can fi ll the void. The indication for reinforcement in tumors should be set by a spine surgeon who is experienced in the operative management of such lesions. Hemangiomas represented the initial indication for this procedure. However, the reinforcement is only indicated if there is an aggressive type of hemangioma. Its fi lling is demanding because the cavity needs to be fi lled as completely as possible in order to avoid recurrence [42].

3.3

• • • • •

CONTRAINDICATIONS

Symptoms that are not related to pathology. Associated spinal stenosis. Infection. Poor visibility with fluoroscopy. If an open procedure is more appropriate.

PREOPERATIVE WORKUP

Clinical

The preoperative evaluation includes a careful assessment of the patient. Information about pain location and pain severity should be obtained. It is important to identify whether or not the patient is able to stay in a prone position for 1 hour. Coumarine-like anticoagulants must be stopped prior to the procedure, however, the intake of aspirin-like medication must not be discontinued. Imaging

The investigation of choice remains a standing x-ray in two planes. If there is a concordance of clinical and imaging investigation, no further examinations are needed. If the fracture pattern or the patient’s history is not clear an MRI investigation is recommended. The reinforcement of osteoporotic vertebrae does not require an intact posterior wall because the cement flow in this direction is easy to control. Often the age of the fracture is difficult to determine. The comparison with older x-rays can be helpful (patients may have had previous chest x-rays). An MRI scan can be helpful in detecting fresh lesions when an edema of the bone may be visible. A Tc-bone scan can be helpful as a screening tool for vertebral fractures. The evaluation of metastatic lesions and tumors with a CT scan can show the exact bony destruction. “Instability” can be identified by comparing a standing x-ray with the MRI scan taken in a supine position. This can also give information about the possibility of achieving some reduction when the patient is positioned prone during surgery. A hyperextension cross table lateral view can be used alternatively.

494

7 7.4

Spinal instrumentation Special techniques and instrumentation

5

STRATEGY FOR REINFORCEMENT

6

Which levels should be reinforced? If there is a fresh fracture in a patient with mild osteoporosis, a single level treatment seems adequate; in such cases a bilateral approach is recommended. If the osteoporosis is more severe than anticipated, injection of the adjacent vertebrae as a prophylactic measure against further fractures is recommended. If multiple fractures are present, the injection of all fractures as well as the adjacent intact vertebrae is recommended. There are clinical situations in which patients sustain multiple fractures with progressive kyphosis over a short period of time. In this situation it is recommended to reinforce the areas of the spine at risk, which are the thoracolumbar junction and the kyphosis of the thoracic spine. In these situations, the reinforcement aims to treat the osteoporotic spine as a whole. This can be achieved within two to three sessions, depending on the number of levels treated each time. The augmentation of 10 to 12 levels allows the most critical part of the spine, ie, from L4 to T5, to be treated. If multiple levels are to be injected, a monolateral approach is chosen, and the injection should be performed initially in the most painful area.

SURGICAL TECHNIQUE

The technique uses a percutaneous transpedicular or a posterolateral approach, and the cement is injected through a cannula. The size of the cannula varies between a 7 and 14 gauge. For the placement of the cannula, CT guidance is recommended by some authors [4]; the injection of the cement is performed under lateral fluoroscopic control, and a biplanar control is recommended. The cement injection requires a specially designed injection gun or a number of small syringes [12, 28, 43]. For the safety of the procedure the cement should be injected in a highly viscous state [44].

6.1

SUGGESTED SURGICAL TECHNIQUE

The presented technique has been refi ned during the last 5 years on more than 2,000 augmented vertebral bodies. The key points of the technique are as follows: • • • •

Use of a guide wire. Use of cannulas with large diameter (8 gauge). Use of radiopaque cement with an adapted viscosity. Direct cement injection using small syringes (2 cc, 1 cc).

A detailed description of the surgical procedure can be obtained from Video 7.4.3-1.

Video 7.4.3-1 Multilevel vertebroplasty in the thoracic spine.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

495

7.4.3 Vertebroplasty

6.2

POSITIONING

The procedure is performed in the operating room. The patient is placed in a prone position resting on a beanbag that allows adjustment and provides maximal comfort. If for any reason a general anesthesia is chosen, the patient can be placed in hyperextension. This may provide a reduction of the compressed vertebrae ( Fig 7.4.3-1).

