UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Manual therapy and the osteoporotic spine Sran, Meena Manindra 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2005-104430.pdf [ 15.14MB ]
Metadata
JSON: 831-1.0092276.json
JSON-LD: 831-1.0092276-ld.json
RDF/XML (Pretty): 831-1.0092276-rdf.xml
RDF/JSON: 831-1.0092276-rdf.json
Turtle: 831-1.0092276-turtle.txt
N-Triples: 831-1.0092276-rdf-ntriples.txt
Original Record: 831-1.0092276-source.json
Full Text
831-1.0092276-fulltext.txt
Citation
831-1.0092276.ris

Full Text

MANUAL THERAPY AND T H E OSTEOPOROTIC SPINE by MEENA MANINDRA SRAN B.Sc. (PT), The University of British Columbia, 1995 M.Phty.St. (Manip. Physio), The University of Queensland, 1998 A thesis submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Experimental Medicine)  THE UNIVERSITY OF BRITISH C O L U M B I A  March 2005 © Meena Manindra Sran, 2005  ABSTRACT Among older adults, osteoporosis and back pain are common alone, and also in combination. These conditions can cause enormous personal suffering and societal burden. A mainstay of physiotherapy treatment of back pain is the category of hands-on treatments known as 'manual therapy'. These are routinely used to assess and treat back pain in various clinical settings but their safety, efficacy, and mechanism of action has not previously been studied in individuals with back pain and osteoporosis.  The objectives of this thesis were to investigate, (1) the evidence for the effectiveness of manual therapy for spinal pain, (2) physiotherapists' perceptions and practice behaviors with respect to the use of manual therapy on individuals with osteoporosis, (2) the safety of posteroanterior (PA) spinal mobilization in the osteoporotic spine, (3) detection and determinants of spinal fracture under a PA load, and (4) whether P A stiffness can predict intervertebral range of motion (ROM) and flexibility in the cadaveric midthoracic spine of older adults.  To achieve these objectives I used a variety of research methods that included; a systematic review, a survey (171 physiotherapists), biomechanical testing, plain radiography, computed tomography (CT), dual energy X-ray absorptiometry, ash weight, and micro-computed tomography (12 cadaveric spine segments, one intact cadaver, 2 physiotherapists, 7 participants).  The systematic review indicated that: (i) physiotherapy that included manual therapy at a dose of 30-45 minutes per session, for 4-8 weeks was effective in adult populations with back or neck pain, and (ii) clinically relevant differences between the manual therapy interventions used in clinical trials may influence the outcomes. M y survey of physiotherapists found that 91% of respondents were concerned about fracture as a complication of treatment when using manual therapy in patients with osteoporosis.  Among the key findings in my study of vertebral biomechanics using cadaveric midthoracic spine segments was that vertebral body injury is an unlikely complication of PA mobilization in the midthoracic spine. Simulated P A mobilization using a mechanical testing machine produced spinous process fractures in every case and no vertebral body fractures. There was a reasonable margin between the failure load, in vitro, and the applied mobilization load, in vivo, for most specimens, however the lowest fracture thresholds (200N) approached the same force as the upper range of the applied loads (223N). Bone mineral density (BMD) of the whole vertebra was not a good predictor of P A failure load (r=0.18, p=0.68) but micro-CT measures of regional bone volume fraction of the spinous process base and middle regions, the sites of fracture, were strongly correlated with P A failure load (base: r=0.74, p=0.01; mid: r=0.73, p=0.01). Plain radiography and CT had poor sensitivity for these spinous process fractures (detected 3/12 and 6/12 fractures, respectively).  D X A scanning was an appropriate surrogate measure for thoracic spine segment bone mineral measurement. Volumetric B M D can be accurately estimated using the elliptical  cylinder method. Trabecular thickness differs significantly between the spinous process and lamina regions, and may have influenced the site of fracture (p=0.003).  Cadaveric studies of spine kinematics during PA mobilization showed that the mobilized thoracic vertebra moves into extension as a result of the mobilization. Further, in cadaveric midthoracic spine segments from older adults, P A stiffness is inversely correlated with R O M and flexibility at the level at which the P A mobilization is applied (r= -0.78 - -0.90).  T A B L E OF CONTENTS ABSTRACT T A B L E OF CONTENTS LIST OF TABLES LIST OF FIGURES PREFACE ACKNOWLEDGEMENTS DEFINITIONS LIST OF ABBREVIATIONS CHAPTER ONE 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 CHAPTER TWO  2.1 2.2 2.3 2.4 2.5 CHAPTER THREE  3.1 3.2 3.3 3.4 3.5 CHAPTER FOUR  4.1 4.2 4.3  ii v viii ix xiii xvi xviii xxviii LITERATURE REVIEW AND INTRODUCTION TO THE THESIS Osteoporosis Vertebral Fractures Back Pain Manual Therapy Spinal Fractures: Detection and Determinants Thoracic Spine Biomechanics Summary Aims, Objectives & Scope of the Thesis References  1 2 5 8 11 19 23 28 31 34  EVIDENCE FOR THE EFFECTIVENESS OF MANUAL THERAPY FOR SPINAL PAIN— A SYSTEMATIC REVIEW Introduction Methods Results Discussion References  59  62 64 68 71 77  PHYSIOTHERAPY AND OSTEOPOROSIS: PRACTICE BEHAVIORS AND CLINICIANS' PERCEPTIONS— A SURVEY Introduction Methods Results Discussion References  79  81 83 85 88 92  FAILURE CHARACTERISTICS OF THE THORACIC SPINE WITH A POSTEROANTERIOR LOAD: INVESTIGATING THE SAFETY OF SPINAL MOBILIZATION Introduction Methods Results  98  100 102 112  v  4.4 4.5  Discussion References  118 123  CHAPTER FIVE  ACCURACY OF DXA SCANNING OF T H E THORACIC SPINE: CADAVERIC STUDIES COMPARING BMC, A R E A L BMD AND GEOMETRIC ESTIMATES OF VOLUMETRIC BMD AGAINST ASH WEIGHT AND CT MEASURES OF BONE V O L U M E Introduction Methods Results Discussion References  129  REGIONAL TRABECULAR BONE MORPHOLOGY IS CORRELATED WITH FAILURE OF THORACIC VERTEBRAE UNDER A POSTEROANTERIOR LOAD Introduction Methods Results Discussion References  156  POSTERO ANTERIOR STIFFNESS AS A PREDICTOR OF INTERVERTEBRAL MOTION IN CADAVERIC THORACIC SPINE SEGMENTS Introduction Methods Results Discussion References  185  5.1 5.2 5.3 5.4 5.5 CHAPTER SIX  6.1 6.2 6.3 6.4 6.5 CHAPTER SEVEN  7.1 7.2 7.3 7.4 7.5 CHAPTER EIGHT 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9  GENERAL DISCUSSION, SUMMARY AND CONCLUSIONS Evidence for the Effectiveness of Manual Therapy for Spinal Pain Physiotherapy and Osteoporosis: Clinicians' Perceptions and Practice Behaviors Safety of Posteroanterior Spinal Mobilization Accuracy of D X A Scanning of the Thoracic Spine Regional Vertebral Trabecular Bone Morphology Posteroanterior Stiffness as a Predictor of Intervertebral Motion Summary and Conclusions Contributions References  132 135 141 149 152  159 162 170 179 182  187 189 198 201 205 209 210 212 214 218 219 220 222 225 228  APPENDIX I APPENDIX II APPENDIX III  Questionnaire and Cover Letter Tekscan Calibration with a Six Axis Load Cell Thermolyne Muffle Furnace: temperature versus time  LIST OF TABLES 65  Table 2.1. Search strategy. Table 2.2. Summary of studies reviewed. MT= manual therapy, LBP= low back pain, PT=physiotherapy, GP= general practitioner  66-67  Table 4.1. Magnitude (°) of rotation [flexion (+) and extension (-)] at T5-6, T6-7, and T7-8 in T5-8 cadaveric spine segments and an intact cadaver during PA mobilization.  117  Table 5.1. Mean, standard deviation (SD), and range for primary outcome variables at the T6 vertebra [AP and lateral B M C (g), A P and lateral aBMD (g/cm ), ash weight (g), v B M D - C T (g/cm ), estimated (est) v B M D using cylinder, cube and elliptical cylinder methods (g/cm )]  143  Table 5.2. Pearson correlation coefficients for primary outcome variables (ash weight (g) with A P and lateral B M C (g), and AP and lateral aBMD (g/cm ); ash weight (g) and v B M D - C T (g/cm ) with estimated (est) v B M D (g/cm ) as a cube, cylinder, and elliptical cylinder * = significant at p< 0.05  144  Table 6.1. Range, mean and standard deviation for trabecular B V / T V , Mean Tb.Th (mm), Mean Tb.N (per mm) and Mean Tb.Sp (mm) for each of the four trabecular regions analysed.  177  Table 6.2. Pearson correlations for the metric T7 trabecular bone measures (BV/TV, mean Tb.Th, mean Tb.N, mean Tb.Sp) in each region (Base= spinous process base; Mid= spinous process middle; Lamina= central lamina; Body= vertebral body centrum) and T6 P A failure Load (N), in cadaveric vertebrae. *significant at p< 0.05  178  Table 7.1. Mean, range and standard deviation range of motion (Deg) in flexion and extension at T5-T6, T6-T7 and T7-T8.  199  Table 7.2. Mean, range and standard deviation (average of T5-6, T6-7 and T78) flexibility (Deg/Nm) for extension, flexion, rotation and lateral bending in T5-T8 cadaveric spine segments. Rotation and lateral bending include motion to the left and right.  199  Table 7.3. Pearson correlation coefficients for T5-T6, T6-T7 and T7-T8 extension and flexion range of motion (Deg), and flexibility (Deg/Nm) in three dimensions [extension, flexion, right and left axial rotation (Rotn.), right and left lateral bending (Lat.Bend)] with posteroanterior (PA) stiffness (kN/Deg). : n = 7 * = significant at p<0.05  200  2  3  3  2  3  3  A  viii  LIST OF FIGURES Figure 1.1. Typical vertebral body compression fracture in the midthoracic spine.  ^  Figure 1.2. Overview of manual therapy techniques. This thesis focuses on joint techniques. PPIVMs= passive physiological intervertebral movements; PAIVMs= passive accessory intervertebral movements  ^  Figure 1.3. Grades of mobilization in a normal range with a hard end-feel.  ^  Figure 1.4. Photograph of a physiotherapist applying PA mobilization to T6. The arrow depicts the direction of the applied load.  18  Figure 1.5. Diagram of hand contact during P A mobilization in the midthoracic spine.  18  Figure 1.6. Load-deformation relationship. The first decrease in load-bearing capacity is termed the failure load.  24  Figure 1.7. Right-handed orthogonal (90° angle) coordinate system. Human motion is typically a combination of translation along any direction and rotation about any axis in space. Figure 3.1. Percentage of respondents who chose each of the treatment modes listed in Question 1: "Which of the following treatment modes would you likely use with an individual with osteoporosis?" Strength= strength exercises; Balance= balance exercises; Flexibility flexibility exercises; Man Ther= manual therapy; Electrother= electrotherapy; US= ultrasound; Laser= laser; Posture= posture reeducation; Ergonomics= ergonomic advice Figure 3.2. Percentage of respondents reporting concerns related to each of the tissues/structures listed in Question 3. vert # = vertebral fracture; other # = other fracture; ligt- ligament injury; tendon= tendon injury; disc = disc injury; muscle= muscle injury Figure 4.1. Lateral view of a thoracic vertebra with typical measures of vertebral body area, spinous process length and spinous process width. (s= intersection of line from anterior inferior vertebral body along the inferior endplate with the anterior border of the spinous process; w= spinous process width; 1= spinous process length; a= anterior height; p= posterior height; h= length of the perpendicular line between the two parallel lines, a and p) Figure 4.2A. Schematic of posteroanterior (PA) loading at the T6 spinous process in T5-T8 cadaveric spine segments. The opto-electronic marker carriers are shown attached to each vertebra.  2g  86  87  104  106  Figure 4.2B. A typical T5-T8 cadaveric specimen secured in the custom testing jig. The opto-electronic marker carriers, base marker fixed to the testing machine (lower left), and the delrin indenter with foam padding (above the specimen) are also shown.  107  Figure 4.3. Marker set up for the intact cadaver test and the Tekscan I-Scan sensor for measuring applied load.  109  Figure 4.4. The physiotherapist applying PA mobilization to T6 of an intact cadaver.  109  Figure 4.5. Load (N) and displacement (mm) for a typical specimen during the failure test, (in vitro cadaveric spine segment) Failure point was defined as the first decrease in load.  ^ ^  Figure 4.6A. Post test radiography detected the site of fracture in only 3 of 12 cases. This radiograph shows a fracture at the base of the T6 spinous process (see arrow).  ^ ^  Figure 4.6B. Post test CT scan detected 6 of the 12 fractures. This is a photograph of the spinous process in an axial slice. The neuroradiologist interpreted the step in the cortical bone (see arrow) as a fracture.  ^ ^  Figure 4.7. Graphical representation of the in vivo applied load data. Participants are listed on the horizontal axis and load in N on the vertical axis. Load ranged from 106-223N, with a mean of 145N. Similar loads were applied in young and old participants. Figure 4.8. A sample of the force data collected with Tekscan during the intact cadaver test. In this example the peak load is approximately 115N and it was reached after almost 8 seconds.  115  115  Figure 4.9. Pearson correlation coefficient for P A failure load (N) and A P bone mineral density (g/cm ): r=0.18.  116  Figure 5.1. Crucible and cover measurement with the Sartorius BP 210 S scale.  138  Figure 5.2. Bone mineral content (g) data: lateral D X A and ash weight for all specimens.  141  Figure 5.3. Limits of Agreement: difference in B M C (g) (difference of lateral D X A B M C and ash weight) vs. average B M C (g) (sum of lateral D X A and ash weight divided by 2).  145  Figure 5.4A. Limits of Agreement: difference in v B M D (g/cm ) (difference of elliptical cylinder v B M D and ash weight/CT vBMD) vs. mean v B M D (g/cm ) (sum of elliptical cylinder and ash/CT v B M D divided by 2). 3  Figure 5.4B. Limits of Agreement: difference in v B M D (g/cm ) (difference of cube estimated v B M D and ash weight/CT vBMD) vs. mean v B M D (g/cm ) (sum of cube and ash/CT v B M D divided by 2).  146  3  Figure 5.4C. Limits of Agreement: difference in v B M D (g/cm ) (difference of cylinder estimated v B M D and ash/CT vBMD) vs. mean v B M D (g/cm ) (sum of cylinder and ash/CT v B M D divided by 2).  147  3  3  Figure 6.1 A . This photograph shows the custom jig, coring bit and drill used to obtain central vertebral body cores from each T7 vertebral body. The vertebral body was held in position by four screws. The platform above the vertebra was adjusted both in tilt and position, to place the drill guide/hole directly over the center of the vertebral body.  ^g  163  Figure 6.1B. Diagram of method of obtaining a central vertebral body core, with the posterior surface in the superior position.  164  Figure 6.2. Low density foam (left) was used to secure the sample in the custom holder (right) during pCT scanning.  164  Figure 6.3. Three dimensional image of the spinous process base, of one sample. Both the trabecular and cortical regions are visualized.  167  Figure 6.4. Sample of a typical cortical thickness measurement at the posterior compartment of the spinous process, at the midline. Cortical thickness was measured as the length of the red line (2.67mm).  168  Figure 6.5A. Pearson correlation for bone volume/total volume (BV/TV) of the base of the T7 spinous process and posteroanterior (PA) failure load (N) of T6 from eleven cadaveric spines. r=0.74, p=0.01  172  Figure 6.5B. Pearson correlation for mean Tb.N (per mm) at the base of the T7 spinous process and posteroanterior (PA) failure load (N) of T6 from eleven cadaveric specimens. r=0.64, p=0.03  173  Figure 6.5C. Pearson correlation for mean Tb.Th (mm) at the base of the T7 spinous process and posteroanterior (PA) failure load (N) of T6 from eleven cadaveric specimens. r=0.64, p=0.03 Figure 6.5D. Pearson correlation for mean Tb.Sp (mm) at the middle of the T7 spinous process and posteroanterior (PA) failure load (N) of T6 from eleven cadaveric specimens. r= -0.61, p=0.048  174  175  Figure 6.6. Mean ± 1 standard deviation Tb.Th (mm) in four regions: spinous process base (Base), spinous process mid (Mid), lamina and vertebral body (Body). Figure 7.1. A typical specimen set-up for the flexibility tests. This photo shows the base marker carrier (below the specimen), the marker carriers attached to the specimen, the specimen embedded in dental cement, fixed in the custom testing jig and secured to the spine flexibility machine. Figure 7.2. Schematic of posteroanterior (PA) loading at the T6 spinous process in T5-T8 cadaveric spine segments. The five optoelectronic marker carriers are shown.  1  _,  192  ^3  Figure 7.3. Diagram of normal alignment of T5, T6 and T7 (top left) and vertebral motion during PA mobilization at T6, where T6 moves into extension relative to T5, and T5 and T7 flex relative to T6 and T8. Figure 7.4. Graphical representation of Range of Motion (ROM; Deg), Neutral Zone (NZ) motion (Deg). Dashed line (—) represents the moment magnitudes of ± 0.2Nm. Figure 7.5. Flexibility was calculated as the slope of the motion (Deg) versus moment (Nm) curve during the loading portion of the curve in each direction (dark solid lines), using all points on that portion of the curve. Dashed line represents the slope.  197  PREFACE Publications Arising from this Thesis Sections of this thesis have been published or submitted as sole or multi-authored papers in refereed journals. Details of the authors' contribution are provided, where relevant. We agree with the stated contributions of the thesis author, as indicated below.  Dr. Karim M . Khan (thesis co-supervisor)  _  Dr. Thomas R. Oxland (thesis co-supervisor)  Published Papers Sran M M . To Treat or Not to Treat—New Evidence for the Effectiveness of Manual Therapy. British Journal of Sports Medicine 2004 38(5):521-525. (presented here as Chapter Two) Sran M M , Khan K M . Physiotherapy and Osteoporosis: Practice Behaviors and Clinicians' Perceptions—a survey. Manual Therapy 2005 10(l):21-27. (presented here as Chapter Three) Authors' contributions: Meena Sran was responsible for the original ideas behind the paper, data collection, analysis, presentation of findings, writing and editing the paper. Karim M . Khan provided editorial assistance and was the key editor on this paper. Sran M M , Khan K M , Zhu Q, McKay H A , Oxland TR. Failure Characteristics of the Thoracic Spine with a Posteroanterior Load: Investigating the Safety of Spinal Mobilization. Spine 2004 29(21):2382-2388. (presented here as Chapter Four) Authors' contributions: Meena Sran was responsible for the original ideas behind the paper, data collection, analysis, presentation of findings, writing and editing the paper. Karim M . Khan and Thomas R. Oxland provided editorial assistance, stimulated discussion, and were the key editors on this paper. Qingan Zhu provided technical assistance and advice on interpretation of the kinematic data. Heather A. McKay provided editorial assistance and stimulated discussion. Sran M M , Khan K M , Keiver K, Chew JB, McKay H A , Oxland TR. Accuracy of D X A scanning of the thoracic spine: Cadaveric studies comparing B M C , areal B M D , and geometric estimates of volumetric B M D against ash weight and CT measures of bone volume. European Spine Journal 2004 Dec.23 (Epub ahead of print). (presented here as Chapter Five) Authors' contributions: Meena Sran was responsible for the original ideas behind the paper, data collection, analysis, presentation of findings, writing and editing the paper.  Thomas R. Oxland and Heather A. McKay provided editorial assistance and stimulated discussion of the results. Karim M . Khan and Thomas R. Oxland were the key editors on this paper. Kathy Keiver provided technical advice and editorial assistance. Jason B. Chew provided technical advice in the interpretation of CT data.  Submitted Papers Sran M M , Khan K M , Zhu Q, Oxland TR. Posteroanterior Stiffness Predicts Sagittal Plane Midthoracic Range of Motion and Three-dimensional Flexibility in Cadaveric Spine Segments. Clinical Biomechanics (presented here as Chapter Seven) Authors' contributions: Meena Sran was responsible for the original ideas behind the paper, data collection, analysis, presentation of findings, writing and editing the paper. Karim M . Khan and Thomas R. Oxland provided editorial assistance, stimulated discussion, and were the key editors on this paper. Qingan Zhu provided technical assistance and advice on interpretation of the kinematic data. Sran M M , Boyd SK, Cooper D M L , Khan K M , Zernicke RF, Oxland TR. Regional Trabecular Morphology is Correlated with Failure of Thoracic Vertebrae under a Posteroanterior Load—a Micro-CT Investigation. Spine (presented here as Chapter Six) Authors' contributions: Meena Sran was responsible for the original ideas behind the paper, data collection, analysis, presentation of findings, writing and editing the paper. Karim M . Khan and Thomas R. Oxland provided editorial assistance, stimulated discussion, and were the key editors on this paper. Steven K . Boyd and David M . L . Cooper provided advice on interpretation of the data, technical advice and helped edit the paper. Ron F. Zernicke provided editorial assistance, stimulated discussion of the results and helped edit the paper.  Abstracts I presented these abstracts orally or as a poster at the conference indicated. Sran M M , Boyd SK, Cooper D M L , Khan K M , Zernicke RF, Oxland TR. Regional Trabecular Morphology by Micro-CT is correlated with Failure of Thoracic Vertebrae under a Posteroanterior Load. NIAMS-ASBMR Scientific Meeting on Bone Quality, Bethesda, 2005. Sran M M , Khan, K M , Zhu Q, Oxland TR. Can a passive accessory movement predict segmental spine motion? - a cadaveric study of posteroanterior mobilization. Canadian Physiotherapy Association, National Congress, Victoria, 2005.  xiv  Sran M M , Khan K M , Cooper D M L , Boyd SK, Zernicke RF, Oxland TR. Regional Trabecular Bone Volume Ratio Predicts Failure of Thoracic Vertebrae under a Posteroanterior Load. 26 Annual Meeting of the American Society for Bone and Mineral Research, Seattle, 2004. th  Sran M M , Khan K M , Zhu Q, Oxland TR. Posteroanterior Stiffness Predicts Sagittal Plane M i d Thoracic Range of Motion in Cadaveric Spine Segments. 5 Combined meeting of the Orthopaedic Research Societies of the USA, Canada, Japan and Europe, Banff, 2004. th  Sran M M , Khan K M , Zhu Q, McKay H A , Oxland TR. Posteroanterior Mobilization of Thoracic Vertebrae with Low Bone Density: Safety and Intervertebral Movements. Canadian Physiotherapy Association Congress, Quebec City, 2004. Sran M M and Khan K M . Physiotherapy and Osteoporosis: Practice Behaviors and Clinicians' Perceptions. Canadian Physiotherapy Association Congress, Quebec City, 2004. Sran M M , Khan K M , Zhu Q, McKay H A , Oxland TR. Spine Biomechanics in Osteoporotic Thoracic Vertebrae: Investigating the Safety and Intervertebral Movements during Posteroanterior Mobilization. Proceedings of the International Federation of Orthopaedic Manipulative Therapists 8 International Congress, Capetown, 2004. th  Sran M M , Khan K M , Zhu Q, Oxland TR, McKay HA. Failure Characteristics ofthe Thoracic Spine with a Posteroanterior Load. Journal of Bone and Mineral Research, 18, Suppl 2, Sept 2003, S375. Sran M M , Khan K M , Zhu Q, McKay H A , Oxland TR. Failure of thoracic vertebrae under a posteroanterior load. 1 Annual Meeting of the Alberta CIHR Bone and Joint Health Training Program, Banff, 2003. st  Sran M M , Khan K M , Zhu Q, McKay H, Oxland T. Failure of thoracic vertebrae with a posteroanterior load. International Society for Study of the Lumbar Spine, Vancouver, 2003. Sran, M . The role of physical therapy in the prevention and management of osteoporosis. 14 International Congress of the World Confederation for Physical Therapy, Barcelona, 2003. th  xv  ACKNOWLEDGEMENTS  'Education is the one thing no one can take away from you', something my dad told me frequently when I was a young girl. Wise words from a man whose family lost everything, including his father, in the partition of India. This thesis is dedicated to my mom, Jagjit Kaur Sran, and my dad, Surinder Singh Sran, who worked incredibly hard and gave up so much to give my siblings and I the opportunity for formal education. I've come to associate a number of words beginning with 'p' with 'PhD', in particular persistence, pain, and privilege. The first two are basically self explanatory, so I'll focus on the 'privilege' of being a doctoral student. When I wasn't enjoying being a doctoral student (which was more than once) I had to remind myself that I chose to take this on, and being able to choose is a privilege that many people in this world don't have. I had the privilege of working with a number of outstanding individuals who, in one way or another, have enriched my life. In particular my co-supervisors, Karim M . Khan and Thomas R. Oxland, and my committee member, Heather A. McKay. It is important to believe in yourself, but it definitely helps to have others believe in you. I want to thank Karim for believing in me. His continued encouragement, confidence in my abilities, and humor were invaluable in surviving the program. Karim is a true clinician-scientist who possesses an incredible combination of enthusiasm, energy, wit and intelligence. I, like many, find juggling clinical practice and research, while creating a bridge between them, is challenging—Karim is an excellent role model. Tom is one of the most ethical, thorough and successful researchers I have come across. As Phil would say, he has 'a brain the size of a planet' which is a tremendous asset in a supervisor and something for which I am grateful. Tom's willingness to collaborate made this thesis possible. I want to thank Tom for many things but in particular his 1:1 teaching, rigorous research, welcoming demeanor and the opportunity to join his lab. Heather has been a source of guidance, support and insight over the past four years, and I want to thank her for her many contributions to my work.  xvi  I am also fortunate to have trained in the Alberta/Canadian Institutes of Health Research Bone and Joint Health Training Program at the University of Calgary for one year of my doctoral studies. My co-supervisors, Ron F. Zernicke and Steven K. Boyd have also contributed significantly to my doctoral thesis by allowing me to train in their labs and to complete a portion of this thesis under their guidance. Both Ron and Steve are strong researchers and have provided excellent mentorship, for which I am grateful. Special thanks to the Zernicke, Hallgrimsson and Kantzas labs. Thanks to my brother, Steve, and my sister, Jasmein, for their ongoing support. Special thanks to my husband, Phil Lawrence, who married me despite the fact that I'm always working, and has been a constant source of encouragement and help over the past two years. Pip's positive energy and enthusiasm for my project, both as a clinician and a friend, have made the roller coaster ride of 'doctoral student life' much easier to take. Thanks to Dot, Gerry and Stuart Brown for everything they did to make my time in Calgary so special. I thank the staff in the Osteoporosis Program at the BC Women's Health Centre, the Allan McGavin Sports Medicine Centre and the Physiotherapy Association of British Columbia for their involvement and support. I have been a part of three research groups and wish to thank: the Division of Orthopaedic Engineering Research (UBC), the Bone Health Research Group (UBC), and the Joint Injury and Arthritis Research Group (UofC). In particular the following individuals have been of great assistance: Qingan Zhu, Marcus Robinson, David Cooper, Jesse Chen, Juay-seng Tan, Simon Sjovold, Kam Sandhu, Patti Janssen, Kathy Keiver, Gilles Galzi, Meghan Donaldson and Jeremy LaMothe. I would also like to thank the following clinicians for their assistance: Rob Labrom, Jason B. Chew, Bruce Forster, John Mayo, Nico Berg, Janet Lundie, Vlad Stanescu, Baht Kaymakci and John Wark. This thesis would not have been possible without the donors and participants. Finally, I'd like to acknowledge scholarships from the Michael Smith Foundation for Health Research, Canadian Institutes of Heath Research and the British Columbia Medical Services Foundation (Vancouver Foundation), and a grant from the British Columbia Sports Medicine Research Foundation.  DEFINITIONS  Activities of daily living: everyday activities such as eating, bathing, dressing, toileting, transferring (for example, getting into or out of a chair) and continence; the activities an individual needs to be able to perform in order to live independently  Accessory movements: movements of joints that patients cannot perform actively, but which can be performed on them by another person  1  Active movement: movement of a mobile segment produced by the muscles that produce movement of the segment  Anisotropic: having different mechanical properties when loaded along different axes  Bone mineral content: total grams of bone mineral as hydroxyapatite within a measured bone region  Bone mineral density (BMD, areal B M D ) : grams of bone mineral per unit of bone area scanned; cannot make volume determinations  Bone volume: the amount of three-dimensional space occupied by a bone  Cortical bone: the dense outer layer of bone  Creep: when a viscoelastic material is subjected to a constant load it deforms with time; the deformation-time curve approaches a steady-state value  This figure is a deformation and load vs. time plot which illustrates deformation over time under a constant load (until the deformation-time curve reaches a steady-state). This phenomenon is termed 'creep'.  Degrees of freedom: the motion of a rigid body in space has six degrees of freedom— three translations and three rotations; when bodies are interconnected certain constraints are placed on their motion, and the number of degrees of freedom decreases  2  Desiccator: an enclosure containing drying agents (desiccants) in which a substance can be kept in a controlled dry atmosphere  Elastic zone (EZ): That part of the physiological intervertebral motion which is measured from the end of the neutral zone up to the physiological limit. Within the EZ, spinal  xix  motion is produced against a significant internal resistance. It is the zone of high stiffness.  Finite Element Modeling (FEM): mathematical modeling of an object that involves dividing it up for analysis, i.e. dividing an object into an array of discrete elements for structural analysis  Functional spinal unit (FSU): two adjacent vertebrae and their intervening soft tissues.  2  Fragility fracture: "a fracture caused by injury that would be insufficient to fracture normal bone: the result of reduced compressive and/or torsional strength of bone".  4  In  the clinic, it is often defined as a fracture that occurs from a fall from standing height or lower, or as the result of minimal or no identifiable trauma.  5  Hypomobility: a clinical term used to describe reduced mobility; normally assessed based on end-feel and comparison with adjacent segments.  Hysteresis: energy loss exhibited by viscoelastic materials when they are subjected to loading and unloading cycles  Isotropic: having uniform properties in all directions independent of the direction of load application  Kinematics: the study of the relationships between positions, velocities, and accelerations of rigid bodies, without concern for how the motions are caused (ignoring forces and moments acting on the body)  Linear Variable Differential Transformer (LVDT): a circuit that produces an electrical output proportional to the displacement of a separate movable core; used to measure position  Manipulation: a small amplitude thrust technique performed with speed  1  Manual therapy: manually performed assessment and treatment techniques; can include joint, neural tissue, and/or muscle techniques  Material properties (of bone): describe the behavior of a specimen of bone tissue without regard to its size (i.e. a size-independent parameter); depends on organic and inorganic components of bone; small uniform specimens are loaded under well-defined conditions  6  Micrometer (urn): unit of length equal to one millionth of a meter, or one micron  Midsagittal plane: the median plane of the body which divides it into two equal halves  Mobilization: a passive movement performed such that it is at all times within the ability of the patient to prevent the movement if he or she so chooses.  1  Mobilization is typically  xxi  performed for the purpose of relieving pain and restoring full-range, pain-free, functional movements or to maintain a functional range of movement in patients who are unconscious or who have an active joint disease such as rheumatoid arthritis.  1  Moment; a quantity equal to the product of a force and the perpendicular distance from that force to the point of interest; unit of measure is the newton meter (Nm)  2  Morphology: the form and structure of an organism  Neutral zone: a region of intervertebral motion around the neutral posture where little resistance is offered by the passive spinal column ; the zone of high flexibility or laxity ; 3  2  typically expressed in meters and degrees (translation and rotation respectively)  Newton (N): the basic metric unit of force giving a mass of one kilogram (2.205 pounds) at an acceleration of one meter (1 yard) per second per second  Nominal: an approximate measurement; the actual measurement may vary slightly  Opto-electronic: the combination of optical and electronics phenomena in a single device; lasers, light-emitting diodes (LEDs), and light detectors of various sorts fall under this definition  xxii  Osteoporosis: a skeletal disease characterized by compromised bone strength predisposing an individual to an increased risk of fracture  7  Passive movement: movement of a mobile segment produced by any means other than by the muscles that produce movement of the segment  1  Passive Accessory Intervertebral Movements (PAIVMs): movements are applied which the individual cannot perform actively, but which can be performed on them by another person  Passive Physiological Intervertebral Movements (PPIVMs): a manual assessment technique for the spine in which the physiological movements of flexion, extension, lateral flexion and rotation are repeated passively in the non-weight bearing position  1  Physiological movements: movements that patients can perform actively by themselves  1  Posteroanterior (PA) spinal mobilization: a mobilization technique where load is applied in the posterior to anterior direction; often performed with the patient lying prone  Radiography: the making of film records (radiographs) of internal structures of the body by exposure of film specially sensitized to x-rays  8  xxiii  Range of Motion: the entire range of the physiological motion (includes both the neutral and elastic zones)  2  Resolution: the number of pixels per square inch on a computer-generated display; the greater the resolution, the better the picture  Rigid Body: an idealization of a real object; assumes that the body is absolutely rigid so that it does not stretch, compress, or otherwise deform no matter how large the forces and moments acting on it.  2  Rotation: spinning or angular displacement of a body about some axis  2  Safe: administered without causing harm or injury  Sagittal plane: any vertical plane parallel to the midsagittal line that divides the body into left and right portions  Servohydraulic: motion is controlled by a servo-controlled valve and high pressure liquids actuate pistons  Stereology: the study of estimating geometrical quantities  xxiv  Stiffness: a term used to describe the force needed to achieve a certain deformation of a structure; a steeper curve represents a stiffer structure  Stress relaxation: decrease in stress in a material subjected to prolonged constant strain at a constant temperature. Stress relaxation behavior is determined in a relaxation test and data is often presented in the form of a stress vs. time plot. The stress relaxation rate is the slope of the curve at any point.  This figure is a stress and deformation vs. time plot to illustrate 'stress relaxation'—the decrease in stress in a material subjected to prolonged constant strain at a constant temperature  Structure: the manner of construction and arrangement of parts; a thing constructed  Structural Properties: describe the behavior of whole bone, which can include reference to bone size, shape, cortical thickness, cross-sectional area, and trabecular architecture  XXV  Three-dimensional motion: most human body joints can move in any of six possible degrees of freedom  Torque: synonymous with 'moment'  Trabecular bone: bone tissue with a spongy honeycomb structure, typically found at the ends of long bones and the middle of vertebrae  Translation: at a given time, all particles in the body have the same direction of motion relative to a fixed point  2  Vertebroplasty: a medical grade cement is injected through a needle into a painful fractured vertebral body with the aim of stabilizing the fracture  Viscoelastic: time dependent property where the deformation of the material is related to the rate of loading, hysteresis, creep, stress relaxation  xxvi  REFERENCES 1.  Maitland GD, Banks K , English K , et al. Maitland's vertebral manipulation. Sixth  ed. Boston: Butterworth Heinemann, 2001. 2.  White A , Panjabi M . Clinical biomechanics of the spine. Second ed. Philadelphia:  JB Lippincott Company, 1990.  3.  Panjabi M M . The stabilizing system of the spine. Part II. Neutral zone and  instability hypothesis. J Spinal Disord 1992;5:390-6.  4.  WHO. Guidelines for preclinical evaluation and clinical trials in osteoporosis.  Geneva, 1998:59.  5.  Brown JP, Josse R G . 2002 clinical practice guidelines for the diagnosis and  management of osteoporosis in Canada. C M A J 2002; 167:S 1-34.  6.  Hayes WC, Bouxsein M L . Biomechanics of Cortical and Trabecular Bone:  Implications for Assessment of Fracture Risk. In Mow V C , Hayes W C eds. Basic Orthopaedic Biomechanics. 2nd ed. Philadelphia: Lippincott-Raven Publishers, 1997.  7.  Osteoporosis prevention, diagnosis, and therapy. J A M A 2001;285:785-95.  8.  O'Toole M ed. Encyclopedia & Dictionary of Medicine, Nursing, & Allied  Health. 5th ed. Toronto: W.B. Saunders Company, 1992.  xxvii  L I S T O F  A B B R E V I A T I O N S  (commonly used abbreviations)  aBMD  areal bone mineral density  AP  anteroposterior  BC  British Columbia  BMC  bone mineral content  BMD  bone mineral density  BV/TV  bone volume/total volume; bone volume fraction  CT  computed tomography  Deg  degrees  DXA.  dual energy X-ray absorptiometry  g  grams  uCT  microcomputed tomography  PA  posteroanterior  ROM  range of motion  SD  standard deviation  SMI  structure model index  Tb.Th  trabecular thickness  Tb.N  trabecular number  Tb.Sp  trabecular separation  UBC  University of British Columbia  Literature Review and Introduction to the Thesis  Chapter One Literature Review and Introduction to the Thesis  Literature Review and Introduction to the Thesis 1.1 O S T E O P O R O S I S At a consensus development conference on osteoporosis in 2000, osteoporosis was defined as ".. .a skeletal disorder characterized by compromised bone strength predisposing to increased risk of fracture. Bone strength reflects the integration of two main features: bone density and bone quality."  1  1.1.1 Prevalence Osteoporosis affects approximately one in four women and one in eight men in Canada and the United States (US). ' Recent data suggest that 16% of Canadian women and 7% of Canadian men over the age of 50 have osteoporosis and at least 1.3 million fractures 4  in the US each year are attributed to osteoporosis. Regions with high trabecular bone 5  content are the most affected by osteoporosis, with spinal compression fractures, proximal femur, distal radius and rib fractures being the most common osteoporotic fractures. " 6  12  1.1.2 Pathophysiology Bone loss is part of normal aging. However in osteoporosis bone mass drops so that fracture risk is elevated (see '1.1.4 Diagnosis of Osteoporosis'). Normal bone physiology requires a balance between resorption and formation, carried out by bone cells under the influence of both the endocrine system and mechanical loading. ' 1 3  1 4  Factors that can  cause an individual to lose bone at a faster than normal rate include hormonal and genetic factors, lifestyle factors including physical inactivity, nutrition and smoking, and the use of medications which affect bone remodeling.  1 5 - 1 8  2  Literature Review and Introduction to the Thesis  1.1.3 Economic and Social Cost of Osteoporosis The estimated Canadian acute care costs for osteoporosis in 1993, which includes admission to hospital, outpatient care and drug therapy, was over 1.3 billion Canadian (CDN) dollars.  