Therapeutic Exercise for Lumbopelvic Stabilisation 2nd edition By Carolyn Richardson

SECOND EDITION THE EUTIC EXERCISE FO LUMBOPELVIC STABILIZATION A Motor Control Approach for the reatment and Prevention of Low Back Pain CAROLYN RICHARDSON PAUL HODGES JULIE HIDES ,/j)CHURCHIUlIVINGSTON� :u

SECOND EDITION THERAPEUTIC EXERCISE FOR LUMBOPELVIC STABILIZATION A Motor Control Approach for the Treatment and Prevention of Low Back Pain Therapeutic Exercise for Lumbopelvic Stabilization (previously entitled Therapeutic Exercise for Spinal Segmental Stabilization in Low Back Pain) is based on the evidence from research undertaken by the authors over a number of years. The significance of these findings to the treatment and prevention of low back pain are now widely acknowledged, not only among researchers but also, and perhaps more importantly, among practitioners concerned with the management and prevention of back pain. In this new edition they have taken the opportunity to extend the scope of the book to accommodate the most recent evidence, which has emerged since the first edition was published in 1999. Key Features • Written by three of the foremost researchers in the field of musculoskeletal dysfunction • Incorporates the very latest research from key workers in the field • Demonstrates the clinical relevance of the research findings to the student and busy practitioner • Presents a new, problem-solving approach to back pain assessment and management based on the latest understanding of the anatomy, physiology and biomechanics involved • Suggests future directions for clinical management and future research • Extensively illustrated with line diagrams and photographs. Therapeutic Exercise for Lumbopelvic Stabilization presents the latest information on the muscle systems involved in the prevention and management of musculoskeletal pain and dysfunction, and introduces a unique approach to clinical management and prevention based on that research. It is an important book in that it not only presents the evidence but also gives practical guidance on how the findings may be applied in everyday / practice. The first edition was widely welcomed and acc,laimed by researchers and clinicians alike. This new This product is appropriate for. edition will continue to provide an indispensable practical • manual therapy reference source for all those working in the field of • physiotherapy musculoskeletal pain and dysfunction. • chiropractic • osteopathy �/��\\ CHURCHILL • orthopaedics LIVINGSTONE • pain management :u • primary care An imprint of Elsevier Ltd Visit our website for ISBN 0-443-07293-0 additional outstanding products II r ELSEVIER www.elsevierhealth.com 9 7804 4 3 7 2 9 32

Therapeutic Exercise for Lumbopelvic Stabilization

For Churchill Livingsione: Commissioning Editor: MalY Law /Saxena Wolfaard Project Development Manager: Mairi McCubbin Project Manager: Samantha Ross Dec;ign: Judith Wright

Thera peu tic Exercise for Lumbopelvic Stabilization A Motor Control Approach for the Treatment and Prevention of Low Back Pain SECOND EDITION Carolyn Richardson BPhty(Hons) PhD Associate Professor and Reader, Division of Physiotheropy, School of Health and Rehabilitation Science, University of Queensland, Brisbane, Australia Pau I W. Hodges BPhty(Hons) MD(Neurosci) PhD NHMRC Senior Research Fellow and Associate Professor, Division of Physiotherapy, School of Health and Rehabilitation Science, University of Queensland, Brisbane, Australia Julie Hides BPhty MPhtySt PhD Clinical Supervisor, The University of Queensland/Mater Hospital Back Stability Clinic, Brisbane, Australia This book has been endorsed by the MACP ma MANIPULATION aD ASSOCIATION OF CHARTERED PHYSIOTHERAPISTS \"\"/�\" �\\ CHURCHILL - LIVINGSTONE EDINBURGH LONDON NEW YORK OXFORD PHILADELPHIA ST LOUIS SYDNEY TORONTO 2004

CHURCHILL LIVINGSTONE An imprint of Elsevier L imited f) Harcourt Brace and Company Limited 1989 �) Harcourt Publishers Limited 1999 G�) Elsevier Limited 2004. All rights reserved. The right of Caroly n Richardson, Paul Hodges and Julie Hides to be identified as a uth ors of this work has been asserted by them in accordance with the Copyright, Designs and Patents Act 1988 No pMt of thi s publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, record ing or otherwise, wi thout either the pr ior permission of the publishers or a licence permitting restricted copy ing in the United Kingdom issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London WIT 4LP. Permissions may be sough t directly from Elsevier's Health Sciences Rights Department in Philadelphia, USA: phone: (+1) 215 238 7869, fax: (+1) 21.5 238 2239, e-mail: [email protected]. You may also comp lete your request on-line via the Elsevier homepa ge (http://www.elsevier.com). by selecting 'Customer Support' and then ' Obtain ing Permissions'. First edition 1999 Second edition 2004 [SBN 0443 07293 0 British Library Cataloguing in Publication Data A catalogue record for this book is ava i lable from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book is available from the Library of Congress Notice Knowledge and best practice in this field are constantly changing. As new research and experience broaden our knowledge, changes in practice, treatment and d rug therapy may become necessary or appropriate. Readers are adv ised to check the most current iJl formation provided (i) on procedures featured or (u) by the manufacturer of each p roduct to be administered, to veri fy the recommended d ose or formula, the method and duration of administration, and contraindica­ tions. [t is the responsibility of the practitioner, relying on their own experience and knowledge of the pa tient, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the author 110r the MACP assume any lia bi lity for any injury and/or damage. your source for books, The Publisher journals and multimedia in the health sciences The Publish.r\"s www.elsevierhealth.com policy is 10 use paper manufachJred Printed in China trom sustainable forests I

v Contents Preface VII 6. The role of weightbearing and non­ Acknowledgements ix weightbearing muscles 93 Carolyn Richardson SECTION 1 Introduction SECTION 3 Impairment in the joint 103 protection mechanisms: concepts --_.._. --= 7. The deload model of injury 105 1. The time to move forward 3 Carolyn Richardson Carolyn Richardson 8. Joint injury 119 SECTION 2 The joint protection 9 Julie Hides mechanisms 9. Pain models 129 PART 1 Introduction 11 Paul Hodges 2. Lumbopelvic stability: a functional model of the SECTION 4 Impairments in the joint biomechanics and motor control 13 protection mechanisms in low Paul Hodges back pain 139 PART 2 Specific joint protection of the 29 10. Abdominal mechanism in low back pain 141 spinal segments Paul Hodges 3. Abdominal mechanism and support of the 11. Paraspinal mechanism in low back pain 149 lumbar spine and pelvis 31 Julie Hides Paul Hodges 12. Impairments in muscles controlling pelvic 4. Paraspinal mechanism and support of the orientation and weightbearing 163 lumbar spine 59 Carolyn Richardson Julie Hides PART 3 The antigravity muscle support SECTI.ON 5 Treatment and prevention system of low back pain 173 5. Stiffness of the lumbopelvic region for 75 load transfer 77 Carolyn Richardson and Julie Hides 13. Principles of the 'segmental stabilization' exercise model 175 Carolyn Richardson, Julie Hides and Paul Hodges

14. Local segmental control 185 Julie Hides, Carolyn Richardson and Paul Hodges 15. Closed chain segmental control 221 Carolyn Richardson and Julie Hides 16. Open chain segmental control and progression into function 233 Carolyn Richardson and Julie Hides References 247 Index 265

vii Preface Over the past decade a major focus of rehabilitation Clinical studies have also conlirmed the presence has turned to exercise to improve the stability of the of muscle dysfunctions involving the deep muscle lumbar spine and pelvis. There are a variety of clin­ system in pelvic pain syndromes. Other develop­ ical and research opinions in this area and many ments include greater understanding of the mech­ methods have become popular in both the clinical anisms for control and coordination of this system arena and fitness industry. This book provides an and the effects of unloading, and pain and injury. overview of our interpretation of the field based on Our recent research involving microgravity envi­ our research and clinical practice. While the first ronments is providing the opportunity to evaluate edition provided an introduction to our new the effects of extreme environments that are likely models of exercise and the state of knowledge at to have an impact on functional environments on that time, this second edition provides an updated Earth. An additional area that is continuing to view that integrates the burgeoning research in this expand is in the assessment of muscle control in field and the clinical advances. lumbopelvic pain. In this field we have made major advances in non-invasive methods to assess In relation to therapeutic exercise in low back this system. pain, we believe that the focus of exercise inter­ ventions by physiotherapists and other health As research physiotherapists in the area of thera­ professionals should be designed to establish the peutic exercise for low back pain we have chosen to optimum interaction of muscles necessary to con­ investigate, in the first instance, the neurophysio­ trol and protect the joints, during the performance logical mechanisms involved in joint protection of the lumbo-pelvic region and the dysfunctions of a great variety of functional body movements. which can occur. Even though many of the ideas and hypotheses presented have not yet undergone In our first edition, we focused on the system of rigorous scientific scrutiny, we feel that we have an deep muscles that our research and clinical evi­ obligation in a text book on therapeutic exercise to dence suggest are vital in control of the lumbar provide details of the new ways to approach exer­ segments: the multifidus, transversus abdominis, cise prescription as well as providing hypotheses for diaphragm and the pelvic floor muscles. So how why certain exercise techniques traditionally used by physiotherapists are likely to be very effective. has our research progressed over the last 5 years? Thus the research findings plus argued hypothe­ Several key aspects have progressed. For instance ses in this text have been used to give some insight long term follow-up data now indicates that the into therapeutic exercise techniques which are likely interventions described in the first edition led to to be effective and also to develop non-invasive reduction of low back pain recurrence rates. Our measures which will reflect the problems in the biomechanical studies have confirmed the import­ ant contribution of the deep muscle system to con­ trol of not just the lumbar spine , but also the pelvis.