6.3

MONITORING/ANESTHESIA

a

The presence of an anesthesiologist is necessary to administer additional sedatives and analgesics. An electrocardiogram and pulse oximetry are applied and an oxygen mask is used. The more painful procedures (placement of the guide wire into the bone, insertion of the cannula, cement injection) require additional intravenous analgesia.

b Fig 7.4.3-1a–b a Positioning of a patient on a beanbag if local anesthesia is considered. b In general anesthesia, positioning in hyperextension can promote spontaneous reduction of the fractured vertebra.

496

7 7.4

Spinal instrumentation Special techniques and instrumentation

6.4

MATERIAL NEEDED TO PERFORM THE PROCEDURE

To perform the vertebroplasty the following material is needed: local anesthetic (mepivacaine 1%), 20 cc syringes for cement distribution and 1/2 cc syringes for cement injection, 2.0 mm/20 cm guide wire, disposable 8 gauge standard bone marrow biopsy cannulas. Bone cement with a high amount of radiopacifier is used to perform the reinforcement ( Fig 7.4.3-2 ). Several specially designed surgical kits are available for this purpose. Vertebroplasty or kyphoplasty kits are available from different companies depending on the procedure chosen.

a

b

Fig 7.4.3-2a–b a Setup of material for vertebroplasty: guide wire (2 mm/20 cm), filling cannulas (8 gauge, 15 cm), bone cement (PMMA) with high radiopacity, mixing cup, 20 cc syringe, 2 cc syringes. b Tools for kyphoplasty: working cannula, reamer, balloon, volume/pressure device, saline enhanced with contrast die (45 cc of saline and 15 cc of iodine contrast die), bone filler devices.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

497

7.4.3 Vertebroplasty

6.5

VISUALIZATION/IMAGING

Free access for the C-arm in the posteroanterior and lateral projection at the level of pathology is mandatory. A high-quality C-arm with a wide distance between tube and camera is essential. It is also possible to install two image intensifiers in order to control both planes simultaneously. The area to be treated is identified before draping and cleaning. If the area of pathology can be seen easily by fluoroscopy, reinforcement can be performed. Limited visualization will prevent the use

a

of this technique. The visualization of the upper thoracic spine is difficult and will vary between T3 and T6 ( Fig 7.4.3-3 ). After draping, the vertebrae to be augmented are identified with the C-arm, which is adjusted in the posteroanterior view so that the view is parallel to the end plates. In this position the pedicles are well visualized. The authors prefer to have a strict posteroanterior view, although it is also possible to use the so-called bulls-eye view. The principle for the orientation of wire insertion is depicted in Fig 7.4.3-4 .

b

Fig 7.4.3-3a–b Free access for the C-arm in the posteroanterior and lateral projections at the area of interest is mandatory. A high-quality C-arm is essential. The area to be treated is examined before draping and the levels to be treated are preliminarily indicated.

498

7 7.4

Spinal instrumentation Special techniques and instrumentation

6.6

INSERTION OF A GUIDE WIRE/CANNULA

Local anesthetic is injected into the skin and the subcutaneous tissue down to the periosteum at the site of insertion of the guide wire. At each site 3–5 cc are administered. In order to avoid repeated changes from the posteroanterior to the lateral view, all vertebrae to be reinforced are injected with local anesthetic. Using the C-arm control, a stab incision is made in the skin and the guide wire is advanced convergent to the projection of the pedicle. In order to penetrate the surface of the bone, some gentle taps with a hammer are necessary. The direction of the guide wire is adjusted as required and advanced continuously under C-arm control. As soon as the tip of the wire reaches the medial border of the pedicle, the position of the wire needs to be verified in the lateral projection ( Fig 7.4.3-4 ).

b

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

a Fig 7.4.3-4a–c a First, align the x-ray beam parallel to end plates. b–c Planning the insertion of the guide wire based on the AP and lateral view: local anesthesia is set, guide wires are advanced under C-arm control. The surgeon needs to combine the AP and lateral view in order to aim the guide wire correctly into the vertebral body. Depending on the size of the pedicle, the guide wire is inserted transpedicularly or parapedicularly. As soon as the tip of the K-wire reaches the medial border of the pedicle on the fluoroscopic AP view, its depth should be checked in the lateral view, so it reaches the posterior border of the vertebral body.