1 9  Data from the United States suggest the economic burden of  osteoporosis has increased tremendously over the past decade, rising to $17-20 billion 20  C D N dollars annually.  The costs will likely increase even further as the prevalence of  osteoporosis in Canada is expected to rise with the aging population—25% of the Canadian population is expected to be over 65 years of age by 2041.  2 1  In addition to the obvious economic impact of osteoporotic fractures, there are serious social costs including reduced quality of life, > " back pain, 6  30 32  ability,  " physical impairments,  22  24  33 36  " and mortality  2 5 - 2 9  decreased functional  37  . This is discussed further in  section 1.2.1 'Consequences of Vertebral Fractures'. 1.1.4 Diagnosis of Osteoporosis Clinical diagnosis of osteoporosis is currently based on two factors, (1) bone mineral density (BMD) by dual energy X-ray absorptiometry (DXA) and (2) evidence of fragility fracture*. A decrease in B M D is associated with an increased risk of fracture.  38  In Canada we use the World Health Organization's (WHO) definitions of osteoporosis.  2 0  (* refers the reader to terms which are defined in the 'Definitions', pages xviii to xxvii)  3  Literature Review and Introduction to the Thesis The WHO definitions are based on comparison of the individual's B M D to the mean for a normal young adult population of the same sex and race. A n individual's results are described as a "T-score" that reflects the number of standard deviations above or below the mean B M D for young normal adults. Specifically, normal B M D is defined as a Tscore above -1.0. Osteopenia, in which B M D is lower than normal but not low enough to be categorized as osteoporosis, is associated with a T-score between -1.0 and -2.5. Osteoporosis is defined as a T-score below -2.5, or evidence of a fragility fracture.  Not  all fractures are associated with an event or incident, but many are. Diagnosis of fragility fracture is determined by a combination of the history of the incident, which is typically fall-related ' 4 0  4 1  and the results of radiography.  1.1.5 Physiotherapy and Osteoporosis Physiotherapy management in the presence of osteoporosis has received little attention to date. Further, much of the currently available evidence relates to exercise prescription for bone health, - " which is only one part of physiotherapy management. 14  42  44  Physiotherapists are health care professionals with specialist skills in musculoskeletal rehabilitation. Physiotherapists use many methods of conservative management in clinical practice, including manual therapy*. Numerous physiotherapeutic treatment methods are effective for reducing pain, improving function and quality of life in a variety of musculoskeletal conditions, ' " 43  45  51  but few have been studied in a population  with osteoporosis. To illustrate this point consider a recent review of musculoskeletal rehabilitation in osteoporosis in which the authors discuss only three methods of rehabilitation after vertebral fracture—exercise, spinal orthoses, and vertebroplasty*.  52  4  Literature Review and Introduction to the Thesis Since other treatment methods have been shown to be effective in populations without osteoporosis, it is possible that they may also be appropriate in osteoporosis. Thus the safety and effectiveness of rehabilitation strategies, besides exercise and bracing, should be tested in the osteoporotic spine.  1.2 V E R T E B R A L F R A C T U R E S Vertebral fractures are the most common type of osteoporotic fracture and they typically manifest as compression fractures of the vertebral body.  5 3 - 5 6  In the US, 5% of 50-year-  old white women and 25% of 80 year-old women have had at least one vertebral fracture.  56  After one vertebral fracture, the risk of a second vertebral fracture increases  Vertebral fractures most commonly occur at the thoracolumbar junction and midthoracic region > > " and prevalence is similar in men and women. ' 7  28  60  63  vertebral fractures are fall-related.  64  66  65  At least 57% of  Most vertebral fractures are related to moderate  trauma, defined as less than or equal to a fall from standing height.  6 6  Minimal trauma  vertebral fractures can include those that occur during lifting and bending activities. It is known that static or dynamic forward bending of the spine, as occurs with an excessive thoracic kyphosis or active flexion of the spine, results in a larger proportion of compressive load on the anterior aspect of the vertebral bodies. This concentration of force, combined with decreased B M D , can lead to vertebral body failure.  6 7  5  Literature Review and Introduction to the Thesis Vertebral body fracture is typically diagnosed by a radiologist and is based on vertebral morphometry, measured on lateral spine radiographs. Vertebral deformity and vertebral fracture are often used interchangeably, but not all vertebral deformities result in height loss sufficient to be diagnosed as fractures (Figure 1.1).  6  Literature Review and Introduction to the Thesis 1.2.1 Consequences of Vertebral Fractures -Economic and Social Impact In 1997 vertebral fracture accounted for over 400,000 total hospital days and generated charges of more than 500 million USD in the United States.  Hospital charges for  individuals admitted for vertebral fracture were between 8000 to 10,000 US dollars (USD) per admission, mean length of stay wasjust under 6 days, and more than 50% of discharged patients required some form of continuing care.  6 8  In addition to the high economic cost, vertebral fractures also have enormous social impact. ' ' ' u  5 6  6 9  7 0  Back pain, physical and functional impairment, reduced quality of life 30 32 37 56 70 71  and increased mortality are common consequences of vertebral fractures.  ' ' ' '  Surprising to many, vertebral fractures are associated with a 16% reduction in expected five-year survival.  Not all vertebral fractures are symptomatic, but both symptomatic  and asymptomatic vertebral fractures are associated with increased morbidity mortality. 56,74,75  and  Back pain resulting from vertebral fracture can be acute or chronic, and is associated with 71  an increased number of physician visits and days missed from work.  Most clinically  diagnosed fractures are detected when the patient seeks medical attention for back pain produced by the fracture.  70  Individuals have been shown to have significant physical impairments post vertebral fracture. These include reduced back muscle strength, impaired thoracic posture and balance, decreased pulmonary function and height loss. > ' > > 34  36 74  76  77  Studies suggest that  7  Literature Review and Introduction to the Thesis individuals also suffer from significant functional impairment post vertebral fracture.  30-32  Functional consequences can include limited ability to perform activities of daily living*, restricted function in employment in addition to social and recreational settings.  31  Functional impairment has been documented in the acute period post fracture, as well as several years post fracture.  A prospective study of men and women found overall  function declined at similar rates among individuals with vertebral or hip fractures and 78  several studies suggest that both back pain and function worsen to a greater extent with the increasing number and severity of fractures.  71 T\ 7Q  ' '  6 22 24  Vertebral fractures also have an important effect on quality of life. ' "  Several studies  found a negative relationship between vertebral fracture and quality of life, ' ' men and women.  in both  2 2  1.3 B A C K P A I N Studies suggest that 60 to 80 percent of individuals in the Western world will experience acute back pain in their lifetime. on  and women countries.  82  RO R9  "  Back pain is a global problem affecting both men  on  " but some studies suggest it is more prevalent in more affluent Although some studies suggest higher prevalence of back pain in individuals R7  with manual or hard labour jobs, also commonly report back pain.  individuals working in office jobs or as homemakers 33  Back pain is also common in older adults. '  88 89  Results from the Canadian National Population Health Survey (1994-1995) found 20.9  8  Literature Review and Introduction to the Thesis percent of persons aged 65 or older, in the province of Ontario, reported having back •  pain.  90  1.3.1 Economic and Social Cost of Back Pain Back pain is associated with high utilization and costs of family physician services, and is one of the top three chronic conditions for which Canadians consult a family physician. Back pain is also one of the ten most costly physical health conditions in the US.  9 0  9 1  Absence, disability and reduced productivity result in enormous cost for employers.  9 1  Q7  Back pain also has enormous social costs.  Back pain is associated with reduced quality  of life, sleeping problems, fatigue, physical limitations and risk of major depression. " 93  97  1.3.2 Back Pain and Osteoporosis For many years, back pain has been discussed as an important clinical sequel of osteoporosis.  "  Both back pain and osteoporosis are prevalent conditions, so it is not  surprising that many individuals have both. > ' 43  98 99  However, with the high prevalence of  back pain in individuals with osteoporosis, many researchers have investigated whether factors such as low B M D , increased kyphosis, or vertebral deformity are associated with an increase in back pain.  Studies have generally found that back pain in a population with osteoporosis is only 7^  significantly related to severe vertebral deformity. For example, Ettinger et al.  found  vertebral deformity was only associated with substantial pain in women with vertebral  9  Literature Review and Introduction to the Thesis height ratios four standard deviations below the population mean and studies suggest that low B M D is not associated with back pain. ' 2 9  1 0 0  One study found that women with  kyphosis reported only slightly more back pain and back disability, which was not statistically significant,  101  while another study found the degree of kyphosis was  associated with severity of thoracic but not lumbar back pain.  1 0 2  With respect to gender, one study found similar prevalence of back pain in men and women over the age of 50, up to the age of 70-79.  1 0 3  After age 79 the prevalence was  higher in women and those with a greater number of previous vertebral fractures.  1 0 3  In a  study of the clinical characteristics of osteoporosis in men, back pain was found to be the chief complaint of 69 of 81 (85%) of participants.  1 0 4  Further, in another recent study the  prevalence of back pain was 75% in a population of older women with osteoporosis, and back pain was a determinant of both balance and functional mobility.  105  1.3.3 Back Pain Management—Physiotherapy Back pain is the most common reason for visiting a physiotherapist (Physiotherapy Association of British Columbia (PABC), October 2001), and many physiotherapists use manual therapy techniques such as spinal mobilization to assess and treat back pain. " 106  108  One study investigated therapy use and costs for patients with back pain for longer than 7 weeks and found physiotherapy was better than exercise at improving quality of life, and use of health care services and absenteeism tended to decrease after a course of physiotherapy.  9 6  10  Literature Review and Introduction to the Thesis 1.4 M A N U A L T H E R A P Y Manual therapy is the umbrella term for manually performed joint, muscle and neural tissue assessment and treatment techniques (Figure 1.2). This thesis focuses on joint techniques. Joint techniques can be divided into assessment and treatment techniques. Assessment techniques aim to measure joint motion and stability. These include passive physiological intervertebral movements* (PPIVMs), movements* (PAIVMs),  1 0 8 - 1 1 0  and stability tests.  1 0 8  passive accessory intervertebral  1 1 1 , 1 1 2  Treatment techniques aim to  restore full-range, pain free, functional movements and relieve pain.  1 0 8  These include  mobilization and manipulation.  In 2002 approximately 40 percent of Canadian registered physiotherapists, with direct patient care as their primary responsibility, identified 'orthopaedics' as their area of 113  clinical practice.  Another seven percent selected sports injuries, which are also  predominantly orthopaedic cases. Further, approximately 25% of physiotherapists registered to practice in British Columbia between the years of 2001 and 2004 participated in postgraduate specialization courses in orthopaedics (based on information from the College of Physical Therapists of British Columbia and the BC Section of the Orthopaedic Division, Canadian Physiotherapy Association). These courses focus on advanced manual assessment and treatment skills. In summary, more Canadian physiotherapists practice in orthopaedics and specialize in orthopaedic manual therapy than any other area of clinical practice or specialization.  11  Literature Review and Introduction to the Thesis 1.4.1 Spinal Mobilization Spinal mobilization* refers to passive movement (typically oscillatory but can be sustained) which can be performed in various parts of the range of motion, as either small or large amplitude movements.  1 0 8 , 1 1 4  Mobilizations are typically graded I through IV.  1 0 8  Each grade is defined as follows and is diagrammatically represented in Figure 1.3, according to the commonly used method of Maitland.  1 0 8  Clinicians routinely use spinal  joint mobilization for clinical assessment of individuals with back pain. Interpretation of findings is based on the degree of stiffness and the motion relative to segments above and below,  1 0 8  as well as the 'end feel', the sensation the examiner feels in the joint at the end  of the range of motion.  1 1 5  Spinal mobilization is also used in treatment, to relieve pain 108  and restore full-range, pain free, functional movements.  Spinal mobilization is  different from manipulation*. Manipulation can be defined as a small amplitude high velocity thrust performed at the limit of a range of movement. Manual Thciap\ Joint Techniques  Muscle Techniqties Assessment Techniques  Treatment Techniques  Ms  Mobilization  PAIVMs  Manipulation  VPW  —J  "L  Neural Tissue Techniques  Slabilin Tests  Figure 1.2. Overview of manual therapy techniques. This thesis focuses on joint techniques. PPIVMs= passive physiological intervertebral movements; PAIVMs= passive accessory intervertebral movements  12  Literature Review and Introduction to the Thesis  Grade I: a small amplitude-movement at the beginning of the range Grade II: a large amplitude movement in that part of the range that is free of stiffness or muscle spasm (i.e. beginning to middle of range) Grade III: a large amplitude movement that moves into stiffness or muscle spasm (i.e. middle to the end of range) Grade IV: a small amplitude movement that stretches into stiffness or muscle spasm, applied at the end of range  I starting * ._. • position -4  III *  ™  d  •  4  ~ <  •  ||  •  IV 108  Figure 1.3. Grades of mobilization in a normal range with a hard end-feel.  A number of studies have investigated therapist inter and intrarater reliability in applying grades of mobilization and measuring P A stiffness. Results are conflicting in that some studies found good reliability  1 1 6 , 1 1 7  and others found poor reliability.  I18  "  120  Of interest,  one group found therapists had much better ability to judge spring stiffness than the P A stiffness of human spines.  The authors suggested that mechanical stiffness may not be  equivalent to the clinical concept of PA stiffness in that PA stiffness of the human spine 121  may be multidimensional.  Another study showed that therapists can accurately judge  spinal stiffness using a matching task.  1 2 2  13  average normal range  Literature Review and Introduction to the Thesis  1.4.2 Effectiveness of Manual Therapy Including Spinal Mobilization In the assessment of spine pain, two studies found that manual examination by a physiotherapist is highly accurate in detecting the segmental level responsible for a 1 9^  patient's complaint when compared against a spinal block. good reliability for pain ratings,  '  1 94  A third study found  1 90  and another study found that results of manual 1 9S  segmental mobility assessment correlated with disability.  In treatment, recent randomized clinical trials found manual therapy to be more effective than other methods of conservative management of back and neck pain.  ' ' 127 129 131  Specifically, manual therapy including spinal mobilization reduces spinal pain, 1 97  and it is more effective  than other physiotherapy modalities that do not include spinal  mobilization (in the treatment of chronic neck pain). Also, a recent RCT found physiotherapy including spinal mobilization was more cost-effective than physiotherapy 1 ^9  without spinal mobilization or routine treatment by a general practitioner. However some older randomized clinical trials analyses  1 4 2  133  "  140  systematic reviews  1 4 1  and meta-  concluded that there was no evidence that spinal manipulative therapy is  superior to other standard treatments for patients with spinal pain. Reviews typically focus on methodological differences, such as the randomization procedure, blinding of patients and outcome assessments, adequate follow-up period, and dropouts,  143  >  144  when  trying to explain the discordant outcomes in clinical trials of manual therapy for back and neck pain. Yet there may be differences between the interventions that constitute  14  Literature Review and Introduction to the Thesis 'manual therapy'. No review has focused on whether such differences exist yet they could have an important effect on the outcome of clinical trials.  1.4.3 Mechanism of Action of Spinal Mobilization The biological mechanisms underpinning the effectiveness of spinal mobilization may be related to it stimulating sympathetic nervous system activity ' > 131  • • 1 3 1  motor activity.  •  145  146 a n  d promoting  •  For example, emerging evidence suggests that spinal joint  mobilization techniques applied to the cervical spine elicit concurrent effects on pain perception, autonomic function, and motor function in patterns that are similar to the patterns of change elicited by stimulation of the periaqueductal gray region of the midbrain.  1 3 1  '  1 4 6  '  1 4 7  Spinal mobilization is also used with the aim of restoring normal joint motion. It is thought that spinal mobilization may act by stretching the joint capsule or reducing muscle tone, thereby reducing stiffness. One study assessed the effect of T4-5 spinal manipulation (high velocity thrust technique) on stiffness in asymptomatic individuals, and found stiffness was not altered after spinal manipulation.  Whether spinal  mobilization or manipulation can change stiffness in individuals with back pain is not known.  1.4.4 Safety of Spinal Mobilization in Osteoporosis There appears to be agreement amongst leading clinicians that spinal manipulation is contraindicated in individuals with osteoporosis, ' " 108 149  151  yet clinical experience  1 5 2  and  15  Literature Review and Introduction to the Thesis published cases  suggest that these techniques are still being used by some  chiropractors. Clinical experience indicates that while some physiotherapists have expressed concern about the safety of manual therapy (including spinal mobilization) other clinicians routinely use manual therapy in older people, a proportion of whom will have osteoporosis. There are no data on the safety of manual therapy in individuals with osteoporosis to guide clinical practice.  1.4.5 Posteroanterior (PA) Spinal Mobilization One spinal mobilization technique used very commonly in clinical practice is posteroanterior (PA) mobilization. process  1 0 8 , 1 1 1  1 5 4 - 1 5 6  It can be applied at an individual spinous  to assess stiffness, quality and range of motion  1 0 8  at that specific level. In  manual assessment, P A mobilization is a passive accessory intervertebral movement (PAIVM) that is typically used in combination with other tests. In treatment, P A 108  mobilization is typically the first technique used in the thoracic spine,  most commonly  to improve sagittal plane* motion. It can be used alone or in combination with other spinal mobilization techniques. Figure 1.4 shows a physiotherapist applying PA mobilization to T6. Figure 1.5 shows the hand contact during P A mobilization in the midthoracic spine. It is routine to assess the thoracic spine in patients presenting with lumbar, thoracic, cervical or shoulder symptoms. Thoracic vertebral motion contributes to end range cervical motion and shoulder elevation through flexion, and greater stiffness in the  16  Literature Review and Introduction to the Thesis thoracic spine may lead to compensatory changes in the more mobile lumbar and cervical regions.  157,158  Structural changes in the vertebral bodies due to osteoporosis and disc degeneration will tend to bring the thorax into flexion. The ensuing decrease in intervertebral and rib cage mobility will limit the potential for thoracic extension.  1 5 7  Application of the P A  mobilization technique, in the straight P A or caudad direction, is thought to produce intervertebral extension in the thoracic spine. Further, Maitland recommends that the central P A mobilization technique should be used first in all cases where symptoms arise from the thoracic vertebrae.  1 0 8  For these reasons the central P A mobilization technique  is relevant to the care of older individuals with back pain and osteoporosis.  Some researchers have studied applied load during spinal mobilization ' ' ' ' 35  manipulation " 162  164  130  159  160  161  or  in vivo or in cadavers. However, no previous study has compared the  load required to fracture a vertebra, in vitro, and the load applied during P A mobilization, in vivo.  17  Literature Review and Introduction to the Thesis  Figure 1.4. Photograph of a physiotherapist applying PA mobilization to T6. The arrow depicts the direction of the applied load.  Figure 1.5. Diagram of hand contact during PA mobilization in the midthoracic spine.  18  Literature Review and Introduction to the Thesis 1.5 SPINAL FRACTURES: DETECTION AND DETERMINANTS Plain radiography is the traditional means of detecting vertebral fractures, clinically. Other methods, such as computed tomography, are also used for fracture detection in some clinical situations. Sensitive methods of vertebral fracture detection are necessary because the presence of fracture is related to future risk of fracture.  57-59  The load at which fracture occurs in any structure is influenced by both material properties* and geometry. Further, external factors such as the loading direction and rate significantly influence the failure load. Bone mineral content (BMC*), areal bone mineral density (aBMD*), volumetric bone mineral density (vBMD), cortical* and trabecular* bone morphology* are parameters that reflect both material properties and geometry of any bone structure* and therefore, influence bone strength. A number of imaging technologies are used to measure these parameters, such as D X A and microcomputed tomography.  1.5.1 Plain Radiography This is a method of X-ray examination in which an X-ray beam is passed through the patient onto a photographic plate. The main role of plain radiography in osteoporosis is in the diagnosis of vertebral compression fractures. Vertebral morphometry is typically measured on lateral spine radiographs. Some approaches have focused on describing the type of vertebral deformity, such as wedge, biconcave or compression, but the semiquantitative grading scale of Genant > 165  166  measures severity of the vertebral fracture  solely on the extent of vertebral height reduction and morphologic change. The degree of  19  Literature Review and Introduction to the Thesis severity is typically graded from 0-4 where 0 corresponds to normal, 1= mild fracture, 2moderate fracture and 3= severe fracture.  1 6 5 , 1 6 6  1.5.2 Dual energy X-ray Absorptiometry (DXA) Bone density is a determinant of bone strength. D X A represents this as an areal bone mineral density. D X A uses X-ray beams of two distinct energy levels and measurements are based on the decrease in energy of the photon beam as it passes through bone and nonmineralized soft tissue. Traditionally, bone mineral data are reported as areal B M D (aBMD) (grams per cm ), calculated by dividing the quantity of bone mineral within the scan area (BMC) by the projected area within the region of interest. Previous studies have shown that lumbar spine B M C measured by D X A correlates closely with the gold standard— ash weight, the weight of the inorganic component of bone. B M D explains up to 90% of the breaking strength of excised bone,  1 7 0  167  "  169  Areal  and a 7% increase  in B M C is equivalent to a three fold increase in bone stiffness and twice the breaking strength.  171  Since D X A is the best predictor of fracture risk among individuals who have  not yet suffered a fragility fracture, it may also accurately predict failure load of the vertebral body in those undergoing manual therapy.  Areal B M D values have clinical utility but can be confounded by changes in bone thickness (or depth). Bones of larger width and height also tend to be thicker yet bone thickness is not factored into estimates of aBMD due to the two-dimensional nature of the measurement.  1 7 2 , 1 7 3  Converting aBMD to 'volumetric bone mineral density' (vBMD)  reduces its dependence on vertebral size.  1 7 4  This concept of v B M D aims to overcome  20  Literature Review and Introduction to the Thesis the limitation of 2-dimensional planar D X A scanning.  1 7 4 - 1 7 6  However, this mathematical  derivation has never been compared with an ash-weight gold standard at the thoracic spine. Previous models for estimating v B M D assume vertebral geometry resembles a cylinder or a cube, yet human thoracic and lumbar vertebral endplate geometry is best estimated as an ellipse.  177 178  Although aBMD explains a large percentage of the breaking strength of bone,  170  bone  172  quantity is not sufficient to fully evaluate fracture risk. bone quality. ' 179  180  Bone strength also relies on  Bone quality can be defined as the characteristics of bone that are  needed in order to perform its function. Quality is influenced by morphologic features such as a bone's cross-sectional geometry, trabecular architecture, and the composition of the matrix.  Thus another limitation of D X A , relevant to this thesis, is its inability to  1 7 2  provide a measure of bone microarchitecture (i.e. trabecular number, trabecular thickness). Osteoporosis involves changes in bone architecture in addition to loss of aBMD.  1 7 9  '  1 8 1  1.5.3 Computed Tomography Tomography is a variation of simple radiography which permits cross-sectional images to be obtained.  1 8 2  Computed tomography (CT) is considered a very good technology for  183 188  imaging bone  "  yet it is not typically used in the care of older adults with suspected  osteoporotic vertebral fractures. To my knowledge, there have been no reports evaluating the sensitivity of radiography and CT scan for the detection of vertebral fractures produced during simulated spinal mobilization.  21  Literature Review and Introduction to the Thesis Further, current software allows for measures of bone volume from CT scans. This measure, used with B M C measured by D X A , could provide a more accurate measure of volumetric bone mineral density than previous methods combining vertebral body geometry and D X A measures.  1 7 4 , 1 7 6 , 1 8 9  1.5.4 Microcomputed Tomography  Bone mineral density is only one factor associated with bone strength.  1 7 9 , 1 9 0  Other  aspects of bone morphology, such as trabecular microarchitecture, also contribute to bone strength and fracture risk.  1 7 9 , 1 8 1  Research suggests a predisposition to loss of the  horizontal trabeculae in osteoporosis,  1 9 1  leading to decreased interconnectivity of the fill  internal scaffolding of the vertebral body and reduced ability to support loads.  Microcomputed tomography (uCT) is a fast, nondestructive, highly accurate measure of microscopic bone structure.  1 9 2  In the spine, uCT has most commonly been used to 193  measure three-dimensional trabecular bone morphology in small bone specimens.  In  the past, most quantitative information on three-dimensional trabecular bone structure was based on applying stereology* to two-dimensional sections. However stereology provides a biased estimate of trabecular bone morphology because it is two-dimensional. The clinical measurement technique, D X A , does not provide any indication of the number, thickness, or orientation of individual trabeculae in a given region, nor does it differentiate between cortical and trabecular bone.  1 9 4  Three-dimensional u.CT image  analysis allows for direct measurement of trabecular thickness (Tb.Th), trabecular number (Tb.N) and trabecular separation (Tb.Sp) which is model-independent (it does  22  Literature Review and Introduction to the Thesis not assume a fixed structure model, i.e. plate or rod model).  1 9 5  determinants of fracture that cannot be measured using D X A .  Thus pCT may highlight 1 9 6  1.6 THORACIC SPINE BIOMECHANICS 1.6.1 Failure Modes— Compression vs. Other Directions Failure load of a structure, the point at which there is a loss in the load-bearing capacity,  197  can be measured using a mechanical testing device. A load-deformation  curve, including the failure load, is shown in Figure 1.6. The load-deformation relationship represents mechanical behavior of a structure.  Bone can be viewed as a material (i.e. a uniform specimen of bone tissue) or a structure (i.e. a vertebra). As a material or a structure, the deformation of bone is dependent on the rate and direction of loading.  Geometry refers to the size and shape of a structure. Vertebral strength is determined not only by the quantity of bone, but also by geometry.  Differences in vertebral size may  increase the risk for vertebral fracture. For example, it is thought by some that Asians have a higher rate of vertebral compression fractures than Caucasians due to their proportionately smaller vertebrae.  Various geometric parameters can be measured to  investigate the influence of vertebral geometry on vertebral strength. These include lowtech methods such as calipers and two-dimensional measures from plain radiographs, to  23  Literature Review and Introduction to the Thesis more complex geometric parameters measured using a variety of current imaging technologies.  Spinal failure or fracture has most commonly been studied with axial compressive loading and a number of studies have investigated the relationship between vertebral 62 198 200  failure load in axial compression and bone mineral measurements using D X A .  '  Based on a search of Medline (1966 to present) and Cinahl (1982 to present) databases, there are no published studies of vertebral failure using modes simulating spinal mobilization or manipulation. Thus, there are no data to guide clinicians on the use of this potentially important therapeutic modality (spinal mobilization). Specifically, is spinal mobilization safe for use with individuals who have osteoporosis?  Deformation Figure 1.6. Load-deformation relationship. The first decrease in load-bearing capacity is termed the failure load ($.).  24  Literature Review and Introduction to the Thesis  1.6.2 Spine Kinematics Kinematics, the study of motion,  is used to describe motion of one bone with respect  to another without reference to the forces that cause the motion. Three-dimensional motion of a rigid body* is typically defined by motion in six directions, three translational* motions and three rotational* motions.  2 0 1  The right-handed orthogonal  (90° angle) coordinate system is typically used for orientation (Figure 1.7).  Skilled clinicians have combined biomechanical theory, knowledge of anatomy and clinical experience to develop numerous manual therapy techniques,  108  > ' 112  150  but the  theory underpinning many of these techniques has not been formally investigated. For example, central P A spinal mobilization is often used to restore thoracic spine motion in the sagittal plane*, yet the magnitude and direction of motion of thoracic vertebrae during this technique have not been measured.  To date, investigations of movements resulting from P A spinal mobilization in the cervical and lumbar spine, in vivo, have been conducted by inferring vertebral translations from the motion of an indenter at the skin surface or by measuring translations and rotations by digitizing radiographs. > > >  203  These studies report  sagittal translations of less than 2 mm (150 N load; in v i v o )  156  up to 3.8 mm (variable  35  loads; spinal model).  35  156  202  Mean anterior displacement of the spinous process was found to  range between 8.8 and 13.0 m m .  156  '  203  Mean sagittal rotation ranged from 1.2-2.4°, and  all lumbar motion segments tended to extend except L5-S1 which tended to flex.  1 5 6  One  group studied cervical spine motion during P A mobilization using interventional  25  Literature Review and Introduction to the Thesis magnetic resonance imaging (iMR), but the analysis of the images was still based on radiographic techniques.  204  They reported little or no rotation and less than 0.5 mm  translation (load not quantified).  205  y  Figure 1.7. Right-handed orthogonal (90° angle) coordinate system. Human motion is typically a combination of translation along any direction and rotation about any axis in 206,207  space.  1.6.3 Spine Stiffness Stiffness is a term used to describe the force needed to achieve a certain deformation of a structure. It is calculated as the slope of the load-deformation curve.  When clinicians use passive accessory intervertebral movements (PAIVMs) they note the stiffness,  122  quantity ' 108  125  to joints above and b e l o w .  and quality 108  1 0 8  '  1 2 3  of motion at one intervertebral joint relative  Three previous studies have measured P A spinal stiffness  in the thoracic spine using measures of displacement at the skin-surface  1 4 8  '  2 0 8  '  2 0 9  and  26  Literature Review and Introduction to the Thesis reported T7 mean P A stiffness measures of 10.7 N/mm, stiffness of 13.6 N/mm.  1 4 8  2 0 8  1 2.5 N/mm,  2 0 9  and T4 P A  Yet, whether or not PA spinal stiffness can predict segmental  spine motion has not been formally investigated.  Of note, some previous in vivo studies of PA stiffness during PA mobilization in the lumbar spine were performed with participants holding their breath during the application of the mobilization to avoid vertebral displacement caused by breathing.  1 5 5 , 2 1 0  This is  problematic in that spinal stiffness has been shown to change throughout the respiratory cycle.  2 1 1  1.6.4 Spine Flexibility  In the spine, flexibility refers to the ratio of the amount of rotation produced to the load applied.  2 0 6  Flexibility can be calculated as the slope of a moment-rotation curve  (Deg/Nm). Range of motion* and neutral zone motion* can also be measured. Panjabi and colleagues  2 1 5  2 1 2 - 2 1 4  measured flexibility in 11 motion segments, one for each  thoracic level from five thoracic spines using a 2-vertebra construct. Applying a 5 N m torque*, they found mean flexibility values (Deg/Nm) of: flexion 0.45, extension 0.35, right lateral bending 0.4, left lateral bending 0.38, right axial rotation 0.39 and left axial rotation 0.33.  2 1 5  Although techniques such as P A mobilization are routinely used in  clinical assessment,  108,216  whether range of motion, three-dimensional flexibility, or  neutral zone motion can be predicted by a passive accessory movement such as P A mobilization has not been tested.  27  Literature Review and Introduction to the Thesis 1.7 SUMMARY Back pain and osteoporosis are common in older adults and each is associated with enormous economic and social costs. In addition, there is a high prevalence of back pain among those who have osteoporosis.  "  Back pain is the most common reason for  visiting a physiotherapist (Physiotherapy Association of British Columbia (PABC), October  2001),  and many physiotherapists use manual therapy techniques such as spinal  mobilization to assess and treat back pain in conventional populations.  My literature review, which included searching Medline and Cinahl databases, highlighted five substantial gaps in knowledge.  Previous systematic reviews focused only on methodologic factors such as the randomization procedure, blinding of patients and outcome assessments, adequate follow-up period, and dropouts,  1 4 3 , 1 4 4  but failed to undertake critical analysis of the  manual therapy intervention itself. Further, previous reviews and meta-analyses, including one published in 2 0 0 3 ,  142  only included studies published in 2000 or earlier.  Since the late 1990s physiotherapy practice has changed considerably based on the results 217 228  of research.  No systematic review has assessed the results of more recent  studies, which are more likely to accurately represent current clinical practice, nor has any review critically analysed the content of the manual therapy intervention used.  28  Literature Review and Introduction to the Thesis Second, recent randomized clinical trials suggest that manual therapy is effective for the SO SI 196 198 990  treatment of spinal pain ' '  " '  so it seems reasonable that clinicians would  consider using such techniques in individuals with osteoporosis who present with back pain. On the other hand, the literature cautions against the use of manual therapy in individuals with osteoporosis. > ' 108  149  216  There are no studies of physiotherapists'  perceptions and current practice patterns with respect to the management of individuals with osteoporosis or the use of manual therapy. Specifically, how many physiotherapists use manual therapy in this population and how many of them have concerns about fracture as a complication of treatment? Third, although manual therapy is effective for back and neck pain in other populations, ' ' 50 51 126  128  there are no data on the safety of manual therapy in the  osteoporotic spine. If patients are not offered manual therapy due to the clinician's fear of fracture then back pain management in this population may be suboptimal. If, however, some patients' skeletons are at risk of fracture because of the forces associated with manual therapy this would be important to know. No studies have been conducted to assess the likelihood of fracture with manual therapy in the osteoporotic spine.  Fourth, PAIVMs are used in spine assessment and clinical dogma suggests these movements provide information about intervertebral motion, but this theory has never been tested. Specifically, there are no data reporting whether PA stiffness predicts intervertebral motion in older spines.  29  Literature Review and Introduction to the Thesis Finally, the breaking strength of a bone depends on a number of factors including B M D and bone geometry. > > 179  l81  190  Knowing whether B M D , geometry, and/or microscopic  bone structure can predict the load at which a vertebra will fracture from a manual therapy technique will further our understanding of the determinants of spinal fracture. There are no published data on the relationship of BMD, vertebral geometry or morphology with vertebral failure load during common manual therapy techniques.  30  Literature Review and Introduction to the Thesis 1.8 AIMS, OBJECTIVES & SCOPE OF T H E THESIS Global Aim ofthe Thesis M y overall aim is to improve the understanding of manual therapy related spine biomechanics and determinants of fracture among older adults with back pain and osteoporosis. This research has the potential to be readily translated to treatment of back pain in this population.  Objectives My thesis focuses on clinically-relevant questions related to manual therapy in the osteoporotic spine. Specifically, I seek new knowledge regarding:  (1) the effectiveness of manual therapy including spinal mobilization for spinal pain What might explain the apparent inconsistencies in clinical trials?  (2) clinicians' perceptions and practice behaviors with respect to manual therapy in patients with osteoporosis Are clinicians using manual therapy in patients with osteoporosis and if so, do they have concerns about its use?  (3) the safety of spinal mobilization in the osteoporotic spine  31  Literature Review and Introduction to the Thesis Is a commonly used spinal mobilization technique, posteroanterior (PA) mobilization, likely to be safe in individuals with osteoporosis? How likely is it that a clinician could cause a vertebral fracture using this technique?  (4) predicting spinal failure during manual therapy Can traditional measures of bone parameters such as B M C , B M D , and/or vertebral geometry predict failure load of thoracic vertebrae under a P A load? If not, can novel measures of vertebral morphology, obtained using uCT, better predict failure load and location under a PA load?  (5) vertebral kinematics during manual therapy and the ability to predict intervertebral motion with a commonly used manual therapy technique Can segmental spine stiffness, during simulated P A spinal mobilization, predict segmental spine motion?  32  Literature Review and Introduction to the Thesis Scope of the Thesis This thesis includes six studies (4 published, 2 submitted) presented in the following six chapters. First, in Chapter Two, I undertook a systematic review of the effectiveness of manual therapy including spinal mobilization for spinal pain. I critically analysed the literature to provide possible explanations for inconsistent outcomes in recent clinical trials of manual therapy. In Chapter Three, I used survey methodology to investigate perceptions about and use of manual therapy by physiotherapists in British Columbia when treating individuals with osteoporosis. Specifically, I investigated what treatment modes are being used, whether or not manual therapy is being used and if there is concern about fracture as a complication of manual therapy. Then, in Chapter Four, I report on an in vitro biomechanical study to measure whether a commonly used spinal mobilization technique (PA mobilization) is safe for use in older midthoracic cadaveric spines with low bone mineral density. This study also investigated vertebral kinematics during simulated PA mobilization and the sensitivity of plain radiography and CT scan to detect fractures produced by simulated P A mobilization. In Chapter Five, I report my investigation of the accuracy of A P and lateral D X A scans in cadaveric midthoracic spine segments and calculations of volumetric bone mineral density against the gold standard of bone content by ashing. Following on from those findings, Chapter Six, reports a study that aimed to measure regional vertebral trabecular morphology, using pCT, as a possible predictor of vertebral failure under a P A load. Finally, Chapter Seven reports a study testing whether PA stiffness can predict intervertebral motion in cadaveric spine segments of older adults.  33  Literature Review and Introduction to the Thesis 1.9  1.  REFERENCES  Osteoporosis prevention, diagnosis and therapy. N I H consensus statements,  2000:1-45.  2.  Hanley D A , Josse R G . Prevention and management of osteoporosis: consensus  statements from the Scientific Advisory Board of the Osteoporosis Society of Canada. 1. Introduction. C M A J 1996;155:921-3.  3.  Melton LJ, 3rd. Epidemiology of spinal osteoporosis. Spine 1997;22:2S-1 IS.  4.  Tenenhouse A , Joseph L, Kreiger N , et al. Estimation of the prevalence of low  bone density in Canadian women and men using a population-specific D X A reference standard: the Canadian Multicentre Osteoporosis Study (CaMos). Osteoporos Int 2000;11:897-904.  5.  Kelsey JL. Osteoporosis: Prevalence and incidence. Osteoporosis, Proceedings of  the NIH Consensus Development Conference, 1984:25-8.  6.  Adachi JD, Loannidis G, Berger C, et al. The influence of osteoporotic fractures  on health-related quality of life in community-dwelling men and women across Canada. Osteoporos Int 2001;12:903-8.  7.  Cooper C, Atkinson EJ, O'Fallon W M , et al. Incidence of clinically diagnosed  vertebral fractures: a population-based study in Rochester, Minnesota, 1985-1989. J Bone Miner Res 1992;7:221-7.  8.  Cuddihy MT, Gabriel SE, Crowson CS, et al. Forearm fractures as predictors of  subsequent osteoporotic fractures. Osteoporos Int 1999;9:469-75.  34  Literature Review and Introduction to the Thesis 9.  Gunter M J , Beaton SJ, Brenneman SK, et al. Management of osteoporosis in  women aged 50 and older with osteoporosis-related fractures in a managed care population. Dis Manag 2003;6:83-91.  10.  Mallmin H , Ljunghall S, Persson I, et al. Fracture of the distal forearm as a  forecaster of subsequent hip fracture: a population-based cohort study with 24 years of follow-up. Calcif Tissue Int 1993;52:269-72.  11.  van der Klift M , de Laet C E D H , McCloskey E V , et al. The incidence of vertebral  fractures in men and women: the Rotterdam Study. J Bone Miner Res 2002;17:1051-6.  12.  Seagger R, Howell J, David H , et al. Prevention of secondary osteoporotic  fractures-why are we ignoring the evidence? Injury 2004;35:986-8.  13.  Riggs B L , Melton LJ, 3rd. The prevention and treatment of osteoporosis. N Engl J  Med 1992;327:620-7.  14.  Khan K , McKay H , Kannus P, et al. eds. Physical Activity and Bone Health.  Champaign, IL: Human Kinetics, 2001.  15.  Van Staa TP, Leufkens HG, Abenhaim L, et al. Use of oral corticosteroids and  risk of fractures. Journal of Bone and Mineral Research 2000;15:993-1000.  16.  Sambrook PN. Glucocorticoid osteoporosis. Current Pharmaceutical Design  2002;8:1877-83.  17.  Sambrook PN. Corticosteroid osteoporosis: practical implications of recent trials.  Journal of Bone and Mineral Research 2000;15:1645-9.  18.  Caling B , Lee M . Effect of direction of applied mobilization force on the  posteroanterior response in the lumbar spine. J Manipulative Physiol Ther 2001;24:71-8.  35  Literature Review and Introduction to the Thesis 19.  Goeree ROB, Pettitt DB, Cuddy L, et al. A n assessment of the burden of illness  due to osteoporosis in Canada. J Soc Obstet Gynaecol Can 1996;18:15-24.  20.  Brown JP, Josse RG. 2002 clinical practice guidelines for the diagnosis and  management of osteoporosis in Canada. C M A J 2002; 167:S 1-34.  21.  Papadimitropoulos E A , Coyte PC, Josse RG, et al. Current and projected rates of  hip fracture in Canada. C M A J 1997;157:1357-63.  22.  Adachi JD, Ioannidis G, Pickard L, et al. The association between osteoporotic  fractures and health-related quality of life as measured by the Health Utilities Index in the Canadian Multicentre Osteoporosis Study (CaMos). Osteoporos Int 2003;14:895-904.  23.  Cockerill W, Lunt M , Silman A J , et al. Health-related quality of life and  radiographic vertebral fracture. Osteoporos Int 2004; 15:113-9.  24.  Kanis JA, Johnell O, Oden A , et al. The risk and burden of vertebral fractures in  Sweden. Osteoporos Int 2004;15:20-6.  25.  Frost H. Managing the skeletal pain and disability of  osteoporosis. Orthopedic Clinics of North America 1972;3:561-70.  26.  Lukert B. Osteoporosis—a review and update. Arch Phys Med Rehabil  1982;63:480-7.  27.  Matthis C, Weber U , O'Neill TW, et al. Health impact associated with vertebral  deformities: results from the European Vertebral Osteoporosis Study (EVOS). Osteoporos Int 1998;8:364-72.  36  Literature Review and Introduction to the Thesis 28.  Ismail A A , Cooper C, Felsenberg D, et al. Number and type of vertebral  deformities: epidemiological characteristics and relation to back pain and height loss. European Vertebral Osteoporosis Study Group. Osteoporos Int 1999;9:206-13.  29.  Ryan PJ, Blake G, Herd R, et al. A clinical profile of back pain and disability in  patients with spinal osteoporosis. Bone 1994;15:27-30.  30.  Galindo-Ciocon D, Ciocon JO, Galindo D. Functional impairment among elderly  women with osteoporotic vertebral fractures. Rehabil Nurs 1995;20:79-83.  31.  Lyles K W , Gold DT, Shipp K M , et al. Association of osteoporotic vertebral  compression fractures with impaired functional status. A m J Med 1993;94:595-601.  32.  Papaioannou A , Watts N B , Kendler D L , et al. Diagnosis and management of  vertebral fractures in elderly adults. A m J Med 2002; 113:220-8.  33.  Dionne CE, Chenard M . Back-related functional limitations among full-time  homemakers: a comparison with women employed full-time outside the home. Spine 2004;29:1375-82.  34.  Lynn SG, Sinaki M , Westerlind K C . Balance characteristics of persons with  osteoporosis. Arch Phys Med Rehabil 1997;78:273-7.  35.  Simmonds MJ, Kumar S, Lechelt E. Use of a spinal model to quantify the forces  and motion that occur during therapists' tests of spinal motion. Phys Ther 1995;75:21222.  36.  Schlaich C, Minne HW, Bruckner T, et al. Reduced pulmonary function in  patients with spinal osteoporotic fractures. Osteoporos Int 1998;8:261-7.  37  Literature Review and Introduction to the Thesis 37.  Cooper C. The crippling consequences of fractures and their impact on quality of  life. A m J Med 1997;103:12S-7S.  38.  Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone  mineral density predict occurrence of osteoporotic fractures. B M J 1996;312:1254-9.  39.  Melton LJ, 3rd, Atkinson EJ, Cooper C, et al. Vertebral fractures predict  subsequent fractures. Osteoporos Int 1999;10:214-21.  40.  Campbell AJ. Preventing fractures by preventing falls in older women. C M A J  2002;167:1005-6.  41.  Kannus P. Preventing osteoporosis, falls, and fractures among elderly people.  Promotion of lifelong physical activity is essential. B M J 1999;318:205-6.  42.  Bennell K, Khan K, McKay H. The role of physiotherapy in the prevention and  treatment of osteoporosis. Man Ther 2000;5:198-213.  43.  Malmros B , Mortensen L, Jensen M B , et al. Positive effects of physiotherapy on  chronic pain and performance in osteoporosis. Osteoporos Int 1998;8:215-21.  44.  Sran M M . Effects of exercise training/mechanical loading on bone -what  physiotherapists need to know. Physiother Can 2002;54:S12.  45.  Cowan S M , Bennell K L , Hodges PW, et al. Simultaneous feedforward  recruitment of the vasti in untrained postural tasks can be restored by physical therapy. J OrthopRes 2003;21:553-8.  46.  Crossley K, Bennell K , Green S, et al. Physical therapy for patellofemoral pain: a  randomized, double-blinded, placebo-controlled trial. A m J Sports Med 2002;30:857-65.  38  Literature Review and Introduction to the Thesis 47.  Dumoulin C, Lemieux M C , Bourbonnais D, et al. Physiotherapy for persistent  postnatal stress urinary incontinence: a randomized controlled trial. Obstet Gynecol 2004;104:504-10.  48.  Haines TP, Bennell K L , Osborne R H , et al. Effectiveness of targeted falls  prevention programme in subacute hospital setting: randomised controlled trial. B M J 2004;328:676.  49.  Hinman RS, Crossley K M , McConnell J, et al. Efficacy of knee tape in the  management of osteoarthritis of the knee: blinded randomised controlled trial. B M J 2003;327:135.  50.  Jull G, Trott P, Potter H, et al. A randomized controlled trial of exercise and  manipulative therapy for cervicogenic headache. Spine 2002;27:1835-43.  51.  Moseley L. Combined physiotherapy and education is efficacious for chronic low  back pain. Aust J Physiother 2002;48:297-302.  52.  Pfeifer M , Sinaki M , Geusens P, et al. Musculoskeletal rehabilitation in  osteoporosis: a review. J Bone Miner Res 2004;19:1208-14.  53.  Osteoporosis: review of the evidence for prevention, diagnosis and treatment and  cost-effectiveness analysis. Introduction. Osteoporos Int 1998;8 Suppl 4:Sl-80.  54.  Genant H K , Cooper C, Poor G, et al. Interim report and recommendations of the  World Health Organization Task-Force for Osteoporosis. Osteoporos Int 1999;10:259-64.  55.  Melton LJ, 3rd, Chrischilles EA, Cooper C, et al. Perspective. How many women  have osteoporosis? J Bone Miner Res 1992;7:1005-10.  39  Literature Review and Introduction to the Thesis 56.  Nevitt M C , Ettinger B, Black D M , et al. The association of radiographically  detected vertebral fractures with back pain and function: a prospective study. Ann Intern Med 1998;128:793-800.  57.  Ettinger B, Black D M , Mitlak B H , et al. Reduction of vertebral fracture risk in  postmenopausal women with osteoporosis treated with raloxifene: results from a 3-year randomized clinical trial. Multiple Outcomes of Raloxifene Evaluation (MORE) Investigators. Jama 1999;282:637-45.  58.  Black D M , Arden N K , Palermo L, et al. Prevalent vertebral deformities predict  hip fractures and new vertebral deformities but not wrist fractures. Study of Osteoporotic Fractures Research Group. J Bone Miner Res 1999;14:821-8.  59.  Lindsay R, Silverman SL, Cooper C, et al. Risk of new vertebral fracture in the  year following a fracture. J A M A 2001;285:320-3.  60.  De Smet A A , Robinson R G , Johnson BP, et al. Spinal compression fractures in  osteoporotic women: patterns and relationship to hyperkyphosis. Radiology 1988;166:497-500.  61.  Harma M , Heliovaara M , Aromaa A , et al. Thoracic spine compression fractures  in Finland. Clin Orthop 1986:188-94.  62.  Moro M , Hecker AT, Bouxsein M L , et al. Failure load of thoracic vertebrae  correlates with lumbar bone mineral density measured by D X A . Calcif Tissue Int 1995;56:206-9.  63.  Sinaki M , Itoi E, Rogers JW, et al. Correlation of back extensor strength with  thoracic kyphosis and lumbar lordosis in estrogen-deficient women. A m J Phys Med Rehabil 1996;75:370-4.  40  Literature Review and Introduction to the Thesis 64.  O'Neill TW, Felsenberg D, Varlow J, et al. The prevalence of vertebral deformity  in european men and women: the European Vertebral Osteoporosis Study. J Bone Miner Res 1996;11:1010-8.  65.  Jackson SA, Tenenhouse A , Robertson L. Vertebral fracture definition from  population-based data: preliminary results from the Canadian Multicenter Osteoporosis Study (CaMos). Osteoporos Int 2000; 11:680-7.  66.  Cooper C, Atkinson EJ, Kotowicz M , et al. Secular trends in the incidence of  postmenopausal vertebral fractures. Calcif Tissue Int 1992;51:100-4.  67.  Bono C M , Einhorn TA. Overview of osteoporosis: pathophysiology and  determinants of bone strength. Eur Spine J 2003; 12 Suppl 2:S90-6.  68.  Gehlbach SH, Burge RT, Puleo E, et al. Hospital care of osteoporosis-related  vertebral fractures. Osteoporos Int 2003;14:53-60.  69.  Burger H, Van Daele PL, Grashuis K, et al. Vertebral deformities and functional  impairment in men and women. J Bone Miner Res 1997;12:152-7.  70.  Ross PD. Clinical consequences of vertebral fractures. A m J Med 1997;103:30S-  42S.  71.  Huang C, Ross PD, Wasnich RD. Vertebral fracture and other predictors of  physical impairment and health care utilization. Arch Intern Med 1996;156:2469-75.  72.  Cooper C, Atkinson EJ, Jacobsen SJ, et al. Population-based study of survival  after osteoporotic fractures. A m J Epidemiol 1993;137:1001-5.  73.  Ettinger B, Black D M , Nevitt M C , et al. Contribution of vertebral deformities to  chronic back pain and disability. J Bone and Miner Res 1992;7:449-56.  41  Literature Review and Introduction to the Thesis  74.  Ensrud K E , Thompson DE, Cauley JA, et al. Prevalent vertebral deformities  predict mortality and hospitalization in older women with low bone mass. Fracture Intervention Trial Research Group. J A m Geriatr Soc 2000;48:241-9.  75.  Kado D M , Duong T, Stone K L , et al. Incident vertebral fractures and mortality in  older women: a prospective study. Osteoporos Int 2003;14:589-94.  76.  Cooper C, Melton JL. Magnitude and Impact of Osteoporosis and Fractures. In  Feldman M K ed. Osteoporosis: Academic Press, Inc., 1996:419-34.  77.  Sinaki M , Wollan PC, Scott RW, et al. Can strong back extensors prevent  vertebral fractures in women with osteoporosis? Mayo Clin Proc 1996;71:951-6.  78.  Greendale G A , DeAmicis TA, Bucur A , et al. A prospective study of the effect of  fracture on measured physical performance: results from the MacArthur Study—MAC. J A m Geriatr Soc 2000;48:546-9.  79.  Huang C, Ross PD, Wasnich RD. Vertebral fractures and other predictors of back  pain among older women. J Bone Miner Res 1996;11:1026-32.  80.  Frank JW, Brooker AS, DeMaio SE, et al. Disability resulting from occupational  low back pain. Part II: What do we know about secondary prevention? A review of the scientific evidence on prevention after disability begins. Spine 1996;21:2918-29.  81.  Kelsey JL, White A A , 3rd. Epidemiology and impact of low-back pain. Spine  1980;5:133-42.  82.  Volinn E. The epidemiology of low back pain in the rest of the world. A review of  surveys in low- and middle-income countries. Spine 1997;22:1747-54.  42  Literature Review and Introduction to the Thesis 83.  Bingefors K , Isacson D. Epidemiology, co-morbidity, and impact on health-  related quality of life of self-reported headache and musculoskeletal pain - a gender perspective. Eur J Pain 2004;8:435-50.  84.  Burchfiel C M , Boice JA, Stafford B A , et al. Prevalence of back pain and joint  problems in a manufacturing company. J Occup Med 1992;34:129-34.  85.  Fillingim RB, Doleys D M , Edwards RR, et al. Clinical characteristics of chronic  back pain as a function of gender and oral opioid use. Spine 2003;28:143-50.  86.  Sheir-Neiss G l , Kruse RW, Rahman T, et al. The association of backpack use and  back pain in adolescents. Spine 2003;28:922-30.  87.  Stranjalis G, Tsamandouraki K, Sakas DE, et al. Low back pain in a  representative sample of Greek population: analysis according to personal and socioeconomic characteristics. Spine 2004;29:1355-60.  88.  Reynolds D L , Chambers L W , Badley E M , et al. Physical disability among  Canadians reporting musculoskeletal diseases. J Rheumatol 1992;19:1020-30.  89.  Badley E M , Tennant A. Changing profile of joint disorders with age: findings  from a postal survey of the population of Calderdale, West Yorkshire, United Kingdom. Ann Rheum Dis 1992;51:366-71.  90.  Iron KS, Manuel D G , Williams J. Using a linked data set to determine the factors  associated with utilization and costs of family physician services in Ontario: effects of self-reported chronic conditions. Chronic Diseases in Canada 2003;24:124-32.  91.  Goetzel RZ, Hawkins K, Ozminkowski RJ, et al. The health and productivity cost  burden of the "top 10" physical and mental health conditions affecting six large U.S. employers in 1999. J Occup Environ Med 2003;45:5-14.  43  Literature Review and Introduction to the Thesis  92.  Hansson TH, Hansson E K . The effects of common medical interventions on pain,  back function, and work resumption in patients with chronic low back pain: A prospective 2-year cohort study in six countries. Spine 2000;25:3055-64.  93.  Fanuele JC, Birkmeyer NJ, Abdu W A , et al. The impact of spinal problems on the  health status of patients: have we underestimated the effect? Spine 2000;25:1509-14.  94.  Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine 2000;25:2940-52.  95.  Faas A , Chavannes A W , van Eijk JT, et al. A randomized, placebo-controlled trial  of exercise therapy in patients with acute low back pain. Spine 1993;18:1388-95.  96.  Hemmila H M . Quality of life and cost of care of back pain patients in Finnish  general practice. Spine 2002;27:647-53.  97.  Currie SR, Wang J. Chronic back pain and major depression in the general  Canadian population. Pain 2004;107:54-60.  98.  Leidig G, Minne HW, Sauer P, et al. A study of complaints and their relation to  vertebral destruction in patients with osteoporosis. Bone Miner 1990;8:217-29.  99.  Patel U , Skingle S, Campbell GA, et al. Clinical profile of acute vertebral  compression fractures in osteoporosis. Br J Rheumatol 1991;30:418-21.  100.  Nicholson PH, Haddaway MJ, Davie M W , et al. Vertebral deformity, bone  mineral density, back pain and height loss in unscreened women over 50 years. Osteoporos Int 1993;3:300-7.  44  Literature Review and Introduction to the Thesis 101.  Ettinger B, Black D M , Palermo L , et al. Kyphosis in older women and its relation  to back pain, disability and osteopenia: the study of osteoporotic fractures. Osteoporos Int 1994;4:55-60.  102.  Ryan PJ, Blake G, Herd R, et al. A clinical profile of back pain and disability in  patients with spinal osteoporosis. Bone 1994;15:27-30.  103.  Finsen V . Osteoporosis and back pain among the elderly. Acta Med Scand  1988;223:443-9.  104.  Peris P, Guanabens N , Monegal A , et al. Aetiology and presenting symptoms in  male osteoporosis. Br J Rheumatol 1995;34:936-41.  105.  Liu-Ambrose T, Eng JJ, Khan K M , et al. The influence of back pain on balance  and functional mobility in 65- to 75-year-old women with osteoporosis. Osteoporos Int 2002;13:868-73.  106.  Boyling JD, Palastanga N , Grieve GP eds. Grieve's Modern Manual Therapy: The  Vertebral Column. 2nd ed: W.B. Saunders Company, 1994.  107.  Giles L G F , Singer KP. Clinical Anatomy and Management of Thoracic Spine  Pained: Butterworth-Heinemann, 2000.  108.  Maitland GD, Banks K, English K, et al. Maitland's vertebral manipulation. Sixth  ed. Boston: Butterworth Heinemann, 2001.  109.  Phillips DR, Twomey LT. A comparison of manual diagnosis with a diagnosis  established by a uni-level lumbar spinal block procedure. Man Ther 2000; 1:82-7.  110.  Riddle D L . Measurement of accessory motion: critical issues and related  concepts. Phys Ther 1992;72:865-74.  45  Literature Review and Introduction to the Thesis 111.  Magarey M E . Examination of the Cervical and Thoracic Spine. In Grant R ed.  Physical Therapy of the Cervical and Thoracic Spine. 2nd ed. New York: Churchill Livingstone, 1994:442.  112.  Lee D. Manual Therapy for the Thorax. Delta: Delta Orthopaedic Physiotherapy  Clinic, 1994.  113.  2002 Human Resource Data (subset). Canadian Alliance of Physiotherapy  Regulators. Toronto, 2003.  114.  Grant R ed. Physical Therapy of the Cervical and Thoracic Spine. 2nd ed. New  York: Churchill Livingstone, 1994.  115.  Cyriax J. Textbook of Orthopaedic Medicine. 8th ed. London: Balliere Tindall,  1982.  116.  Jorgensson A . The iliopsoas muscle and the lumbar spine. Aust J Physiother  1993;39:125-32.  117.  M l G, Bullock M . Motion profile of the lumbar spine in an ageing population  assessed by manual examination. Physiotherapy Practice 1987;3:70-81.  118.  Binkley J, Stratford PW, Gill C. Interrater reliability of lumbar accessory motion  mobility testing. Phys Ther 1995;75:786-92.  119.  Lindsay D M , Meeuwisse W H , Mooney M E , et al. Interrater reliability of manual  therapy assessment techniques. Physiother Can 1995;47:173-80.  120.  Maher C, Adams R. Reliability of pain and stiffness assessments in clinical  manual lumbar spine examination. Phys Ther 1994;74:801-9.  46  Literature Review and Introduction to the Thesis 121.  Maher C, Adams R. Is the clinical concept of spinal stiffness multidimensional?  Phys Ther 1995;75:854-60.  122.  Chiradejnant A , Maher CG, Latimer J. Objective manual assessment of lumbar  posteroanterior stiffness is now possible. J Manipulative Physiol Ther 2003;26:34-9.  123.  Jull G, Bogduk N , Marsland A. The accuracy of manual diagnosis for cervical  zygapophysial joint pain syndromes. Med J Aust 1988;148:233-6.  124.  Phillips DR, Twomey LT. A comparison of manual diagnosis with a diagnosis  established by a uni-level lumbar spinal block procedure. Man Ther 1996;2:82-7.  125.  Lundberg G, Gerdle B. Correlations between joint and spinal mobility, spinal  sagittal configuration, segmental mobility, segmental pain, symptoms and disabilities in female homecare personnel. Scand J Rehab Med 2000;32:124-33.  126.  Aure OF, Nilsen JH, Vasseljen O. Manual therapy and exercise therapy in  patients with chronic low back pain: a randomized, controlled trial with 1-year follow-up. Spine 2003;28:525-31.  127.  Hoving JL, Koes B W , de Vet HC, et al. Manual therapy, physical therapy, or  continued care by a general practitioner for patients with neck pain. A randomized, controlled trial. Ann Intern Med 2002;136:713-22.  128.  Giles L G , Muller R. Chronic spinal pain: a randomized clinical trial comparing  medication, acupuncture, and spinal manipulation. Spine 2003;28:1490-502.  129.  Farrell JP, Twomey LT. Acute low back pain. Comparison of two conservative  treatment approaches. Med J Aust 1982; 1:160-4.  47  Literature Review and Introduction to the Thesis 130.  Goodsell M , Lee M , Latimer J. Short-term effects of lumbar posteroanterior  mobilization in individuals with low-back pain. J Manipulative Physiol Ther 2000;23:332-42.  131.  Sterling M , Jull G, Wright A. Cervical mobilisation: concurrent effects on pain,  sympathetic nervous system activity and motor activity. Man Ther 2001;6:72-81.  132.  Korthals-de Bos IB, Hoving JL, van Tulder M W , et al. Cost effectiveness of  physiotherapy, manual therapy, and general practitioner care for neck pain: economic evaluation alongside a randomised controlled trial. B M J 2003;326:911.  133.  Andersson G B , Lucente T, Davis A M , et al. A comparison of osteopathic spinal  manipulation with standard care for patients with low back pain. N Engl J Med 1999;341:1426-31.  134.  Bronfort G, Evans R, Nelson B, et al. A randomized clinical trial of exercise and  spinal manipulation for patients with chronic neck pain. Spine 2001;26:788-97.  135.  Cherkin DC, Deyo R A , Battie M , et al. A comparison of physical therapy,  chiropractic manipulation, and provision of an educational booklet for the treatment of patients with low back pain. N Engl J Med 1998;339:1021-9.  136.  Curtis P, Carey TS, Evans P, et al. Training primary care physicians to give  limited manual therapy for low back pain: patient outcomes. Spine 2000;25:2954-60.  137.  David J, Modi S, Aluko A A , et al. Chronic neck pain: a comparison of  acupuncture treatment and physiotherapy. Br J Rheumatol 1998;37:1118-22.  138.  Hsieh C Y , Adams A H , Tobis J, et al. Effectiveness of four conservative  treatments for subacute low back pain: a randomized clinical trial. Spine 2002;27:1142-8.  48  Literature Review and Introduction to the Thesis 139.  Hurwitz E L , Morgenstern H , Harber P, et al. A randomized trial of medical care  with and without physical therapy and chiropractic care with and without physical modalities for patients with low back pain: 6-month follow-up outcomes from the U C L A low back pain study. Spine 2002;27:2193-204.  140.  Jordan A , Bendix T, Nielsen H, et al. Intensive training, physiotherapy, or  manipulation for patients with chronic neck pain. A prospective, single-blinded, randomized clinical trial. Spine 1998;23:311-8.  141.  Gross A R , Kay T, Hondras M , et al. Manual therapy for mechanical neck  disorders: a systematic review. Man Ther 2002;7:131-49.  142.  Assendelft WJ, Morton SC, Y u EI, et al. Spinal manipulative therapy for low  back pain. A meta-analysis of effectiveness relative to other therapies. Ann Intern Med 2003;138:871-81.  143.  van Tulder M W , Koes B W , Bouter L M . Conservative treatment of acute and  chronic nonspecific low back pain. A systematic review of randomized controlled trials of the most common interventions. Spine 1997;22:2128-56.  144.  Koes B W , Assendelft WJ, van der Heijden GJ, et al. Spinal manipulation for low  back pain. A n updated systematic review of randomized clinical trials. Spine 1996;21:2860-71.  145.  McGuiness J, Vicenzino B, Wright A. Influence of a cervical mobilization  technique on respiratory and cardiovascular function. Man Ther 1997;2:216-20.  146.  Vicenzino B , Collins D, Benson H, et al. A n investigation of the interrelationship  between manipulative therapy-induced hypoalgesia and sympathoexcitation. J Manipulative Physiol Ther 1998;21:448-53.  49  Literature Review and Introduction to the Thesis 147.  Wright A , Sluka K A . Nonpharmacological treatments for musculoskeletal pain.  Clin J Pain 2001;17:33-46.  148.  Lee M , Latimer J, Maher C. Manipulation: investigation of a proposed  mechanism. Clin Biomech 1993;8:302-6.  149.  Tobis JS, Hoehler F. Musculoskeletal Manipulation: Evaluation of the Scientific  Evidence. Springfield, 111: Charles C Thomas Publisher, 1986.  150.  Grieve G. Common Vertebral Joint Problems. 2nd edition. New York: Churchill  Livingstone, 1988.  151.  Ernst E. Chiropractic spinal manipulation for back pain. Br J Sports Med  2003;37:195-6.  152.  Sran M M . Chiropractic spinal manipulation for back pain; commentary. Br J  Sports Med 2003;37:196.  153.  Haldeman S, Rubinstein S M . Compression fractures in patients undergoing spinal  manipulative therapy. J Manipulative Physiol Ther 1992;15:450-4.  154.  Bjornsdottir SV, Kumar S. Posteroanterior spinal mobilization: state of the art  review and discussion. Disabil Rehabil 1997;19:39-46.  155.  Lee M , Liversidge K. Posteroanterior stiffness at three locations in the lumbar  spine. J Manipulative Physiol Ther 1994; 17:511-6.  156.  Lee R, Evans J. A n in vivo study of the intervertebral movements produced by  posteroanterior mobilization. Clin Biomech 1997;12:400-8.  50  Literature Review and Introduction to the Thesis 157.  Edmondston SJ, Singer KP. Thoracic spine: anatomical and biomechanical  considerations for manual therapy. Man Ther 1997;2:132-43.  158.  Itoi E. Roentgenographic analysis of posture in spinal osteoporotics. Spine  1991;16:750-6.  159.  Harms M C , Bader DL. Variability of forces applied by experienced therapists  during spinal mobilization. Clin Biomech 1997;12:393-9.  160.  Harms M C , Innes S M , Bader DL. Forces measured during spinal manipulative  procedures in two age groups. Rheumatology (Oxford) 1999;38:267-74.  161.  Chiradejnant A , Latimer J, Maher C G . Forces applied during manual therapy to  patients with low back pain. J Manipulative Physiol Ther 2002;25:362-9.  162.  Gal J M , Herzog W, Kawchuk G N , et al. Forces and relative vertebral movements  during SMT to unembalmed post-rigor cadavers: peculiarities associated with joint cavitation. J Manip Physiol Ther 1995;18:4-9.  163.  Herzog W, Conway PJ, Kawchuk G N , et al. Forces exerted during spinal  manipulative therapy. Spine 1993;18:1206-12.  164.  Kirstukas SJ, Backman JA. Physician-applied contact pressure and table force  response during unilateral thoracic manipulation. J Manip Physiol Ther 1999;22:269-79.  165.  Genant H K , Jergas M . Assessment of prevalent and incident vertebral fractures in  osteoporosis research. Osteoporos Int 2003;14 Suppl 3:S43-55.  166.  Genant H K , Wu C Y , van Kuijk C, et al. Vertebral fracture assessment using a  semiquantitative technique. J Bone Miner Res 1993;8:1137-48.  51  Literature Review and Introduction to the Thesis 167.  Edmondston SJ, Singer KP, Price R l , et al. Accuracy of lateral dual energy X-ray  absorptiometry for the determination of bone mineral content in the thoracic and lumbar spine: an in vitro study. Br J Radiol 1993;66:309-13.  168.  Ho CP, K i m RW, Schaffler M B , et al. Accuracy of dual-energy radiographic  absorptiometry of the lumbar spine: cadaver study. Radiology 1990;176:171-3.  169.  Sabin M A , Blake G M , MacLaughlin-Black SM, et al. The accuracy of volumetric  bone density measurements in dual x-ray absorptiometry. Calcif Tissue Int 1995;56:2104.  170.  Martin RB. Determinants of the mechanical properties of bones. J Biomech  1991 ;24 Suppl 1:79-88.  171.  Currey JD. The mechanical consequences of variation in the mineral content of  bone. J Biomech 1969;2:1-11.  172.  Judex S, Boyd S, Yi-Xian Q, et al. Combining High-resolution Micro-computed  Tomography with Material Composition to Define the Quality of Bone Tissue. Current Osteoporosis Reports 2003 ;1:11-9.  173.  Katzman D K , Bachrach L K , Carter DR, et al. Clinical and Anthropometric  Correlates of Bone Mineral Acquisition in Healthy Adolescent Girls. J Clin Endocrinol Metab 1991;73:1332-39.  174.  Kroger H , Vainio P, Nieminen J, et al. Comparison of different models for  interpreting bone mineral density measurements using D X A and M R I technology. Bone 1995;17:157-9.  175.  Carter DR, Bouxsein M L , Marcus R. New approaches for interpreting projected  bone densitometry data. J Bone Miner Res 1992;7:137-45.  52  Literature Review and Introduction to the Thesis 176.  Kroger H , Kotaniemi A , Kroger L , et al. Development of bone mass and bone  density ofthe spine and femoral neck- a prospective study of 65 children and adolescents. Bone Miner 1993;23:171-82.  177.  Panjabi M M , Takata K , Goel V , et al. Thoracic human vertebrae. Quantitative  three-dimensional anatomy. Spine 1991;16:888-901.  178.  Panjabi M M , Goel V , Oxland T, et al. Human lumbar vertebrae. Quantitative  three-dimensional anatomy. Spine 1992;17:299-306.  179.  Rosen CJ. Pathophysiology of osteoporosis. Clin Lab Med 2000;20:455-68.  180.  Parfitt A M , Matthews C, Villaneuva A . Relationships between surface, volume,  and thickness of iliac trabecular bone in aging and osteoporosis. J Clin Invest 1983;72:1396-409.  181.  Guglielmi G, Gluer CC, Majumdar S, et al. Current methods and advances in  bone densitometry. Eur Radiol 1995;5:129-39.  182.  Sutton D. Radiology and Imaging for Medical Students. 7th ed. New York:  Churchill Livingstone, 1998.  183.  Kiuru M J , Haapamaki V V , Koivikko M P , et al. Wrist injuries; diagnosis with  multidetector CT. Emerg Radiol 2004;10:182-5.  184.  Dai L Y . Imaging diagnosis of thoracolumbar burst fractures. Chin Med Sci J  2004;19:142-4.  185.  Catalano L W , 3rd, Barron OA, Glickel SZ. Assessment of articular displacement  of distal radius fractures. Clin Orthop 2004:79-84.  53  Literature Review and Introduction to the Thesis 186.  Nakstad PH, Server A , Josefsen R. Traumatic cervical injuries in ankylosing  spondylitis. Acta Radiol 2004;45:222-6.  187.  Falchi M , Rollandi GA. CT of pelvic fractures. Eur J Radiol 2004;50:96-105.  188.  Sizensky JA, Marks R M . Imaging of the navicular. Foot Ankle Clin 2004;9:181-  209.  189.  Faulkner R A , McCulloch R G , Fyke SL, et al. Comparison of areal and estimated  volumetric bone mineral density values between older men and women. Osteoporos Int 1995;5:271-5.  190.  Genant H K , Gordon C, Jiang Y , et al. Advanced imaging of the macrostructure  and microstructure of bone. Horm Res 2000;54 Suppl 1:24-30.  191.  Einhorn TA. Bone strength: the bottom line. Calcif Tissue Int 1992;51:333-9.  192.  Muller R, Van Campenhout H, Van Damme B , et al. Morphometric analysis of  human bone biopsies: a quantitative structural comparison of histological sections and micro-computed tomography. Bone 1998;23:59-66.  193.  Muller R. Bone microarchitecture assessment: current and future trends.  Osteoporos Int 2003; 14 Suppl 5:89-99.  194.  Genant H K , Engelke K, Fuerst T, et al. Noninvasive assessment of bone mineral  and structure: state of the art. J Bone Miner Res 1996;11:707-30.  195.  Hildebrand T, Ruegsegger P. A new method for the model independent  assessment of thickness in three-dimensional images. J Microsc 1997;185:67-75.  54  Literature Review and Introduction to the Thesis 196.  Muller R, Bauss F, Smith S, et al. Mechano-structure relationships in normal,  ovariectomized and Ibandronate treated aged macaques as assessed by microtomographic imaging and biomechanical testing. Transactions of the 47th Annual Meeting of the Orthopaedic Research Society. San Francisco, California, 2001:66.  197.  Hayes WC, Bouxsein M L . Biomechanics of Cortical and Trabecular Bone. In  Mow V C , Hayes WC eds. Basic Orthopaedic Biomechanics. Second ed. Philadelphia: Lippincott Williams & Wilkins, 1997.  198.  Cheng X G , Lowet G, Boonen S, et al. Prediction of vertebral and femoral strength  in vitro by bone mineral density measured at different skeletal sites. J Bone Miner Res 1998;13:1439-43.  199.  Burklein D, Lochmuller E, Kuhn V , et al. Correlation of thoracic and lumbar  vertebral failure loads with in situ vs. ex situ dual energy X-ray absorptiometry. J Biomech 2001;34:579-87.  200.  Singer K , Edmondston S, Day R, et al. Prediction of thoracic and lumbar vertebral  body compressive strength: correlations with bone mineral density and vertebral region. Bone 1995;17:167-74.  201.  Andriacchi TP, Natarajan R N , Hurwitz DE. Musculoskeletal Dynamics,  Locomotion, and Clinical Applications. In Mow V C , Hayes W C eds. Basic Orthopaedic Biomechanics. Philadelphia: Lippincott Williams & Wilkins, 1997.  202.  Frobin W, Leivseth G, Biggemann M , et al. Sagittal plane segmental motion of  the cervical spine. A new precision measurement protocol and normal motion data of healthy adults. Clin Biomech (Bristol, Avon) 2002;17:21-31.  203.  Lee R, Evans J. Load-displacement-time characteristics of the spine under  posteroanterior mobilization. Aust J Physiother 1992;38:115-23.  55  Literature Review and Introduction to the Thesis 204.  Dupuis PR, Yong-Hing K , Cassidy JD, et al. Radiologic diagnosis of degenerative  lumbar spinal instability. Spine 1985;10:262-76.  205.  McGregor A H , Wragg P, Gedroyc W M . Can interventional M R I provide an  insight into the mechanics of a posterior-anterior mobilisation? Clin Biomech (Bristol, Avon) 2001;16:926-9.  206.  White A , Panjabi M . Clinical biomechanics of the spine. Second ed. Philadelphia:  JB Lippincott Company, 1990.  207.  Panjabi M M , White A A , 3rd, Brand R A , Jr. A note on defining body parts  configurations. J Biomech 1974;7:385-7.  208.  Edmondston SJ, Allison GT, Althorpe B M , et al. Comparison of ribcage and  posteroanterior thoracic spine stiffness: an investigation of the normal response. Man Ther 1999;4:157-62.  209.  Lee M , Steven GP, Crosbie J, et al. Variations in posteroanterior stiffness in the  thoracolumbar spine: preliminary observations and proposed mechanisms. Phys Ther 1998;78:1277-87.  210.  Shirley D, Ellis E, Lee M . The response of posteroanterior lumbar stiffness to  repeated loading. Man Ther 2002;7:19-25.  211.  Shirley D, Hodges PW, Eriksson A E , et al. Spinal stiffness changes throughout  the respiratory cycle. J Appl Physiol 2003;95:1467-75.  212.  Oxland TR, Lund T, Jost B, et al. The relative importance of vertebral bone  density and disc degeneration in spinal flexibility and interbody implant performance. A n in vitro study. Spine 1996;21:2558-69.  56  Literature Review and Introduction to the Thesis 213.  Oxland TR, Panjabi M M . The onset and progression of spinal injury: a  demonstration of neutral zone sensitivity. J Biomech 1992;25:1165-72.  214.  Panjabi M M . The stabilizing system of the spine. Part II. Neutral zone and  instability hypothesis. J Spinal Disord 1992;5:390-6.  215.  Panjabi M M , Brand R A , Jr., White A A , 3rd. Three-dimensional flexibility and  stiffness properties of the human thoracic spine. J Biomech 1976;9:185-92.  216.  Grieve GP. Mobilisation of the Spine. Edinburgh: Churchill Livingstone, 1984.  217.  Moseley L. Unraveling the barriers to reconceptualization of the problem in  chronic pain: the actual and perceived ability of patients and health professionals to understand the neurophysiology. J Pain 2003;4:184-9.  218.  Hides JA, Richardson C A , Jull GA. Magnetic resonance imaging and  ultrasonography of the lumbar multifidus muscle. Comparison of two different modalities. Spine 1995;20:54-8.  219.  Hides JA, Richardson C A , Jull GA. Multifidus muscle recovery is not automatic  after resolution of acute, first-episode low back pain. Spine 1996;21:2763-9.  220.  Hides JA, Stokes MJ, Saide M , et al. Evidence of lumbar multifidus muscle  wasting ipsilateral to symptoms in patients with acute/subacute low back pain. Spine 1994;19:165-72.  221.  Hodges P, Kaigle Holm A , Holm S, et al. Intervertebral stiffness of the spine is  increased by evoked contraction of transversus abdominis and the diaphragm: in vivo porcine studies. Spine 2003;28:2594-601.  57  Literature Review and Introduction to the Thesis 222.  Hodges P, Richardson C, M l G. Evaluation of the relationship between  laboratory and clinical tests of transversus abdominis function. Physiother Res Int 1996;1:30-40.  223.  Hodges PW. Is there a role for transversus abdominis in lumbo-pelvic stability?  Man Ther 1999;4:74-86.  224.  Hodges PW, Cresswell A G , Daggfeldt K, et al. In vivo measurement of the effect  of intra-abdominal pressure on the human spine. J Biomech 2001;34:347-53.  225.  Hodges PW, Richardson CA. Inefficient muscular stabilization of the lumbar  spine associated with low back pain. A motor control evaluation of transversus abdominis. Spine 1996;21:2640-50.  226.  Hodges PW, Richardson C A . Feedforward contraction of transversus abdominis  is not influenced by the direction of arm movement. Exp Brain Res 1997;114:362-70.  227.  Hodges PW, Richardson C A . Delayed postural contraction of transversus  abdominis in low back pain associated with movement of the lower limb. J Spinal Disord 1998;11:46-56.  228.  Hodges PW, Richardson C A . Transversus abdominis and the superficial  abdominal muscles are controlled independently in a postural task. Neurosci Lett 1999;265:91-4.  229.  Niemisto L, Lahtinen-Suopanki T, Rissanen P, et al. A randomized trial of  combined manipulation, stabilizing exercises, and physician consultation compared to physician consultation alone for chronic low back pain. Spine 2003;28:2185-91.  58  Effectiveness of Manual Therapy  Chapter Two Evidence for the Effectiveness of Manual Therapy for Spinal Pain— a systematic review.  