viii PREFACE musculoskeletal system in relation to joint protec­ using therapeutic exercise as a preventative meas­ tion. The principles presented can be applied to any ure and to promote a change in l i fe style, not only as a treatmen t after problems have occurred. This region of the body. One of the main aims of our past as well as our is a major focus for our ongoing work. We hope that you find reading the second edition future research is to demonstrate that the pre­ of this textbook thought provoking and enjoyable. scribed therapeutic exercises are resulting in We are excited that the work continues to evolve improvement in the joint protection systems, and hence demonstrate that changes in these mecha­ and grow in new directions. We hope that this book nisms are closely linked to the resolution and even will be useful to clinicians, students and researchers prevention of painful symptoms. As a final note, we also would like to emphasize the importance of alike, and may stimulate new ideas which will ultimately help those w ith lumbopelvic pain.

ix Acknowledgements The authors wish to first of all acknowledge the of a neutral spine position for both the testing and past efforts of Gwend olen Jull whose exceptional knowledge of physiotherapy rehabilitation have treatment of spinal and trunk muscles . made a signi ficant contribution to this book. Thanks to the staff, clinicians and students of With the expansion of the research, each the Joint Stability Research Unit Warren Stanton, Alison Grimaldi, Ruth Sapsford, Sally Hess, Chris researcher has their own group of people to thank. Hamilton, Daniel Belavy, Nathan Stewart, Quentin Caroline and Julie extend their thanks to the primary overseas research collaborator, Professor Scott, Helen Flem ing, Sue Roll, Sue Kelley, Jan Chris Snij ders from Erasmus University in Smith, Mark Comerford, Rowena Toppenberg, Holland, whose biomechanical models blended Heidi Keto and the staff at the Back Stability Clinic well with their therapeutic exercise models for low for their contribution to the development of the back pain. In addition it was Professor Snijders exercise model. who invited them onto the Topical Team for Low The editors' knowledge and expertise in real­ Back Pain, an initiative of the European Space time ultr asound imaging has only d eveloped and Agency (ESA). evolved with the assistance of Dr Dav id Cooper, Thanks also to Benny Elmann Larsen, Senior an obstetrician and gynaecologist who specializes Physiologist for ESA, and Richard Lilmehan, NASA in ultrasound imaging. His encouragement and astronaut, who have given Caroline and Julie the guidance have been invaluable . Introducing some­ confidence to pursue the ideas on the 'Deload thing new into a profession is never easy, and evi­ Model of Injury'. Dr Steve Wilson and Daniel dence of the success of the medium can be seen as Belavy deserve special mention with their excep­ it develops into being a routinely used rehabilita­ tional expertise in instrument and sof tware devel­ tion and assessment tool in many physical therapy opment for the measures used for the University of practices around the world. Queensland's contribution to ESA's Berlin bed rest A S PECIAL THANKS TO OUR FAM ILIES study at the Free University in Berlin. Caroline thanks her husband, Bren, for his never The collaborative research with Dr. Joseph Ng, failing support with this very difficult but exciting of the Hong Kong Poly technic University , has led task of writing a book and his numerous trips to Europe with her to meetings and conferences. to new discoveries of the dysfunc tions present in the trunk muscles of low back pain patients and Julie sends special thanks to her husband, assisted with the development of the exercise Damian, and ch ildren Emma, Jonathon and model. It was physiotherapist, Christine Hamilton Cameron for their patience and encouragement. who alerted us, many y ears ago, of the importance

Without the support of her parents, Jill and David of the neural control of the spine. They have Cooper (and provision of extensive childcare taught him the rigor of scientific endeavour. assistance) she never would have completed the 'last yards'. Thanks particularly to the international col­ laborators. The team in the Biomechanics and PH motor control laboratory of the Department of Neuroscience at Karolinska Institute (Alf Paul thanks the collaborators in Australia and Thorstensson, Andy Cresswell, Karl Daggfeldt overseas for their contribution to this work. At the and Anatoli Grigorenko) have provided important University of Queensland he thanks the team of guidance in biomechanics and motor control and the Human Neuroscience Unit (Michel Coppieters, stimulated Paul's pursuit of a second doctorate, in Catharina Bexander, Andrew Chapman, Rebecca neuroscience. On the other side of Sweden the Mellor, Angela Chang, Sallie Cowan, Paulo collaborators at the Sahlgrenska University Ferreira, Manuela Ferreira, Joanna Knox, Linda­ Hospital (Allison Kaigle Holm, Sten Holm, Lars Joy Lee, David McDonald, Nicola Mok, Steven Ekstrom, Tommy Hansson) have provided an Saunders, Michelle Smith, Donna Urquhart, unrivalled opportunity to test the mechanisms of Richard Yang) for their stimulating debate and spinal control when the methods go beyond the contribution to the work presented here. In partic­ limits of human experimentation. ular, thanks to Lorimer Moseley who has con­ tributed significantly to the work dealing with He would also like to thank his collaborators at pain and its effect on the motor system. TIUs work the University of Melbourne (Kim Bennell, Sallie is providing new insights into the possible reasons Cowan, Kay Crossley) for giving him the opportu­ for development of recurrent pain. nity to be involved in testing the model of motor learning, albeit in another part of the body. This He is also grateful to his colleagues at the Prince research is providing a foundation for under­ of Wales Medical Research Institute in Sydney standing the mechanisms for efficacy of exercise. (Simon Gandevia, Jane Butler, Janet Taylor) who opened his eyes to the delights of neuroscience Finally, he thanks his family (Merryn, Freya, and challenged him to test the limits of human Finn and Sofia) who have provided unrelenting eXDerimentation to gain a greater understanding support, despite the fact that Paul has been away often to undertake researd1 and present the outcomes.

SECTION 1 Introduction SECTION CONTENTS 1. The time to move forward 3

3 Chapter 1 The time to move forward Carolyn Richardson CHAPTER CONTENTS INTRODUCTION Introduction 3 Painful musculo-skeletal health problems such as Emphasis on motor control problems as a basis low back pain contribute significantly to morbidity in the general popula tion and form a maj or part of of exercise 3 the high costs of health care in the industrialized Exercise based on impairments in the world. Ironically, back pain is very prevalent in the neurophysiological mechanisms of joint general 'health-focused' population who exercise to protection 6 prevent health problems i.n the cardiovascular sys­ The future 7 tem, and it is also a major problem for those who train and compete at a high level in sports and ath­ letic events. Until recently, the prevention and treatment of insidious-onset mechanical low back pain have relied on the premise that the cause of mechanical low back pilin is a graduill breakdown (i.e. 'wear a.nd teilr') of the j oint structures and associated soft tissues over periods of time. Biomechanicill and ergonomic research has successfully focused on ways of minimizing high forcl�s on the spine and has highlighted to the community till' value of such fa ctors as safe working postures and furniture design in the prevention of low bilck pain. EMPHASIS ON MOTOR CONTROL PROBLEMS AS A BASIS OF EXERCISE The first edition of this book addressed, for the first time, the deep muscles close to the lumbar spine and pel vis, their possible function i.n protecting the joints from injury and their dysfunction in low back pain. Front this new information, a new pMa­ digm of exercise was devised that addressed the motor control problems in the muscles and focused