499

7.4.3 Vertebroplasty

Before changing the projection, the wires are preliminarily inserted at all levels where cement injection is planned. Once the wires are preliminarily inserted, a picture of their position is stored in the image intensifier and the C-arm is then switched to the lateral position. In the lateral projection, the tip of the wires must be beyond the posterior wall of the spinal canal, otherwise the wire needs to be relocated by switching back to the posteroanterior view. The guide wire is then advanced cautiously with gentle hammer taps and if necessary, redirected in order to reach the center of the vertebral body. The fi lling cannulas are inserted over the guide wires by rotating movements. This procedure can be painful and the anesthesiologist should be informed to provide appropriate analgesia. The tip of the cannula should be advanced until the anterior half of the vertebral body is reached ( Fig 7.4.3-5 ).

The guide wire must not be pushed forward while inserting the cannula. After insertion of the cannula the guide wire is removed and a blunt trocar is used to clear the tip of the cannula; this should be performed with the hammer. Anterior perforation must be avoided. Probing with the trocar also confi rms the position of the cannula inside the bone.

Fig 7.4.3-5a–c After preliminary insertion of the guide wires, the cannulas are slid over and placed in the center of the vertebral body. This is done under fluoroscopic control. The tip of the cannula is then cleared with a blunt trocar which in turn confirms the location in the bone. If there is any doubt about the location, the C-arm should be switched to the posteroanterior projection.

a

b

c

500

7 7.4

Spinal instrumentation Special techniques and instrumentation

6.7

CEMENT PREPARATION

The fi lling of the vertebral body is accomplished with PMMA cement. At the present time, there is no alternative, such as a CaP cement, available. Different cements can be used for reinforcement. Conventional PMMA cement does not show sufficient radiopacity, however, formulas are available for the reinforcement of osteoporotic bone with a high amount of barium or zirconium. These cements are highly radiopaque and provide a sufficient working time. The mixed cement is fi lled into 2 cc syringes and only injected when the viscosity is adequate—which means producing long threads of cement with no dripping when pushed from the syringe. Because of the potential deleterious extravasation risk, low-viscous cement must not be injected. Each surgeon must be familiar with the cement with respect to mixing, working time, and hardening; for example, the vertebroplastic formula needs at least 7 minutes before injection can be started.

a

b

c

d

e

f

The use of injection guns can increase the injection force, however, the flow control is not as direct, especially when long and thin connecting tubes are used ( Fig 7.4.3-6 ).

Fig 7.4.3-6a–f Preparation of cement: the components are mixed in a bowl with the spatula for 30 seconds, after another 30 seconds of rest the cement is decanted into a 20 cc syringe and another minute of rest is given. The cement is then filled into 2 cc syringes. These syringes provide adequate power for the cement injection (1 cc syringes can be used alternatively). The cement is not injected until the viscosity is adequate, which is when it does not drip out any more when pushed from the syringe but appears with long threads instead. For the vertebroplastic formula that is used most commonly, injection is started at least 7 minutes after mixing is started.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

501

7.4.3 Vertebroplasty

6.8

CEMENT INJECTION

Cement injection is done under continuous lateral C-arm control. Before any cement is injected, the images of the vertebrae to be reinforced are stored in the image intensifier and should be available on the second screen as a reference. The anesthesiologist needs to be informed of the injection in order to provide adequate analgesics. Once the cement is ready to be applied the cannulas are fi lled. The void volume of a standard injection cannula takes 1.5 cc of cement. With high radiopaque cement the cement flow can already be seen in the cannula. Once the cement appears in front of the cannula, injection is stopped. If a bilateral approach is used, the contralateral side is fi lled as well. It is important to see the flow behavior of both cannulas. Once the fi lling of one side is accomplished, the other side is hidden by the cement and the flow is more difficult to monitor. Therefore, when a bilateral approach is chosen the injection is performed stepwise and simultaneously. Often, if one side shows an outflow of cement, fi lling via the opposite pedicle is still possible. The injection is continued after the cement appears at the opening of the cannula. The cement should behave like a “growing cloud”. The flow of the cement needs to be monitored with continuous fluoroscopic control. If a “spider-like” behavior occurs, one should wait until further setting of the cement; the cannula can eventually be pulled back a little and the fi lling is then restarted and the flow monitored. If it does not behave as expected or if the cement cannot be seen clearly, the injection must be stopped. The cement flow toward the posterior wall of the vertebral body can be monitored more reliably than the lateral flow. The cement flow follows the path of least resistance. When a fresh fracture is reinforced, the flow along a fracture gap toward the disc space is frequent. In such a situation a