59  Effectiveness of Manual Therapy ABSTRACT Background. In recent randomized clinical trials (RCTs) manual therapy was more effective than other methods of conservative management of spinal pain, yet earlier randomized controlled trials, systematic reviews and meta-analyses concluded manual therapy was not superior to other standard treatments for patients with back or neck pain. Objectives. To systematically review randomized clinical trials comparing manual therapy including either spinal joint mobilization (with or without manipulation) or manipulation alone, with other conservative treatments for back or neck pain. A secondary goal was to explore possible explanations for apparently inconsistent findings. Methods. I searched Medline, Cinahl, and Embase databases for randomized clinical trials comparing manual therapy including either spinal joint mobilization (with or without manipulation) or manipulation alone with other conservative treatments for back or neck pain. Only studies published as full papers, in English, between January 1, 1998 and December 31, 2003 were included. Results. Thirteen studies met the inclusion criteria and five reported positive results. Four of the five studies with positive results used manual therapy in combination with another physiotherapy treatment mode. Assessment and treatment protocols used in RCTs did not always mirror clinical practice guidelines and the dose of manual therapy used varies greatly among studies. Physiotherapy including manual therapy at a dose of 30-45 minutes per session, for 4-8 weeks was found to be effective in adult populations with back or neck pain.  60  Effectiveness of Manual Therapy Discussion. There are clinically relevant differences between studies reporting positive results of manual therapy and those reporting no significant difference over other conservative treatments. Conclusions. Interventions based on 'best practice' guidelines or textbooks written by experts appear to be more successful, and physiotherapy that includes manual therapy at a dose of 30-45 minutes per session, for 4-8 weeks is effective in adult populations with back or neck pain.  61  Effectiveness of Manual Therapy 2.1  INTRODUCTION  In recent randomized clinical trials (RCTs), manual therapy was more effective than other methods of conservative management for low back and neck pain. hand, some older randomized clinical trials  6 - 1 3  systematic reviews  1 4  1-5  On the other  and meta-analyses  15  found no evidence that spinal manual therapy was superior to other standard treatments for patients with low back or neck pain.  A n updated systematic review on this topic is necessary for two reasons. First, previous reviews and meta-analyses, including one published in 2003, only included studies 15  published in 2000 or earlier. Clinical practice evolves in response to new scientific evidence and since the late 1990s physiotherapy practice has changed considerably. For example, spinal segmental motor control and stabilization research has particularly influenced the physiotherapy management of low back pain through the addition of specific low load exercise therapy.  1 6 - 2 5  Another example is the improved understanding  of pain physiology and the influence of neurophysiology education.  These clinical  advances were not incorporated into clinical trials undertaken before the late 1990s. Thus, there is a need for a systematic review of more recent trials of spinal manual therapy that are more likely to accurately represent current clinical practice.  With good reason, previous systematic reviews critically evaluated methodologic factors such as the randomization procedure, appropriate blinding of patients and outcome 77 78  assessments, adequate follow-up period, and dropouts.  '  Although these factors are  undoubtedly important, the quality of the intervention itself must also receive close  62  Effectiveness of Manual Therapy scrutiny. Yet analysis of the intervention is a significant omission among previous systematic reviews.  Therefore, the primary objective of this review was to systematically identify, select and critically appraise randomized clinical trials, published between 1998 and 2003, that compared spinal manual therapy including spinal joint mobilization (with or without manipulation) or manipulation alone, with other conservative treatments for back or neck pain. The secondary objective was to explore possible reasons for the apparently inconsistent findings among various manual therapy trials.  63  Effectiveness of Manual Therapy 2.2 METHODS Definitions The term 'manual therapy' is used in a wide range of ways with a wide range of meanings, but for this review it includes manually performed assessment and treatment methods (which can include joint, neural tissue, and/or muscle techniques).  2 9 - 3 3  Inclusion criteria The systematic review was restricted to randomized clinical trials comparing manual therapy including either spinal joint mobilization (with or without manipulation) or manipulation alone with other conservative treatments for acute or chronic back or neck pain. Only studies published as full papers, in English, between January 1, 1998 and December 31, 2003 were included. Pilot studies were not included.  Search strategy I searched Medline, Cinahl, and Embase databases for randomized clinical trials comparing manual therapy including either spinal joint mobilization (with or without manipulation) or manipulation alone with other conservative treatments for back or neck pain. Search strategies for each database are outlined in Table 2.1.  64  Effectiveness of Manual Therapy Table 2.1. Search strategy.  Database  MeSH Headings  Limits  Medline  Manipulation, Orthopedic  Human  Manipulation, Chiropractic  English  Manipulation, Osteopathic  1998-2003  Physical Therapy Techniques Musculoskeletal Manipulations Comparative Study (back or neck) and pain  Cinahl  manual therapy  English  chiropractic  Clinical Trial  chiropractic manipulation  1998-2003  Manipulation, Orthopedic osteopathy (back or neck) and pain  Embase  manipulative medicine  Human  (back or neck) and pain  English 1998-2003  Search results Thirteen studies met the inclusion criteria (Table 2.2). One study of bone-setting by Finnish folk healers who lacked formal education was excluded as all other studies 34  involved formally educated professionals.  65  Table 2.2. Summary of studies reviewed. MT= manual therapy, LBP= low back pain, PT=physiotherapy, GP= general practitioner Dose  Results  Effect Size for Positive Studies  Population (n)  MT limited to Manipulation Only  Manual Treatment Delivered By:  Clinically Relevant/Guideline Based Manual Treatment  Aure OF, Nilsen JH, Vasseljen 0. Spine 2003;28:525-31.  20-60 yrs; chronic LBP>8 wks, less than 6 mos (49)  No  Physical Therapists  Yes  1. Manual Therapy plus Exercise Therapy 2. Exercise Therapy alone  No control group  45 min. (15 min MT);2 sessions/wk, 8 wks  Significantly larger improvements in MT group (maintained at 1 year follow-up)  .78  Hoving JL, Koes BW, deVetHC, et al. Ann Intern Med 2002;136:713-22  18-70 yrs; pain or stiffness in the neck for at least 2 wks (183)  No  Physical Therapists  Yes  1. Manual therapy plus specific exercise training 2. Active exercise focused physical therapy 3. Continued care by GP  Continued care by a GP  45 min.; 1 session/wk for up to 6 wks  Physical Therapy including MT more effective than Physical Therapy without MT or continued care by a GP  Not given  Jull G, Trott P, Potter H, etal. Spine 2002;27:183543.  18-60 years; cervicogenic headache at least 1X/wkfor a period of 2mos-10yrs (200)  No  Physical Therapists  Yes  1. Manual Therapy 2. Exercise Therapy (low load endurance training) 3. Combined Manual Therapy and Exercise Therapy 4. Control  No treatment control group  30 min, 8-12. sessions,6 wks  MT as effective as ET and both significantly better than control  .80  Moseley L Aust J Physiother 2002;48:297-302  Chronic low back pain> 2 mos. (57)  No  Physical Therapists  Yes with respect to clinical relevance (individualized and variety of techniques allowed) but no references cited for manual therapy techniques.  1. Manual therapy, specific exercise training, and neurophysiology education 2. Medical management by GP  Management byaGP  2x/wk, 4 wks  Combined physiotherapy treatment including MT, specific exercise training, and neurophysiology education resulted in improved function and pain at land 12 mos.  Not given  Giles LG, Muller R. Spine 2003;28:1490502  17 years or older; mechanical back or neck pain for a minimum of 13 wks (115)  Yes  Chiropractors  Yes (for low back pain). No (for neck pain).  1. Spinal manipulation 2. Sports physician follow up (limited) and medication 3. Acupuncture (needle)  No control group  20 min, 2x/wk, maximum 9 wks  Greater short term benefit for back pain with manipulation, but not for neck pain. Acupuncture more effective for neck pain.  Not given  Andersson GB, Lucente T, Davis AM, etal. N Engl JMed 1999:341:142631.  20-59 years; LBP lasting at least 3 weeks but less than 6 mos (178)  No  Osteopaths  Yes  'standard care' by physicians which could include active PT  1x/wkfor4wks then 1x/2wksfor 8 wks  No significant difference between groups. Both groups improved.  66  Interventions/Groups  Control Group  Author, yr  1. osteopathic treatment 2. 'standard care' by physicians  (Manual Therapy or Manipulation)  Author, yr  Population (n)  MT Limited to Manipulation Only  Manual Treatment Delivered By:  Clinically Relevant/Guideline Based Manual Treatment  Bronfort G, Evans R, Nelson B, etal. Spine 2001;26:788-97  20-65 years; mechanical neck pain for at least 12 weeks (191)  Yes (but this group also received 45 min of sham microcurrent therapy)  Chiropractors  No. A reference for the use of spinal manipulation for low back pain is cited, but only cervical and thoracic spine techniques were used.  Cherkin DC, Deyo RA, Battie M, etal. N Engl J Med 1998;339:1021-9.  20-64 years; LBP minimum 7 days after seeing physician, (321)  Yes  Chiropractors  Curtis P, Carey TS, Evans P, et al. Spine 2000;25:2954-60  21-65 yrs; acute LBP of less than 2 mos (295)  No (Manipulation plus muscle energy techniques)  Hsieh CY, Adams AH, Tobis J, etal. Spine 2002;27:1142-8.  18 yrs or older; LBP>3 wks and less than 6 mos (200)  Hurwitz EL, Morgenstern H, HarberP, etal. Spine 2002;27:2193-204  Control Group  Dose (Manual Therapy or Manipulation)  Results  1. Spinal manipulation plus upper body and neck strengthening exercise 2. Aerobic exercise plus MedX cervical extension and rotation machine 3. Spinal manipulation  No control group  20 1 hour sessions over 11 wks  No significant difference between groups with respect to pain, neck disability, medication use.  No, side lying only  1. Chiropractic manipulation 2. Education Booklet 3. McKenzie exercises  No control group  Up to 9x over 1 month  No significant difference between groups.  Physicians with limited training (18 hrs) in manual therapy  No  1. Manipulation and Muscle Energy Techniques plus Enhanced Care 2. Enhanced care alone  No control group  Initial plus 4 followups; 2x/wkfor2wks  Only 43% of patients in the MT group actually received the planned treatment; no significant difference between groups  Yes  Chiropractors  No, limited techniques  1. Back School 2. Myofascial Therapy 3. Joint Manipulation 4. Combined Joint Manipulation & Myofascial Therapy  No control group  3x/wk for 3 wks  All groups improved; no significant between-group differences at 3 or 6 months  18 years or older; LBP (681)  Yes  Chiropractors  Yes  1. Medical care only 2. Chiropractic care only 3. Medical care with limited physical therapy 4. Chiropractic care with modalities  No control group  Treatment dose not prescribed  Chiropractic no better than other groups; Physical therapy plus medical care group had less pain at 6 wks and 6 mos than medical care only  Jordan A, Bendix T, Nielsen H, etal. Spine 1998;23:311-8  20-60 yrs; chronic neck pain at least 3 mos (167)  No  Mobilization by Physical Therapists; Manipulation by Chiropractors  Yes with respect to clinical relevance (individualized) but no references cited for mobilization or manipulation techniques.  1. Manipulation 2. Physiotherapy without manipulation 3. Strength training (with a focus on neck muscle training)  No control group  David J, Modi S, Aluko AA, et al. Br J Rheumatol 1998;37:1118-22.  18-75 yrs; neck pain> 6 wks duration (70)  No  Physical Therapists  Not Clear  1. Physiotherapy 2. Acupuncture  No control group  67  Interventions/Groups  Physiotherapy: 30 min, 2x/wk, 6 wks  No significant difference between groups.  Chiropractic: 15-20 min, 2x/wk, 6 wks 1x/wk, 6 wks (maximum)  No significant difference between acupuncture and physiotherapy groups. Both groups improved.  Effectiveness of Manual Therapy 2.3 RESULTS Table 2.2 summarises descriptive information on the studies. Examining the trials for homogeneity revealed that the mean age of participants was similar amongst the studies and most participants were of Caucasian origin (with the exception of two studies ' ). 1  2  One study had a 'no treatment' control group. Two studies had a 'standard care' control group, consisting of treatment by a general practitioner. ' The other ten studies did not 4 5  have a 'no treatment' or 'standard care' control group (Table 2.2).  Less than one third of the studies reviewed reported prospective power calculations. ' " 3 6  One study reported what appears to be retrospective power.  8  9  Primary Analysis: Effectiveness of Manual Therapy In seven of the thirteen studies analysed, there was no difference between the manual therapy group and other treatment groups (Table 2.2). Four of these seven studies used manipulation only, one compared physiotherapy with acupuncture, and in one, the practitioners administering the intervention were physicians with 18 hours training in manual therapy.  Five of the thirteen studies in the past six years (1998-2003), however, reported that manual therapy was superior to exercise therapy alone, continued care by a general practitioner, sports physician follow-up plus medication, acupuncture and no treatment. " 3  6,10  68  Effectiveness of Manual Therapy Secondary Analysis: Differences Between Effective and Ineffective Interventions Manual therapy vs. manipulation onlv Four of the thirteen studies reported better results in the manual therapy group as compared with the other group(s). " Five of the remaining nine studies used manipulation only and all but one reported no significant difference or a poorer 10  response as compared with the other group(s). ' ' ' 2  7  9  n  Time per session Only six studies reported the time per session. Time varied from 20-60 minutes per treatment. Three of the five studies with positive results allowed between 30 and 45 minutes per treatment. One (of the studies with positive results) did not report treatment time and the other had mixed results (positive for back pain but not for neck pain) and 5  allowed 20 minutes per treatment.  1 0  Total number of sessions The total number of sessions varied from five to twenty, with a frequency of between once per week and three times per week. Some studies did not prescribe a maximum or minimum number of sessions per week (Table 2.2).  Number of weeks of treatment The number of weeks of treatment varied from three to twelve. The five studies with positive results used between four and nine weeks of treatment. " ' 3  6 10  69  Effectiveness of Manual Therapy Combining manual therapy with other physiotherapy treatment modes Four ofthe five studies with positive results used manual therapy in combination with another aspect of physiotherapy treatment (exercise therapy,  4 , 6  specific exercise  training, ' and neurophysiology education ). 3 5  5  70  Effectiveness of Manual Therapy 2.4. DISCUSSION This systematic review confirmed that a number of recent studies found that manual therapy, in particular physiotherapy that included spinal mobilization, was superior to a range of comparison therapies. Physiotherapy including spinal mobilization was found to reduce pain, improve function, improve spinal range of motion, improve general health, and achieve earlier return to w o r k .  3-6  While this supports many clinicians' impression  that manual therapy is effective in certain patients, there are sufficient data to permit critical analysis of at least five differences between effective and ineffective interventions, and to speculate as to how these differences may underpin the reported results.  Trial Intervention: Manual Therapy or Manipulation Alone This systematic review revealed that interventions that consisted of a variety of manual therapy techniques, rather than joint manipulation alone, appeared to yield better results. For example, Jull et a l compared the effectiveness of (i) manual therapy delivered by 3  physiotherapists, (ii) specific exercise therapy delivered by physiotherapists, (iii) combined manual and specific exercise therapy, and a (iv) no treatment control group, for treatment of cervicogenic headache. At 12-month follow up both manual therapy groups and also the specific exercise group had significantly reduced headache frequency and intensity, neck pain, and disability. In this study, manual therapy included both lowvelocity cervical joint mobilization techniques and high-velocity manipulation techniques. These results are relevant to physiotherapists with postgraduate certification in manual therapy as they receive extensive training in both of these techniques.  71  Effectiveness of Manual Therapy Similarly, Hoving et al compared physiotherapy including manual therapy with 4  physiotherapy without manual therapy for individuals with chronic neck pain. Of note, these authors allowed the use of low-velocity joint mobilizations but no high-velocity low amplitude thrust techniques (manipulation).  Trial Intervention: Guideline Based or Apparently Empirical? Assessment and treatment protocols used in RCTs do not always mirror clinical practice guidelines, which are typically set out in textbooks or published as guidelines written by experts in the field and based on current available evidence. Some studies tested protocols that differed substantially from clinical practice. For example, Andersson et al. compared osteopathic treatment (including manual therapy) with 'standard care' by 1  physicians. However the reported 'standard care' included medication, active physiotherapy, ultrasonography, diathermy, hot or cold packs (or both), use of a corset, or transcutaneous electrical nerve stimulation (TENS)! Clearly health maintenance organization (HMO) physicians have neither the time (45 minutes), equipment, or skills (i.e. active physiotherapy) to provide this treatment. Further, two studies of manipulation by chiropractors included participants with back or neck pain, yet the intervention protocol they employed appeared to be directed exclusively toward low back pain management.  1 0  '  U  Three studies used very restricted manual assessment and/or treatment  techniques " which do not reflect'best practice'. Three of the five studies with positive results used manual treatment (by physiotherapists) based on published guidelines or clinical textbooks written by experts in the field. ' ' 3  4  6  72  Effectiveness of Manual Therapy Association Between the Dose of Manual Therapy or Manipulation (minutes, sessions, weeks) and Outcome As with pharmacological therapy, the dose-response of therapy warrants examination. Time devoted to therapy per treatment session, the number of treatment sessions, and the number of weeks of treatment, are all important factors for therapists, patients and payers.  Treatment duration would influence not only cost-effectiveness, but may also impact the effectiveness of manual therapy treatment. Despite the potential clinical importance of dose of therapy, doses varied substantially between protocols. One study compared chiropractic care only, medical care only, medical care with limited physiotherapy, and chiropractic care with modalities but did not prescribe a treatment dose.  The authors  monitored utilization of the various treatment modes and time per session and found that 1/3 of patients randomly assigned to medical care with physiotherapy had no physiotherapy visits, and 20% of patients in the chiropractic groups received concurrent medical care, whereas only 7% of patients in the medical care groups received concurrent chiropractic care. They also reported that chiropractors and medical providers in their study spent an average of 15 minutes with patients at each visit, and physiotherapists averaged 31 minutes per patient visit.  Single or Combined Therapy and Outcome A number of the studies I reviewed investigated a combination of therapies such as care provided by two health care professionals or a combination of manual therapy or manipulation with another mode of treatment. Of note, four of the five studies with  73  Effectiveness of Manual Therapy positive results used manual therapy in combination with another aspect of physiotherapy treatment (exercise therapy ' , specific exercise training ' , and neurophysiology 4  6  3 5  education ). Similar positive results were not seen in chiropractic studies of spinal 5  manipulation combined with exercise ' or modalities. 9 11  2  Methodological Factors This critical appraisal also examined two key methodological factors that can influence RCT findings. First, the presence or absence of a control group is an important factor. Of the five studies with positive results, one had a 'no treatment' control and two used 3  'standard care' by a general practitioner as the control group (Table 2.2). In one of these studies the 'continued care by a GP' group received more than routine GP care, including ergonomic advice, advice on psychosocial issues and home exercises.  4  In the second  study, 6/28 participants in the control group received weekly manipulations from their general practitioner.  5  One study reported using a 'standard care by physicians' control group but close scrutiny revealed that this group also received active physiotherapy, ultrasonography, diathermy, hot or cold packs (or both), use of a corset, and/or transcutaneous electrical nerve stimulation (TENS).  1  Lack of a robust control group is a common problem among clinical trials in manual therapy, resulting in many sources of bias. Many factors that are unrelated to the treatment, such as the therapist's enthusiasm, patient-provider rapport and patient and  74  Effectiveness of Manual Therapy  health care provider expectations are all thought to contribute to the therapeutic effect.  12  15  Second, an important issue when examining discordant outcomes of RCTs is power  1 6  as  underpowered studies can lead to Type II error. Fewer than one third of the studies reported prospective power calculations ' " and one study reported what appears to be retrospective power . Experts suggest that it is illogical to calculate retrospective power, 9  and recommend that researchers report effect sizes for both significant and nonsignificant findings to evaluate whether the sample size might have been too small to 17  detect a real effect.  Conclusions  My primary objective was to systematically identify, select and critically appraise randomized clinical trials comparing spinal manual therapy including spinal joint mobilization (with or without manipulation) or manipulation alone, with other conservative treatments for back or neck pain. I found clinically relevant differences between studies reporting positive results of manual therapy and those reporting no significant difference over other conservative treatments. Specifically, those interventions in which the treatment protocol reflected what therapists do in clinical practice (i.e. using more than one manual therapy technique or combining manual therapy with other modes of treatment such as specific exercise training) appeared to be consistently more effective than those using manipulation or exercise therapy alone. Interventions based on 'best practice' guidelines were effective, and physiotherapy 75  Effectiveness of Manual Therapy including manual therapy at a dose of 30-45 minutes per session, for 4-8 weeks was effective in reducing pain and improving function.  76  Effectiveness of Manual Therapy  2.5.  1.  R E F E R E N C E S  Andersson G B , Lucente T, Davis A M , et al. A comparison of osteopathic spinal  manipulation with standard care for patients with low back pain. N Engl J Med 1999;341:1426-31.  2.  Hurwitz E L , Morgenstern H, Harber P, et al. A randomized trial of medical care  with and without physical therapy and chiropractic care with and without physical modalities for patients with low back pain: 6-month follow-up outcomes from the U C L A low back pain study. Spine 2002;27:2193-204.  3.  Jull G, Trott P, Potter H, et al. A randomized controlled trial of exercise and  manipulative therapy for cervicogenic headache. Spine 2002;27:1835-43.  4.  Hoving JL, Koes B W , de Vet HC, et al. Manual therapy, physical therapy, or  continued care by a general practitioner for patients with neck pain. A randomized, controlled trial. Ann Intern Med 2002;136:713-22.  5.  Moseley L. Combined physiotherapy and education is efficacious for chronic low  back pain. Aust J Physiother 2002;48:297-302.  6.  Aure OF, Nilsen JH, Vasseljen O. Manual therapy and exercise therapy in  patients with chronic low back pain: a randomized, controlled trial with 1-year follow-up. Spine 2003;28:525-31.  7.  Cherkin DC, Deyo RA, Battie M , et al. A comparison of physical therapy,  chiropractic manipulation, and provision of an educational booklet for the treatment of patients with low back pain. N Engl J Med 1998;339:1021-9.  8.  Curtis P, Carey TS, Evans P, et al. Training primary care physicians to give  limited manual therapy for low back pain: patient outcomes. Spine 2000;25:2954-60.  77  Effectiveness of Manual Therapy 9.  Hsieh C Y , Adams A H , Tobis J, et al. Effectiveness of four conservative  treatments for subacute low back pain: a randomized clinical trial. Spine 2002;27:1142-8.  10.  Giles L G , Muller R. Chronic spinal pain: a randomized clinical trial comparing  medication, acupuncture, and spinal manipulation. Spine 2003;28:1490-502.  11.  Bronfort G, Evans R, Nelson B, et al. A randomized clinical trial of exercise and  spinal manipulation for patients with chronic neck pain. Spine 2001;26:788-97.  12.  Blumenfeld A , Tischio M . Center of excellence for headache care: group model at  Kaiser Permanente. Headache 2003;43:431-40.  13.  Carosella A M , Lackner J M , Feuerstein M . Factors associated with early discharge  from a multidisciplinary work rehabilitation program for chronic low back pain. Pain 1994;57:69-76.  14.  Egbert L D , Battit GE, Welch CE, et al. Reduction of Postoperative Pain by  Encouragement and Instruction of Patients. A Study of Doctor-Patient Rapport. N Engl J Med 1964;270:825-7.  15.  Shapiro A K , Shapiro E. Patient-provider relationships and the placebo effect. In  Matarazzo JD, Weiss S M , Herd JA, et al. eds. Behavioral Health: A Handbook of Health Enhancement and Disease Prevention. New York: Wiley-Interscience, 1984:371-83.  16.  Glantz SA. Primer of biostatistics. Fifth Edition ed. New York: McGraw-Hill,  2001.  17.  Zumbo B, Hubley A. A note on misconceptions concerning prospective and  retrospective power. Statistician 1998;47:385-8.  78  Practice Behaviors and Clinicians' Perceptions  Chapter Three Physiotherapy and Osteoporosis: Practice Behaviors and Clinicians' Perceptions—a survey.  79  Practice Behaviors and Clinicians' Perceptions ABSTRACT Background. Physiotherapists typically use a variety of modes to treat their clients, including manual therapy. The literature cautions against the use of manual therapy in individuals with osteoporosis, yet clinical experience and published cases suggest that these techniques are used by at least some, if not many, clinicians. Objectives. To measure the most common treatment modes used by a random sample of physiotherapists practicing in the province of British Columbia (BC) in the treatment of individuals with osteoporosis. To assess whether physiotherapists in B C have concerns about the use of manual therapy in individuals with osteoporosis, particularly whether physiotherapists have concerns about fracture as a complication of treatment. Methods. I developed a questionnaire and sent it to 171 randomly selected physiotherapists working in the province of B C . Results. The response rate was 39% (67/171). Ninety-seven percent of respondents reported using strength exercises and postural reeducation, while 45% reported using manual therapy in this population. Ninety-one percent of respondents reported having concerns about the use of manual therapy. Vertebral fracture and rib fracture were the most commonly reported concerns. Conclusions. Most physiotherapists practicing in B C , Canada use evidence based methods (i.e. strength training) when treating individuals with osteoporosis, a large number use manual therapy, and most have concerns about its use. Physiotherapists are most concerned about fractures, in particular vertebral fracture, but injury to other musculoskeletal tissues is also of concern. Studies of safety and effectiveness of manual therapy in this population are needed to guide clinical practice.  80  Practice Behaviors and Clinicians' Perceptions 3.1 INTRODUCTION Osteoporosis, characterized by low bone mass and increased fracture risk, affects 1 in 5 postmenopausal women. ' Given the population prevalence, and also the various secondary causes of osteoporosis, it is likely that physiotherapists in all areas of practice see patients with compromised bone health.  Further, an individual may present to  physiotherapy for any number of problems, related or unrelated to osteoporosis. For example, back pain is also common in older adults ' and is associated with reduced 4  5  mobility, independence, and health related quality of life. ' Back pain is also the most common reason for visiting a physiotherapist (Physiotherapy Association of British Columbia (PABC), October 2001) and there are many individuals with both osteoporosis and back p a i n .  7-9  Physiotherapists can use a variety of modes to treat clients with osteoporosis. Pain relief, increased strength, improved posture and improved range of motion are a few common goals of therapy for such individuals. various populations,  1 2 - 1 7  1 0 , 1 1  Given the effectiveness of manual therapy in  it seems reasonable that clinicians would consider using such  techniques in individuals with osteoporosis. The biological mechanisms underpinning this effectiveness may be related to spinal mobilization stimulating sympathetic nervous system activity  1 5 , 1 7 > 1 8  and promoting motor activity.  19  For example, emerging evidence  suggests than spinal joint mobilization techniques applied to the cervical spine elicit concurrent effects on pain perception, autonomic function, and motor function in patterns that are similar to the patterns of change elicited by stimulation of the periaqueductal gray  81  Practice Behaviors and Clinicians' Perceptions region ofthe midbrain. ' ' 15  17  20  However, the literature cautions against the use of manual 91 9T  therapy in individuals with osteoporosis.  There appears to be agreement amongst leading clinicians that spinal manipulation (high velocity thrust) techniques are contraindicated in individuals with osteoporosis, clinical experience  2 5  and published cases  2 6  2 1 - 2 4  yet  suggest that these techniques are still being  used by chiropractors. Clinical experience indicates that while some colleagues have expressed concern about the safety of manual therapy (including spinal mobilization) other clinicians routinely use manual therapy in older people, a proportion of whom will have osteoporosis.  There are no previous studies of clinicians' practice behaviors and perceptions with respect to the management of individuals with osteoporosis. Although there is vast literature on the effects of mechanical loading on bone (which physiotherapists can use to prescribe appropriate exercises) there are no data on the safety of manual therapy in this population. ' 2 4  2 5  For these reasons the aims of this study were 1) to measure the most  common treatment modes used by a random sample of physiotherapists practicing in the province of British Columbia (BC) for treating individuals with osteoporosis and 2) to assess whether physiotherapists in B C have concerns about the use of manual therapy in individuals with osteoporosis, such as fear of fracture as a complication of treatment.  82  Practice Behaviors and Clinicians' Perceptions 3.2 M E T H O D S Design  This cross-sectional study was approved by the University of British Columbia Clinical Research Ethics Board and the Research Review Committee at Children's & Women's Health Centre of British Columbia (BC Women's Health Centre).  Materials  I developed a brief questionnaire (Appendix I) to (1) measure the most common treatment modes used by a random sample of physiotherapists practicing in the province of B C in the treatment of individuals with osteoporosis, and (2) to assess whether physiotherapists in B C have concerns about the use of manual therapy in individuals with osteoporosis, particularly whether physiotherapists have concerns about fracture as a complication of treatment.  Subjects and procedures  The questionnaire and accompanying cover letter (Appendix I) was faxed to a random sample of physiotherapists in the province of B C . The fax was sent to every fifth member with a fax number, from a list of members of the provincial association. The survey was sent to physiotherapists working in all areas of practice. A total of 208 questionnaires were sent but 37 were not transmitted. Thus a total of 171 questionnaires were both sent and transmitted. Sixty-seven individuals responded by completing the questionnaire and returning it to the B C Women's Health Centre by fax or mail within three weeks.  83  Practice Behaviors and Clinicians' Perceptions  Data analysis The response rate was calculated by dividing the number of respondents by the number of questionnaires that were both sent and transmitted. The number of respondents who 1) used each treatment mode (Appendix I, Question 1) 2) had concerns about the use of manual therapy 3) had concerns about injury to each of the tissues/regions listed (i.e. vertebral fracture, other fracture, disc injury, muscle injury) (Appendix I, Question 3) was expressed as a percentage of the total respondents.  84  Practice Behaviors and Clinicians' Perceptions 3.3 RESULTS The response rate (67/171) was 39%. The percentage of all respondents who selected each treatment mode is presented in Figure 3.1. Two respondents (3%) reported that they do not treat individuals with osteoporosis. Thirty-nine percent of respondents reported using treatment modes 'other' than those listed. A wide variety of responses were received in the 'other' category, including dietary calcium, weight bearing activity, fall prevention education, pain and time management, energy conservation, endurance training, referral to physician for medications, hydrotherapy, support/bracing and education of personal trainers.  Forty-five percent of respondents reported using manual therapy in this population (Figure 3.1). In Question 2, 91% of respondents reported having concerns about the use of manual therapy techniques on individuals with osteoporosis, 6% did not have concerns and 3%> stated that they do not treat individuals with osteoporosis. With the exception of one individual, respondents who reported using manual therapy (in Question 1, Appendix I) also reported concern about its use (in Question 2, Appendix I).  The percentage of respondents concerned about each of the tissues/structures listed is presented in Figure 3.2. Vertebral fracture and other fracture were the most commonly reported concerns. Of the respondents who were concerned about 'other fractures' (Figure 3.2), 13% reported concern about rib fracture. Hip, wrist, and humerus fracture were of concern for a small number of respondents (3%, 1% and 2% respectively).  85  Practice Behaviors and Clinicians' Perceptions  Other  39%  Ergonomics  88%  Posture  97%  Laser  18%  a>  •a o  US  119%  0)  £  Electrother  46%  Man Ther  45%  Flexibility  92%  Balance  79%  Strength  97% 10  20  30  40  50  60  70  80  90  100  Percentage of Respondents  Figure 3.1. Percentage of respondents who chose each of the treatment modes listed in Question 1: "Which of the following treatment modes would you likely use with an individual with osteoporosis? " Strength^ strength exercises; Balance= balance exercises; Flexibility=flexibility  exercises; Man Ther= manual therapy; Electrother=  electrotherapy; US= ultrasound; Laser = laser; Posture=posture reeducation; Ergonomics= ergonomic advice  86  Practice Behaviors and Clinicians' Perceptions  3%  muscle  9%  disc  tendon  1  6%  18%  ligt  63%  other #  83%  vert#  10  20  30  40  50  60  70  80  90  100  P e r c e n t a g e of R e s p o n d e n t s  Figure  3.2. Percentage of respondents reporting concerns related to each of the  tissues/structures listed in Question 3. vert # = vertebral fracture; other # = other fracture; ligt= ligament injury; tendon= tendon injury; disc = disc injury; muscle= muscle injury  87  Practice Behaviors and Clinicians' Perceptions 3.4 DISCUSSION This study presents novel data about current practice behaviors and perceptions of physiotherapists with respect to the management of individuals with osteoporosis in BC. Most respondents reported concern about the use of manual therapy in this population. Despite this concern, the results suggest that many clinicians (45% of the sample) use manual therapy in individuals with osteoporosis. Studies suggest that manual therapy can relieve pain " and improve motor control ' 12  17  1 5  2 7  so it seems reasonable that  physiotherapists would consider the potential benefits of manual therapy for individuals even if they have osteoporosis.  The questionnaire did not specifically state whether the individual was being treated for osteoporosis or an unrelated condition. However, two respondents reported that they do not treat osteoporosis. One stated that 'we are a hand-only clinic'. This response is of interest as one would speculate that a hand clinic would treat some individuals during recovery from wrist fracture. Colles' fracture is a common sentinel osteoporotic fracture and a strong predictor of future fracture. ' 2 8  2 9  Treatment modes  As expected, most respondents marked a number of treatment modes. Strength exercises, postural reeducation, flexibility exercises, and ergonomic advice were utilized by a vast majority of respondents (Figure 3.1), while balance training was less commonly used yet is also thought to be an important factor in preventing falls and subsequent fractures. ' ~ 7 30  88  Practice Behaviors and Clinicians' Perceptions 3 2  Manual therapy was used by almost half the respondents (Figure 3.1) yet few data  exist with respect to its safety or efficacy in this population.  Concern Regarding Fracture or Other Tissue Injury The results suggest that physiotherapists are concerned about fracture as a complication of manual therapy treatment, in particular vertebral fracture and rib fracture. Surprisingly, ligament, tendon and disc injury were also of concern, albeit for a much smaller number of respondents (Figure 3.3). This may reflect a lack of data about secondary tissue changes associated with osteoporosis, or lack of knowledge on the part of physiotherapists with respect to whether or not they should be concerned about these tissues.  Response Rate The response rate in this study is among the upper range found in surveys among other health care professionals. '  33 34  response rates as high as 5 3 %  While some previous studies of physiotherapists report 35  and 65%, 1 feel the 39%) response rate in my study is 36  acceptable for numerous reasons. First, the study included a random sample of all physiotherapists who were members of their professional association in a specific jurisdiction. Research topic has been shown to affect response rates  yet few clinicians  treat primarily individuals with osteoporosis. Clinicians may be more motivated to respond to a study about direct-access  (which more obviously affects their caseload  and earning potential) or a questionnaire specific to their area of special interest. '  38 39  89  Practice Behaviors and Clinicians' Perceptions Some previous studies with higher response rates only involved clinicians in a specific area of practice  3 9  or education.  4 0  Second, while incentives have been shown to affect response rates 1 did not offer any 4 1  in this study. A study evaluating the use of prepaid incentives suggested that some incentives increase response rate yet introduce further bias because some incentives will appeal more to certain individuals.  