4 INTRODUCTION on improving the mechanical support of the spinal always been considered important concepts in the joints through specific deep-muscle contraction exer­ development of mechanical low back pain. The cises. This 'segmental stabilization training' tech­ complexity of this mechanism through the bones, nique aimed at relieving the pain that had resulted muscle and soft tissue has attracted the interest of from irritation of pain-sensitive structures, subse­ many researchers. Load transfer has been shldied in quent to tissue injury. This i.nitial breakthrough relation to biomechanical models dealing with the was possible because of astute clinical observation stresses across the bones of the pelvis (Dalstra and of patients with low back pain and through novel research approaches, which allowed the morph­ Huiskes 1995); other biomechanical models predict ology and control of the deep muscles to be inves­ that muscle forces and associated soft tissues (e.g. tigated scientifically for the first time. We also explained the clinical tests that could be used to fascias and ligaments) have a stabilizi.ng effect on assess the problems in the deep muscles in patients load transfer and decrease the stress on the pelvic joints. with low back pain. This new paradigm of deep muscle function and Some models have focused more on the way muscles protect the spine for weightbearing in terms dy sfunction, and the ty pe of exercise required for of their effect on neutral spinal curves in relation to the management of low back pain, was developed the pelviS. The biomechanical models of Keifer with recognition of the close relationship between muscle function and the biomechanics of spinal et al (1997, 1998), which defined spinal stability in stability. The models of the sp.inal stability are overviewed in detail in Chapter 2. terms of the compressive load-bearing capacity of the spine, included pelvic rotation in the model of Widening understanding of joint neutral postures. This model has contributed to our protection mechanisms in relation to understanding of ho\"v a lordotic posture enhances exercise for the prevention and the compressive loading of the spine. In addition, treatment of low back pain it demonstrated that the global muscles are suffi­ cient to stabilize and maintain equilibrium for small An expanded model for therapeutic exercise for sagittal movements, although the addition of the the prevention and treatment of low back pain has local muscles, most importantly multifidus, could evolved through a deeper understanding of the decrease the forces in the global muscles and fur­ joint protection mechanisms from two different ther enhance stability. but essential perspectives. Chapters 2, 3 and 4 dis­ cuss, based on research findings, the role of the It has been the extensive biomechanical model­ local muscle system in the support of the spinal segments, and its role in complex functions where ling and anatomical studies from Erasmus Univer­ the control of the spine must be matched to the demands of internal and external forces. To pro­ sity in the Netherlands that has alerted researchers vide a further basis for exercise to promote the in low back pain to innovative ideas of the role of integration of the local and global muscle systems muscles and associated soft tissues in decreasing into function, another aspect of the joint protection stress on the structures of the lumbopelvic region mechanisms is introduced in this edition. This aspect is related to the role of the antigravity muscle during load transfer. Snijders and colleagues (1995 support system. (review), 1998) have carried out many studies on The antigravity muscle support system the important interaction between gravity, muscle andjoint protection forces, load transfer and the stability of the sacro­ iliac joints. Wcightbearing mechanisms and the way load is specifically transferred through the pelvis have Erasmus University also pioneered the load transfer concept by studying the loading patterns in terms of the effect on the posterior layer of the tho­ racolumbar fascia (Vleeming et al 1995) and the effects of the loading patterns on the ligaments of the pelvis, which are often painful in low back pain (Vleeming et aI1996). These models and anatomical studies highlighted the importance of particular muscles, not only the transverses abdominis and the erector spinae but also the large superficial muscles

The time to move forward that attach to the fascia. Gluteus maximus and More convincing scientific support for the delin­ the contralateral latissimus dorsi are considered eation of skeletal muscle function into these two important in the mechanism of load transfer diag­ onally from arms to legs. These models have pro­ categories has been provided by our involvement vided the impetus for strengthening programmes for patients with low back pain, involving a tnmk in microgravity (space) research. The microgravity environment, with minimal gravitational load cues extension-rotation action (Mooney et al 2001). present but where body movement remains import­ ant, provides the ideal model to test the theories of While all these models for weightbearing func­ the delineation of muscle function into functional tion of muscles are important in our understand­ categories. Results of animal and human research to date strongly support the concept that lack of ing of the biomechanical action of muscles, it is information to the body about gravity differentially our contention that the way in which the central and peripheral neural system links specific affects the antigravity (one-joint) muscles, which muscles from each segment of the kinetic chain to change their patterns of use, and display changes in give an effective antigravity support system forms physiology. Opposite changes occur in the multi­ a basis for optimal exercise for the integration of joint, multifunction muscles, which experience increased use in the microgravity environment; the local and global muscles within the framework this results in a lack of atrophy and indications are that they may even increase their levels of recruit­ of an effective prevention and rehabilitative exer­ ment. This change in muscle physiology occurs cise programme for low back pain. through a process known as neuromuscular pl<'ls­ ticity, where the physiological structure of muscles For an understanding of the antigravity muscle support system, it will be argued that skeletal is determined by the pattern of neural impulses muscles can be classified, from a neurological point of view, into weightbearing and non-weightbearing delivered to it. categories. This delineation of skeletal muscle fw1C­ tion relies on the premise that weightbearing is the This evidence has resulted in the change of ter­ entity that separates muscles into two distinct func­ tional categories. In essence, the minimization of minology for muscles from an anatomjcal to a motor weightbearing (deloading) promotes activity in control perspective, into weightbearing and non­ the non-weightbearing muscles and reduces the weightbearing muscle categories, to allow for a contribution of the weightbearing muscles, while clearer understanding of their role in therapeutic increasing weightbearing promotes activity in the exercise and in the methods used to prevent and weightbearing muscles and reduces the contribu­ treat low back pain. Interestingly, from a 'global' tion of the non-weightbearing muscles. An under­ perspective, the microgravity (deloaded) environ­ standing of this classification is essential both for ment results in problems to astronauts when they prescribing preventative exercise for low back pain return to earth. The impairments in the antigravity and for generating a well-balanced management joint-support mechanisms (i.e. reduced weightbear­ approach. ing muscle function) that develop in microgravity would result in their weightbearing joints being The scientific significance of the delineation of unprotected on their retmn to a gravitational envir­ skeletal muscle function into weightbearing and onment. For this reason, astronauts could be prone non-weightbearing categories came, in part, from to significant musculo-skeletal injuries, especially a study of muscle fW1ction by Richardson and low back pain, on their return to Earth, emanating from injury to the joint structures of the lumbopelvic Bullock (1986), where the effects of gravity and region. It will be argued that a similar process occurs on Earth as a resuit of lack of weightbearing weightbearing, including the gravitational load (deloading) during many functional and recre­ cues, were minimized in a rapid, non-weightbearing ational activities. motor task. This situation of reduced weightbear­ ing and minimal sensory input resulted in reduced The argwnent for an emphaSis on the antigravity, use of the antigravity (one-joint) musculature and weightbearing muscle system i.n the prevention and higher levels of use of the multijoint, multifunc­ rehabilitation strategies for low back pain is devel­ tion muscles, which were facilitated in the non­ oped in four separate chapters withjn this text. weightbearing motor tasks.