bilateral approach can be helpful. The most difficult part is the control of cement flow laterally and avoiding cement flow into the draining veins. Therefore, it is recommended to switch to a posteroanterior projection during injection, especially when the fi lling pattern is not uniform. If the cement has hardened to a state where injection with the syringe is no longer possible, the trocar can be used to carefully push the cement volume of the cannula forward. This technique allows the insertion of more cement at a higher viscosity, which can be helpful, particularly when an initial extravasation has occurred. In the lumbar spine, the risk of extravasation into veins seems increased. The image is limited due to the pelvis and soft tissues. Therefore, the author advises the use of a posteroanterior control after starting the injection in order to check the flow behavior. The newer PMMA formulas allow enough time to perform this check. The amount of cement injected depends upon the individual situation. It is possible to achieve a cement support from the upper to the lower end plate. The more pronounced the osteoporosis the easier the injection. On average 3–6 cc of PMMA is applied per level ( Fig 7.4.3-7 ). If multiple levels are injected, a monolateral approach is usually chosen, alternating left and right in order to avoid interactions of the cannulas. If there is an asymmetric collapse of the vertebra, the site of greater collapse is initially reinforced in order to prevent further deformation. It is possible to fi ll two cannulas simultaneously; if multilevel reinforcement is necessary, up to six levels can be injected in two to three steps. The cement should be allowed to harden before removal of the cannula in order to avoid dragging cement into the soft tissues. Bleeding at the puncture site is common and a tight closure of the skin incision is required.

502

7 7.4

Spinal instrumentation Special techniques and instrumentation

7

POSTOPERATIVE PROCEDURE

After the procedure the patient remains in a supine position for 1 hour for wound compression. Subsequently, mobilization is allowed according to the patient’s wish. The effect of the procedure can be evaluated immediately as the usual pain level should be reduced and the only discomfort present is due to the puncture.

a

b

c

d

Fig 7.4.3-7a–d The cement is injected under continuous lateral C-arm control. The appearance of the cement at the tip must be observed very carefully. If the cement disperses in any direction far from the tip this means direct connections to vascular channels in the bone. One must then wait for 45 seconds before applying further small amounts of cement. The cement should behave like a growing cloud. Cement is injected stepwise. After the first 2 cc are applied, a posteroanterior control with the C-arm should be performed. Afterward continuation of injection with the C-arm in the lateral projection is done. Once growing resistance to the injection is felt, the trocar can be used to push in the cement; this allows a very controlled application of the cement. At every sign of leakage one must stop the injection.

AOSPINE MANUAL—PRINCIPLES AND TECHNIQUES

503

7.4.3 Vertebroplasty

8

PITFALLS

The major complications related to percutaneous cement reinforcement are problems of cement extravasation, embolization, and cement leakage toward the spinal canal. These complications must be avoided. Furthermore, systemic reactions during cement injection can occur as a consequence of the extravasation of the toxic cement monomer. In the literature, several reports on complications can be found [32, 45–51]. In order to minimize the risk of extravasation, it is strongly recommended to follow the instructions described: • The use of large diameter cannulas. • Injection of cement with enhanced radiopacity. • Injection at the correct viscosity. The use of small syringes allows good control of the cement flow. Any suspicious cement flow behavior must lead to a discontinuation of the injection. The fi lling behavior changes with increasing viscosity—if the cement flow does not behave as expected, one should pause for 45 seconds and reinject a small amount of cement. Reinforcement of the osteoporotic vertebral bone means substitution of the bone marrow with cement. The fatty bone marrow is spilled into the circulation and cleared in the lungs. Therefore, the maximal amount of cement injected per session is restricted to 25–30 cc. In other words, no more than six levels should be reinforced per session [36, 38]. The risk of a fracture at adjacent levels appears to be increased after cement reinforcement [27, 39, 52]. Therefore, patients and their doctors should be made aware that if new pain does occur, a new fracture may have been sustained and should be injected. Patients with osteoporosis should be evaluated and treated systemically with vitamin D and bisphosphonates and, therefore, be followed up by the family doctor or osteologist.

The placement of the cannula carries the potential risk of damaging neural structures. Therefore, if this technique is performed, familiarity with the spinal anatomy and experience with open surgery are mandatory.

504

7 7.4

Spinal instrumentation Special techniques and instrumentation

9

BIBLIOGRAPHY

Galibert P, Deramond H, Rosat P, et al (1987) [Preliminary note on the treatment of vertebral angioma by percutaneous a