4 2  Finally, using data from the Community Tracking  Study's physician survey, researchers examined how survey estimates and data quality changed as additional respondents completed the survey. They found that differences in response rates are unlikely to significantly impact the quality of data collected unless one achieves a response rate significantly above 65%.  4 3  A reason to consider the response rate is to avoid drawing biased conclusions. M y response rate indicates that a substantial proportion of clinicians have concerns about the use of manual therapy in patients with osteoporosis. Even in the unlikely event that all clinicians who did not reply had no concern about the use of manual therapy in this population, my results would still suggest that 36% of clinicians have concerns. This would still warrant further research.  Clinical Implications These findings suggest that a large percentage of physiotherapists practicing in B C use evidence based methods (specifically strength training - ' ) when treating individuals 32  44  45  with osteoporosis. Many also use manual therapy in this population and most have  90  Practice Behaviors and Clinicians' Perceptions concerns about its use. Given the similarities in physiotherapy training across the various provinces in Canada, ' 4 6  4 7  it is likely that these B C data would generalize nationwide.  Further, manual therapy is an internationally practiced and researched treatment ' ' " 12 13 15  17 91 99 4R ' ' '  so these data may even have international relevance. This suggests a need for  appropriately designed studies to provide data to address the safety concerns reported in this study. Physiotherapists are most concerned about fractures, in particular vertebral fracture, but injury to other musculoskeletal tissues is also of concern. Clinical leaders agree that manipulation is contraindicated in this population,  21-24  but consensus on other manual  therapy techniques has not been reached. The results of this study suggest that a significant number of physiotherapists use manual therapy in this population. Evidence suggests manual therapy is effective for some conditions, so it seems appropriate that physiotherapists would consider using it in this population. However, there are no data with respect to the safety of manual therapy for individuals with osteoporosis. As clients with osteoporosis could potentially benefit from manual therapy, trials are needed to examine its safety and effectiveness in this population.  91  Practice Behaviors and Clinicians' Perceptions 3.5  1.  REFERENCES  Melton LJ, 3rd, Chrischilles E A , Cooper C, et al. Perspective. How many women  have osteoporosis? Journal of Bone and Mineral Research 1992;7:1005-10.  2.  Melton LJ, 3rd. Epidemiology of spinal osteoporosis. Spine 1997;22:2S-1 IS.  3.  Sran M M . Effects of exercise training/mechanical loading on bone -what  physiotherapists need to know. Physiother Can 2002;54:S12.  4.  Badley E M , Tennant A. Changing profile of joint disorders with age: findings  from a postal survey of the population of Calderdale, West Yorkshire, United Kingdom. Ann Rheum Dis 1992;51:366-71.  5.  Reynolds D L , Chambers L W , Badley E M , et al. Physical disability among  Canadians reporting musculoskeletal diseases. J Rheumatol 1992;19:1020-30.  6.  Hansson TH, Hansson E K . The effects of common medical interventions on pain,  back function, and work resumption in patients with chronic low back pain: A prospective 2-year cohort study in six countries. Spine 2000;25:3055-64.  7.  Malmros B , Mortensen L, Jensen M B , et al. Positive effects of physiotherapy on  chronic pain and performance in osteoporosis. Osteoporos Int 1998;8:215-21.  8.  Leidig G, Minne HW, Sauer P, et al. A study of complaints and their relation to  vertebral destruction in patients with osteoporosis. Bone Miner 1990;8:217-29.  9.  Patel U , Skingle S, Campbell GA, et al. Clinical profile of acute vertebral  compression fractures in osteoporosis. Br J Rheumatol 1991;30:418-21.  92  Practice Behaviors and Clinicians' Perceptions 10.  Larsen J. Osteoporosis and the Physiotherapist. In Sapsford R, Bullock-Saxton J,  Markwell S eds. Women's health : a textbook for physiotherapists. London: W B Saunders, 1998.  11.  Bennell K , Khan K, McKay H. The role of physiotherapy in the prevention and  treatment of osteoporosis. Man Ther 2000;5:198-213.  12.  Hoving JL, Koes BW, de Vet HC, et al. Manual therapy, physical therapy, or  continued care by a general practitioner for patients with neck pain. A randomized, controlled trial. Ann Intern Med 2002;136:713-22.  13.  Niemisto L, Lahtinen-Suopanki T, Rissanen P, et al. A randomized trial of  combined manipulation, stabilizing exercises, and physician consultation compared to physician consultation alone for chronic low back pain. Spine 2003;28:2185-91.  14.  Farrell JP, Twomey LT. Acute low back pain. Comparison of two conservative  treatment approaches. Med J Aust 1982;1:160-4.  15.  Sterling M , Jull G, Wright A. Cervical mobilisation: concurrent effects on pain,  sympathetic nervous system activity and motor activity. Man Ther 2001;6:72-81.  16.  Goodsell M , Lee M , Latimer J. Short-term effects of lumbar posteroanterior  mobilization in individuals with low-back pain. J Manipulative Physiol Ther 2000;23:332-42.  17.  Vicenzino B, Collins D, Benson H, et al. A n investigation of the interrelationship  between manipulative therapy-induced hypoalgesia and sympathoexcitation. J Manipulative Physiol Ther 1998;21:448-53.  18.  McGuiness J, Vicenzino B , Wright A. Influence of a cervical mobilization  technique on respiratory and cardiovascular function. Man Ther 1997;2:216-20.  93  Practice Behaviors and Clinicians' Perceptions 19.  Sterling M , Jull G, Wright A. Cervical mobilisation: concurrent effects on pain,  sympathetic nervous system activity and motor activity. Man Ther 2001;6:72-81.  20.  Wright A , Sluka K A . Nonpharmacological treatments for musculoskeletal pain.  Clin J Pain 2001;17:33-46.  21.  Tobis JS, Hoehler F eds. Musculoskeletal Manipulation: Evaluation of the  Scientific Evidence. Springfield, 111: Charles C Thomas Publisher, 1986.  22.  Maitland GD, Banks K, English K, et al. eds. Maitland's vertebral manipulation.  Sixth ed. Boston: Butterworth Heinemann, 2001.  23.  Grieve G. Common Vertebral Joint Problems. 2nd edition. New York.: Churchill  Livingstone, 1988.  24.  Ernst E. Chiropractic spinal manipulation for back pain. Br J Sports Med  2003;37:195-6.  25.  Sran M M . Chiropractic spinal manipulation for back pain; commentary. Br J  Sports Med 2003;37:196.  26.  Haldeman S, Rubinstein SM. Compression fractures in patients undergoing spinal  manipulative therapy. J Manipulative Physiol Ther 1992;15:450-4.  27.  Jull G, Trott P, Potter H , et al. A randomized controlled trial of exercise and  manipulative therapy for cervicogenic headache. Spine 2002;27:1835-43.  28.  Cuddihy MT, Gabriel SE, Crowson CS, et al. Forearm fractures as predictors of  subsequent osteoporotic fractures. Osteoporos Int 1999;9:469-75.  94  Practice Behaviors and Clinicians' Perceptions 29.  Mallmin H, Ljunghall S, Persson I, et al. Fracture of the distal forearm as a  forecaster of subsequent hip fracture: a population-based cohort study with 24 years of follow-up. Calcif Tissue Int 1993;52:269-72.  30.  Campbell A J , Robertson M C , Gardner M M , et al. Randomised controlled trial of  a general practice programme of home based exercise to prevent falls in elderly women. B M J 1997;315:1065-9.  31.  Myers TA, Briffa N K . Secondary and tertiary prevention in the management of  low-trauma fracture. Aust J Physiother 2003;49:25-9.  32.  Liu-Ambrose T, Khan K M , Eng JJ, et al. Both resistance and agility training  reduce fall risk in 75-85 year old women with low bone mass: A six-month randomized controlled trial. J A m Geriatr Soc 2004;52:657-65.  33.  Bhandari M , Devereaux P, Swiontkowski M F , et al. A randomized trial of opinion  leader endorsement in a survey of orthopaedic surgeons: effect on primary response rates. Int J Epidemiol 2003;32:634-6.  34.  Stone P, Ream E, Richardson A , et al. Cancer-related fatigue—a difference of  opinion? Results of a multicentre survey of healthcare professionals, patients and caregivers. Eur J Cancer (Engl) 2003;12:20-7.  35.  Robinson AJ, McCall M , DePalma MT, et al. Physical therapists' perceptions of  the roles of the physical therapist assistant. Phys Ther 1994;74:571-82.  36.  Crout K L , Tweedie JH, Miller DJ. Physical therapists' opinions and practices  regarding direct access. Phys Ther 1998;78:52-61.  37.  de Wit NJ, Quartero A O , Zuithoff AP, et al. Participation and successful patient  recruitment in primary care. J Fam Pract 2001;50:976.  95  Practice Behaviors and Clinicians' Perceptions 38.  Mantle J, Versi E. Physiotherapy for stress urinary incontinence: a national  survey. B M J 1991;302:753-5.  39.  Barry S, Wallace L , Lamb S. Cryotherapy after total knee replacement: a survey  of current practice. Physiother Res Int 2003 ;8:111-20.  40.  Walker J M . Curricular content on urinary incontinence in entry-level physical  therapy programmes in three countries. Physiother Res Int 1998;3:123-34.  41.  Halpem SD, Ubel PA, Berlin JA, et al. Randomized trial of 5 dollars versus 10  dollars monetary incentives, envelope size, and candy to increase physician response rates to mailed questionnaires. Med Care 2002;40:834-9.  42.  Berry S, Kanouse D. Physician response to a mailed survey: A n experiment in  timing payment. Public Opin Q 1987;51:102-16.  43.  Schoenman JA, Berk M L , Feldman JJ, et al. Impact of differential response rates  on the quality of data collected in the CTS physician survey. Eval Health Prof 2003;26:23-42.  44.  Kerr D, Morton A , Dick I, et al. Exercise effects on bone mass in postmenopausal  women are site-specific and load-dependent. J Bone Miner Res 1996;11:218-25.  45.  Kohrt W M , Ehsani A A , Birge SJ, Jr. Effects of exercise involving predominantly  either joint-reaction or ground-reaction forces on bone mineral density in older women. J Bone Miner Res 1997;12:1253-61.  46.  CUP A C , The Alliance. Competency Profile for the Entry-Level Physiotherapist in  Canada. Canadian University Physical Therapy Academic Council (CUPAC) and the Canadian Alliance of Physiotherapy Regulators (The Alliance) 1995.  96  Practice Behaviors and Clinicians' Perceptions 47.  CPA. Description of Physiotherapy in Canada: 2000 and Beyond. Canadian  Physiotherapy Association (CPA) 2000.  48.  Kotoulas M . The use and misuse of the terms "manipulation" and "mobilization"  in the literature establishing their efficacy in the treatment of lumbar spine disorders. Physiother Can 2002;Winter:53-61.  97  Safety of Posteroanterior Spinal Mobilization  Chapter Four Failure Characteristics of the Thoracic Spine with a Posteroanterior Load: Investigating the Safety of Spinal Mobilization  98  Safety of Posteroanterior Spinal Mobilization  ABSTRACT Background. Osteoporosis and back pain are common alone and in combination among older adults. Spinal mobilization techniques have been shown to relieve back pain and improve function in various clinical settings. However, whether controlled spinal mobilization can cause vertebral fracture in individuals with osteoporosis is not known. Objectives. To quantify failure load and pattern of midthoracic vertebrae under a posteroanterior (PA) load and to compare failure load, in vitro, with applied load, in vivo. Methods. Twelve T5-8 cadaveric specimens (mean age 77 yrs) were scanned using D X A , radiographed, and measured for bone size. I measured failure load, failure site, and intervertebral motion (using a precision opto-electronic camera system) when a PA load was applied at the spinous process of T6 using a servohydraulic material testing machine. Post test radiography and CT scan were used to verify failure site. These tests were repeated in an intact cadaver, using a Tekscan I-Scan sensor to measure applied loads. I also quantified in vivo applied loads during PA mobilization during seven trials by two experienced physiotherapists. Results. Mean (SD) in vitro failure load of 479 N (162 N) was significantly higher than the mean (SD) in vivo applied load of 145N (38 N) (p=.0004). Macroscopic observation revealed a fracture at the T6 spinous process in eleven specimens, and one at the T7 spinous process. These fractures were detected by plain radiography in three of twelve cases and by CT scan in six of twelve cases. The mobilized vertebra (T6) extended with respect to the inferior vertebra and the vertebra above typically flexed during P A mobilization. Conclusions. The results of this study suggest a reasonable margin between failure load, in vitro, and applied mobilization load, in vivo.  99  Safety of Posteroanterior Spinal Mobilization  4.1  INTRODUCTION  Back pain is common in older adults ' and is associated with reduced mobility, l  2  independence, and health related quality of life. ' 3  4  Spinal mobilization is a widely used  manual therapy technique among physiotherapists and osteopaths to treat back pain. There is evidence that manual therapy including spinal mobilization may be therapeutic in that it can reduce spinal p a i n , motor activity.  5-8  stimulate sympathetic nervous system activity, ' ' and promote 6 7 9  6  However, there is concern that manual therapy could cause fracture in individuals with osteoporosis (Chapter Three), a condition characterized by low bone mass and increased fracture risk that affects 1 in 5 postmenopausal women. painless until complicated by a fracture,  12-16  1 0 , 1 1  Osteoporosis is generally  so it can be present without the patient's  knowledge. Thus, unless a clinician treating back pain recognizes the major clinical risk factors for osteoporosis (such as age, sex, family history), he or she may unwittingly use spinal mobilization in a patient with osteoporosis. My clinical experience suggests that many practitioners treat older people without considering osteoporosis, while others eschew the potential therapeutic effect of spinal mobilization in patients with osteoporosis for fear of causing fracture.  PA mobilization is a commonly used spinal mobilization technique  1 7 - 1 9  yet previous  studies do not provide data on P A spinal fracture thresholds. While some investigators reported applied load with this technique, ' ' or spinal stiffness, > " they did not also 5 20 21  18  22  24  measure spinal failure or fracture. Spinal failure has most commonly been studied with  100  Safety of Posteroanterior Spinal Mobilization  axial compressive loading " (mimicking the compressive and bending loads associated with trauma), but not the force generated by therapeutic spinal mobilization. To my knowledge, spinal failure with a PA load, and the relationship of bone quality and geometry to P A failure load, has not been previously investigated. Investigation of the intervertebral movements resulting from P A spinal mobilization in the lumbar spine, in vivo, have been conducted using indirect measures, specifically radiography  1 9  as well as a load cell, two  linear variable differential transformers (LVDTs) and a motor driven force applicator to measure displacement of the skin over the spinous process.  M y primary objective was to quantify failure load and pattern of midthoracic vertebrae, in vitro, when a PA load was applied to T6 in T5-8 cadaveric spine segments of older spines and compare this in vitro failure load with the load applied in vivo. M y secondary goal was to measure the amount and direction of motion produced during cyclic P A loading of T6 in vitro, and to validate these in vitro measures by comparison with kinematics during P A mobilization in an intact cadaver. M y final aim was to examine the relationship of bone density and geometry with P A failure load.  101  Safety of Posteroanterior Spinal Mobilization  4.2 METHODS Study Design I performed (1) failure load measurements and kinematic analyses on cadaveric thoracic spine segments, (2) kinematic analysis and applied load measurement in an intact cadaver, and (3) applied load measurements in human volunteers.  Sample Size Calculation Sample size for this study is based on previous studies of applied load during P A mobilization and porcine pilot tests of PA failure load and assumes beta equal to 0.2 (power equal to 0.80) and alpha equal to 0.05. Assuming 2-samples of unequal variance, with a mean (SD) in group 1 (in vivo applied load) of 200 N (60 N) and, in group 2 (in vitro failure load) of 800 N (200 N), I require 2 samples in group 1 and 7 samples in group 2. As there are no available data on P A failure load in cadaveric vertebrae, I estimated the mean (SD) failure load based on my porcine pilot experiments and previous studies of axial compressive failure load in cadaveric spine segments. Due to the uncertainty in the estimate for the mean and standard deviation I included 7 samples in group 1 and 12 samples in group 2.  In Vitro Failure Load Specimen preparation Fresh-frozen (unembalmed) human cadaveric spines were obtained from the U B C Department of Anatomy. Donors included 6 females, 5 males and 1 unknown. Age at death ranged from 62-93 years, mean 77 years. Each specimen was assigned an  102  Safety of Posteroanterior Spinal Mobilization  identification number to maintain anonymity. This study was approved by the U B C Clinical Ethics Review Board.  Each specimen was dissected to isolate a segment consisting of T5-T8. The fifth through ninth ribs were removed. A n axial cut was made to separate T4-T5 and T8-T9 at the intervertebral disc. Incisions were made between the facet joints and spinous processes at the same levels. The ligaments and discs between T5 and T8 were left intact. The isolated segments were stored frozen at -20 degrees C until testing.  Pretest measurements Bone area (BA, cm ), bone mineral content (BMC, grams) and areal bone mineral density 2  (aBMD, grams/cm ) were assessed in the A P and lateral orientation for each T5-T8 segment using a Hologic QDR 4500W bone densitometer (DXA). A qualified technician analyzed all scans using standardized procedures as outlined in the Hologic User's Guide. A P acquisition parameters were used for both A P and lateral scans. For lateral scans, the posterior elements were excluded from the region of interest. Members of my research group had conducted a short-term precision study, in vivo, with 17 subjects measured twice, using this instrument. The coefficient of variation for B M C and aBMD at the lumbar spine (LS) (L1-L4) was less than 0.7%. A spine anthropomorphic phantom was scanned daily for quality assurance of the QDR 4500. As in other studies,  31-33  bags of rice were used to  simulate soft tissue surrounding the spine segment. Lateral and A P radiographs were taken of each specimen prior to testing.  103  Safety of Posteroanterior Spinal Mobilization  Figure 4.1. Lateral view of a thoracic vertebra with typical measures of vertebral body area, spinous process length and spinous process width. (s= intersection of line from anterior inferior vertebral body along the inferior endplate with the anterior border of the spinous process; w= spinous process width; 1= spinous process length; a- anterior height; p=posterior height; h= length of the perpendicular line between the two parallel lines, a and p)  Measures of vertebral body area, spinous process length, and spinous process width were obtained from the lateral radiographs. In the sagittal plane, T6 spinous process dimensions were measured by first drawing a line from the most anterior inferior point of the vertebral body along the inferior endplate. The intersection of this line with the anterior border of the spinous process was labeled (s). Spinous process length was measured as the distance from (s) to the inferior tip of the T6 spinous process. Spinous process width was measured as the distance from (s) across the spinous process. Vertebral body dimensions included anterior height, posterior height, superior endplate depth and inferior endplate depth. Assuming the  104  Safety of Posteroanterior Spinal Mobilization  anterior and posterior borders were parallel, the area of the T6 vertebral body was calculated for each specimen using the following equation: A= [(a+p)/2] h  where a=  anterior height, p=posterior height, and h= the length of the perpendicular line drawn between the two parallel lines, a and p (Figure 4.1).  Mechanical testing Steel (24 gauge) wire was secured to the pedicles of T5 and T8 of each spine segment. Each specimen was embedded in dental cement such that half of the vertebral bodies of T5 and T8 and the attached steel wire were fixed in cement. Thus, the T5 transverse processes were fully embedded and half of the T8 transverse processes were embedded. The T5-6 and T7-8 facet joints remained free, as did all parts of T6 and T7. Marker carriers with four infrared light emitting diodes (LEDS) were secured to the vertebral body of T5 (middle) and the transverse processes of T6-8 with a 3.5 mm cancellous bone screw. In cases where the transverse process was too weak to sufficiently secure the marker carrier, the middle of the vertebral body was used. Each marker carrier was positioned to be in clear view of the cameras while avoiding contact with dental cement or another marker carrier during the test (Figures 4.2A and 4.2B).  105  Safety ofPosteroanterior Spinal Mobilization  Posteroanterior Load  Figure 4.2 A. Schematic of posteroanterior (PA) loading at the T6 spinous process in T5T8 cadaveric spine segments. The opto-electronic marker carriers are shown attached to each vertebra.  The specimen was oriented horizontally in the testing machine (Instron 8874, Instron Corp. Canton, M A ) to facilitate the P A load application (Figure 3.2A and 3.2B). The T5 and T8 specimen mounts were clamped rigidly, such that the T6 spinous process was aligned with the linear actuator of the machine. Load was applied through a circular delrin indenter (20 mm diameter, Young's Modulus 3.1 GPa), mounted on the end of the actuator. This indenter had a 7 mm diameter groove for the spinous process and 3 mm foam (PPT, Langer Biomechanics Group Inc, NY) was attached to the bottom of the indenter to enable a distributed load transmission and to prevent slipping of the spinous process. A very low  106  Safety ofPosteroanterior Spinal Mobilization  load (5 N) functional test was used to verify that the intended P A load did not produce noticeable coupled axial rotation.  Figure 4.2B. A typical T5-T8 cadaveric specimen secured in the custom testing jig. The opto-electronic marker carriers, base marker fixed to the testing machine (lower left), and the delrin indenter with foam padding (above the specimen) are also shown.  A cyclic P A load of 50-200 N was applied to the most posterior point of the T6 spinous process at 0.5 Hz for 30 seconds. Kinematic data were recorded during the test at 10 Hz using a precision opto-electronic camera system (Optotrak 3020, Northern Digital, Waterloo, Ontario). The accuracy of this system exceeds 0 . 1 ° .  34  After the cyclic test, the  specimen was loaded to failure at a displacement rate of 2 mm/sec.  Post-test measurements Immediately after failure, we used macroscopic observation, gentle dissection, and digital  107  Safety of Posteroanterior Spinal Mobilization  photography to identify and document the fracture. I repeated lateral and A P radiography using identical settings as in the pre-test films. The radiographs were reported by the same musculoskeletal radiologist for evidence of fracture. Each specimen was scanned with CT using 1.3 mm helical scans at an image interval of 0.6 mm on bone algorithm (display field of view: 16 cm). Together with the radiologist, I generated sagittal and coronal reconstructions on the scanner to simulate the plain radiographs. A l l CT scans were reported by one radiologist.  Intact Cadaver A fresh whole cadaver (male, age 74 years) without significant scoliosis (on examination) was tested at the U B C Department of Anatomy. K-wires (2.4 mm diameter) were secured to the T5, T6, T7 transverse processes and the T8 spinous process by a spine surgeon. I attached opto-electronic marker carriers to the wires (Figure 4.3). I performed P A mobilization to T6 (Figure 4.4) and the applied load was measured using a Tekscan I-Scan sensor (Tekscan 5051, Boston, M A , Appendix II) (Figure 4.3). I followed a metronome to ensure mobilization was applied at a rate of 0.5 Hz for 15 cycles. Kinematics were recorded simultaneously at 10 Hz using a precision opto-electronic camera system (Optotrak 3020, Northern Digital, Waterloo, Ontario). This cyclic mobilization was followed by three attempts to fracture the spine with a maximal manual P A load.  108  Safety of Posteroanterior Spinal Mobilization  Figure 4.3. Marker set up for the intact cadaver test and the Tekscan I-Scan sensor for measuring applied load.  Figure 4.4. The physiotherapist applying PA mobilization to T6 of an intact cadaver.  109  Safety of Posteroanterior Spinal Mobilization  In Vivo Applied Load For comparison with clinical mobilization, I measured applied load during grade IV P A mobilization (see sections 1.4.1 and 1.4.5) of T6 in human volunteers during seven trials by two experienced physiotherapists (including the author). Applied vertical load to the spine was estimated indirectly by the decrease in force on a floor mounted force platform while the therapists performed P A mobilization at the T6 spinous process.  Data Analysis In the spine segment tests and in the intact cadaver, I determined vertebral kinematics for each spinal level via the opto-electronic marker carriers. The anterior inferior vertebral body was the origin in the T5-T8 spine segments and the spinous process was the origin in the intact cadaver. A global coordinate system was used for all tests. For the spine segment tests a base marker was fixed to the jig. The x axis was medial-lateral and the y axis cephalad-caudad. For the intact cadaver test axes were identified by digitizing the left and right scapulae (directly opposite the T6 spinous process), and the T l and L I spinous processes (x and y axes respectively). I assumed each vertebra to be a rigid body for measurement, and used custom software (KIN 2000, written in LabVIEW 6.0, National Instruments, USA).  For each in vitro failure test, there was a deviation from the real-time load-displacement curve and an audible 'crack'. Failure was defined as the first decrease in load (Figure 4.5).  I used Pearson correlation coefficients to assess the relationship between failure load, B M D  110  Safety of Posteroanterior Spinal Mobilization  and bone geometry measures, and Student's t-test to compare failure load, in vitro, and applied load, in vivo. Statistical significance was set at p< 0.05.  H1064 Failure test- load/displacement  T3  re o  14  displacement (mm)  Figure 4.5. Load (N) and displacement (mm) for a typical specimen during the failure test, (in vitro cadaveric spine segment) Failure point was defined as the first decrease in load.  111  Safety of Posteroanterior Spinal Mobilization  4.3 R E S U L T S Pretest Bone Measures  Bone mineral density (g/cm ) ranged from 0.463-1.020 (mean=0.686, SD=0.165) with the anteroposterior scan, and 0.270-0.636 g/cm with the lateral scan (mean 0.386, SD=0.098). Vertebral body area ranged from 460-988 mm (mean= 642 mm , SD=145 mm ). Spinous 2  2  2  process width was, on average 1.1 cm (SD=0.2, range= 0.8 to 1.7 cm) while spinous process length ranged from 2.6 to 4.5 cm (mean= 3.5, SD=0.5). No specimen showed any sign of malignancy. One specimen had a 20% wedge compression fracture of T8.  Failure Load and Failure Site, In Vitro, and Applied Load, In Vivo  In vitro failure load ranged from 200-728 N with a mean of 479 N+/-162 N(SD). Macroscopic observation revealed a fracture at the T6 spinous process in eleven specimens, and one at the T7 spinous process. The fractures identified post testing were evident on plain radiographs in only three of twelve specimens despite very close scrutiny (Figure 4.6A). CT scans detected the fracture in two of the three detected on plain radiographs, and four other specimens (Figure 4.6B). Thus, five fractures identified macroscopically remained undiagnosed by either imaging method. There was no evidence of damage to the vertebral body in any specimen.  Applied loads, in vivo, ranged from 106-223 N , with a mean of 145+/-38 N(SD) (Figure 4.7). There was a significant difference between failure load, in vitro, and applied load, in vivo (p=.0004).  112  Safety of Posteroanterior Spinal Mobilization  Figure 4.6A. Post test radiography detected the site offracture in only 3 of 12 cases. This radiograph shows a fracture at the base of the T6 spinous process (see arrow).  113  Safety of Posteroanterior Spinal Mobilization  Figure 4 . 6 B . Post test CT scan detected 6 of the 12 fractures. The neuroradiologist interpreted the step in the cortical bone (see arrow) as a fracture.  114  Safety of Posteroanterior Spinal Mobilization  250 - r -  •SET  ...  -  ...  1BW  _ 200 -  5. g  150 100 50  \J  I  I  1  I  I  2  3  I  4  I  5  I  6  I  7  participant Figure 4.7.  Graphical representation of the in vivo applied load data. Participants are  listed on the horizontal axis and load in N on the vertical axis. Load rangedfrom106-223 N, with a mean of 145 N. Similar loads were applied in young and old participants.  The physiotherapist was unable to produce a fracture in the intact cadaver. Figure 4.8 shows a sample of the force data collected in the intact cadaver tests. The maximum load applied was 135 N . Calibration of the Tekscan sensor with a six axis load cell showed that Tekscan consistently underestimated applied loads by 15-20% (see Appendix II).  120 -  time (seconds)  Figure 4.8. A sample of the force data collected with Tekscan during the intact cadaver test. In this example the peak load is approximately 115 N and it was reached after almost 8 seconds.  115  Safety of Posteroanterior Spinal Mobilization  Failure Load and Bone Measures There was a weak, non-significant, relationship between failure load at the spinous process and vertebral areal B M D by D X A (Pearson r=0.18 with A P D X A , p=0.57 (Figure 4.9); r=0.13 with lateral D X A , p=0.68). Non-significant correlations were found between bone size and failure load (r= -0.14 spinous process length; r= -0.34 spinous process width; r= -0.27 vertebral body bone area). 850 750  o  650  ,—  i  r\  o  — 550 •o  _  fi  I  2  o~~  o  450  c  3 «  o  l  —  350  o  250 :  o  150 0.4  0.5  0.7 A P BMD (g/cm  0.8 2  0.9  1.0  1.1  )  Figure 4.9. Pearson correlation coefficient for PA failure load (N) and AP bone mineral density (g/cm ): r=0.18. 2  116  Safety of Posteroanterior Spinal Mobilization  Intervertebral Movements During PA Mobilization When a 50-200 N cyclic P A load was applied to the most posterior part of the T6 spinous process, 10 of 12 segments flexed at T5-6 and 11 extended at T6-7. At T7-8 the T7 vertebra flexed in nine specimens and extended in two (one excluded because T8 marker carrier not secured). In the intact cadaver there was flexion at T5-6, extension at T6-7 and flexion at T7-8. The magnitudes of rotation at T5-6, T6-7, and T7-8 (cadaveric segments and intact cadaver) are presented in Table 4.1.  Table 4.1. Magnitude (°) of rotation [flexion (+) and extension (-)] at T5-6, T6-7, and T78 in T5-8 cadaveric spine segments and an intact cadaver during PA mobilization. Spine Segments Mean (Range) Standard Deviation Intact Cadaver  T5-6 0.11 (-0.18-0.39) 0.14 0.20  T6-7 -0.38 (-1.04-0.03) 0.30 -0.20  T7-8 0.14 (-0.26-0.43) 0.20 0.22  117  Safety of Posteroanterior Spinal Mobilization  4.4 DISCUSSION In this chapter, I report novel data of failure load and failure site with spinal mobilization using a cadaveric spine segment model. This provides insight into whether spinal mobilization may have a margin of safety when performed in older people. However, before discussing the clinical relevance of my findings I outline biomedical engineering issues of the measurement technique, my findings of the determinants of failure load, the kinematic data as well as factors that determine the likely generalizability of my cadaver model to a patient population.  Failure Load, Site and Detection in the Cadaveric Thoracic Spine Model To my knowledge, the failure load of vertebrae under a posteroanterior load has not previously been reported, despite the long-term and widespread use of this technique in clinical practice to treat back pain.  '  It appears that thoracic posteroanterior failure load  ranges from 200-727 N in cadaveric specimens. This failure load is much lower than that 9 S 97 9R  reported for the midthoracic spine with axial loading.  ' '  This is likely due to the  cantilever loading of the spinous process during PA mobilization. M y data suggest that aBMD of the whole vertebra is not a good predictor of P A failure load, despite its strong 97 98  correlation with axial compressive failure load in previous studies. ' ' z  z o  This may be due to  the fact that aBMD is an integrated measure of the entire vertebra, whereas the fractures occurred in the posterior elements. Although macroscopic observation and load-displacement data clearly indicated structural failure in every test, plain radiography had poor sensitivity for revealing any evidence of  118  Safety of Posteroanterior Spinal Mobilization  fracture. CT scan was superior to plain radiography in this study, but it nevertheless failed to detect 50% of spinous process fractures. These data, while novel in the setting of spinal mobilization and fracture detection, are consistent with imperfect sensitivity of plain radiography in detecting pars intra-articularis stress fractures.  37  I note that small fractures  TO  can also elude CT scan.  Kinematic Analysis in Spine Segments and Intact Cadaver It appears that real-time intervertebral motion at the thoracic spine may be different from that reported for the lumbar spine. Using a cyclic load, I found less motion than reported in lumbar spine studies that used a static load. A strength of my study was the direct measurement of vertebral motion using bone pins, compared with indirect measurement (e.g., radiographs ) in other studies. 19  1 9 , 2 9  In both of my in vitro studies I found that the  mobilized vertebra (T6) extended with respect to the inferior vertebra and the vertebra above typically flexed (relative to the mobilized vertebra) during P A mobilization. This behavior is due mainly to the length of the thoracic spinous process which results in the load application point being more inferior than would occur in the lumbar spine. This may explain differing results in the lumbar spine where all lumbar motion segments tended to extend except L5-S1 which tended to flex.  1 9  I note that previous studies of intervertebral  motion, in vivo, during P A mobilization were performed with participants holding their breath during the application of the mobilization to avoid vertebral displacement caused by breathing. ' 1 8  2 4  This is problematic in that spinal stiffness has been shown to change  throughout the respiratory cycle.  39  119  Safety of Posteroanterior Spinal Mobilization  Clinical Implications Before drawing conclusions about my study, I discuss the clinical relevance, and limitations, of the thoracic spine segment model. I used fresh-frozen cadaveric specimens that retain their biomechanical properties despite freezing formalin-fixed specimens.  4 1  4 0  and thus, are preferable to  M y specimens were of the appropriate age and I studied T6 as  it appears to be the most vulnerable of the thoracic vertebrae to osteoporotic compression fractures. ' 2 7  4 2 - 4 5  The spine segment kinematics were similar to those seen in the intact  cadaver, which supports the contention that the segment mirrors whole-body behavior. Furthermore, I applied cyclic PA mobilization with loads similar to those applied, in vivo, ' ' and at a frequency of 0.5 Hz, consistent with clinical guidelines. The loads I 5  19  21  36  measured, in vivo, were similar to loads measured by other research groups using a custom instrumented treatment couch (60-230 N) and a load cell 5  2 1  (58-178 N). M y applied load  tests measured only the vertical force vector of the PA mobilization technique with the patient lying prone, but vertical forces contribute between 90 and 99.8% of the total force during PA mobilization. ' 2 0  4 6  My primary research question was to determine PA failure load in a spine segment model. I found that while the mean in vitro failure load was greater than the load a physiotherapist applied to a whole cadaver and to volunteers, one specimen failed at a level that was lower than the maximum in vivo load applied. While my novel data provide a baseline for comparing in vitro failure loads with applied in vivo loads, I must be cautious when comparing data obtained from a spine segment with data obtained in vivo. Nevertheless, my in vivo data are corroborated by other studies ' ' 5  19  21  and my spine segment model has  120  Safety of Posteroanterior Spinal Mobilization  strengths as outlined above. While muscle contraction, the rib cage, intraabdominal organs and intraabdominal pressure are likely to increase spinal stiffness in vivo, " there is no 47  50  evidence to suggest they would increase strength of the spinous process. Thus, if pressed to give an opinion, I would speculate that failure load of the spinous process in a patient, in vivo, would be similar to the failure loads we measured in spine segments.  M y results suggest that vertebral body injury is an unlikely complication of P A mobilization in the midthoracic spine. My finding that fractures occurred at the spinous process is novel and suggests that clinicians may wish to remain alert to the potential for injury to this structure when P A mobilization is used. I could find no case report of a spinous process fracture associated with mobilization. My data, and that of others  5 1  suggest that i f such pathology were to occur, plain radiography would have poor sensitivity to detect it. Also CT scan had only 50% sensitivity for the fracture, raising the question of whether patients complaining of back pain, that may be associated with spinous process fracture, should be investigated with radioisotopic bone scan.  I note one technical point about the terminology of spinal mobilization techniques. Physiotherapists, chiropractors, and osteopaths all learn a variety of manual therapy techniques with which to mobilize spines. > " 36  52  55  it is crucial to note that my data are  limited to posteroanterior mobilizations performed by physiotherapists. My data do not necessarily generalize to other techniques performed by other professionals. Future research should test a variety of mobilization and manipulation techniques to determine failure load using each of them.  121  Safety of Posteroanterior Spinal Mobilization  In summary, a cadaveric thoracic spine segment can withstand upwards of 200 N but physiotherapists can deliver up to 223 N with thoracic posteroanterior spinal mobilization. Kinematic data suggest that the mobilized thoracic vertebra moves into extension as a result of the mobilization. Although there was a reasonable margin between the failure load, in vitro, and the applied mobilization load, in vivo, for most specimens, the lowest fracture thresholds were around the same force as the upper range of the applied loads.  122  Safety of Posteroanterior Spinal Mobilization  4.5  1.  R E F E R E N C E S  Reynolds DL, Chambers L W , Badley E M , et al. Physical disability among  Canadians reporting musculoskeletal diseases. J Rheumatol 1992;19:1020-30.  2.  Badley E M , Tennant A. Changing profile of joint disorders with age: findings from  a postal survey of the population of Calderdale, West Yorkshire, United Kingdom. Ann Rheum Dis 1992;51:366-71.  3.  Hansson TH, Hansson E K . The effects of common medical interventions on pain,  back function, and work resumption in patients with chronic low back pain: A prospective 2-year cohort study in six countries. Spine 2000;25:3055-64.  4.  Malmros B , Mortensen L, Jensen M B , et al. Positive effects of physiotherapy on  chronic pain and performance in osteoporosis. Osteoporos Int 1998;8:215-21.  5.  Goodsell M , Lee M , Latimer J. Short-term effects of lumbar posteroanterior  mobilization in individuals with low-back pain. J Manipulative Physiol Ther 2000;23:33242.  6.  Sterling M , Jull G, Wright A. Cervical mobilisation: concurrent effects on pain,  sympathetic nervous system activity and motor activity. Man Ther 2001;6:72-81.  7.  