6 INTRODUCTION L Chapter 5 provides more detail of how the local Deloadingldecrease in proprioceptive input system operates wi th in a framewor k of j oint pro­ Joint injury tection for weightbearing to provide stiffness and Pain sensory input locally for the lumbopelvic region. �� ii In Chapter 6, the argument will be developed that Development of impairments the local muscles form part of a separate, larger in the joint protection mechanism antigravity muscle system, which links the joints � of the entire functional kinetic chain including both Movement impairment syndromes the upper and lower limbs. Chapter 7 describes the Musculo-skeletal pain syndromes impairments to the joint protective mechanisms Figure 1.1 The continuous cycle of increasing disability in the joint protection system and its relationship to that can develop in the antigravity muscle system impairments in movement. with deloading (i.e. a reduction in weightbearing), musculo-skeletal system. Thus a vicious cycle while Chapter 12 describes the impa irments that results, which eventuilily leads to progressive i:lJ1d can develop in pelvic orientation and weightbearing increasing disability. These impairments in the joint protection mechanisms will eventually res u lt ill function in patients with low back pain. . impaired movement patterns ilnd musculo-skeletal Althou gh the case is put forward for the import­ pain syndromes (Fig. 1.1). ance of the spinal stability mechanisms as well as A recurring theme, wluch must be addressed in the a l1 tigravity muscle support system in the design clinical treatment trials in the future, is that these of therapeutic exercise for the prevention and treat­ three factors (deloading, injury, pain) may all lead to changes in motor control and motor function ment of low back pain, the impairments that occur in these systems also have a significant i nfl uence and that the impairments may vary considerably on exercise design. between individuals with similar low back pain EXERCISE BASED ON IMPAIRMENTS symptoms. Therefore, patients with low back pain IN THE NEUROPHYSIOLOGICAL ret}uirc individ ua lized exerci se ma nagement, based MECHANISMS OF JOINT PROTECTION on individu,ll clinical assessment of the impair­ ments within the three stages of the tr,lining model. The exercise model explained in this text is based The segmenta l stabilization training model not onl y on the joint protective mechanisms but This expanded segmental stabilization training a lso on impairments in the neurophysiological management approach is based on principles of mechanisms of joint protection. New perspectives prevention and trea tment (summarized in Ch. 13) of m.otor control i mpairments are included, not only developed through our increased understanding of for 'dcloading' (Ch. 7) but also for 'injury' (Ch. 8) the complex neurophysiolllgiGll processes involved and 'p'lin' (Ch. 9). These have been added to this edition to provide essential information for improv­ in joint p rotection and the imp airments that arc ing the efficiency of the conservative treatment of involved in the development of painful symptoms. low back pain as well as to provide the basis for The change in perspective that divides muscles new guidelines for the prevt'ntion of chronjc, dis­ into local, weightbearing ,md non-weightbl'aring categories rather than into groups based on motor a bling spinal pain. control categories (or on anatomkal descriptions), and the clOst' reliltionship between local and weight­ These effects of deloading, injury and pain on the bearing categories, has resulted in an expanded protective muscle system have led to new mecha­ n i stic models that explain how the d evelopment of impairments in the joint protecti on mechanism can result in a con tinuous cycle of increasing dis­ ability. It will be argued that these three factors are closely linked with the impairments that develop in the joint protection mecharusms. In turn, a lack of joint protection would lead to further joint injury and pain as well as de10ading of the

The time to move forward 7 view of segmentill stabilization exercise. While local Stage 3: 0PClI chain seglllental control The aim is to continue to develop segmental cont rol at individ­ muscles 'upport the joints (i.e. individual segments) ual joints in relation to open kinetic chain move­ of the s pine and pelvis, the weightbearing (usually ment of adjacent segments (Ch. 16). This final step one-joint) muscles play an important role, together directs progression so that all muscles (i.e. the local, with the l ocal muscles, in linking each segment o f weightbearillg and non-weightbeari.ng) are inte­ grated into functional movement tasks in a formal the kinetic chain t o give a n effective antigravity sup­ way, so that compensations by more active (i.e. port system. Therefore, segmental control is also non-weightbearing) muscles can be detected. required for the large joints of the girdles and limbs within a framework of antigravity weightbearing THE FUTURE control. A more clearly defined model of exercise Clinical screening assessments must be validated management of low back pain has been devised that that reflect the mechanisms of jOint protection and has been expanded to include three levels of pro­ the level of impairment. Evidence must be pro­ vided that an improvement in the joint protection gressive segmental control. mechanisms correlates well with the reduction in Stage �l: local St!gl/lr.'lItlll colltrol The aim is to develop pain and disabiLity. From this, evidence-based treat­ segmental control via activation and training of ments and prevention strategies that are based on the impairments present in an individual with pain the local muscle system (Ch. 14). The deep local can be expected to be developed ill the near future. muscles, which form the most basic dement of fn addition, new non-invasive laboratory-based the joint protection system, are activated and assessment procedures of the antigravity, weight­ exercises are given to enhance kinaesthetic aware­ bearing function of muscles are currently being ness and muscular control. developed at the University of Queensland to help Stllge 2: closed c/zaill scgl/lelltlll cOlltrol The aim is to continue to develop segmental control ilt individ­ to understand the possible causes of low back pain. ual joints through activation and training of the local muscles in conjunction with the antigravity These procedures, which would be suitable for use system that I in ks th.e specific weightbearing mus­ by all health professionals, are now being investi­ cles from each segment of the kinetic chain to give gated as part of the European Space Agency Bedrest an effectiVl' <lIltigravity support system (Ch. 15). Study, which is being undertaken at the Free This step in the functional rehabilitation process UniverSity in Berlin. is devised with a knowledge of the optimal pat­ terns of muscle activation required for weight­ bearing ,md joint support.

SECTION 2 I9 The joint protection mechanisms SECTION CONTENTS Part 1 Introduction 11 2. Lumbopelvic stability: a functional model of the biomechanics and motor control 13 Part 2 Specific joint protection of the spinal segments 29 3. Abdominal mechanism and support of the lumbar spine and pelvis 31 4. Paraspinal mechanism and support of the lumbar spine 59 Part 3 The antigravity muscle support system 75 5. Stiffness of the lumbopelvic region for load transfer 77 6. The role of weightbearing and non­ weightbearing muscles 93

11 PART 1 Introduction PART CONTENTS 2. Lumbopelvic stability: a functional model of the biomechanics and motor control 13

13 Chapter 2 Lumbopelvic stability: a functional model of the biomechanics and motor control Paul Hodges CHAPTER CONTENTS INTRODUCTION Introduction 13 Many contemporary approaches to therapeutic Biomechanical considerations of lumbopelvic exercise for the spine are based on the premise that control and stability 13 the low back pain results from, and is perpetuated Conceptual model of control of lumbopelvic by, repetitive microtrauma to the spinal structures stability 15 resulting from poor control of spinal stability Muscle system considerations for lumbopelvic (Farfan 1975). Although the neurobiology of pain control 16 Motor control mechanisms for lumbopelvic indicates that this simple biomechanical hypoth­ esis is unlikely to explain the complexity of pain, control 20 Functional model of instability 25 there is considerable evidence to validate the bio­ Clinical application 26 mechanical model. One factor that complicates the debate regarding the validity of 'stability exercises' is the complexity of the biomechanics and motor control of stability. A key issue is what authors mean by the term 'stability'. TI1e aim of this chapter is to develop a clinical model of stability that can be used to guide exercise intervention. To develop the model, it is necessary to consider the general requirements for spinal stabihty, the muscle systems that may contribute to this c ontrol and the strate­ gies used by the central nervous system (eNS) to meet the demands of spinal control. BIOMECHANICAL CONSIDERATIONS OF LUMBOPELVIC CONTROL AND STABILITY Lumbopelvic stability is often regarded to be a static principle. For instance, exercise interventions that aim to i mprove stability commonly involve training patients to maintain a static trunk posture during function. However, this is an oversimplified notion of stability. Instead, stability and control should be thought of as a dynamic process of controlling static