Vicenzino B , Collins D, Benson H, et al. A n investigation of the interrelationship  between manipulative therapy-induced hypoalgesia and sympathoexcitation. J Manipulative Physiol Ther 1998;21:448-53.  8.  Farrell JP, Twomey LT. Acute low back pain. Comparison of two conservative  treatment approaches. Med J Aust 1982;1:160-4.  9.  McGuiness J, Vicenzino B , Wright A. Influence of a cervical mobilization  technique on respiratory and cardiovascular function. Man Ther 1997;2:216-20.  123  Safety of Posteroanterior Spinal Mobilization  10.  Melton LJ, 3rd, Chrischilles E A , Cooper C, et al. Perspective. How many women  have osteoporosis? J Bone Miner Res 1992;7:1005-10.  11.  Melton LJ, 3rd. Epidemiology of spinal osteoporosis. Spine 1997;22:2S-1 IS.  12.  Haczynski J, Jakimiuk A. Vertebral fractures: a hidden problem of osteoporosis.  Med Sci Monit 2001;7:1108-17.  13.  Epstein S, Goodman GR. Improved strategies for diagnosis and treatment of  osteoporosis. Menopause 1999;6:242-50.  14.  Hawker GA. The epidemiology of osteoporosis. J Rheumatol Suppl 1996;45:2-5.  15.  Nichols K J . Evaluation of osteoporosis. J A m Osteopath Assoc 2000;100:S4-7.  16.  Taxel P. Osteoporosis: detection, prevention, and treatment in primary care.  Geriatrics 1998;53:22-3.  17.  Bjornsdottir SV, Kumar S. Posteroanterior spinal mobilization: state of the art  review and discussion. Disabil Rehabil 1997;19:39-46.  18.  Lee M , Liversidge K. Posteroanterior stiffness at three locations in the lumbar spine.  J Manipulative Physiol Ther 1994; 17:511-6.  19.  Lee R, Evans J. A n in vivo study of the intervertebral movements produced by  posteroanterior mobilization. Clin Biomech 1997;12:400-8.  20.  Harms M C , Innes S M , Bader DL. Forces measured during spinal manipulative  procedures in two age groups. Rheumatology (Oxford) 1999;38:267-74.  124  Safety of Posteroanterior Spinal Mobilization  21.  Simmonds M J , Kumar S, Lechelt E. Use of a spinal model to quantify the forces  and motion that occur during therapists' tests of spinal motion. Phys Ther 1995;75:212-22.  22.  Edmondston SJ, Allison GT, Althorpe B M , et al. Comparison of ribcage and  posteroanterior thoracic spine stiffness: an investigation of the normal response. Man Ther 1999;4:157-62.  23.  Lee M , Steven GP, Crosbie J, et al. Variations in posteroanterior stiffness in the  thoracolumbar spine: preliminary observations and proposed mechanisms. Phys Ther 1998;78:1277-87.  24.  Shirley D, Ellis E, Lee M . The response of posteroanterior lumbar stiffness to  repeated loading. Man Ther 2002;7:19-25.  25.  Burklein D, Lochmuller E, Kuhn V , et al. Correlation of thoracic and lumbar  vertebral failure loads with in situ vs. ex situ dual energy X-ray absorptiometry. J Biomech 2001;34:579-87.  26.  Cheng X G , Lowet G, Boonen S, et al. Prediction of vertebral and femoral strength  in vitro by bone mineral density measured at different skeletal sites. J Bone Miner Res 1998;13:1439-43.  27.  Moro M , Hecker A T , Bouxsein M L , et al. Failure load of thoracic vertebrae  correlates with lumbar bone mineral density measured by D X A . Calcif Tissue Int 1995;56:206-9.  28.  Singer K , Edmondston S, Day R, et al. Prediction of thoracic and lumbar vertebral  body compressive strength: correlations with bone mineral density and vertebral region. Bone 1995;17:167-74.  125  Safety of Posteroanterior Spinal Mobilization  29.  Lee R, Evans J. Load-displacement-time characteristics of the spine under  posteroanterior mobilization. Aust J Physiother 1992;38:115-23.  30.  Hologic. Model QDR-4500 User's Guide. Waltham, M A : Hologic, Inc., 1996.  31.  Grant JP, Oxland TR, Dvorak M F . Mapping the structural properties of the  lumbosacral vertebral endplates. Spine 2001;26:889-96.  32.  Belkoff S M , Mathis J M , Jasper L E , et al. The biomechanics of vertebroplasty. The  effect of cement volume on mechanical behavior. Spine 2001;26:1537-41.  33.  Belkoff S M , Mathis J M , Fenton DC, et al. A n ex vivo biomechanical evaluation of  an inflatable bone tamp used in the treatment of compression fracture. Spine 2001 ;26:1516.  34.  Hamming JA, Goertzen DJ, Oxland TR. The effect of marker configuration and  placement on kinematic accuracy. The Canadian Orthopaedic Association 54th Annual Meeting, 1999:59.  35.  Cyriax J. Textbook of Orthopaedic Medicine. 7th ed. London: Balliere Tindall,  1978.  36.  Maitland GD, Banks K, English K, et al. Maitland's vertebral manipulation. Sixth  ed. Boston: Butterworth Heinemann, 2001.  37.  Anderson K, Sarwark JF, Conway JJ, et al. Quantitative assessment with SPECT  imaging of stress injuries of the pars interarticularis and response to bracing. J Pediatr Orthop 2000;20:28-33.  38.  Link T M , Meier N , Rummeny EJ, et al. Artificial spine fractures: detection with  helical and conventional CT. Radiology 1996;198:515-9.  126  Safety of Posteroanterior Spinal Mobilization  39.  Shirley D, Hodges PW, Eriksson A E , et al. Spinal stiffness changes throughout the  respiratory cycle. J Appl Physiol 2003;95:1467-75.  40.  Panjabi M M , Krag M , Summers D, et al. Biomechanical time-tolerance of fresh  cadaveric human spine specimens. J Orthop Res 1985;3:292-300.  41.  Wilke HJ, Krischak S, Claes L E . Formalin fixation strongly influences  biomechanical properties of the spine. J Biomech 1996;29:1629-31.  42.  Ismail A A , Cooper C, Felsenberg D, et al. Number and type of vertebral  deformities: epidemiological characteristics and relation to back pain and height loss. European Vertebral Osteoporosis Study Group. Osteoporos Int 1999;9:206-13.  43.  Sinaki M , Itoi E, Rogers JW, et al. Correlation of back extensor strength with  thoracic kyphosis and lumbar lordosis in estrogen-deficient women. A m J Phys Med Rehabil 1996;75:370-4.  44.  De Smet A A , Robinson RG, Johnson BP, et al. Spinal compression fractures in  osteoporotic women: patterns and relationship to hyperkyphosis. Radiology 1988;166:497500.  45.  Harma M , Heliovaara M , Aromaa A , et al. Thoracic spine compression fractures in  Finland. Clin Orthop 1986:188-94.  46.  Viner A , Lee M . Direction of manual force applied during assessment of stiffness in  the lumbosacral spine. J Manipulative Physiol Ther 1995;18:441-7.  47.  Cholewicki J, Juluru K, Radebold A , et al. Lumbar spine stability can be augmented  with an abdominal belt and/or increased intra-abdominal pressure. Eur Spine J 1999;8:38895.  127  Safety of Posteroanterior Spinal Mobilization  48.  Cholewicki J, Juluru K, McGill SM. Intra-abdominal pressure mechanism for  stabilizing the lumbar spine. J Biomech 1999;32:13-7.  49.  Hodges PW, Cresswell A G , Daggfeldt K, et al. In vivo measurement of the effect of  intra-abdominal pressure on the human spine. J Biomech 2001;34:347-53.  50.  Shirley D, Lee M , Ellis E. The relationship between submaximal activity of the  lumbar extensor muscles and lumbar posteroanterior stiffness. Phys Ther 1999;79:278-85.  51.  Kattan K R , Pais MJ. The spinous process: the forgotten appendage. Skeletal Radiol  1981;6:199-204.  52.  Grant R M . Physical therapy of the cervical and thoracic spine. 3rd ed. New York:  Churchill Livingstone, 2002.  53.  Grieve GP. Mobilisation of the spine: a primary handbook of clinical method. 5th  ed. New York: Churchill Livingstone, 1991.  54.  Hartman LS. Handbook of osteopathic technique. 3rd ed. London: Chapman & Hall,  1997.  55.  Twomey LT, Taylor JR. Physical therapy of the low back. 3rd ed. New York:  Churchill Livingstone, 2000.  128  Accuracy of DXA scanning of the thoracic spine  Chapter Five Accuracy of DXA Scanning of the Thoracic spine: Cadaveric Studies Comparing BMC, Areal BMD and Geometric Estimates of Volumetric BMD against Ash Weight and CT Measures of Bone Volume  129  Accuracy of DXA scanning of the thoracic spine ABSTRACT Background. Biomechanical studies of the thoracic spine often scan cadaveric segments using D X A to obtain measures of bone mass. Only one study reported accuracy of lateral scans of thoracic vertebral bodies. The accuracy of D X A scans of thoracic spine segments and of anteroposterior (AP) thoracic scans has not been investigated. Objectives. To (1) investigate the accuracy of A P and lateral thoracic D X A scans by comparison with ash weight, the gold standard for measuring bone mineral content; (2) to compare three methods of estimating volumetric bone mineral density (vBMD) with a novel standard— ash weight (g)/bone volume (cm ) as measured by computed 3  tomography (CT). Methods. Twelve T5-8 spine segments were scanned with D X A (AP and lateral) and CT. The T6 vertebrae were excised, posterior elements removed, and then the vertebral bodies were ashed in a muffle furnace. I proposed a new method of estimating v B M D and compared it with two previously published methods. Results. B M C values from lateral D X A scans displayed the strongest correlation with ash weight (r=0.99) and were on average 12.8% higher (p<0.001). As expected, B M C (AP or lateral) was more strongly correlated with ash weight than aBMD [AP (r=0.54) or lateral (r=0.71)] or estimated v B M D . Estimates of v B M D with any of the three methods were strongly correlated with v B M D calculated by dividing ash weight by CT-derived volume, but the mean difference was lowest when using the novel elliptical cylinder method.  130  Accuracy of DXA scanning of the thoracic spine Conclusions. These data suggest that readily available D X A scanning is an appropriate surrogate measure for thoracic spine bone mineral and the elliptical cylinder method should be used when calculating v B M D of the vertebral body.  131  Accuracy of DXA scanning of the thoracic spine 5.1 INTRODUCTION Thoracic spine osteoporosis often results in compression fractures with significant personal and societal cost.  1 - 3  In spine research, bone mass (aBMD and B M C ) is often  quantified by D X A . Previous studies have shown that lumbar spine B M C measured by D X A correlates closely with the gold standard— ash weight, the weight ofthe inorganic component of bone. vitro,  7-10  4 6  In addition, lumbar bone mass by D X A predicts failure load in  and fracture risk, in vivo.  11  The same relationship has not been established  (and therefore cannot be assumed) for the thoracic spine.  Ash weight is the most accurate measure of B M C . This method involves removing water and the organic matrix of bone, leaving only mineral. Bone is composed of water, mineral, proteins and other macromolecules such as lipids and sugars. The mineral or inorganic phase makes up to 60% of bone tissue. Most of the organic matrix, approximately 90%, is collagen while a much smaller 5 to 8% is noncollagenous proteins. Anatomic site, age, dietary history and presence of disease can result in 12  differences in bone composition.  Although numerous researchers have used D X A methodology to assess bone mineral in biomechanical studies of the thoracic spine,  13-18  the accuracy of D X A must be tested  specifically at the thoracic spine for several reasons. First, soft tissue mass (i.e. ligament, disc) and composition of nonmineralized connective tissue can limit the accuracy of aBMD results  1 9  and the thoracic and lumbar regions differ in this respect. Second, some  laboratories use bags of rice to simulate soft tissue,  ' ' as opposed to the typically  132  Accuracy of DXA scanning of the thoracic spine recommended water bath. There are no published data describing how this technical variant may influence the relationship between thoracic B M C measured by D X A scans and ash weight. Third, researchers often examine spine segments that include two or 14  17 7 0  77  more complete vertebrae and accompanying ligaments and intervertebral discs. ' ' ' In the thoracic spine, such specimens have not been scanned with D X A and then compared with a gold standard. Finally, an accuracy study is needed because more recent equations to estimate v B M D have never been tested against ash weight gold-standard. Traditionally, bone mineral data are reported as aBMD (grams per cm ), calculated by dividing the quantity of bone mineral within the scan area (BMC) by the projected area within the region of interest. Areal B M D values have clinical utility but can be confounded by changes in bone thickness (or depth). Bones of larger width and height also tend to be thicker yet bone thickness is not factored into estimates of aBMD. dependence on vertebral size is reduced.  2 4  2 3  By converting aBMD to v B M D the  This concept of v B M D aims to overcome the 24 27  limitation of 2-dimensional planar D X A scanning. "  However, this mathematical  derivation has never been compared with an ash weight gold standard at the thoracic spine. Further, previous methods assume vertebral geometry resembles a cylinder or a cube, yet human thoracic vertebral endplate geometry is best estimated as an ellipse. Thus, I proposed a new method of estimating v B M D , assuming vertebral body geometry resembles an elliptical cylinder.  133  Accuracy of DXA scanning of the thoracic spine To provide a solid methodological foundation for the use of D X A scan in thoracic spine research, my primary aim was to test the accuracy of A P and lateral thoracic spine D X A measures when rice provided the surrogate for soft tissues. The objective was to correlate D X A measures of the sixth thoracic vertebra (T6) from twelve fresh-frozen cadaveric spine segments with the ash weight of this vertebral body. M y secondary aim was to compare three different methods of estimating v B M D and their relationships with v B M D calculated by dividing ash weight by bone volume measured by CT.  134  Accuracy of DXA scanning of the thoracic spine 5.2 M E T H O D S I obtained fresh-frozen (unembalmed) human cadaveric spines from the U B C Department of Anatomy. Specimen donors included six females, five males, and one unknown. Age at death ranged from 62-93 years, mean 77 years. This study was approved by the U B C Clinical Ethics Review Board. Specimens were stored at -20° C until dissection.  Scanning with D X A Thoracic vertebrae T5-8 were separated as a segment from the rest of the spine (see Chapter Four, 'Specimen preparation'). Bone area (BA, cm ), B M C (g) and aBMD, 2  (g/cm ) were assessed in the AP and lateral orientation for each vertebra in the segment (T5-T8) using a Hologic QDR 4500W bone densitometer (DXA). A qualified technician analyzed all scans using standardized procedures as outlined in the Hologic Users 90  Guide.  For lateral scans, the posterior elements were excluded from the region of  interest. Members of my research group had conducted a short-term precision study, in vivo, with 17 subjects measured twice, using this instrument. The coefficient of variation for B M C and aBMD at the lumbar spine (L1-L4) was less than 0.7%. A spine anthropomorphic phantom was scanned daily to maintain quality assurance of the QDR 4500. As in other studies, ' ' bags of rice were used to simulate soft tissue 14 20 21  surrounding the spine segment. Scanning with C T and Measuring Bone Volume Each specimen was scanned with CT using 1.3 mm axial helical scans at a slice interval of 0.6 mm on bone algorithm (display field of view: 16 cm). Together with a radiologist,  135  Accuracy of DXA scanning of the thoracic spine I generated sagittal and coronal reconstructions on the scanner. Bone volume was measured on a G E Advantage Workstation 4.1 using the Fast Volume Rendering application by isolating the T6 vertebral body on a 3D model. Test-retest reliability was conducted with the same radiologist calculating bone volume for each specimen twice, with the second measure being made two weeks after the first. Reliability ranged from 0.2-9.7% (mean 2.7%, SD 2.7%).  Determining Ash Weight I dissected each T6 vertebra from its T5-8 segment and cleaned it of soft tissues after immersion in distilled water (75° C for 12 hours). The vertebral body was then separated from the posterior elements at the junction of the pedicle and body using a 200 pm diamond saw ( E X A K T 300, Norderstedt, Germany).  Each specimen was placed in a Coorstek (Fisher Scientific, Nepean, ON) ashing crucible (High Form 250 ml) with a cover (90 mm). The crucibles and covers were washed with distilled water and dried (at 95° C) for 48 hours prior to use. Vertebral bodies were dried (95° C for 48 hours) then ashed at 600° C for 42 hours in a Thermolyne 30400 muffle furnace. The maximum temperature of this muffle furnace is 900° C and the crucibles can be used to a maximum temperature of 1150° C. The furnace was turned up and down in increments of 200° C per hour to avoid rapid heating or cooling of the crucibles or samples (total time equaled 48 hours but time at 600° C was 42 hours). The temperature versus time profile for this furnace near the time of testing is presented in Appendix III.  136  Accuracy of DXA scanning of the thoracic spine Following each step the crucibles were transferred to a desiccator* for 24 hours and then weighed using a Sartorius BP 210 S scale (4 decimal places, Figure 5.1). Then they were returned to the desiccator for another 24 hours, after which they were weighed again to ensure a stable weight. The only exception was the ash weight measure, which was conducted three times (24 hours, 48 hours, and 72 hours) after transfer to the desiccator.  The following measurements and calculations were made (all from the desiccator): a) weight of the crucible and lid b) weight of the crucible c) weight of crucible and dried vertebra d) weight of crucible and vertebra after ashing e) weight of ash (d-a)  Volumetric B M D Volumetric B M D of the vertebral body was calculated as the ash weight (g)/ bone volume (cm ) from the CT scans. One specimen was excluded from v B M D measurement 3  due to missing CT scan data.  137  Accuracy of DXA scanning of the thoracic spine  Figure 5.1. Crucible and cover measurement with the Sartorius BP 210 S scale.  138  Accuracy of DXA scanning of the thoracic spine Estimated vBMD Volume was estimated using two previously reported methods. The Kroger et al. method  9 4 97  ' assumes vertebral body geometry resembles a cylinder using the equation: 4  F  v B M D = aBMD  Tlxdepth  where depth= vertebral body depth on a lateral radiograph (average of middle and 9r»  inferior width). The AP/lateral pair method  assumes the shape of the vertebral body  approximates a cube: aBMD vBMD= width Vertebral body width was measured on A P radiographs (average of middle and inferior width). Lateral aBMD values were used in both equations. A third method was proposed and compared with the two previously described methods. This method assumes vertebral body geometry resembles an elliptical cylinder: vBMD=  aBMD  where a= vertebral body width from the A P radiographs, and aBMD was taken from the lateral D X A scans, as in the previous two methods.  139  Accuracy of DXA scanning of the thoracic spine Statistical Analysis One specimen was excluded from statistical analysis as severe degenerative changes artificially elevated bone mass values obtained by D X A (i.e. lateral B M C of 5.25 g), and the T6 vertebra could not be dissected from the segment without removal of some osteophytes. A paired t-test was used to measure whether there was a statistical difference between ash weight (g) and B M C (g) by lateral D X A . Pearson correlation coefficients were used to assess the association between the primary outcome variables [ash weight (g), B M C (g) and aBMD (g/cm ) by D X A , v B M D [ash weight (g)/ CT bone 2  volume (cm )], estimated v B M D (g/cm )]. I calculated standard error of the estimate 3  3  (SEE) for the ash weight (g) and lateral D X A B M C (g) regression equation as well as the v B M D and estimated v B M D regression equations. I compared two methods of measuring vertebral body B M C , lateral D X A and ash weight, using the Bland-Altman method to determine the level of agreement and detect any outliers. " 30  32  I also used the  Bland-Altman method to determine the level of agreement between each of the three methods of estimating v B M D (elliptical cylinder, cube and cylinder) and v B M D by ash weight and CT. Analysis was conducted using standard statistical software (SPSS). Statistical significance was set at p< 0.05.  140  Accuracy of DXA scanning of the thoracic spine 5.3 R E S U L T S Range, mean, and standard deviation for all primary outcome variables are presented in Table 5.1. Ash weight (g) and B M C by lateral D X A (g) for individual specimens is shown in Figure 5.2.  • lateral DXA • ash weight  0  2  4  6  8  10  12  specimen number  Figure 5.2. Bone mineral content (g) data: lateral DXA and ash weight for all specimens.  Ash weight (g) and lateral D X A B M C (g) were statistically different (pO.OOl) yet highly correlated (r= 0.99, p<0.05). Pearson correlation coefficients for primary outcome variables are presented in Table 5.2. Lateral B M C (g) displayed the strongest correlation with ash weight (g), followed by A P B M C (g), lateral aBMD (g/cm ), A P aBMD (g/cm ), 2  2  and finally v B M D (g/cm ). The mean difference between ash weight (g) and lateral 3  141  Accuracy of DXA scanning of the thoracic spine D X A B M C (g) was 0.31 g +/- 0.10 (SD) (0.37 - 0.24 = 95% CI ofthe difference), which was a 12.8% mean difference [mean difference (g)/mean ash weight (g) X 100]. The standard error of the estimate (SEE) for the ash weight and lateral D X A B M C regression equation was 0.09 g. The SEE for the v B M D (ash weight/CT bone volume) and elliptical cylinder, cube, and cylinder regression equations was 0.01 g/cm . Figure 5.3 shows the 3  level of agreement between two methods of assessing vertebral body B M C (lateral D X A and ash weight). The mean difference between the methods is 0.31 g (SD=0.10 g).  Estimating v B M D as a cube, cylinder, or elliptical cylinder was similarly correlated with volumetric B M D measured by ash weight and CT (r=0.94 for cube, r=0.95 for cylinder, r=0.92 for elliptical cylinder), and statistically significant (Table 5.2). However, BlandAltman plots of each method of estimating v B M D against v B M D measured using ash weight and CT suggest some differences between the three methods. The mean difference between estimated v B M D using the elliptical cylinder method and ash/CT derived v B M D was -0.001 g/cm (SD=0.014), whereas the mean difference was -0.027 g/cm (SD=0.014) for the cube method, and 0.013 g/cm (SD=0.011) for the 3  3  cylinder method. Figures 5.4 A , B and C show the three Bland-Altman plots for the v B M D measures.  142  Accuracy of DXA scanning of the thoracic spine  Table 5.1. Mean, standard deviation (SD), and range for primary outcome variables at the T6 vertebra [AP and lateral BMC (g), AP and lateral aBMD (g/cm ), ash weight (g), 2  vBMD-CT (g/cm ), estimated (est) vBMD using cylinder, cube and elliptical cylinder 3  methods (g/cm )] 3  Mean AP BMC (g) Lateral BMC (g)  SD  Range  4.99  1.42  2.83-6.83  2.74  0.71  1.89-4.22  0.66  0.13  0.46-0.93  Lateral BMD (g/cm )  0.36  0.06  0.27-0.48  Ash weight (g)  2.43  0.68  1.58-3.84  vBMD-CT (g/cm )  0.14  0.03  0.09-0.23  est vBMD (g/cm ) (cylinder)  0.16  0.03  0.11-0.24  est vBMD (g/cm ) (cube)  0.12  0.03  0.08-0.18  est vBMD (g/cm ) (elliptical cylinder)  0.14  0.03  0.09-0.21  AP BMD (g/cm ) 2  2  3  3  3  3  143  Accuracy of DXA scanning of the thoracic spine  Table 5.2. Pearson correlation coefficients for primary outcome variables (ash weight (g) with AP and lateral BMC (g), and AP and lateral aBMD (g/cm ); ash weight (g) and 2  vBMD-CT (g/cm ) with estimated (est) vBMD (g/cm ) as a cube, cylinder, and elliptical cylinder * = significant at p< 0.05  ash weight (g)  vBMD (g/cm )  AP BMC (g)  *r=0.87, p<0.01  Lateral BMC (g)  *r=0.99, p<0.01  3  Lateral aBMD (g/cm )  *r=0.71, p=0.01  — — — —  est vBMD (g/cm ) (cylinder)  r=0.09, p=0.80  *r=0.95, p<0.01  est vBMD (g/cm ) (cube)  r=0.24, p=0.48  *r=0.94, p<0.01  est vBMD (g/cm ) (elliptical cylinder)  r=0.37, p=0.26  *r=0.92, p<0.01  AP aBMD (g/cm )  r=0.54, p=0.09  2  2  3  3  3  144  Accuracy of DXA scanning of the thoracic spine  0.6 +2 SD  0.5  3 0  s m  0.4 H  1 o c o  03  mean  • •  • •  0.2 -2 SD  0.1  2  3  Average (mean) BMC (g)  Figure 5.3. Limits of Agreement: difference in BMC (g) (difference of lateral DXA BMC and ash weight) vs. average BMC (g) (sum of lateral DXA and ash weight divided by 2).  145  Accuracy of DXA scanning of the thoracic spine  0.040H  +2SD 0.020H  O.OOOH  mean  -0.020 H -2SD -0.040 H  —I— 0.080  —I 0.100  1—  0.120  1  0.140  1  1  0.160  0.180  1  0.200  I  0.220  Mean vBMD (g/cm ) Elliptical Cylinder and Ash/CT methods 3  Figure 5.4A. Limits of Agreement: difference in vBMD (g/cm ) (difference of elliptical cylinder estimated vBMD and ash weight/CT vBMD) vs. mean vBMD (g/cm ) (sum of 3  elliptical cylinder and ash/CT vBMD divided by 2).  146  Accuracy of DXA scanning of the thoracic spine  +2SD  o.ooo -+  -0.010 H  -0.020 H  mean  -0.030 H  -0.040 H  -0.050 H  -2SD -0.060 H  —I— 0.080  1  0.100  0.120  0.140  1  1  0.160  0.180  I  0.200  0.220  Mean vBMD (g/cm ) Cube and Ash/CT methods 3  Figure 5.4B. Limits of Agreement: difference in vBMD (g/cm ) (difference of cube 3  estimated vBMD and ash weight/CT vBMD) vs. mean vBMD (g/cm ) (sum of cube and ash/CT vBMD divided by 2).  147  Accuracy of DXA scanning of the thoracic spine  +2SD 0.030 H  to  o  II B Q  E  mean  CQ  S < 0)  o c  0.020 H  0.01 O H  •D C  <D  0.000 -  C  -2SD -0.010 H  0.100  0.150  0.200  Mean vBMD (g/cm ) Cylinder & Ash/CT methods 3  Figure 5.4C. Limits of Agreement: difference in vBMD (g/cm ) (difference of cylinder estimated vBMD and ash/CT vBMD) vs. mean vBMD (g/cm ) (sum of cylinder and 3  ash/CT vBMD divided by 2).  148  Accuracy of DXA scanning of the thoracic spine 5.4 DISCUSSION M y results suggest that thoracic spine researchers may obtain the most accurate representation of midthoracic vertebral body bone mineral by using the lateral D X A scan B M C analysis. When the A P scan was used, the correlation of B M C and ash weight remained acceptably high (Table 5.2). These data, taken together with the previous study of lateral scanning of thoracic vertebrae, suggest that lateral scanning may be preferable 4  to A P scanning i f investigators are aiming to estimate B M C of thoracic vertebrae using D X A . This finding extends those previously limited to the lumbar spine.  6  D X A measures of aBMD did not correlate as closely with ash weight as did B M C (Table 5.2). This difference between B M C and aBMD in assessing ash weight of mineral is consistent with B M C being a closer biological correlate of mineral than aBMD which includes a measure of bone area, not mineral alone.  There was a strong statistically significant (p<0.01) correlation between estimated v B M D and v B M D calculated by dividing ash weight by vertebral body volume measured by CT (Table 5.2). Previous estimates of v B M D were performed on lumbar vertebrae and assumed that vertebral body geometry resembled a cylinder or a cube. Yet quantitative three dimensional anatomic studies suggest thoracic  2 8  and lumbar vertebral body 33  endplate dimensions are best approximated as an ellipse. I proposed a novel method of estimating v B M D that incorporated information gained from these anatomical studies."'" I found the mean difference in v B M D was lowest when using the novel elliptical cylinder method (Figures 5.4 A , B , C). The standard deviation of the difference was similar between the methods.  149  Accuracy of DXA scanning of the thoracic spine Laboratory researchers investigating spine biomechanics often scan a whole segment of the spine or functional spinal unit with D X A prior to testing, > ' > 13  14 17  22 n o  t only  disarticulated vertebrae. The experimental setup for these in vitro tests commonly excludes the ribs and thus, does not mirror the situation in vivo. I scanned each specimen as a segment (T5-8) as did Sabin et al.  6  Other authors have separated the posterior  elements from the vertebral body prior to scanning with D X A , with the goal of ashing precisely the bone that was scanned. ' Despite this difference, I report a correlation 4  5  between lateral B M C and ash weight that is very similar to that found in the only other study of thoracic vertebrae.  4  The consistent results may be related to similar D X A data  acquisition and analysis protocols, and/or similar methods for sectioning vertebrae. While soft tissues, such as the ligaments and discs, affect accuracy of bone mineral measurements, my results are similar to Edmondston et al. who scanned only the 4  vertebral body.  There was high agreement between the two methods of measuring B M C (ash weight and lateral D X A ; Figure 5.2). I would expect even higher agreement if the specimens were scanned after sectioning.  The relevance of the present study is in providing a methodologic foundation for present and future researchers who study the thoracic spine and use D X A as a method of determining bone mass. This is a field of increasing importance because of the increasing incidence and prevalence of osteoporosis. the one other study that addressed this question.  4  34  My data extend the findings of  Together, these studies allow  150  Accuracy of DXA scanning of the thoracic spine researchers to base their interpretation of thoracic D X A scan accuracy on data rather than relying on assumptions based upon lumbar spine experiments.  I conclude that, for testing, in vitro, lateral or AP B M C by D X A is highly correlated with ash weight of the vertebral body. This has practical implication in that thoracic spine researchers should use lateral D X A results if the vertebral body is their primary focus. The posterior elements are better represented in the A P scan but failure of the spinous process is predicted by neither AP nor lateral D X A measures (Chapter Four). Calculations of v B M D assuming that vertebral geometry resembles an elliptical cylinder, cube, or cylinder, as may be used in mathematical models, are all highly correlated with v B M D measured by ash weight and CT but the mean difference is lowest when using the novel elliptical cylinder method.  151  Accuracy of DXA scanning of the thoracic spine 5.5 R E F E R E N C E S 1.  Adachi JD, Loannidis G, Berger C, et al. The influence of osteoporotic fractures  on health-related quality of life in community-dwelling men and women across Canada. Osteoporos Int 2001;12:903-8.  2.  Matthis C, Weber U , O'Neill TW, et al. Health impact associated with vertebral  deformities: results from the European Vertebral Osteoporosis Study (EVOS). Osteoporos Int 1998;8:364-72.  3.  van der Klift M , de Laet C E D H , McCloskey E V , et al. The incidence of vertebral  fractures in men and women: the Rotterdam Study. J Bone Miner Res 2002;17:1051-6.  4.  Edmondston SJ, Singer K P , Price R l , et al. Accuracy of lateral dual energy X-ray  absorptiometry for the determination of bone mineral content in the thoracic and lumbar spine: an in vitro study. Br J Radiol 1993;66:309-13.  5.  Ho CP, K i m RW, Schaffler M B , et al. Accuracy of dual-energy radiographic  absorptiometry of the lumbar spine: cadaver study. Radiology 1990;176:171-3.  6.  Sabin M A , Blake G M , MacLaughlin-Black S M , et al. The accuracy of volumetric  bone density measurements in dual x-ray absorptiometry. Calcif Tissue Int 1995;56:2104.  7.  Edmondston SJ, Singer K P , Day RE, et al. Ex vivo estimation of thoracolumbar  vertebral body compressive strength: the relative contributions of bone densitometry and vertebral morphometry. Osteoporos Int 1997;7:142-8.  8.  Cheng X G , Lowet G, Boonen S, et al. Prediction of vertebral and femoral strength  in vitro by bone mineral density measured at different skeletal sites. J Bone Miner Res 1998;13:1439-43.  152  Accuracy of DXA scanning of the thoracic spine 9.  Moro M , Hecker AT, Bouxsein M L , et al. Failure load of thoracic vertebrae  correlates with lumbar bone mineral density measured by D X A . Calcif Tissue Int 1995;56:206-9.  10.  Shepherd DE, Leahy JC, Mathias KJ, et al. Spinous process strength. Spine  2000;25:319-23.  11.  Cummings SR, Black D M , Nevitt M C , et al. Bone density at various sites for  prediction of hip fractures. The Study of Osteoporotic Fractures Research Group. Lancet 1993;341:72-5.  12.  Einhorn TA, Simon SR, Surgeons. Orthopaedic Basic Science: biology and  biomechanics of the musculoskeletal system. Rosemont, 111: American Academy of Orthopaedic Surgeons, 2000.  13.  Belkoff S M , Mathis J M , Jasper L E , et al. A n ex vivo biomechanical evaluation of  a hydroxyapatite cement for use with vertebroplasty. Spine 2001;26:1542-6.  14.  Belkoff S M , Mathis J M , Jasper L E , et al. The biomechanics of vertebroplasty.  The effect of cement volume on mechanical behavior. Spine 2001;26:1537-41.  15.  Breeze SW, Doherty BJ, Noble PS, et al. A biomechanical study of anterior  thoracolumbar screw fixation. Spine 1998;23:1829-31.  16.  Butler TEJ, Asher M A , Jayaraman G, et al. The strength and stiffness of thoracic  implant anchors in osteoporotic spines. Spine 1994;19:1956-62.  17.  Coe JD, Warden K E , Herzig M A , et al. Influence of bone mineral density on the  fixation of thoracolumbar implants. A comparative study of transpedicular screws, laminar hooks, and spinous process wires. Spine 1990;15:902-7.  153  Accuracy of DXA scanning of the thoracic spine 18.  Hollowell JP, Vollmer DG, Wilson CR, et al. Biomechanical analysis of  thoracolumbar interbody constructs. How important is the endplate? Spine 1996;21:10326.  19.  Svendsen OL, Hassager C, Skodt V , et al. Impact of soft tissue on in vivo  accuracy of bone mineral measurements in the spine, hip, and forearm: a human cadaver study. J Bone Miner Res 1995;10:868-73.  20.  Belkoff S M , Mathis J M , Fenton DC, et al. A n ex vivo biomechanical evaluation  of an inflatable bone tamp used in the treatment of compression fracture. Spine 2001;26:151-6.  21.  Grant JP, Oxland TR, Dvorak M F . Mapping the structural properties of the  lumbosacral vertebral endplates. Spine 2001;26:889-96.  22.  Butler TE, Jr., Asher M A , Jayaraman G, et al. The strength and stiffness of  thoracic implant anchors in osteoporotic spines. Spine 1994;19:1956-62.  23.  Katzman D K , Bachrach L K , Carter DR, et al. Clinical and Anthropometric  Correlates of Bone Mineral Acquisition in Healthy Adolescent Girls. J Clin Endocrinol Metab 1991;73:1332-39.  24.  Kroger H , Vainio P, Nieminen J, et al. Comparison of different models for  interpreting bone mineral density measurements using D X A and M R I technology. Bone 1995;17:157-9.  25.  Carter DR, Bouxsein M L , Marcus R. New approaches for interpreting projected  bone densitometry data. J Bone Miner Res 1992;7:137-45.  154  Accuracy of DXA scanning of the thoracic spine 26.  Faulkner R A , McCulloch RG, Fyke SL, et al. Comparison of areal and estimated  volumetric bone mineral density values between older men and women. Osteoporos Int 1995;5:271-5.  27.  Kroger H, Kotaniemi A , Kroger L, et al. Development of bone mass and bone  density of the spine and femoral neck- a prospective study of 65 children and adolescents. Bone Miner 1993;23:171-82.  28.  Panjabi M M , Takata K, Goel V , et al. Thoracic human vertebrae. Quantitative  three-dimensional anatomy. Spine 1991;16:888-901.  29.  Hologic. Model QDR-4500 User's Guide. Waltham, M A : Hologic, Inc., 1996.  30.  Bland D G , Altaian J M . Measurement in medicine: analysis of method comparison  studies. Statistician 1983;32:307-17.  31.  Glantz SA. Primer of biostatistics. Fifth Edition ed. New York: McGraw-Hill,  2001.  32.  Kaymakci B , Wark JD. Precise accurate mineral measurements of excised sheep  bones using X-ray densitometry. Bone Miner 1994;25:231-46.  33.  Panjabi M M , Goel V , Oxland T, et al. Human lumbar vertebrae. Quantitative  three-dimensional anatomy. Spine 1992;17:299-306.  34.  Cummings SR, Melton LJ. Epidemiology and outcomes of osteoporotic fractures.  Lancet 2002;359:1761-7.  155  Regional vertebral morphology  Chapter Six Regional Trabecular Bone Morphology is Correlated with Thoracic Vertebral Failure under a Posteroanterior Load  156  Regional vertebral morphology ABSTRACT Background. Failure load of thoracic cadaveric vertebrae under a posteroanterior (PA) load (where fracture occurred at the base or middle of the spinous process)—and which may be linked to fracture during spinal mobilization—is not predicted by lateral or A P D X A or geometry of the spinous process or vertebral body. In Chapter Four I hypothesized that this may be due to B M D being an integrated measure of the entire vertebra, whereas these particular fractures occur in the spinous process. Objectives. The primary objective of this study was to measure trabecular bone morphology, including bone volume ratio (BV/TV), trabecular number (Tb.N), thickness (Tb.Th) and separation (Tb.Sp) using microcomputed tomography (pCT) in four regions of thoracic vertebrae and to correlate those measures with PA failure load at the adjacent vertebra. The secondary objectives were to (a) compare trabecular B V / T V , Tb.N, Tb.Th, Tb.Sp and structure model index (SMI) of the spinous process base with that found at the central lamina and middle spinous process regions, as a possible predictor of the fracture site; (b) assess the relationship of trabecular B V / T V at the vertebral body centrum and the spinous process base; and (c) measure cortical thickness in the posterior and anterior compartments of the spinous process base, and to correlate those measures with PA failure load. Methods. The T7 vertebra was dissected from 11 cadaveric midthoracic spine segments and sectioned to produce regional samples of the spinous process, the central lamina, and a central vertebral body core (8 mm diameter). Each sample was scanned with pCT (SkyScan 1072, 15pm nominal isotropic resolution). I segmented and analysed four trabecular regions (spinous process base, spinous process middle, central lamina and  157  Regional vertebral morphology vertebral body centrum). I measured cortical thickness in the posterior (tension) and anterior (compression) compartments of the spinous process base. Results. B V / T V at the base or middle of the T7 spinous process (fracture sites), mean Tb.N and mean Tb.Th at the base were strongly correlated with P A failure load of T6 (BV/TV base: r=0.74, p=0.01; B V / T V middle: r=0.73, p=0.01; mean Tb.N base: r=0.64, p=0.03; mean Tb.Th base: r=0.65, p=0.03). Mean Tb.Th of the central lamina was significantly greater than mean Tb.Th of the spinous process base (p=0.002). Conclusions. These novel data extend the findings of my previous study (Chapter Four) where A P and lateral B M D measures by D X A were not correlated with failure load of the spinous process. B V / T V of the base and middle regions of the spinous process were correlated with failure at those sites, and differences in Tb.Th at the base (compared with the lamina) may have influenced the site of fracture.  158  Regional vertebral morphology 6.1 I N T R O D U C T I O N In Chapter Four I found that B M D of the whole thoracic vertebra was not a good predictor of P A failure load, despite its strong correlation with axial compressive failure load in previous studies.  1,2  I hypothesized that this may be due to B M D being an  integrated measure of the entire vertebra, whereas these particular fractures occur in the posterior elements (Chapter Four, page 118). Also, B M D is only one factor associated with bone strength. '  3 4  Trabecular microarchitecture also contributes to bone strength and fracture risk " and 4  7  researchers have improved prediction of bone strength by including certain measures of Q  microarchitecture.  In the past, most quantitative information on three-dimensional  trabecular bone morphology was based on applying stereology* to two-dimensional sections. Disadvantages of this method are that the serial sectioning technique does not allow for subsequent mechanical testing or other secondary measurements since the samples are destroyed during preparation, both the preparation and analysis is time consuming, and stereology provides a biased estimate of trabecular bone morphology because it is two-dimensional.  Also, another limitation of calculating trabecular morphology from two-dimensional images is that their derivation indirectly assumes a fixed structure model, typically an ideal plate or rod model. With age there is a typical progression of trabecular structure from a more plate-like to a more rod-like structure.  9  If a sample deviates from the  assumed model it will result in a bias in the parameters measured.  9  Day et al  1 0  159  Regional vertebral morphology investigated the errors created when the parallel plate model was used to calculate morphometric parameters in the human proximal tibia, by comparison with direct thickness measures. They found trabecular thickness was consistently underestimated using the plate model and that use of the plate model resulted in a volume-dependent bias in thickness measures. They attributed this to the more rod-like structure of samples of low volume fraction.  1 0  In recent years high-resolution digital imaging techniques, such as microcomputed tomography (uCT), have permitted detailed quantitative nondestructive analysis of three11 12  dimensional microscopic bone structure.  '  Three-dimensional uCT image analysis  allows for direct measurement of trabecular thickness (Tb.Th), trabecular number (Tb.N) and trabecular separation (Tb.Sp) which is model-independent.  1 3  Three-dimensional  data sets represent the complex structure of trabecular bone for visual or biomechanical analysis, as well as for finite element modeling* (FEM) to predict mechanical properties.  12,14  To my knowledge, no studies have used this novel technology to measure  regional vertebral trabecular morphology, including regions of the posterior elements. During P A loading at the most posterior aspect of the spinous process of T6 (as in Chapter Four), the posterior side of the spinous process base is under tension and the anterior side is compressed. The thickness of the cortex in these two compartments (posterior and anterior) may be related to P A failure load.  160  Regional vertebral morphology Thus the primary objective of this study was to measure morphological parameters, including trabecular bone volume ratio (BV/TV), mean Tb.N, mean Tb.Th, and mean Tb.Sp using pCT in four regions of thoracic vertebrae and to correlate those measures with PA failure load at the adjacent vertebra. The secondary objectives were to (a) compare trabecular B V / T V , mean Tb.N, mean Tb.Th, mean Tb.Sp and structure model index (SMI) of the spinous process base with that found at the central lamina and middle spinous process regions, as a possible determinant of the fracture site; (b) assess the relationship between B V / T V ofthe vertebral body centrum with the spinous process base; and (c) measure cortical thickness in the posterior and anterior compartments of the spinous process base, and to correlate those measures with P A failure load.  161  Regional vertebral morphology 6.2 METHODS Eleven fresh-frozen (unembalmed) human cadaveric spines obtained from the U B C Department of Anatomy were used in this study (also used in Chapter Four and Chapter Five). Specimen donors included five females, five males, and one unknown. Age at death ranged from 62-93 years, mean 78 years. This study was approved by the University of British Columbia and the University of Calgary Institutional Clinical Ethics Review Boards. Each specimen was dissected to isolate a segment consisting of T5-T8 with intact disc and ligaments (see Chapter Four, 'Specimen preparation'). The isolated segments were stored frozen at - 2 0 ° C until testing.  Posteroanterior (PA) Failure Load Each T5-8 cadaveric spine segment was loaded to failure at a displacement rate of 2 mm/sec. Load was applied in the P A direction at the most posterior part of the spinous process of T6 using a servohydraulic material testing machine (Instron 8874), simulating the P A spinal mobilization technique. The procedure is detailed in Chapter Four, under the headings 'Mechanical Testing' and 'Post-test Measurements', pages 105-108.  Specimen Preparation The T7 vertebra was dissected and sectioned with a rotating diamond blade saw (ISOMET Low Speed Saw, Buehler, Lake Bluff, IL, U S A ; blade thickness= 0.012 inches) to produce regional samples of the spinous process and the central lamina.  162  Regional vertebral morphology A custom jig, coring bit and a drill were used to extract a central vertebral body core (8 mm diameter), in the posterior to anterior direction from each specimen (Figures 6.1 A and 6.IB). Each sample was placed in a custom plastic specimen holder and secured in position using low density foam prior to scanning (Figure 6.2).  Figure 6.1 A. This photograph shows the custom jig, coring bit and drill used to obtain central vertebral body cores from each T7 vertebral body. The vertebral body was held in position by four screws. The platform above the vertebra was adjusted both in tilt and position, to place the drill guide/hole directly over the center of the vertebral body.  163  Regional vertebral morphology  | | Drill  Figure 6.1B. Diagram of method of obtaining a central vertebral body core, with the posterior surface in the superior position.  Figure 6.2. Low density foam (left) was used to secure the sample in the custom holder (right) during pCT scanning.  164  Regional vertebral morphology Scanning with Microcomputed Tomography (pCT) Each sample was scanned with a high-resolution pCT scanner (SkyScan 1072; SkyScan, Aartselaar, Belgium), providing a nominal* isotropic* resolution of 15pm. The samples were rotated through 180 degrees at a rotation step of 0.9 degrees. The X-ray settings were standardized to lOOkV and 98pA with an exposure time of 6.0 seconds per frame. Two-frame averaging was used to improve the signal to noise ratio. A 1 mm-thick aluminum filter was employed to minimize beam hardening artifacts. The scan time for each sample was approximately one hour. Given the necessary field of view of 16 mm, the maximum resolution was 15 pm. Each scan was reconstructed using a cone-beam two dimensional reconstruction algorithm with superimposed ring-artifact reduction (SkyScan software). Each image series was passed through a median filter (5x5 cubic kernel using Image J v. 1.31) to partly suppress the noise in the original data. I segmented four trabecular regions (spinous process base, spinous process middle, central lamina, and vertebral body centrum) using a global threshold value to extract the mineralized bone phase.  Data Analysis Bone volume fraction (BV/TV) was calculated as the proportion of trabecular bone tissue volume with respect to total bone volume.  1 5  Trabecular thickness (Tb.Th) is determined  as an average of the local thickness at each voxel representing bone.  1 6  A series of  spheres is fit inside the bone phase. The largest sphere associated with each bone voxel is found and used to calculate an average thickness.  1 3  Both a single mean trabecular  thickness calculation and a histogram of trabecular thicknesses are produced. ' Trabecular separation (Tb.Sp), a measure ofthe distance between trabeculae, is measured  165  Regional vertebral morphology directly from the images using the same method used to measure Tb.Th but by applying the analysis to the marrow spaces instead of bone. Trabecular number (Tb.N) is the number of trabeculae per mm. Structure model index (SMI) is a dimensionless index estimating the relative prevalence of rods and plates in trabecular bone where an ideal plate and rod have SMI values of zero and three respectively.  15 17  '  B V / T V , mean Tb.Th (mm) mean Tb.N (per mm), mean Tb.Sp (mm) and SMI for each region was measured using standard CT Analyser software (v. 1.2.35.3 SkyScan, Aartselaar, Belgium). Studies report high correlations between morphometric analysis 11 18  performed with uCT and conventional histology.  '  Figure 6.3 shows a three-  dimensional image of the spinous process base for one specimen. Both the trabecular and cortical compartments are shown. Cortical thickness of the spinous process base was measured from two-dimensional images extracted from the three-dimensional data volume. I measured cortical thickness of the posterior compartment of the spinous process base (the side under tension during P A loading) and also at the anterior compartment (the side being compressed during P A loading) (Figure 6.4). Twenty-one images, each ten slices apart, were measured and the mean cortical thickness was calculated. Thus this measurement spanned 200 slices, the same 200 slices that were used in the trabecular analysis. Test-retest reliability was conducted by performing three measurements at each of five sites in one specimen. Reliability ranged from 0.004-0.045 mm (mean 0.021 mm, SD 0.014 mm).  166  Regional vertebral morphology  167  Regional vertebral morphology  Posterior  Anterior  Figure 6.4. Sample of a typical cortical thickness measurement at the posterior compartment of the spinous process, at the midline. Cortical thickness was measured as the length of the red line (2.67 mm).  Statistical Analysis Statistical analysis was conducted using SPSS 12.0 for Windows. A l l variables were normally distributed. Pearson correlations were computed to measure the relationship between P A failure load (N) and B V / T V , mean Tb.Th, mean Tb.N and mean Tb.Sp in each of the four regions, as well as P A failure load (N) and cortical thickness at the posterior and anterior spinous process base, at the midline. I used one-way repeated  168  Regional vertebral morphology measures A N O V A to compare trabecular B V / T V , mean Tb.Th, mean Tb.N, mean Tb.Sp and SMI in three regions (spinous process base, spinous process middle, and lamina) with region as the factor. Pearson correlation coefficient was computed to assess the relationship between B V / T V of the vertebral body centrum and the spinous process base. I set the significance level at 0.05 a priori. I used the paired t-test to compare Tb.Th in the spinous process base, middle and lamina, and applied a Bonferroni correction to correct for multiple pairwise comparisons. The Bonferroni correction gave me a new significance level of 0.016.  169  Regional vertebral morphology 6.3 R E S U L T S Range, mean and standard deviation for trabecular B V / T V , mean Tb.N, mean Tb.Th and mean Tb.Sp in each of the four regions (spinous process base, spinous process middle, central lamina, central vertebral body) are presented in Table 6.1. The mean (SD), range for SMI in each of the four regions was: base 1.53 (0.27), 1.23-2.07; middle 1.60 (0.45), 1.21-2.75; lamina 1.40 (0.29), 0.98-1.86; vertebral body 1.65 (0.38), 0.86-2.02. Mean cortical thickness of the anterior and posterior compartments of the spinous process base ranged from 0.18-1.34 mm (anterior) and 0.64-2.47 mm (posterior). Mean (SD) thickness was 0.65 mm (0.39 mm) and 1.64 mm (0.52 mm) for the anterior and posterior cortex respectively.  Pearson correlation coefficients for all metric trabecular bone measures and P A failure load are presented in Table 6.2. B V / T V at the base or middle of the T7 spinous process (fracture sites) showed a strong significant correlation with P A failure load of T6 (base: r=0.74, p=0.01 (Figure 6.5A); middle: r=0.73, p=0.01). Mean Tb.N and mean Tb.Th at the spinous process base was significantly correlated with P A failure load (mean Tb.N: r=0.64, p=0.03, Figure 6.5B; mean Tb.Th: r=0.65, p=0.03, Figure 6.5C). Mean Tb.Sp at the middle of the spinous process was inversely correlated with P A failure load (r= -0.61, p=0.048) (Figure 6.5D). B V / T V of the vertebral body centrum was not significantly correlated with B V / T V of the spinous process base (r=0.60, p=0.05). Anterior cortical thickness (compressed side) of the spinous process base was more strongly correlated with PA failure load than cortical thickness of the posterior  170  Regional vertebral morphology compartment, but neither correlation was statistically significant (anterior: r=0.51, p=0.11; posterior: r=0.13, p=0.69).  Mean Tb.Th was significantly lower at the spinous process than at the lamina (p=0.003, A N O V A ) (Figure 6.6). B V / T V between the three regions (spinous process base, middle and lamina) approached significance (p=0.06). There was a trend toward decreased B V / T V from the spinous process to the lamina (linear contrast p=0.03) Further comparison of Tb.Th with the paired t-test revealed a significant difference between the spinous process base and lamina (p=0.002).  171  Regional vertebral morphology  i 0.08  i 0.12  i 0.16  i 0.20  BV/TV Base  Figure 6.5A. Pearson correlation for bone volume/total volume (BV/TV) of the base of the T7 spinous process and posteroanterior (PA) failure load (N) ofT6 from eleven cadaveric spines. r=0.74, p=0.01  172  Regional vertebral morphology  i  i  i  i  i  0.60  0.70  0.80  0.90  1.00  Tb.N Base (per mm)  Figure 6.5B. Pearson correlation for mean Tb.N (per mm) at the base of the T7 spinous process and posteroanterior (PA) failure load (N) ofT6 from eleven cadaveric specimens. r=0.64,p=0.03  173  Regional vertebral morphology  0.140  0.160  0.180  0.200  0.220  Mean T b . T h (mm) B a s e  Figure 6 . 5 C . Pearson correlation for mean Tb. Th (mm) at the base of the Tl spinous process and posteroanterior (PA) failure load (N) ofT6 from eleven cadaveric specimens. r=0.64, p=0.03  174  Regional vertebral morphology  0.900  1.000  1.100  1.200  1.300  M e a n T b . S p (mm) Mid  Figure 6.5D. Pearson correlation for mean Tb.Sp (mm) at the middle of the T7 spinous process and posteroanterior (PA) failure load (N) ofT6 from eleven cadaveric specimens. r= -0.61, p=0.048  175  Regional vertebral morphology  Figure 6.6. Mean ± 1 standard deviation for Tb. Th (mm) in four regions: spinous process base (Base), spinous process mid (Mid), lamina and vertebral body (Body).  176  Regional vertebral morphology  Table  6.1. Range, mean and standard deviation for trabecular BV/TV, Mean Tb. Th (mm), Mean Tb.N (per mm) and Mean Tb.Sp (mm)  for each of the four trabecular regions analysed.  Region  Range  Mean  SD  Range  Mean  SD  Base Middle  0.068-0.217 0.080-0.211 0.115-0.265 0.041-0.222  0.131 0.144  0.046 0.034  0.134-0.224 0.134-0.227  0.167 0.174  0.032  0.160 0.108  0.039 0.059  0.144-0.243 0.104-0.222  0.195 0.153  Lamina Body  177  Mean Tb.Sp (mm)  Mean Tb.N (per mm)  Mean Tb.Th (mm)  BV/TV  0.028 0.027 0.035  Range 0.5330.5330.6000.400-  1.067 1.000 1.333 1.133  Mean  SD  Range  Mean  SD  0.770 0.836  0.159  0.827-1.231 0.824-1.291 0.710-1.276 0.591-1.203  1.010 0.961  0.135 0.130 0.141 0.202  0.818 0.667  0.161 0.202 0.237  0.991 0.929  Regional vertebral morphology  Table 6.2. Pearson correlations for the metric T7 trabecular bone measures (BV/TV, mean Tb.Th, mean Tb.N, mean Tb.Sp) in each region (Base= spinous process base; Mid- spinous process middle; Lamina= central lamina; Body= vertebral body centrum) and T6 PA failure Load (N), in cadaveric vertebrae, ^significant at p< 0.05  Variable and Region Base BV/TV Mean Tb.Th Mean Tb.N Mean Tb.Sp Mid BV/TV  Correlation with PA Failure Load (N) r 0.74*  P 0.01  0.65* 0.64* -0.40  0.03  -  -  0.03 0.23  0.73*  0.01  Mean Tb.Th  0.50  0.12  Mean Tb.N  0.46  0.12  Mean Tb.Sp  -0.61*  0.048  Lamina BV/TV  -  -  0.57  0.07  Mean Tb.Th  0.32  0.33  Mean Tb.N Mean Tb.Sp  0.43 -0.51  0.19 0.11  Body  -  -  BV/TV  0.44  0.10  Mean Tb.Th  0.55  0.08  Mean Tb.N  0.32  0.33  Mean Tb.Sp  -0.52  0.10  178  Regional vertebral morphology 6.4 DISCUSSION This investigation using uCT suggests that B V / T V of the base and middle regions of the spinous process is correlated with failure when the spinous process is subjected to a PA load. Further, the lower mean Tb.Th of the spinous process base compared with the lamina may have influenced the site of fracture.  I remind the reader that integral vertebral body B M D measured by D X A was not correlated with failure load of the spinous process (Chapter 4). That finding, taken together with the results of this investigation, suggests that scientists and clinicians should not rely on bone mass measures from the vertebral body if they seek information about the posterior elements. My results are relevant in the case of spinal fixation and implant research, where D X A may be used to gain insights into bone strength. M y results extend the findings of Coe et al.  1 9  who found a significant correlation between  B M D and loads to failure for transpedicular screws and spinous process wires, but no significant correlation between B M D and load to failure for laminar hooks. Furthermore, the results of this study suggest that vertebral body trabecular B V / T V is not strongly correlated with trabecular B V / T V of the posterior elements.  A strength of this study was the use of three-dimensional image analysis by pCT which allowed direct measurement of trabecular thickness, number and separation that is modelindependent.  1 5  This avoids model-dependent errors associated with two-dimensional  analysis, ' and avoids errors that occur with planar methods such as D X A . 10 15  179  Regional vertebral morphology Cortical thickness of the spinous process did not predict P A failure load. It is possible that other spinous process cortical measures, such as cortical porosity, may predict P A failure load even though cortical thickness did not. I could not measure cortical porosity because the large size of my samples provided inadequate resolution of cortical pores. Cortical thickness in the compressed anterior compartment was more strongly correlated with P A failure load than posterior cortical thickness, although this was not statistically significant, and warrants further investigation with a larger sample size. It is also possible that PA failure load is related to a combination of trabecular and cortical bone mass and structure, and this could be tested in a larger study. Finally, although my sample size was normally distributed, it was small and thus, conclusions must remain limited.  A limitation of this study was that I studied P A failure of T6 vertebrae and then investigated whether regional vertebral morphology of T7 vertebrae predicted T6 failure load. Clearly, mechanical testing and measures of morphology on the same vertebrae would have been ideal but this was not possible because T6 was damaged. The presence of this damage to the bone precluded a meaningful measurement of the bone morphology. Although vertebral compressive strength has been shown to increase caudally ' the increase in the cross-sectional area of the vertebral end-plate,  due to  there is no evidence  to suggest that PA failure load would be significantly different between T6 and T7. B M C and B M D , by D X A , also gradually increases caudally in the midthoracic spine but, as I found in Chapter Four, lateral and AP D X A are not good predictors of spinous process failure load.  180  Regional vertebral morphology At present, our uCT scanner is unable to scan an entire cadaveric vertebra therefore I sectioned each vertebra and scanned each region individually. While quantitative computed tomography (QCT) can accommodate whole vertebrae and spine segments, the resolution (100-400 urn depending on the machine) is not sufficient for 3D analysis of trabecular microarchitecture. New high-resolution uCT scanners have recently become available (XtremeCT, Scanco Medical, Basserdorf, Switzerland) that will allow measurement of entire vertebral bodies at resolutions sufficient to establish trabecular morphology (30-40 micrometers).  Whereas previous data indicate that B M D by D X A is not a good predictor of PA failure load,  2 1  regional B V / T V of the spinous process base and middle regions, the sites of  fracture, is strongly correlated with PA failure load. It is noteworthy that trabecular thickness differs significantly between the spinous process base and lamina regions, and may have influenced the site of fracture in Chapter Four.  181  Regional vertebral morphology 6.5 R E F E R E N C E S 1.  More- M , Hecker AT, Bouxsein M L , et al. Failure load of thoracic vertebrae  correlates with lumbar bone mineral density measured by D X A . Calcif Tissue Int 1995;56:206-9.  2.  Singer K , Edmondston S, Day R, et al. Prediction of thoracic and lumbar vertebral  body compressive strength: correlations with bone mineral density and vertebral region. Bone 1995;17:167-74.  3.  Genant H K , Gordon C, Jiang Y , et al. Advanced imaging of the macrostructure  and microstructure of bone. Horm Res 2000;54 Suppl 1:24-30.  4.  Rosen CJ. Pathophysiology of osteoporosis. Clin Lab Med 2000;20:455-68.  5.  Parfitt A M . Implications of architecture for the pathogenesis and prevention of  vertebral fracture. Bone 1992;13 Suppl 2:S41-7.  6.  Kleerekoper M , Villanueva AR, Stanciu J, et al. The role of three-dimensional  trabecular microstructure in the pathogenesis of vertebral compression fractures. Calcif Tissue Int 1985;37:594-7.  7.  Guglielmi G, Gluer CC, Majumdar S, et al. Current methods and advances in  bone densitometry. Eur Radiol 1995;5:129-39.  8.  Muller R, Bauss F, Smith S, et al. Mechano-structure relationships in normal,  ovariectomized and Ibandronate treated aged macaques as assessed by microtomographic imaging and biomechanical testing. Transactions of the 47th Annual Meeting of the Orthopaedic Research Society. San Francisco, California, 2001:66.  9.  Muller R. Bone microarchitecture assessment: current and future trends.  Osteoporos Int 2003;14 Suppl 5:89-99.  182  Regional vertebral morphology 10.  Day JS, Ding M , Odgaard A , et al. Parallel plate model for trabecular bone  exhibits volume fraction-dependent bias. Bone 2000;27:715-20.  11.  Muller R, Van Campenhout H, Van Damme B , et al. Morphometric analysis of  human bone biopsies: a quantitative structural comparison of histological sections and micro-computed tomography. Bone 1998;23:59-66.  12.  Borah B, Gross GJ, Dufresne TE, et al. Three-dimensional microimaging  (MRmicroI and microCT), finite element modeling, and rapid prototyping provide unique insights into bone architecture in osteoporosis. Anat Rec 2001;265:101-10.  13.  Hildebrand T, Ruegsegger P. A new method for the model independent  assessment of thickness in three-dimensional images. J Microsc 1997;185:67-75.  14.  van Rietbergen B, Weinans H, Huiskes R, et al. A new method to determine  trabecular bone elastic properties and loading using micromechanical finite-element models. J Biomech 1995;28:69-81.  15.  Sky Scan. Structural parameters of trabecular bone measured by the Sky Scan  micro-CT [SkyScan, Aartselaar, Belgium], 2004.  16.  Ulrich D, van Rietbergen B, Laib A , et al. The ability of three-dimensional  structural indices to reflect mechanical aspects of trabecular bone. Bone 1999;25:55-60.  17.  Judex S, Boyd S, Yi-Xian Q, et al. Combining High-resolution Micro-computed  Tomography with Material Composition to Define the Quality of Bone Tissue. Current Osteoporosis Reports 2003;1:11-9.  18.  Uchiyama T, Tanizawa T, Muramatsu H, et al. A morphometric comparison of  trabecular structure of human ilium between microcomputed tomography and conventional histomorphometry. Calcif Tissue Int 1997;61:493-8.  183  Regional vertebral morphology 19.  Coe JD, Warden K E , Herzig M A , et al. Influence of bone mineral density on the  fixation of thoracolumbar implants. A comparative study of transpedicular screws, laminar hooks, and spinous process wires. Spine 1990;15:902-7.  20.  Biggemann M , Hilweg D, Brinckmann P. Prediction of the compressive strength  of vertebral bodies of the lumbar spine by quantitative computed tomography. Skeletal Radiol 1988;17:264-9.  21.  Sran M M , Khan K M , Zhu Q, et al. Failure Characteristics of the Thoracic Spine  with a Posteroanterior Load: Investigating the Safety of Spinal Mobilization. Spine 2004;29:2382-8.  184  Posteroanterior Stiffness and Intervertebral Motion  Chapter Seven Posteroanterior Stiffness as a Predictor of Intervertebral Motion in Cadaveric Thoracic Spine Segments  185  Posteroanterior Stiffness and Intervertebral Motion ABSTRACT Background. Spinal joint mobilization is a mainstay of clinical assessment of individuals with back pain. The clinician manually assesses stiffness and joint motion relative to segments above and below. Although clinical theory suggests that manually performed techniques can predict or detect intervertebral motion, this hypothesis remains untested. Objectives. To correlate stiffness measured during mechanically simulated central posteroanterior (PA) mobilization with (1) flexion and extension range of motion (ROM; Deg), (2) flexion/extension neutral zone (NZ) motion (Deg) and (3) flexibility (Deg/Nm) in three dimensions, in midthoracic cadaveric spine segments of older adults. Methods. Using a precision opto-electronic camera system and a custom spine testing machine, I measured intervertebral R O M , N Z motion and three-dimensional flexibility in eight T5-8 cadaveric specimens (mean age 81 yrs). I then measured stiffness when a cyclic P A load was applied at the spinous process of T6 using a servohydraulic material testing machine (Instron 8874), simulating the P A spinal mobilization technique. Results. There was a strong significant inverse relationship between stiffness during cyclic P A loading of T6 and flexion or extension R O M of T6 relative to T7 (r= -0.88, p O . O l , extension; r= -0.81, p=0.01, flexion), and T6-7 flexibility in all six directions. Conclusions. Stiffness during simulated central cyclic P A mobilization in the cadaveric midthoracic spine is inversely correlated with flexion and extension R O M and threedimensional flexibility at the level at which the technique is applied. These findings provide biomechanical support for the inclusion of specific joint mobilization in the assessment of older adults with back pain.  186  Posteroanterior Stiffness and Intervertebral Motion INTRODUCTION When evaluating patients with back pain, musculoskeletal health care practitioners including physiotherapists commonly use spinal joint mobilization to assess the stiffness and joint motion relative to segments above and below.  1  Stiffness is a term used to  describe the force needed to achieve a certain deformation of a structure. One very common clinical spinal mobilization technique is P A mobilization.  2 - 4  It can be applied  at an individual spinous process ' to assess stiffness at a specific segmental level. 5  6  In the assessment of spine pain, studies have found that manual examination by a physiotherapist is highly accurate in detecting the segmental level responsible for a patient's complaint when compared against a spinal block, ' and is correlated with disability.  9  Clinical dogma suggests that the diagnostic assessment technique of spinal  joint mobilization also provides a measure of spinal motion ' ' but this hypothesis has 5  6  10  never been tested formally. Three studies have previously measured P A spinal stiffness in the thoracic spine by inferring vertebral translations from the motion of an indenter at the skin-surface,  IM3  but no previous study has examined the association between  stiffness during P A mobilization and an objective measure of intervertebral motion.  For this reason, my primary objective was to correlate stiffness measured during mechanically simulated central P A mobilization of T6 with (1) flexion and extension range of motion (ROM; Deg), (2) flexion/extension neutral zone (NZ) motion (Deg) and (3) flexibility (Deg/Nm) in three dimensions (flexion, extension, right and left lateral  187  Posteroanterior Stiffness and Intervertebral Motion bending, right and left axial rotation), in midthoracic cadaveric spine segments of older adults.  188  Posteroanterior Stiffness and Intervertebral Motion METHODS Eight fresh-frozen (unembalmed) human cadaveric spines were obtained from the U B C Department of Anatomy. Donors included five females and three males. Age at death ranged from 70-93 years, mean 81 years. Each specimen was given a number for identification to maintain anonymity. This study was approved by the U B C Clinical Ethics Review Board.  Specimen Preparation Each specimen was dissected to isolate a segment consisting of T5-T8 with intact discs and ligaments (see Chapter Four, 'Specimen preparation'). The isolated segments were stored frozen at -20° C until testing.  Lateral and A P radiographs were taken of each specimen prior to testing. No specimen showed any sign of malignancy. One specimen had a 20% wedge compression fracture of T8 and another specimen had signs of severe degenerative change (i.e. large osteophytes).  Steel (24 gauge) wire was secured to the pedicles of T5 and T8 of each spine segment. Each specimen was embedded in dental cement such that half of the vertebral bodies of T5 and T8 and the attached steel wire were fixed in cement. Thus, the T5 transverse processes were fully embedded and half of the T8 transverse processes were embedded. The T5-6 and T7-8 facet joints remained free, as did all parts of T6 and T7.  189  Posteroanterior Stiffness and Intervertebral Motion Measuring Intervertebral Range of Motion, Neutral Zone Motion and Flexibility A custom spine testing machine  1 4  was used to load the specimens in flexion/extension,  axial rotation and lateral bending (Figure 7.1). This machine consists of a DC motor, a planetary reduction gearhead, an articulated arm of two universal joints, a ball spline and a torque cell (model TRT-200, Transducer Techniques, C A , USA). Pure moments (no other external forces were applied to the spine) were applied to the upper vertebra at 2.0 degrees per second to a moment maximum of 4.0 Nm, so as to be non-destructive. I applied three complete loading cycles, as described by W i l k e ,  15  in flexion/extension,  axial rotation and lateral bending. Marker carriers with four infrared light emitting diodes (LEDS) were secured to the vertebral body of T5 (middle) and the transverse processes of T6-8 with 3.5 mm cancellous bone screws. In cases where the transverse process was too weak to sufficiently secure the marker carrier, the middle of the vertebral body was used. Each marker carrier was positioned to be in clear view of the cameras while avoiding contact with dental cement or another marker carrier during the test. A n opto-electronic camera system (Optotrak 3020, Northern Digital, Waterloo, Ontario, Canada) monitored the three-dimensional positions of the marker carriers. The accuracy of this system exceeds 0.1°.  16  Cyclic PA Loading (simulated PA mobilization) Next, the spine segment was oriented horizontally in the testing machine (Instron 8874, Instron Corp. Canton, M A ) to facilitate the P A load application (Figure 7.2). The T5 and T8 specimen mounts were clamped rigidly, such that the T6 spinous process was aligned with the linear actuator of the machine. Load was applied through a circular delrin  190  Posteroanterior Stiffness and Intervertebral Motion indenter (20 mm diameter, Young's Modulus 3.1 GPa), mounted on the end of the actuator. I chose a 20 mm diameter to simulate pisiform hand contact (Figure 1.5) and to ensure that the load was only applied at the T6 spinous process. This indenter had a 7 mm diameter groove for the spinous process and 3 mm foam (PPT, Langer Biomechanics Group Inc, N Y ) was attached to the bottom of the indenter to enable a distributed load transmission and to prevent slipping of the spinous process. A very low load (5 N) functional test was used to verify that the intended P A load did not produce noticeable coupled axial rotation.  A cyclic P A load of 50-200 N was applied to the most posterior point of the T6 spinous process at 0.5 Hz for 30 seconds. Kinematic data were recorded during the test at 10 Hz using a precision opto-electronic camera system (Optotrak 3020, Northern Digital, Waterloo, Ontario). Figure 7.3 shows vertebral motion during PA mobilization at T6, as found in Chapter Four.  191  Posteroanterior Stiffness and Intervertebral Motion  Figure 7.1. A typical specimen set-up for the flexibility tests. This photo shows the base marker carrier (below the specimen), the marker carriers attached to the specimen, the specimen embedded in dental cement, fixed in the custom testing jig and secured to the spine flexibility machine.  192  Posteroanterior Stiffness and Intervertebral Motion  Posteroanterior Load  Figure 7.2. Schematic of posteroanterior (PA) loading at the T6 spinous process in T5T8 cadaveric spine segments. The five opto-electronic marker carriers are shown.  193  Posteroanterior Stiffness and Intervertebral Motion  Figure 7.3. Diagram of normal alignment ofT5, T6 and T7 (top left) and vertebral motion during PA mobilization at T6, where T6 moves into extension relative to T5, and T5 and T7flex relative to T6 and T8.  Data Analysis For each spinal level, I determined vertebral kinematics via the opto-electronic marker carriers. The anterior-inferior point on the vertebral body was the origin and a global coordinate system that was aligned with a base marker fixed to the machine (Figure 7.1 and 7.2) was used for all tests. The x-axis was in the anterior-posterior direction in the cyclic P A mobilization tests and the positive x-axis was to the left in the flexibility tests  194  Posteroanterior Stiffness and Intervertebral Motion (Figure 1.7). The y-axis was in the cephalad-caudad direction in all tests and the z-axis was the cross-product of the x and y axes. I assumed each vertebra to be a rigid body for measurement and I used custom software (KIN 2000, written in LabVIEW 6.0, National Instruments, USA).  For the tests conducted with the spine machine, kinematic data were collected from three cycles, but only the third cycle was used for statistical analysis. The R O M was defined as angular deformation in flexion and extension at the peak moments  15  '  17  (Figure 7.4).  N Z motion was defined as the maximum difference in rotation within the momentrotation curve between moment magnitudes of ± 0.2 N m (Figure 7.4). Flexibility was calculated as the slope of the motion (Deg) versus moment (Nm) curve during the loading portion of the curve in each direction, using all points on that portion of the curve (Figure 7.5). I calculated flexibility at all three spinal levels (T5-6, T6-7 and T7-8).  Stiffness (kN/Deg) during P A mobilization was determined as the slope of the P A force versus T6-7 rotation angle curve for the loading portion of the curve between the two force limits (50-200 N). The slope was calculated using a simple linear regression using all points on the curve. I also determined P A stiffness using indenter motion (mm) (with PA force, as above) for comparison with previous studies of PA stiffness in the midthoracic spine.  195  Posteroanterior Stiffness and Intervertebral Motion One specimen had signs of severe degenerative change and was found to have small flexibility values (Table 7.2). However, this specimen's R O M and flexibility values fell within ± 3 SD of the group mean so I did not remove it from the statistical analysis.  Statistical Analysis Pearson correlation coefficients were used to assess the relationship between T6-7 stiffness during cyclic P A loading of T6 and (1) flexion and extension R O M of the T5-6, T6-7, and T7-8 functional spinal units, (2) N Z motion in flexion/extension at T6-7, and (3) flexibility (Deg/Nm) of T5-6, T6-7 and T7-8 in extension, flexion, left and right axial rotation, and left and right lateral bending. Statistical significance was set at p<0.05.  -5  -4  -3  Extension  -2  -1  0  1  Moment (Nm)  2  3  4  5  Flexion  Figure 7.4. Graphical representation of Range ofMotion (ROM; Deg), Neutral Zone (NZ) motion (Deg). Dashed line (—) represents the moment magnitudes of ±0.2 Nm.  196  Posteroanterior Stiffness and Intervertebral Motion  -5  -4  -3  Extension  -2  -1  0  1  M o m e n t (Nm)  2  3  4  5  Flexion  Figure 7.5. Flexibility was calculated as the slope of the motion (Deg) versus moment (Nm) curve during the loading portion of the curve in each direction (dark solid lines), using all points on that portion of the curve. Dashed line represents the slope.  197  Posteroanterior Stiffness and Intervertebral Motion RESULTS T6-7 P A stiffness ranged from 0.25 to 2.02 kN/Deg (mean 0.99, SD 0.70). Mean, range and SD for flexion and extension R O M at T5-6, T6-7 and T7-8 are presented in Table 7.1. The mean, range, and SD flexibility for extension, flexion, axial rotation, and lateral bending of T5-6, T6-7, and T7-8 (average of all three intervertebral levels) is presented in Table 7.2. There was a strong significant inverse relationship between stiffness during cyclic P A loading of T6 and flexion or extension R O M of T6 relative to T7 [r- -0.88, p<.01, extension; r= -0.81, p= 0.01, flexion]. The relationship between T6-7 P A stiffness and T6-7 N Z motion in flexion/extension was not statistically significant (r= -0.60, p=0.11). There was a strong statistically significant relationship between T6-7 PA stiffness and T6-7 flexibility (Deg/Nm) in all directions (extension, flexion, left and right axial rotation, left and right lateral bending). There was no relationship between T6-7 P A stiffness and flexion/extension R O M or three-dimensional flexibility at T5-6 or T7-8 (Table 7.3). One specimen's right lateral bending flexibility data were excluded due to contact with dental cement during the test.  PA stiffness (N/mm) using the motion of the indenter at the surface of the spinous process (mm), ranged from 91-245 N/mm (mean= 148, SD= 42).  198  Posteroanterior Stiffness and Intervertebral Motion Table 7.1. Mean, range and standard deviation range of motion (Deg) in flexion and extension at T5-T6, T6-T7 and T7-T8. Flexion  Mean  Range Standard Deviation  Extension  T5-T6  T6-T7  T7-T8  T5-T6  T6-T7  T7-T8  0.69  1.25  0.83  0.80  1.31  0.99  0.02-2.06  0.17-2.54  0.17-2.31  0.07-1.92  0.15-2.24  0.28-2.52  0.66  0.87  0.66  0.73  0.81  0.74  Table 7.2. Mean, range and standard deviation (average ofT5-6, T6-7 and T7-8) flexibility (Deg/Nm) for extension,flexion,rotation and lateral bending in T5-T8 cadaveric spine segments. Rotation and lateral bending include motion to the left and right. T5-T8 Flexibility (Deg/Nm) Mean Range Standard Deviation  Extension  Flexion  Axial Rotation  Lateral Bending  0.31  0.27  0.62  0.43  0.00-0.88  0.01-0.67  0.04-1.58  0.03-1.02  0.25  0.18  0.43  0.26  199  Posteroanterior Stiffness and Intervertebral Motion  Table 7.3. Pearson correlation coefficients for T5-T6, T6-T7 and T7-T8 extension and flexion range of motion (Deg), andflexibility (Deg/Nm) in three dimensions [extension, flexion, right and left axial rotation (Rotn.), right and left lateral bending (hat.Bend)] with posteroanterior (PA) stiffness (kN/Deg).  A  Range of Motion (Deg)  Stiffness (kN/Deg)  T5-6 Extension T5-6 Flexion  r= - 0 . 4 5 , p= 0.26 r= -0.48, p= 0.22 r= -0.88*, p< 0.01  T6-7 Extension T6-7 Flexion T7-8 Extension T7-8 Flexion Flexibility (Deg/Nm) T5-6 Extension T5-6 Flexion T5-6 Right Lat. Bend T5-6 Left Lat. Bend T5-6 Right Axial Rotn. T5-6 T6-7 T6-7 T6-7 T6-7 T6-7 T6-7 T7-8  Left Axial Rotn. Extension Flexion Right Lat. Bend Left Lat. Bend Right Axial Rotn., Left Axial Rotn. Extension  T7-8 Flexion  * = significant at p<0.05  r= -0.81*, p= 0.01 r= -0.38, p= 0.35 r= -0.32, p= 0.43  -  r= - 0 . 6 5 , r= -0.62, r= -0.17, r= -0.32, r= - 0 . 3 1 ,  p= p= p= p=  0.08 0.10 0.69 0.44  p=0.45 r= -0.26, p=0.53 r= -0.90*, p< 0.01 r= -0.87*, p< 0.01 r= -0.78*, p= 0.02 r= -0.82*, p= 0.01 r= -0.82*, p= 0.01 r= -0.79*, p= 0.02 r= - 0 . 4 3 , p= 0.28 r= - 0 . 5 3 , p=0.18  T7-8 Right Lat. Bend T7-8 Left Lat. Bend  r= - 0 . 6 8 , p= 0.09 r= - 0 . 2 1 , p= 0.62  T7-8 Right Axial Rotn.  r= - 0 . 5 1 , p= 0.20 r= 0.04, p=0.92  T7-8 Left Axial Rotn.  :n =7  A  200  Posteroanterior Stiffness and Intervertebral Motion DISCUSSION M y findings suggest that stiffness during simulated central cyclic P A mobilization in the cadaveric midthoracic spine is inversely correlated with flexion and extension R O M (Deg) at that spinal level. Further, I found that T6-7 P A stiffness is a strong predictor of flexibility (Deg/Nm) in all six directions (flexion, extension, left and right lateral bending, left and right axial rotation) at T6-7 (Table 7.3). The significant results at T6-7 only may be related to T5-6 and T7-8 intervertebral motion being somewhat limited by my methodology, which may account for the lower mean R O M at these levels compared with T6-7 (Table 7.1). Intervertebral motion is also influenced by differences in intervertebral disc height spinal level to the next.  19  and morphology and orientation of the facet joints from one In Chapter Four I found that the mobilized thoracic vertebra  moves into extension under a PA load (Figure 7.3). This may explain the slightly stronger relationship between T6-7 extension R O M (Deg) and flexibility (Deg/Nm) with T6-7 P A stiffness (kN/Deg), as compared with flexion R O M or flexibility.  20 21  Many clinicians and scientists consider N Z motion to be analogous to joint laxity.  '  My results suggest that P A stiffness is not a good predictor of N Z motion thus clinicians should likely use other techiniques, rather than P A mobilization, if the aim is to assess 22  laxity in the midthoracic spine. In this study I measured P A stiffness, intervertebral R O M , N Z motion and threedimensional flexibility in cadaveric midthoracic spine segments consisting of four vertebrae. A previous study  2 3  measured flexibility of the cadaveric thoracic spine using  201  Posteroanterior Stiffness and Intervertebral Motion a two-vertebra construct that included intervening ligaments and the head and neck of articulating ribs. Despite these methodological differences, our extension and lateral bending results are similar except that I found greater flexibility in axial rotation and less in flexion. The greater flexibility in axial rotation may be related to my having removed the ribs  2 4  or anatomical differences between the thoracic regions tested in the two  studies. Note that I studied 24 motion segments from the midthoracic region of eight thoracic spines, while Panjabi et al. tested 11 motion segments, one for each thoracic level from five thoracic spines.  