14 THE JOINT PROTECTION MECHANISMS position when appropriate in the functional con­ proportion of the mass of the body, t runk move­ text, but allowing the trunk to move with control ment is important for the control of postural eq ui­ in other situations. Any intervention that focuses librium with respect to imposed forces (Oddsson solely on one extreme of this spectrum is unlikely to 1988). If the equilibrium of the body is disturbed lead to an optimal functional outcome. It is impor­ by external (e.g. unexpected movement of the sup­ tant to consider a functional definition of stability port surface) or internal (e.g. by reactive forces and then to consider the elements that may con­ from limb movement) forces, movement of the tribu te to its con trol. trunk occurs to move the centre of mass (COM) over the new base of support or to ellter the orien­ Spinal stability is commonly m odelled in the tation of the body (e.g. Keshner et ell ] 98<,)). It is context of Euler mechanics. In this context, stability important to consider thi::; function of the trunk as is considered in terms of buckling from compres­ the demands for control of equilibrium may con­ sive forces (Crisco and Panjabi 1991, Gardner-Morse flict with the requirements for control of spinal et al 1995, Cholewicki and McGill 1996). In this orientation or intervertebral motion. For instJnce, highly developed model, it is considered that the trunk alignment cannot be maintained if l1'love­ spine is inherently unstable to compressive forces. ment of the trunk is required to move the COM In line with this hypothesis, in vitro studies have indicLlted thLlt coJlapse of the lumbar spine (with over a new base of support (Huang et aI2001). The all of the passive elements rem oved) occurs with next level in the hierarchy of spinJI control is the compressive loading of as little as 90 N. In view of control of orientation of the spine and pelvis. At this, the model argues that antagonistic muscle this level, it is important to consider the control of Jctivity is required to maintain the lumbar spine the curvature and posture of the spine. It is at this in a mechanically stable equilibrium and prevent level that control of buckling is most critiG11. In buckling (Crisco and Panjabi 1991, Gardner-Morse function, <I co m plex array of interna l and external et a11995, Cholewicki and McGill 1996). Although forces, incl udin g gravity, are imposed on the body. it is clear that this model explains a component of For instance, in a simple task such as rapid flexion of the arm, the reactive moments from the move­ lumbopelvic stability and control, it can be criti­ ment generate il flexion moment at the trunk, cized from several perspectives. First, it likens the while at the same time the forward posi tion of the spine to a mJst of J yacht, which must be main­ <lrm moves the COM of the body forward, again tained upright without buckling; clearly, this cannot causing the spine to flex (Houissl't and /'JttMa L'xpl<1in the breadth of human functional rcquire­ ll)H1, Hodges et (111900). If tl1L' goal is to mJintJin mcnts. Second, it does not emphasize the control upright posture, these perturbations must be over­ of movement. For instJJ1ce, as the spinc is moved come. Similar! y, when a vveigh t-Iiftl'r ra ises a mass from flexion to extension, J controlled sequence of from the floor, buckling of the spine must be pn'­ vented by muscle activity (Cholewicki et alllJ91). intervertebral rotation and translation is required At the most basic level of spinal control is the con­ (Bogduk d al 1995). This component of stability trol of intervertebrJI translation and rot1tion; how­ requires a fine-tuned system to coordinate stJbility ever, this G1nnot lx' compktl>ly separated from the Llnd movement. Sever<ll Juthors have considered control of spin,ll orientation ( l'anjJ bi et aI19Sl)). Th,1t is, buckl Lng can occur Jt the intl'rvcrkbralll'vel the requirement for control during movel1lcnt. changes in spinal orientation involve intclvl'rtebral motion. However, separak attention mList be p,1id to pJnjabi (1992a,b) hJS recognized that (lfOll11li the control of trans lJtions and rotations. For instance, nelltrJI position, where the spine exhibits least stiff­ during an Jrc of movement it is importJnt to con­ ness, the requirement for spinal control is increased. trol the coordination between tr,mslation and rot,1- Towards the end of range, incre<lsing support is tion betwecn segml'l1ts (Bogduk ct al 19(5). It h<ls provided by the passive elements. been shown that if the spine is modelled with ,111 An <ldditional considl'ration is th<lt lumbopelvic segments crossed by mLlsciL>, but with one vertebrae stability must be considered at several interdepend­ with no muscle attJchment, then thL' spine is JS ent levels: intervertebral control, control of IUl1l­ bopelvi c orientation and the control of whole-body equilibrium (Hodges and Jull 20(3) (Fig. 2.1). At the most general leveL <IS the trunk forms J IMge

Lumbopelvic stability: a functional model of the biomechanics and motor control 15 (a) (b) (c) Figure 2.1 Lumbopelvic stability at interdependent levels. (a) Control of whole-body equilibrium; (b) control of lumbopelvic orientation; (c) intervertebral control. stable as having no muscle at all (Crisco and Panjabi of control, from the control of 'whole-body equilib­ 1991). Consequently, segmental control is an essen­ rium to the control of intervertebral motion. As yet, no single biomechanical model considers each of tial component for spinal stability, and this control is these elements, and any criteria that aim to optimize particularly relevant in the context of low back pain stability based on these separate models is unlikely to be adequate. In ZI clinicZlI context, it is important ZlS it is the control of this dement th'lt appears to be compromised in this population. to cons ider all elements, although evidl'ncl' sug­ The same principles of control of orientation gests that emphasis should be placl'd on control of intervertebral motion, at least initially. The next Zlnd intl'Ivl'rtebral motion also apply to the pelvis. byer of complexity is to consider how thesl' multi­ At one level, there is the need to control orientation ple d emands of stability may be controlled. of tIll' pelvis dround tl1L' three orthogonal dxes. CONCEPTUAL MODEL OF CONTROL Howevl'r, there is also the requirement to control OF LUMBOPELVIC STABILITY the relationship bl'tvVl'l'n segml'nts of the pelvis. In Panjabi (J 992a,b) introduced Zln innovativl' model upright positions, the sacro-iliZlc joints Zlre subjected of the spinal stabilization sy stem, which Sl'rVt'S as an Zlppropriate model for understanding the mainten­ to considerZlble shear forcl' as the mass of the upper ance of spinal stability, the entity of inst1bility Zlnd the cliniGll paradigm for the assessment and trec1t­ body must be tTZlnsfl'rrl'd to the lower limbs via ment of the muscle dysfunction in thl' pMicnt with the ilia (Snijders et all <;)<;)3, 1995). The body hZls two low bZlck pain. The model incorporates a passive subsystem, an active subsystem Zlnd ,1 neural con­ mechanisms to overcome this; one mechanism trol subsystem (Fig. 2.2). The first considL'fation is thM passive structures of the spine and pelvis depends on the wedge shape of the sacro-iliac joints (,form c1osurl\"), and the other involves com­ pression of the sacrn-iliac joints via muscle con­ traction ('force closure') (Snijders et al 1 <;)93, 1 <;)%). In summary, a composite model is required that considers tIll' spl'ctrum of demands for stability; this will include control of buckling forces and con­ trol d u rin g movement, ,1S 'Nell ,1S the multiple levels

16 THE JOINT PROTECTION MECHANISMS Figure 2.2 The three systems that contribute to lumbopelvic stability. (Adapted from Panjabi 1992a.) �/I \\\\- Passive contribute, to some extent, to the control of aU of must activate muscles at the right time, by the right the elements of stability described in the previous amount, in the correct sequence and then turn section. This passive subsystem incorporates the muscles off appropriately. Based on this mode\\, osseous and articular struchlres and the spinalliga­ Panjabi contended that the three subsystems are ments, all of which contr ibute to the control of interdependent components of the spinal stabiliza­ spinal movement and stability. While being integral tion system with one capable of compens,)ting for components of the spinal stabilization system, the deficits in another (Panjabi 1992a). From the purely passive elements offer most restraint towards the biomechanical perspective, back pain may occur end of the range of movement, but they do not as ,) consequence of deficits in control of the spinal provide substantial support around the neutral segment when abnormally largl' segmental motions position, where the spine exhibits least stiffness cause compression/stretch on neural structures (I'anjabi 1992b). or abnormal deformation of ligaments and pain­ sensitive structures. These deficits may potentially The active subsystem refers to the force­ be caused by a dysfw1ction in any of the three sys­ generating capacity of the muscles themselves, tems, vvruch cannot be com pensated for by the other which provides the mechanical ability to stabilize systems. Instability will be considered in greater the spinal segment. However, the muscle system is detail later in this chapter. only as good as the system that drives it, the control subsystem. This latter system must sense the MUSCLE SYSTEM CONSIDERATIONS requirements of stability and plan strategies to meet those demands. This model recognizes that the FOR LUMBOPELVIC CONTROL neural control subsystem must coordinate muscle activity in advance of predictable challenges to As described by Panjabi (1992b), the active sub­ stability - and coordinate responses to afferent feed­ system, or muscles, provides the mechanism by back from unpredictable chilllenges. The system which the control system mily modulate the