Three previous studies have measured P A spinal stiffness in the thoracic spine by inferring translation from the motion of an indenter at the skin-surface " u  T7 mean P A stiffness measures of 10.7 N/mm  11  1 3  and reported  and 12.5 N/mm ; T4 P A stiffness of 12  1 T  13.6 N/mm.  1J  For comparison only, I also determined P A stiffness using motion of the  indenter (mm) at the surface of the spinous process and the P A force data. I reported much higher stiffness than these previous studies (mean 148 N/mm) using this method. This can be attributed to methodological differences between the studies. I studied midthoracic cadaveric spine segments consisting of four vertebrae and their intervening discs and ligaments. With a longer specimen (i.e. more than four vertebrae) I would expect lower stiffness and increased flexibility. Further, measuring P A stiffness by inferring vertebral translation from the motion of an indenter, which is separated from the bone by 3 mm thick foam, similar to the skin and soft tissues in vivo studies, is likely to affect the accuracy of the stiffness measure. Thus I would expect P A stiffness, in vivo, to be reduced due to the length of the spine (compared to my study) and increased by the rib  202  Posteroanterior Stiffness and Intervertebral Motion cage. Another important difference between the studies, which might explain some of the variability in P A stiffness measures, is that my specimens were much older than the individuals included in previous investigations (mean age 28.6 years, years,  12  11  mean age 26.6  age range 20-42 years ). 13  There are a number of strengths of this study. First, I measured vertebral motion using bone pins, whereas previous studies calculated stiffness based on vertebral translations inferred from the motion of an indenter at the skin surface.  1 1 - 1 3  Further, the accuracy of  the motion measurement system used in my study exceeds 0.1°, I applied cyclic P A mobilization with loads similar to those applied in vivo, ' 2  0.5Hz, consistent with clinical guidelines.  25>26  and at a frequency of  6  I used a moment maximum of 4 Nm to avoid damage to the specimen during testing. Panjabi et al.  2 3  used a moment maximum of 5 N m but they also tested the lower thoracic  spine. Studies of the cervical and lumbar (and thoracolumbar) spine have typically used maximum moments between 1.5-2.0 Nm and 7.5-10.0 N m respectively, so the 4.0 N m used in my study falls in between these values and is similar to a previous study in the thoracic spine.  Clinical Implications These novel data provide biomechanical support for a technique that clinicians use in daily practice to assess spinal joint motion.  4  - ' 6  1 0  -  1 2  .  2 5  >  2 7  -  3 3  T  f  ounc  } that stiffness under a  cyclic PA load, simulating the PA mobilization technique in cadaveric midthoracic spine  203  Posteroanterior Stiffness and Intervertebral Motion segments of older adults, reflects spinal R O M and flexibility at the particular level at which it is applied. These data support the clinical practice of PA mobilization in the assessment of patients with spine pain, if the goal is to gain specific information about intervertebral motion.  204  Posteroanterior Stiffness and Intervertebral Motion REFERENCES 1.  Cyriax J. Textbook of Orthopaedic Medicine. 8th ed. London: Balliere Tindall,  1982.  2.  Lee R, Evans J. A n in vivo study of the intervertebral movements produced by  posteroanterior mobilization. Clin Biomech 1997;12:400-8.  3.  Lee M , Liversidge K. Posteroanterior stiffness at three locations in the lumbar  spine. J Manipulative Physiol Ther 1994; 17:511-6.  4.  Bjornsdottir SV, Kumar S. Posteroanterior spinal mobilization: state of the art  review and discussion. Disabil Rehabil 1997;19:39-46.  5.  Magarey M E . Examination of the Cervical and Thoracic Spine. In Grant R ed.  Physical Therapy of the Cervical and Thoracic Spine. 2nd ed. New York: Churchill Livingstone, 1994:442.  6.  Maitland GD, Banks K, English K, et al. Maitland's vertebral manipulation. Sixth  ed. Boston: Butterworth Heinemann, 2001.  7.  Jull G , Bogduk N , Marsland A . The accuracy of manual diagnosis for cervical  zygapophysial joint pain syndromes. Med J Aust 1988;148:233-6.  8.  Phillips DR, Twomey L T . A comparison of manual diagnosis with a diagnosis  established by a uni-level lumbar spinal block procedure. Man Ther 1996;2:82-7.  9.  Lundberg G, Gerdle B. Correlations between joint and spinal mobility, spinal  sagittal configuration, segmental mobility, segmental pain, symptoms and disabilities in female homecare personnel. Scand J Rehab Med 2000;32:124-33.  205  Posteroanterior Stiffness and Intervertebral Motion 10.  Grieve GP. Mobilisation of the spine: a primary handbook of clinical method. 5th  ed. New York: Churchill Livingstone, 1991.  11.  Edmondston SJ, Allison GT, Althorpe B M , et al. Comparison of ribcage and  posteroanterior thoracic spine stiffness: an investigation of the normal response. Man Ther 1999;4:157-62.  12.  Lee M , Steven GP, Crosbie J, et al. Variations in posteroanterior stiffness in the  thoracolumbar spine: preliminary observations and proposed mechanisms. Phys Ther 1998;78:1277-87.  13.  Lee M , Latimer J, Maher C. Manipulation: investigation of a proposed  mechanism. Clin Biomech 1993;8:302-6.  14.  Goertzen DJ, Lane C, Oxland TR. Neutral zone and range of motion in the spine  are greater with stepwise loading than with a continuous loading protocol. A n in vitro porcine investigation. J Biomech 2004;37:257-61.  15.  Wilke HJ, Wenger K , Claes L. Testing criteria for spinal implants:  recommendations for the standardization of in vitro stability testing of spinal implants. Eur Spine J 1998;7:148-54.  16.  Hamming JA, Goertzen DJ, Oxland TR. The effect of marker configuration and  placement on kinematic accuracy. Canadian Orthopaedic Association, 54th Annual Meeting, 1999:59.  17.  White A , Panjabi M . Clinical biomechanics of the spine. Second ed. Philadelphia:  JB Lippincott Company, 1990.  206  Posteroanterior Stiffness and Intervertebral Motion 18.  Kulak RF, Schultz A B , Belytschko T, et al. Biomechanical characteristics of  vertebral motion segments and intervertebral discs. Orthopedic Clinics of North America 1975;6:121-33.  19.  Edmondston SJ, Singer KP. Thoracic spine: anatomical and biomechanical  considerations for manual therapy. Man Ther 1997;2:132-43.  20.  Panjabi M M . The stabilizing system of the spine. Part II. Neutral zone and  instability hypothesis. J Spinal Disord 1992;5:390-6.  21.  Richardson C, Jull G, Hodges P, et al. Therapeutic Exercise for Spinal Segmental  Stabilization in Low Back Pained. London: Churchill Livingstone, 1999.  22.  Lee DG. Manual Therapy for the Thorax - a Biomechanical Approach. Delta:  Delta Orthopaedic Physiotherapy Clinic, 1994.  23.  Panjabi M M , Brand R A , Jr., White A A , 3rd. Three-dimensional flexibility and  stiffness properties of the human thoracic spine. J Biomech 1976;9:185-92.  24.  Oda I, Abumi K , Lu D, et al. Biomechanical role of the posterior elements,  costovertebral joints, and rib cage in the stability of the thoracic spine. Spine 1996;21:1423-9.  25.  Goodsell M , Lee M , Latimer J. Short-term effects of lumbar posteroanterior  mobilization in individuals with low-back pain. J Manipulative Physiol Ther 2000;23:332-42.  26.  Simmonds MJ, Kumar S, Lechelt E. Use of a spinal model to quantify the forces  and motion that occur during therapists' tests of spinal motion. Phys Ther 1995;75:21222.  207  Posteroanterior Stiffness and Intervertebral Motion 27.  Lee M , Liversidge K. Posteroanterior stiffness at three locations in the lumbar  spine. J Manipulative Physiol Ther 1994; 17:511-6.  28.  Lee M , Steven GP, Crosbie J, et al. Towards a theory of lumbar mobilisation-the  relationship between applied manual force and movements of the spine. Man Ther 2000;1:67-75.  29.  Lee R, Evans J. Load-displacement-time characteristics of the spine under  posteroanterior mobilization. Aust J Physiother 1992;38:115-23.  30.  Lee R, Evans J. A n in vivo study of the intervertebral movements produced by  posteroanterior mobilization. Clin Biomech 1997;12:400-8.  31.  Lee R Y , Evans JH. The role of spinal tissues in resisting posteroanterior forces  applied to the lumbar spine. J Manipulative Physiol Ther 2000;23:551-6.  32.  Shirley D, Ellis E, Lee M . The response of posteroanterior lumbar stiffness to  repeated loading. Man Ther 2002;7:19-25.  33.  Chiradejnant A , Maher C G , Latimer J. Objective manual assessment of lumbar  posteroanterior stiffness is now possible. J Manipulative Physiol Ther 2003;26:34-9.  208  General Discussion, Summary and Conclusions  Chapter Eight General Discussion, Summary and Conclusions  209  General Discussion, Summary and Conclusions GENERAL DISCUSSION In this final chapter I discuss and summarise the findings ofthe six studies that comprise this thesis. I discuss the results of my studies on (1) the effectiveness of manual therapy for spinal pain, (2) physiotherapists' perceptions and practice patterns with respect to the management of individuals with osteoporosis, (3) the safety of manual therapy in older spines with low bone density, (4) the accuracy of D X A in the thoracic spine and differences between three methods of estimating v B M D , (5) pCT measures of regional bone volume ratio and trabecular microarchitecture as predictors of vertebral failure under a P A load, and (6) P A stiffness as a predictor of intervertebral motion. For each study, I also discuss limitations and recommend future studies that would be a logical extension of my research to date. I close this chapter with a summary, conclusions and my contributions to the field.  8.1 Evidence for the Effectiveness of Manual Therapy for Spinal Pain My systematic review (Chapter Two) suggested that there were clinically relevant differences between studies reporting positive results of manual therapy and those reporting no significant difference over other conservative treatments. In particular, key aspects of the manual therapy intervention had received little research attention to date. M y analysis indicated that: (1) manual therapy techniques, rather than joint manipulation alone, appeared to yield better results; (2) interventions in which the treatment protocol reflected clinical practice appear to be consistently more successful (i.e. protocols where a number of manual therapy techniques were used or manual therapy was combined with  210  General Discussion, Summary and Conclusions another mode of treatment; (3) interventions based on 'best practice' guidelines were successful; (4) physiotherapy including manual therapy at a dose of 30-45 minutes per session for 4-8 weeks was effective in reducing pain and improving function. M y conclusions differ from previous systematic reviews and meta-analyses that included older studies and failed to critically assess the elements of the manual therapy intervention itself. ' I limited my review to a recent six year period to reflect the l 2  dramatic changes in clinical practice, and thus clinical trials, due to research published in the late 1990s.  To extend this research, a future review might include all clinical trials published on the topic, but also include a further subanalysis within the years I studied to scrutinize the impact of changes in clinical practice and an increase in relevant physiotherapy research. However, at present there may be insufficient data to warrant such a study. Also, this type of review will likely be confounded by improvements in methodological factors over time.  A limitation of my review is the lack of statistical analysis. Further information may be gained by conducting a meta-analysis, but the large variability of the manual therapy interventions used in clinical trials (i.e. type, dose), as identified in my review, should be taken into consideration.  Future research should seek to better identify those patients who have a high likelihood of improving with manual therapy treatment. For example, Flynn et al identified five  211  General Discussion, Summary and Conclusions variables to form a clinical prediction rule for patients with low back pain who are likely to respond favorably to a specific manipulative technique.  Manual therapy is not only used in the treatment of low back and neck pain. Pilot studies have been conducted in individuals with thoracic pain and cervico-brachial pain 4  syndrome. The next step is for researchers to conduct well designed RCTs to determine 5  the effectiveness of manual therapy for pain and disability in individuals with these conditions, or patients with pain in other regions or with different diagnoses.  To date, clinical trials of manual therapy have been conducted in adults. Future studies should consider investigating the effectiveness of manual therapy in other age groups (i.e. teenagers, older adults) and special populations, such as those with osteoporosis.  8.2 Physiotherapy and Osteoporosis: Clinicians' Perceptions and Practice Behaviors I found that physiotherapists reported using evidence-based therapies when treating individuals with osteoporosis. A number also reported using manual therapy in this population. Physiotherapists reported concern about fracture as a complication of manual therapy treatment, in particular vertebral fracture and rib fracture.  The brief questionnaire used in this study provides novel information on clinical practice but also has limitations. I did not measure how often manual therapy was used, nor did I provide an opportunity for therapists to clarify or expand on their concerns. I surveyed  212  General Discussion, Summary and Conclusions physiotherapists working in all areas of practice (public and private workplaces) with or without any specific postgraduate qualifications. A survey of only those with postgraduate qualifications in manual therapy may provide different data and insights. It is not unreasonable to speculate that physiotherapists may also have concerns about other modes of treatment in those with osteoporosis, such as exercise prescription. A further survey would need to be administered to investigate this question. As my research focused on manual therapy, I asked the specific question about the use of manual therapy.  In the opening question I stated that the individual had osteoporosis, but did not state what the individual was being treated for (i.e. osteoporosis or an unrelated condition), or whether treatment was being given in an area where osteoporotic fractures commonly occur. While this may have been helpful for the participants, I felt that this additional information could have alerted therapists that I was interested in their knowledge of osteoporosis (i.e. it is a systemic condition mainly affecting trabecular bone) and common sites of osteoporotic fracture (Appendix I, Question 3).  Using quantitative research alone was a limitation. I met my specific objectives but this small survey justifies further investigation of therapists' beliefs and concerns using qualitative methodology. For example, it is possible that barriers to care exist, such as treatment time, which could limit a clinician's ability to provide optimal care. Specifically, conventional appointment times may be relevant for younger adults attending physiotherapy but may not be adequate for older individuals who present with various comorbidities and/or restricted mobility.  213  General Discussion, Summary and Conclusions  8.3 Safety of Posteroanterior Spinal Mobilization There were five key findings from my study in Chapter Four: (1) vertebral body injury is an unlikely complication of PA mobilization in the midthoracic spine; (2) failure load, in vitro, and applied load, in vivo, were significantly different for most specimens, however the lowest fracture thresholds were around the same force as the upper range of the applied loads; 3) P A failure typically occurs at the spinous process; (4) plain radiography and CT scan have poor sensitivity for that spinous process fracture; and (5) vertebral kinematics in the midthoracic spine during PA mobilization differ from previous reports in the lumbar spine.  The limitations of this study relate to the spine segment model, which is devoid of muscle contraction, the rib cage, intraabdominal organs and intraabdominal pressure. A l l of these are likely to increase spinal stiffness in vivo. Nevertheless, my purpose was to investigate the failure load and failure site of midthoracic vertebrae under a P A load and there is no evidence to suggest that the associated skeleton and organs would increase strength of the spinous process, in vivo. Further, I validated my spine segment kinematics with an intact cadaver and found both the magnitude and direction of vertebral motion to be similar. I could not find any reports of spinous process fracture associated with mobilization and my data suggest that if this pathology were to exist then traditional methods of fracture diagnosis, such as plain radiography and CT scan, would have poor sensitivity to detect it. The inability of two radiological methods to detect anatomically proven fractures suggests that a trial of more intensive investigation (bone  214  General Discussion, Summary and Conclusions scan, MRI) may be warranted in patients who present with a clinical history that strongly suggests vertebral fractures but in whom x-ray is normal.  Bone is a viscoelastic material and thus its biomechanical behavior varies with the rate of loading. Bone is stiffer and is able to sustain a higher load to failure when loads are applied at a higher rate. However, at very high strain rates (>1 per second) bone 6  becomes more brittle, such as in impact trauma. Clinically, the loading rate influences the fracture pattern and the amount of soft tissue damage that occurs. Stored energy is released when a bone fractures. At a low loading rate, such as that used in my study, the energy can dissipate through the formation of a single crack so bone and soft tissues remain relatively intact. This is consistent with the fractures produced in my study. At a high loading rate, as occurs in impact trauma, the greater stored energy cannot dissipate quickly enough through a single crack, and thus a comminuted fracture and extensive soft tissue damage is often the result.  6  I used a loading rate of 2 mm/second. Manual therapy  techniques can be applied slowly or more quickly, as with high velocity thrust techniques (manipulation). M y aim was to study the P A spinal mobilization technique that can often be applied at a faster rate but it is still slow as compared with manipulation. However, my aim was not only to measure the failure load but also to investigate the site of failure under a PA load. Due to the short delay in stopping the test once fracture occurred, a faster loading rate would have resulted in further damage between the time of initial failure and the time that the machine was stopped. Thus, my loading rate allowed for simulation of the mobilization technique but also allowed sufficient control of the experiment.  215  General Discussion, Summary and Conclusions  The World Health Organization classification of osteoporosis is based on a history of fragility fractures or D X A based B M D values obtained from the proximal femur and lumbar spine. Thus, it is not appropriate to classify my specimens as being osteoporotic based on their thoracic spine B M D . Also, the WHO classification applies to in vivo scans, which clearly differs from scanning cadaveric specimens. Nevertheless, the lumbar spines of the specimens used in all of my cadaveric studies were also scanned by D X A at U B C . The t-scores for the lumbar vertebrae of these specimens were all below -2.5.  Of note, German researchers found that both in situ and ex situ lumbar AP D X A B M C and B M D were similarly correlated with the failure load of L 3 .  7  This suggests that D X A  scan results from spine segments are unlikely to show significantly different correlations with failure load than scan results from intact cadavers.  I measured the load applied by two experienced physiotherapists, one male and one female, both trained in the same undergraduate and postgraduate programs. The range of applied loads in my study (106-223 N) was smaller than some previous studies (i.e. 60230 N ). Differences in the number of participants, the region studied and the experience and training of the therapists can contribute to variability in applied force.  I used the Tekscan system to measure applied load in the intact cadaver tests. Calibration of the Tekscan unit with a six-axis load cell revealed consistent underestimation of  216  General Discussion, Summary and Conclusions applied load data (Appendix II). In future, other alternatives to the Tekscan should be considered for such research. For example, Herzog and colleagues used a thin flexible pressure mat to measure applied load. Future studies requiring this type of load 9  measurement may consider comparing the accuracy of such a system with that of the Tekscan system.  The result of my Chapter Four research has implications for physiotherapy practice and training. My results provide an indication of typical loads applied by experienced therapists and the failure load of vertebrae under a PA load and suggests that there is only a small amount of overlap between physiotherapists' applied load and the failure load in the cadaveric specimens. M y study provides specific data that could be used to help train physiotherapists so they can be made aware of the loads they are applying. A previous study found that physiotherapists can be trained to apply specific forces.  1 0  This type of  training may improve therapist confidence for the use of manual therapy in individuals with osteoporosis, and increase the safety margin.  The pattern of fracture can provide valuable information about the forces involved in this loading mode. I photographed the failure site after each failure test. A limitation of this method was that the poor resolution in some photos made it difficult to clearly see the whole fracture line. High resolution CT could be used to better assess fracture patterns. High-resolution CT images may help guide researchers on which bone measures (i.e. cortical thickness, which region of the cortex) might be associated with this fracture pattern.  217  General Discussion, Summary and Conclusions The information gained through this study, specifically the applied load and P A failure load data, could be used to develop accurate mathematical models so that the influence of anatomical variations could be investigated. For example, a common postural fault in individuals with osteoporosis is an excessive midthoracic kyphosis. Modeling a spine with an increased kyphosis, along with structural and material characteristics of the vertebrae, such as the spinous process length, the size of the vertebral body, and the bone mineral density of the vertebra, may provide additional information on the effect of these factors on P A failure load. Such biomechanical findings could then be correlated with clinical measures of kyphosis, to provide a guide to clinicians as to what degree of kyphosis constitutes a major risk for vertebral compression fracture. The analytical process of Receiver Operator Curve development could be used to identify appropriate thresholds for risk, and with that, potential interventions. Interventions could include strength training, spinal orthoses  11  or manual therapy, but each of these would need to be  tested in the relevant population.  8.4 Accuracy of DXA Scanning of the Thoracic Spine In Chapter Five I found that lateral or AP B M C of spine segments by D X A was highly correlated with the ash weight of the vertebral body. M y research extended the one previous study of D X A accuracy in thoracic vertebrae by evaluating both A P and lateral D X A scans. Also I scanned spine segments, as this is often the unit tested in thoracic spine research, rather than the vertebral body itself. Previously published methods of estimating v B M D have never been tested against an ash weight gold standard. I tested  218  General Discussion, Summary and Conclusions the two previous published methods and I also proposed a new method of calculating v B M D (elliptical cylinder). I found that calculations of v B M D assuming that vertebral body geometry resembles an elliptical cylinder, cube, or cylinder, as may be used in mathematical models, were all highly correlated with v B M D measured by ash weight and CT. However the mean difference was lowest when using the new elliptical cylinder method.  8.5 Regional Vertebral Trabecular Bone Morphology  In Chapter Four I reported that bone mineral density of the whole vertebra was not a good predictor of P A failure load, despite its known correlation with axial compressive failure load. ' 12  13  I then hypothesized that this may be due to B M D being an integrated measure  of the entire vertebra, whereas the fractures from P A loading occurred in the spinous process. Thus the purpose of the study reported in Chapter Six was to test whether regional trabecular bone morphometry, measured using pCT, could predict failure at this site. I found that bone volume fraction of the base and middle regions of the spinous process predicted failure at those sites. Further, the lower mean trabecular thickness of the spinous process base compared with the lamina may have influenced the site of fracture. Whether a difference in trabecular thickness influences vertebral body fracture warrants future consideration. The question of what causes vertebral bodies to fail remains a germane question in the field. As mentioned previously (Chapter Six), it is only recently that pCT scanners have become able to scan whole vertebral bodies.  219  General Discussion, Summary and Conclusions I measured cortical thickness from two-dimensional images due to limitations of the SkyScan software. Future studies of cortical thickness and its relationship with failure load should consider using a scanner and software which allows for three-dimensional cortical thickness measures. Beyond their interest in my study, these findings provide important evidence for the role of trabecular bone in maintaining structural strength.  8.6 Posteroanterior Stiffness as a Predictor of Intervertebral Motion Although clinical theory has long suggested that manually performed spinal techniques can predict or detect intervertebral motion, this hypothesis has not been previously tested. I found that stiffness during central cyclic P A mobilization in the cadaveric midthoracic spine of older adults was a strong predictor of flexion and extension R O M and threedimensional flexibility at that spinal level. M y findings provide biomechanical support for the inclusion of specific joint mobilization in the assessment of older individuals with back pain.  However, the limitations of this study must also be acknowledged. First, the cadaver specimens were all from older adults, so the generalizability of my findings to younger clinical populations must be considered with caution. Also, my cadaveric study results do not account for muscle contraction, the rib cage, intraabdominal organs and intraabdominal pressure, all of which are likely to increase spinal stiffness, in vivo. " 14  17  I  did, however, previously validate the cadaveric spine segment kinematics against those of an intact cadaver (Chapter Four) and found both the magnitude and direction of vertebral  220  General Discussion, Summary and Conclusions motion was similar. Further, the length of the specimen can also influence P A stiffness, but this was kept constant for all specimens and tests. However, use of longer spine segments (i.e. six vertebrae) may provide a better indication of whether P A stiffness at one level can predict R O M and flexibility at adjacent levels.  Although my results suggest that P A stiffness is correlated with intervertebral motion at the spinal level at which the technique is applied, I did not address the extent to which therapists can perceive spinal stiffness. Maher and colleagues investigated therapist inter- and intrarater reliability in measuring P A stiffness. In one study they found therapists had much better ability to judge spring stiffness than the P A stiffness of human 18  spines  but another study showed that therapists can accurately judge spinal stiffness  using a matching task.  1 9  Future research might consider whether there is a threshold  between degrees of stiffness (mild, moderate, severe) which therapists can detect and other more subtle variations in stiffness that are difficult to perceive with manual testing.  221  General Discussion, Summary and Conclusions  8.7 Summary and Conclusions Summary 1. Interventions based on 'best practice' guidelines or textbooks written by experts appear to be more successful, and physiotherapy including manual therapy at a dose of 30-45 minutes per session, for 4-8 weeks is effective in adult populations with back or neck pain.  2. Many physiotherapists practicing in British Columbia, Canada use evidence based methods (i.e. strength training) when treating individuals with osteoporosis. A large number use manual therapy and most have concerns about its use in individuals with osteoporosis. Physiotherapists are concerned about fracture as a complication of treatment, in particular vertebral fractures. Studies on the safety and effectiveness of manual therapy in this population are needed to guide clinical practice.  3. Although there was a reasonable margin between the vertebral failure load, in vitro, and the applied PA mobilization load, in vivo, for most specimens, the lowest fracture thresholds were around the same force as the upper range of the applied loads. Vertebral body injury is an unlikely complication of PA mobilization in the midthoracic spine. Plain radiography and CT scan had poor sensitivity for the spinous process fractures produced. Kinematic data suggest that the mobilized thoracic vertebra moves into extension as a result of the mobilization.  222  General Discussion, Summary and Conclusions 4. D X A scanning is an appropriate surrogate measure for thoracic spine segment bone mineral measurement. Calculations of v B M D assuming that vertebral body geometry resembles an elliptical cylinder, cube, or cylinder, as may be used in mathematical models, were all highly correlated with v B M D measured by ash weight and CT but the mean difference was lowest when using the novel elliptical cylinder method.  5. B M D measures by D X A did not correlate with failure load of the spinous process. However, further regional investigation using pCT showed that B V / T V of the base and middle regions of the spinous process correlated with failure at those sites. The lower mean Tb.Th of the spinous process compared with the lamina may have influenced the site of fracture.  6. Stiffness under a cyclic P A load in vitro, simulating the P A mobilization technique in the midthoracic spine, reflected spinal R O M and flexibility at the particular level at which it was applied. These data support the clinical practice of specific P A mobilization in the assessment of patients with spine pain, if the goal is to gain specific information about intervertebral motion.  223  General Discussion, Summary and Conclusions Conclusions 1. Physiotherapy including manual therapy at a dose of 30-45 minutes per session, for 48 weeks is effective in adult populations with back or neck pain. There are clinically relevant differences between the manual therapy interventions used in clinical trials that may influence the outcomes.  2. Many physiotherapists in British Columbia, Canada use manual therapy in patients with osteoporosis and most have concerns about its use in this population. Physiotherapists are concerned about fracture as a complication of treatment, in particular vertebral fracture.  3. Vertebral body injury is an unlikely complication of PA mobilization in the midthoracic spine. P A mobilization produces spinous process fractures. Areal B M D of the whole vertebra is not a good predictor of PA failure load, despite its strong correlation with axial compressive failure load in previous studies. The mobilized thoracic vertebra moves into extension as a result of the mobilization.  4. D X A scanning is an appropriate surrogate measure for thoracic spine segment bone mineral measurement. The elliptical cylinder method should be used when calculating v B M D of the vertebral body.  5. Regional B V / T V of the spinous process base and middle regions, the sites of fracture, is strongly correlated with P A failure load. Trabecular thickness differs significantly  224  General Discussion, Summary and Conclusions between the spinous process and lamina regions, and may have influenced the site of fracture.  6. In midthoracic spine segments, P A stiffness predicts segmental R O M and flexibility at the level at which the PA mobilization is applied.  8.8 Contributions  I made the following novel contributions to the field:  1. M y systematic review (Chapter Two) identified four factors that differed among the interventions that constituted 'manual therapy' that may explain the conflicting outcomes in clinical trials and previous reviews.  2. M y survey of clinicians' perceptions and practice behaviors with respect to the management of individuals with osteoporosis (Chapter Three) is the first study published on this topic. This study provides quantitative data on the most common treatment modes used by physiotherapists in the province of British Columbia, Canada when treating individuals with osteoporosis, including the percentage that are using manual therapy and are concerned about its use in this population. This survey also highlights the type of injuries clinicians are concerned about.  225  General Discussion, Summary and Conclusions 3. M y biomechanical study on the safety of PA mobilization in the midthoracic spine (Chapter Four) provides novel data on the safety of manual therapy in older spines with low bone density. The results suggest that some manual therapy techniques may be safe for treating back pain in individuals with low bone density and that vertebral body injury is unlikely with this technique. The imaging techniques used in this study were not sensitive to the fractures produced and bring to the surface the issue of possible false negatives with plain radiography and CT scan when imaging vertebral fractures. M y kinematic analysis, both in spine segments and an intact cadaver, provide accurate data on the magnitude and direction of motion during PA mobilization in the midthoracic spine—which has not been previously measured using bone pins.  4. My methodological study on the accuracy of D X A scanning of thoracic spine segments (Chapter Five) will allow researchers to base their interpretation of A P and lateral thoracic D X A scan accuracy on data rather than relying on assumptions based upon lumbar spine experiments. I proposed a new accurate method of calculating v B M D based on the results of quantitative three dimensional anatomical studies which suggest that thoracic vertebral endplate dimensions are best approximated as an ellipse.  5. The results of my regional investigation of vertebral bone volume fraction and trabecular microarchitecture (Chapter Six), using pCT, explain the poor association between PA failure load and B M D by D X A . To my knowledge, this is the first investigation of cadaveric regional bone microarchitecture to include regions outside the vertebral body.  226  General Discussion, Summary and Conclusions 6. My novel data (Chapter Seven) indicate that P A stiffness can predict intervertebral motion. This provides support for a technique that clinicians use in daily practice to assess intervertebral motion in patients with back pain.  227  General Discussion, Summary and Conclusions 8.9 REFERENCES 1.  Assendelft WJ, Morton SC, Y u EI, et al. Spinal manipulative therapy for low-  back pain. A meta-analysis of effectiveness relative to other therapies. Ann Intern Med 2003;138:871-81.  2.  Koes B W , Assendelft WJ, van der Heijden GJ, et al. Spinal manipulation for low  back pain. A n updated systematic review of randomized clinical trials. Spine 1996;21:2860-71.  3.  Flynn T, Fritz J, Whitman J, et al. A clinical prediction rule for classifying  patients with low back pain who demonstrate short-term improvement with spinal manipulation. Spine 2002;27:2835-43.  4.  Schiller L. Effectiveness of spinal manipulative therapy in the treatment of  mechanical thoracic spine pain: a pilot randomized clinical trial. J Manipulative Physiol Ther 2001;24:394-401.  5.  Allison GT, Nagy B M , Hall T. A randomized clinical trial of manual therapy for  cervico-brachial pain syndrome — a pilot study. Man Ther 2002;7:95-102.  6.  Nordin M , Frankel V H . Basic Biomechanics of the Musculoskeletal System.  Third ed. Philadelphia: Lippincott Williams & Wilkins, 2001.  7.  Burklein D, Lochmuller E, Kuhn V , et al. Correlation of thoracic and lumbar  vertebral failure loads with in situ vs. ex situ dual energy X-ray absorptiometry. J Biomech 2001;34:579-87.  8.  Goodsell M , Lee M , Latimer J. Short-term effects of lumbar posteroanterior  mobilization in individuals with low-back pain. J Manipulative Physiol Ther 2000;23:332-42.  228  General Discussion, Summary and Conclusions 9.  Herzog W, Conway PJ, Kawchuk G N , et al. Forces exerted during spinal  manipulative therapy. Spine 1993;18:1206-12.  10.  Keating J, Matyas T A , Bach T M . The effect of training on physical therapists'  ability to apply specified forces of palpation. Phys Ther 1993;73:45-53.  11.  Kaplan RS, Sinaki M , Hameister M D . Effect of back supports on back strength in  patients with osteoporosis: a pilot study. Mayo Clin Proc 1996;71:235-41.  12.  Moro M , Hecker A T , Bouxsein M L , et al. Failure load of thoracic vertebrae  correlates with lumbar bone mineral density measured by D X A . Calcif Tissue Int 1995;56:206-9.  13.  Singer K , Edmondston S, Day R, et al. Prediction of thoracic and lumbar vertebral  body compressive strength: correlations with bone mineral density and vertebral region. Bone 1995;17:167-74.  14.  Cholewicki J, Juluru K , McGill S M . Intra-abdominal pressure mechanism for  stabilizing the lumbar spine. J Biomech 1999;32:13-7.  15.  Cholewicki J, Juluru K , Radebold A , et al. Lumbar spine stability can be  augmented with an abdominal belt and/or increased intra-abdominal pressure. Eur Spine J 1999;8:388-95.  16.  Hodges PW, Cresswell A G , Daggfeldt K , et al. In vivo measurement of the effect  of intra-abdominal pressure on the human spine. J Biomech 2001;34:347-53.  17.  Shirley D, Lee M , Ellis E. The relationship between submaximal activity of the  lumbar extensor muscles and lumbar posteroanterior stiffness. Phys Ther 1999;79:27885.  229  General Discussion, Summary and Conclusions 18.  Maher C, Adams R. Is the clinical concept of spinal stiffness multidimensional?  Phys Ther 1995;75:854-60.  19.  Chiradejnant A , Maher C G , Latimer J. Objective manual assessment of lumbar  posteroanterior stiffness is now possible. J Manipulative Physiol Ther 2003;26:34-9.  230  Appendix I Questionnaire and Cover Letter  231  Appendix II Tekscan Calibration with a Six Axis Load Cell  Tekscan vs Six Axis Load Cell: Trial "heavy 3 "  measurement number Load for three trials (measurement numbers 1-3) during manual loading of a human forearm, as measured by Tekscan and a six axis load cell. Dashed line represents the six axis load cell and the solid line represents Tekscan.  235  Tekscan vs Six Axis Load Cell: Trial "forearm 2"  1  2  3  4  5  measurement number  Load (N) for five trials (measurement numbers 1-5) during manual loading of a human forearm, as measured by Tekscan and a six axis load cell. Dashed line represents the six axis load cell and the solid line represents Tekscan.  236  Appendix III  Thermolyne Muffle Furnace: Temperature vs. Time Profile  237  Temperature versus time. Profile Thermolyne muffle furnace in room 148 C  furnace t e m p / time relationship  O 400 o |  300  £ 200  5  10  15  20  25  30  35  Time in Minutes  Gilles Galzi Tuesday, July 23, 2002 C:\xxx\Temperature versus time.doc  40  45  50  55  60  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092276/manifest

Comment

Related Items