L u m bope l v i c stab i l ity: a fu n ctio n a l m odel of the b i o m echa nics and m otor co n tro l 17 stability of the spine. Consideration of the need to (a) . ..� modulate stability is important . Many biomechan­ ical models argue for optimal stability of the spine. Fig ure 2.3 Muscles of the lumbo pel v i c reg ion: (a) loca l In this context, it may be considered that stability and (b) g l obal. would be optimal if stifhless was maximized and no lumbopelvic movement was allowed. However, the more lateral muscles acted as guy ropes sup­ in reality, the muscle system modulates or changes porting the vertebrae, as they would the mast of a stiffness to match the demands of internal and ship, and were more concerned with bending the external forces. Why do we not just stiffen the neck (i.e. the control of neck orientation). It has spine? The answer lies in the fact that movement is been realized over succeeding years that the way important for optim al spinal health. Movement is in which muscles support and stabilize the spine is required to assist in the dissipation of forces and to far more intricate than this simple model. Never­ minimize energy expenditure. For example, energy theless, it is pertinent to address this issue of local expenditure in gait is increased if pelvic motion is (central) and global (guy ropes) muscles systems reduced (Perry 1992). If it is important to match the in an attempt to understand muscle function in demands for stability, which muscles are involved? relation to the st<1bility of the spine. A large number of muscles cross the spine and may contribute to modulation of lumbopelvic stability. Bergmark (1989) categorized the trunk muscles All of these muscles can contribute to stability to into local and global muscle systems based on som(' extent. Although considerable effort has been architectural properties (Fig. 2.3). The local muscle placed on identification of the muscles that con­ system included deep muscles and the deep por­ tribute the 'most' to stability, this is the wrong ques­ tions of some muscles that have their origin or tion. With consideration of the complexity of insertion on the lumbar vertebrae. These muscles stabiLity described above, it can be seen that no sin­ control the stiffness and intervertebral rebtionship gle muscle could provide the greatest contribution of the spinal segments and the posture of lumbar to all elements of stability. Consider the complexity segments. Although this system of muscles is of muscles required to ensure the dextrous move­ essential for stability, it is not sufficient for stability ments of the hand, which contains multiple layers as the muscles are ineffective for control of spinal of muscle, to fine-tLuIe the control motion of indi­ orientation. The lumbar multifidus muscle, with vidual segments of the fingers. Estimation of which its vertebrae-to-vertebrae attachments (Macintosh muscle(s) provides the greatest force for finger flex­ and Bogduk 1986), is a prime example of a muscle ion hardly l'ncapsulates thl' requirements for fine­ of the local system. The smaller intersegmental tUI1L'd control. Instead, it is important to consider muscles, such as the intertransversarii and inter­ the differential control of the st'parate elements of spinales, may not predominate as mechanical the st<lbility. Several classification systems have stabilizers but have a proprioceptive role instead been developed that ascribe different muscles to the (Bogduk 1997). In the (lbdominal group, Bergmark control of individua.l elements of stability. These a.re (1989) suggested that the posterior fibres of the described in the next section. Muscular control of segmental motion and spinal orientation The first suggestion that some muscles surround­ Lng the spine Me primMily concerned with control of intersegmental motion is ascribed to Leonardo da Vinci (Crisco and Panjabi 19(1). In describing muscles of the neck, he suggested that the more central muscles st,lbilized the spin<11 segment (i.e. provided intersegment<11 control of the neck) while

18 THE JOINT PROTECTION MECHANISMS obliquus internus abdominis, which insert into the Figure 2.4 The coord i na tion between the local and global muscles of the trunk is analogous to the thoracolumbar fascia, form part of the local sys­ coordination of musical instruments in an orchestra. Like the trunk muscles, all instruments contribute to the final tem. However, the significance of this insertion is output, but the contribution of each is specialized and all are needed for optimal function. uncertain, as it is only present in a minority of indi­ muscles provide the optimal control of buckling viduals (Bogduk and Macintosh 1984). The deepest forces (McGill et aI1996), training those muscles is muscle, the transversus abdominis, with its direct unlikely to resolve the deficits in muscle control. attachments to the lumbar vertebrae through the To put this in perspective, it can be useful to con­ thoracolumbar fascia and the decussations with its sider the spine as an orchestra. At one extreme we opposite in the midline, can also be considered a have loud instruments th,lt give volume with east', local muscle of the abdominal muscle group. such as a tuba Wig. 2.4). This is akin to the super­ ficial muscles, which dficicntly provide control of The global muscle systetn encompasses the buckling forces and stiffL'n the spine. At the other large, superficial muscles of the trunk that do not extremc, we havl' the instruments that contribute have direct attachment to the vertebrae aJ'\\d cross to thl' fineT elements of melody, such as a violin or multiple segments. These muscles are the torque flute. This is similar to the contribution of the deep segmental muscles, which provide minimal con­ generators for spinal motion and act like guy ropes tribution to the controillf buckling forcL's but pro­ to control spinal orientation, balance the external vide <In efficient mL'chanism to finL'-tune the control of intervertebral motion and the segments loads applied to the trunk and transfer load from the thorax tn the pelvis (Bergmark 1989). In this of the pelvis. Neither system alone can proviLlL­ optimal spinal control, and both ciemL'nts must be way, the large variations in external loads that occur with nonnal daily ftmctioll are accommo­ coordinated to meet thL' dL'mands for spi.nal health. This dol'S not appear to be the case in Imv back dated by the global muscles so that the resulting load on the lumbar spine and its segments is con­ pain. We will return to this topic in Chapter 10. tinually minimized. Consequently, this system is A final consideration is that local muscle control is critical for lumbopelvic stability but cannot fine­ required over thL' spectrum of functional dl'm,lnds from light tasks such i1S reaching or moving while tune the control of intervertebral motion. Notably, sitting to weight-lifting tasks. This is p<1rticubrly data from a biomechanical in vivo model indicate that, although the large muscles linking the pelvis to the rib cage provided a Significant amOlmt of stiffness to the spinal column, activity of the local muscle system was vital in providing stability of the spinal segments (Cholewicki et al 1'1(7). Even whL'n forces generated by the brge global muscles wen' substantial, the spine was unstable if thert' was no activ ity in the 10ca I muscle system. A small increase in the level of activity of the muscles of the IOG11 system could prevent spinal instability. Muscles that mily be considcfL'd as part of the global sy stem are thl' obliquus intcrnus abdominis, the obliquus extl'rnus abdominis, the rectus abdo­ minis, thL' lateral fibres of the quadratus lumborum and portions of thL' erector spinaL'. Although this system is likely <In oversimplifi­ cation of the complt'x control of spinal stability, it provides ,1 useful model to consid<::'. r clinically. An important consideration is that evidence suggests that it is the local system which is most impaired in low back pain, although both systems ,He necessary to meet thl' demands of spinal stability. ThereforL', while modelling studies may ,lrgue that thl' global

L u m bope l v i c sta bil i ty : a fu n ct i o n a l mode l of the b i o m echa n i cs a n d motor contro l 1 9 nota b le as the req u i rement fo r s t rong globa l mus­ p rogr(lmmes for low back pain that incorpor(l te cle acti v i ty is l i kely to be m in i m a l i Jl l igh t activities, h igh levels of tr u nk muscle cO-<l ctiva tion. yet the muscles of the local system a re needed for sa fe fu nction (It the segmen tili level. Second , global m uscles can o n ly p rovide a non­ specific contri b u t i on to s p i n a l con t ro l . Panjab i et <I I Lim itations of t h e global system (1989) a rg u e d tha t a major a d v a n tage of the m u l t i­ fid u s m uscle was tha t i ts segmental organi za tion I n the previous section, it was argued tha t the p rovid es a n ideal mech an i s m for the nerv o u s sys­ gl oba l muscles cannot con tribute to the con tro l of tem to con trol i n d i v id u a l segments. This is n o t i.n tervertebril l mo ti o n . nt is i s not completely correct as the g l ob (l l s y s te m c a n i n fl u e nce i n te rv e r te b r a l possible with the g lobal muscles. motion as a res u l t of com pressive forces exerted by co-activa tion of a n tilgonist globa l muscl es. While Thi rd, global m u scles have a l im i ted a b il i ty to compression can ilssist in the con tro l of shea r and rotation fo rces, th i s i s <l ssocia ted w i th <I 'cos t ' . contro l shear forces. Th i s h a s been argued b io m e­ cha n ically ( Bogd u k 1 997) <l nd from i n v i v o s t u d ­ First, gl oba l co-activa tion i ncreases the compres­ ies . For ex a m ple, i f shea r forces a re i m posed on the s i v e l oa d on l u m b(l r segm e n ts (Ga rd ne r-Morse a nd spine, there is no cha nge i n acti v i ty of the globa l Stokes 1 998 ) . The s u perfi c i a l tru n k m u sc l es gener­ muscles, implying tha t d eeper local muscles m ust ate torq ue a t the tru nk. Th i s torq ue must be over­ con trol t h i s elemen t ( Raschke ,md Ch<l ffn I 99 6 ) . come by a ntagonist activa tion in order to keep the sp in e upright, and this cO-<lctiva tion res u l ts in a A si mi lar situation may exist in the sacro-iliac compressiv e load on the spi ne (Lavender et a l joints. As men tioned above, sta b i l i ty o f the sacro­ i liac j o i n ts i s d ep e nd ent on thei r compression 1 992, Mirka (l n d M a r r<l s \"1';)93, Thelen e t a l 1 995, (Snij ders e t <1 1 1 99 5 ) . Altho ugh i t i s a rgued thil t this G a rd ner-M orse cllld Sto kes 1 998) . Excessi ve com­ compressi o n force i s , to a l a rge e xtent, p rovid ed b y pressi on, w h ic h res u l ts i n increased in trildiscal the large global m u scles working in disc rete syn­ p ressu re a nd loading through the posterior elem­ e n ts o f the s p i ne, has long been cons i d e red to be ergies (e.g. the contraction of glu teu s mClximus w i th a risk fac tor for spi.n a l degl:neration and pain ( Nachem son and Morris 1 964) . If greater dema n d the d i a gon<l lly o pposed l a ti s s i m u s dors i ) , this i s i s pi <Iced o n t h e s u perfi c i a l m usc l e system, th e likely to be i neffec t i v e in lig h t tasks i n which these IOil d i n g mily be i.nc re(lsed . W h i le i ncreilsed muscles are rela ti ve l y inacti ve . Instea d, h o r i zonta l co-con tra c tion i s expected d u ri n g l i fting <l c ti v i t i es forces p rod uced by the local abd o m ina l muscles and with i nc rea sed tru n k a c c e l e ra tion ( M a rra s a n d M i rkil 1 99 0 ) a n d u n p redic tability ( v a n D i een a n d (e. g . transv ersus abd o m i n.is) w i l l compress a n d sta­ de Looze 1999), increased co-contraction of the bilize the sacro-iliac joints (Snijd ers et al 1 995) . globa l m u scles h(l s bee n d e tec ted i n pa t i en ts who devel op l o w back p<lin compared w i th norma l Fourth, a n tagonist gl oba l m uscle co-a ctivation p<l l n-free subjec ts (Ra d ebo l d et a l 2000) . Excessive globa l m u scle co-contraction d u ri ng light func­ res u l ts i n (l restriction o f sp i n <l l mo t i on, tha t i s i n ti on<l l ta sks m <l y even be i n d i c a tive o f i n appro p r i ­ ' rigid ity ' o f the spin e . I t is k nown tha t I n hea l thy <l te t ru n k m u s c l e c o n t r o l in patients ,v i th b a c k p a i n subjects the CNS uses movemen t ra ther than simple (Rad ebold et a l 2000 ) . These clinical findings sup­ stiffening of the spine to overcome cha l lenges to s ta­ port the hypothesi.s of Chole\" vicki e t al ( 1 997), who bility ( Hodges et al 1 999, 2000a ) (Fig. 2 . 5) and red uce studied the stabi li zing fu nction of the trunk flexors ene rgy expend i ture (Perry 1 992) . A s tra tegy of h\"u n k and ex tensors a ro u n d a neu tra l spine posture. Their hypo thesis \"va s that <I d ysfunction i n the stiffenj ng, al though req u i ring less-complex neu.ra l pass ive s ta b i l izing system may be i ndica ted by control, m<ly com promi se opti ma l spi nal functi on. increased levels of trunk muscle co-activa tion. This hy po th eSiS c h a l l e nges m a n y c ur ren t exercise Fina l ly, trun k m u scles a re i nvo l v ed i n fu n c t i on s o t h e r thil n spina l con tro l a nd movement ( Hodges and Candevi<l 2000a ) . A s th e s u pe r fici (l l a b d o m ­ i n a l m u s c l e s depress t h e r i b cage a n d a re inv o l v e d in forced exp i ra t ion ( DeTroyer a n d Esten ne 1 988), increased ac tiv i ty o f these muscles in indiv i d u a ls w i th pa i n may le<ld to com p rom ised respi r a tory fu n c tion , fo r exa m p l e res tricted movem ent of the chest w<l l I . In c on tra st, loca l m u sc l t:s h a v e l i m i ted effec t on rib cage motion ( DeTroyer and Fs tenne 1 988) . Therefore, re l i a nce on g lobal mu cles fo r

2 0 T H E JOINT PROTECT I ON MEC H ANISMS (a) (b) (e) 0 Sh 20°I Tl�2 S I 2°T-L loP S1 S1 ;U SopL3 L L-P Sop T- L P H-L L 1 20000...000011 H-L ] COPAP m ]] COMAP 1m m COMv lAP mmHg F i g u re 2 . 5 Post u ra l responses u se movement rather than simply ma k ing the spine rig i d . The placemen t o f m ark ers t o measure t runk motion a nd the a n g les that a re m e a sured a re shown in ( a ) a n d (bl . respectively. (c) The onset o f a rm moveme n t is show n w ith a solid vertical li n e. The d a ta i n d ica te t h a t when the a rm is flexed ra p i dly at the shoulder (dow n wa rd motion in (el l , the spine moves in the opposite d i rection initi a lly and this spina l motion s ta r ts befo re the on set of movement. A P, anteroposterior; C , cerv ica l ; COM, centre of m a s s ; C O P, centre of p ressu re ; H - L, hip-lumb ar an gle (a n gle between the thigh a n d lumba r s p i nel ; L, lumba r; PS I S , posterior superior il i a c s p ine ; S, sacral; S h , sho u lder ; T, thora c i c . con trol m<l)' be p robl e m a t i c f r o m a system ic p o i n t systems, compare these req u i rements a ga i n s t a n of v i ew. In co n tra s t, l o c a l musc les a l low contro l l e d spinal m otion and have the ability to control indi­ ' i n ternal model o f b o d y d yna m ics' a n d tht'll gener­ v i d u al segments, w i th m inimal e ffec t o n the r ib a te a coord i n a ted res ponse of tht' trunk musclt's so cage, t h u s m i n i m i z i ng c o n flict w i th respi ration. th at the muscle activity occurs at t he righ t t i m e , a t M OTO R CO NTRO L M ECHAN I S M S the right a moun t a n d s o o n . To com p l ica te this i ssue F O R L U M B O P E LV I C C O N T R O L f u r ther, muscle activity m u s t be coord ina ted to The ch a l le n ge i s i m mense fo r the C N S to move <l n d con tro l t h e spine, d esp i te con s ta n t cha n ges in i nter­ ma intain control of the spine w i thi n the h i e ra rchy n al and external fo rces . The CNS must con tinua ll y of i n terdependent levels: con trol o f i n tL'rvertebral i nterpret t h e s ta tu s o f sta bi l i ty, p la n mech-an isms transla tion and rota tion, control o f s p i na l pos t u re / to overcome predictable challenges and rapidly o r i en ta tion, <lnd co n t rol o f b o d y w i th res pec t t o th e i n i tiate ac tiv i ty i n response to unexpec ted chal­ env iron men t. N o tabl y, under the e ffec ts of gravi ty, lenges . I t m u s t i n terpret t h e a ffere n t in p u t from the the CNS m u s t in tegra tl' the c on t ro l of e x ternal peri p h e r a l mech a n orecepto rs, a n d o th e r sensory fo rces fo r we ightbea r i ng and con t rol o f COM . The specific c h arac teri stics o f th a t contro l of a re consid ­ e red in more deta i l i n C ha p ters 5 a nd 6. In a d d i tion, unlike the muscles of the l imb, trunk muscks per­ form a variety of homeosta ti c fu n c t i o n s as w e l l a s

L u m b o p e l v i c sta b i l i ty : a fu n c t i o n a l m o d e l of t h e b i o m e c h a n i cs a n d motor c o n tro l 2 1 Onset Onset Onset Onset TrA d e l t o i d TrA deltoid D e l t o i d �___--�-' Flexion Abduction 1 00 ms Extension Fi g u re 2.6 El ectro myog ra p h ic activi ty of the a bdo m i n a l (rectus a b do m i n i s (RA) , ob l i q u u s exte r n u s a b do m i n is (OE) , o b l i q u u s i n te rn u s a bd o m i n i s ( 0 1 ) a n d tra nsve rs u s a b d o m i n i s (TrA)) , s u p erfi c i a l m u l ti fi d u s ( M F) a n d d e l to i d m u sc l e s fo r s h o u l d e r fl ex i o n , a bd u c t i o n a n d exte n s i o n i n a r e p rese n t a t i ve s u bject. T h e ti m e of a l i g n me n t of t h e traces a t t h e o n se t of el ectro myog ra p h i c activity of the d e l toid is noted, a n d the o nset of a ctivi ty of the TrA is shown by the d ashed l i n e . N ote the o n set o f a c t i v i ty of th e TrA p ri o r to th a t of th e d e l to i d and t h e o t h e r t r u n k m u sc l es , and t h e c o n s i s t e n t period b e t w e e n t h e o n set of a c t i vi ty o f t h e TrA a n d d e l t o i d . A l s o n o te t h e c h a n g e i n seq u e n ce of a c t i v i ty o n set of the R A , E O , 1 0, a n d M F a s a fu n c t i o n of l i m b - m ove m e n t d i rec t i o n . (R e p rod u ced w i t h p e r m i s s i o n fro m H o d ges a n d R i ch a rd so n 1 9 97b, p. 3 64.) move men t a nd control o f the tru n k, i n c l u d ing res­ over a l i fetime of moveme n t experience a nd hold­ piration a nd con tinence . This section will consider i ng i n fo r m a tion o f the i n teraction between inter­ the s tra tegies Llsed b y the CNS to u n dertake th is nal and ex terna l force (Cahery a nd Mass ion 1 981, con t ro l . G u r fi n keI 1 9 94 ) . An i m p o r ta n t fea t u re of th is feed­ Feedforwa rd co n tro l o f l u m b o p e l v i c forward con tro l o f the spine is tha t it p ro v i des sta b i l i ty insigh t i n to the d i fferential s tra teg ies used by the L u m b opelv i c s tab i l ity is con tro l l e d in a d v a n c e of i mposed forces ( i . e . feed forw ard) when the CNS to control each of the elements of stability perturbation to the tru n k is p redictable . Fo r i n s ta n ce, a c tivi ty of the t r u nk m uscles occurs in a n d h ow these may be i n tegra ted . C on s i s te n t w i t h a d v a nce of the m u s c l e respon s i b l e fo r m o v ement of the lower (Hodges a nd Ric ha rdson 1 997a) and t h e a rchi tec tu r a l p rope rties o f t h e t ru n k muscles upper (Belen 'kii et al 1 967, Bouisset and Zatta ra 1981, Aruin a nd Latash 1 995, Hodges and described a bove, the temporal and s p a t i a l p a r a m­ Richa rdson 1 997b) limbs and prior to loading when eters of a c tivity of the s u p e r fici a l or glob a l tru n k a m a s s is a d d ed to the t ru n k i n a p red icta b l e m a n ­ ner (Cresswell et a l 1 994) ( Fig. 2.6) . I n this type of muscles i s lin ked t o the d i rec tion o f forces a c tin g t a s k , the C N S pred i c ts the effec t tha t th is move­ men t w i l l have on the body a nd p la ns a sequence on the spine ( i . e . superfi ci a l trun k muscle ac tiv ity of muscle activ i ty to overcome this pertu rba tion. is earlier and of larger a m p l i tude when the i r activ­ This p rediction in v o lves a n ' i n ternal system of body dynamics', which is a n abstract construct built up ity opposes the d i rec t i on of rea c t i v e forces ) , w h i c h is consistent \\v i th tbe control o f o r i e n t a t i o n o f t h e spi ne ( A ruin a nd La tash 1 995, H odges a nd Ri cha rd son 1 997c, Hodges e t a l 1 999) . In associ­ a tion with l i mb movements, this activity has been show n also to be cons i s tent w i th the control of the d is t ur b a nce to eq u i l ib ri u m a nd to move the COM in a manner consistent w i th the maintenance of upright stance (Aru i n and Latash 1 995, Hodges e t a I 1 9(9) . In con tra st, acti v i ty o f the deep i n trinsic

2 2 T H E J O INT P R OT E CTIO N M E C H AN I S M S L B e fo r e Rapid arm movement Fi g u re 2.7 Activity of the tru n k S h o u l d e r _________ ]ifWIVV\\� 1 5\" Flex m uscles measured with movement 1 5' Ext e l ectromyogra phy (EMG) d u r i n g ]F l e x o i d EMG re petitive arm movement. Standing --- � -t tt+-»+i-ittk#tt 1 mV s u bj e c t s ra p i d l y a n d r e p e t i t i v e l y moved th ei r arm. Activi ty of Extensor /\"Hfi t+H-t++W -t + h+H-� ] 1 mV tra nsversus a b d o m i n i s (TrA) a n d t h e EMG d i a phra g m occu rred ton ica l ly ]ES EMG --- �·�-H+<t-t-H4-*H+t+l-H- \" 500 I-l-V th ro u g h ou t the movemen t, w h i l e a c t i v i ty o f t h e e recto r s p i n a e ( ES) Rib cage I� ' n s p m uscles occurred phasical ly with move m e n t e a c h m ov e m e n t o f t h e a r m . Pya , �.�'!\".�'tjljll II. ]���DiaPh i n t ra - a bd o m i n a l p res s u re ; Pd ; , ---M���'h''.r\"�__ •• 400 I-l-V tra nsd i a p h ra g m p ress u re. (Ada p ted fro m Hodges a n d Ga ndevia 2000a.) ]TrA ( R ) EMG _________ �''-'-'r'1\"'l'j.',(k 4 0 0 I-l-V ��.r'\" ] 20 cm H20 Pga --- v.� ] 20 cm H20 '--' '--' 1s 1s m u sc l es (both t ransversus abdomin i s and mu l t i ­ Repetitive l i m b movements may also prov ide fi dus) is ind ependen t of the d i rec tion o f rea ctive an e X Cl m p i e o f open loop con tro L H o w e v e r, as the forces ( H o d ges a nd Richa rd son 1 9 97c, Mosel ey movement is ongoing, it is not possi ble to exclude the contrib u tion of a ffere n t i np u t t o t h e o rga n i za­ et al 200 2 ) . T h i s is c o n s i s ten t w i th the a rchi tec t u ra l tion o f the tru n k m u sc l e a c ti v i ty, a n d stud ies have p ro pe rt ies of t h ese m u scles to p rov i d e a genera l su ggested that sp i na l mecha nisms d epende n t on i n c rease i n i n terv e r tebra l con troL There fore, the a ffere n t feed back m a y be i m po r ta n t for t h i s con tro l d a t a s u ggest th a t the eNS u s e s fee d fonv a rd non­ (Zed ka and Prochazka 1 9(7) . A l though the mechan­ i s m for co n trol o f repe t i t i ve movement is n o t com­ d i rec t i on spec i fic ac tiv i ty o f the i n tr i n s i c local mus­ pldL' l y u n d erstood , t h e rL' i s ev i d enCL' o f d i f fe re n t i a l c les to con tro l in tervertebral motion, and it u ses a c ti v i ty o f the d ccp a n d s u pe r f i c i i1 ] m u s c les t h ,l t i s con s i s tent w i th the di ffe re n t roles Df these muscles . t LUlL'd d irL'ction-specific responses of the s u perfi c i a l For in s tance, tunic ac t i v i ty of the i n tr i n s ic s p ir\\il l global mu scles t o con trol spinal orien ti1 tion (Hodges m u sc l es ucc u rs in Cl ssoci a t i o n w i th rep etitive u p per L't al 1 999) . Recen t d i1 ta suggest thil t the eNS uses l i m b movcmen t ( transversus abdom i n i s (Hodges d i sc re te s t ra teg ies to con trol each fac tor. W hen the a n d C i m d evi,l 2000a ), m u l ti fi d u s ( M oscley t't a l p repar<l t i o n for movem e n t i s ma n i p u lCl ted or su b­ 2(02); F i g . 2 . 7), repL' ti t i v e ]own l i m b m o v e lll c n t jects perfo rm an a t te n tion-dem i1 n d ing task, the d ur ing ga i t ( H odges ,m d S<lLlI1 d e rs 2UO J ) a n d re pet­ la tency for limb movement an d the pos t u ra l ac tiv i ty itive trunk m ovement (CfL'sswL'i l et i11 1 9Y2a ) . I n of the su perfic ial m uscles is d el ayed, b u t there is no con trCl s t, s u �X'rfi c i a l m u sc l e il C t i v ity occ u rs in a cha n gL� in the l a tency o f the deep m u s c l e rL'sponsc : phasic m anner lin kl'd tu t h e d i rt'ction uf l i m b IllllVt'­ men t ( H odges and Gandev i ,l 20ll0Cl ) . tra nsversus abdo ll1inis (Hodges and Ri c h a rd son Feed b a c k co n tro l of l u m b o p e l v i c sta b i l i ty 1 99<)b) and deep fi bres o f m u l tifidus (Moseley ct a l W h e n t h e sp i ne is perturbed u n p red ictil b l y, the 2Um). Th i s suggests that the deep muscle response n e rvo us system m u s t rcsp u n d ra p i d l y. At the mo re b a s i c end of the spec tru m , fec d b a c k-l1w diil te d is m o re ru d i m e n tiuy i1 n d may be control l ed by i1 co n tro l m a y operCl te il t a r d l e x l e v e l . These !lIme bClsic mec h a n i sm by the e N S . I m porta n t l y, these responses h,l \\IC' been shown to be I in ked to the s peed of limb movement (H odges a n d Ri c h i1 rdson J 997c) and t he mClSS o f the l imb (Hod gl's a n d R i c h a rdson I 997b,c ), sugges ting tha t tht' eNS p re­ d i cts the a m p l i tude of the reactivC' forces a nd adj usts the fCL'd forwa rd responses a ccord i n g l y.