DESIGN AND BIOMECHANICAL EVALUATION OF A RODENT SPINAL FIXATION DEVICE by MARYAM SHAHROKNI B.A.Sc., University of Ottawa, 2007 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Biomedical Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) October 2010 © Maryam Shahrokni ii ABSTRACT Previous experimental studies of spinal cord injury (SCI) in rodents established the importance of fixation of the spine in survival models following a mechanical injury. However, no fixation device has been designed to provide spinal stabilization, prevent additional damage to the cord, and promote fusion at the site of injury. The present study aims to design a novel rat spinal fixation device, which will be used in future survival studies and investigates its biomechanical effectiveness in stabilizing the spine up to eight weeks post injury. A custom-made magnetic resonance imaging (MRI) compatible fixation device was designed to stabilize the C5/C6 joint. This was achieved in an animal model by creating a 1.5 mm fracture- dislocation injury between C5 and C6 spinal segments of Sprague-Dawley rats using a multimechanism SCI test system. A biomechanical evaluation of the device-spine system was conducted at these segments. Cycles of stepwise directed shear forces with a maximum of 0.98 N were applied at a known distance from the injured site producing flexion and extension bending moments, while the resulting two-dimensional motions between C5 and C6 were measured and presented in the form of load-displacement curves. This was implemented at two time points: immediately (n = 6), and eight weeks post-injury (n = 9) and the results were compared to an intact group (n = 6). Average ± S.D. flexion/extension ranges of motion (ROM) were 18.1 ± 3.3º, 19.9 ± 7.5º, and 1.5 ± 0.7º, and neutral zones (NZ) were 3.4 ± 2.8º, 5.0 ± 2.4º, and 0.7 ± 0.5º, respectively for the intact, injured/fixed, and injured/8-week groups. The results show that there is a significant difference in ROM and NZ between the injured/fixed and injured/8-week groups (p-values = 0.0002, and 0.006, respectively). The device acutely stabilizes the spine by restoring its stiffness to the initial stiffness of the intact specimen. It also proves that along with the biological factors over time, fusion is promoted at the site of injury. This study presents the design and evaluation of a novel well-characterized spinal fixation device for rats, which will be used in future experimental SCI survival models. iii TABLE OF CONTENTS ABSTRACT .................................................................................................................................... ii TABLE OF CONTENTS ............................................................................................................... iii LIST OF TABLES ......................................................................................................................... vi LIST OF FIGURES ...................................................................................................................... vii ACKNOWLEDGEMENTS ......................................................................................................... xiv DEDICATION .............................................................................................................................. xv 1 INTRODUCTION .................................................................................................................. 1 1.1 Background ...................................................................................................................... 1 1.2 Motivation ........................................................................................................................ 3 1.3 Cervical Spine Anatomy Of The Rat ............................................................................... 4 1.3.1 Cervical Vertebrae .................................................................................................... 5 1.3.2 Ligaments And The Intervertebral Disks .................................................................. 7 1.3.3 The Spinal Cord ........................................................................................................ 8 1.4 Cervical Spine Injury ....................................................................................................... 8 1.4.1 Spinal Column Injury ................................................................................................ 8 1.4.2 Spinal Cord Injury..................................................................................................... 9 1.5 Spine Stabilization.......................................................................................................... 11 1.5.1 Current State Of Research ...................................................................................... 11 1.5.2 Concept Of Load Sharing ....................................................................................... 15 1.5.3 Definition Of Spinal Stabilization And Biomechanical Assessment ...................... 16 1.6 Objectives And Hypothesis ............................................................................................ 18 1.7 Project Scope .................................................................................................................. 19 2 METHODS ........................................................................................................................... 20 iv 2.1 Implant Design ............................................................................................................... 20 2.1.1 Design Criteria ........................................................................................................ 20 2.1.2 Design Prototypes ................................................................................................... 22 2.2 Biomechanical Evaluation Of Implant ........................................................................... 24 2.2.1 Experimental Protocol ............................................................................................ 25 2.2.1.1 Custom Loading Apparatus ............................................................................. 25 2.2.1.2 Kinematic Analysis.......................................................................................... 32 2.2.1.3 Biomechanical Evaluation ............................................................................... 35 2.2.1.4 Statistical Methods .......................................................................................... 49 2.2.2 Establishing A Dislocation Magnitude In A Pilot Study ........................................ 50 3 RESULTS ............................................................................................................................. 52 3.1 Implant Final Design ...................................................................................................... 52 3.2 Biomechanical Evaluation.............................................................................................. 58 3.2.1 Dislocation Injury Parameter Repeatability ............................................................ 58 3.2.2 Kinematic Analysis Of In Vitro Specimens ............................................................ 59 3.2.2.1 Intact Specimens .............................................................................................. 60 3.2.2.2 Injured/Fixed Specimens ................................................................................. 61 3.2.3 Kinematic Analysis Of In Vivo Specimens ............................................................. 64 3.2.3.1 Injured/3-Week Specimens Qualitative Analysis ............................................ 64 3.2.3.2 Injured/8-Week Specimens .............................................................................. 65 3.2.4 Statistical Summary Result Of All Groups ............................................................. 65 3.2.4.1 Degree Of Intervertebral Motion Across Groups ............................................ 65 3.2.4.2 ROM ................................................................................................................ 67 3.2.4.3 NZ .................................................................................................................... 69 3.3 Pilot Study Results In Establishing A Dislocation Magnitude ...................................... 70 v 4 DISCUSSION ....................................................................................................................... 71 4.1 Implant Performance ...................................................................................................... 71 4.2 Limitations ..................................................................................................................... 71 4.2.1 Design Limitations .................................................................................................. 71 4.2.2 Biomechanical Evaluation ...................................................................................... 72 4.2.3 Statistical Limitations ............................................................................................. 73 4.3 Biomechanical Fixation.................................................................................................. 74 4.3.1 Effect Of Implant Clamping Force On Specimen Kinematics ............................... 76 5 CONCLUSIONS AND FUTURE WORK ........................................................................... 78 5.1 Conclusions .................................................................................................................... 78 5.2 Contributions .................................................................................................................. 78 5.3 Future Directions ............................................................................................................ 78 REFERENCES ............................................................................................................................. 79 APPENDICES .............................................................................................................................. 92 Appendix A: The 21-Point Basso, Beattie, Bresnahan Locomotor Rating Scale And Operational Definitions Of Categories And Attributes [133] ................................................... 92 Appendix B: Histological Slides Of Pilot Study....................................................................... 94 Appendix C: Force-Displacement Curves For Injured Specimens ........................................... 97 Appendix D: Specimens Rotation-Moment Plots Corresponding To The 3 rd Cycle .............. 104 Appendix E: Matlab Script For Calibration Of The Instrumented Forceps............................ 120 Appendix F: Matlab Script For Clamping Force Data Acquisition ........................................ 123 Appendix G: Drawings Of The Fixation Device .................................................................... 133 Appendix H: Animal Monitoring Sheets ................................................................................ 134 Appendix I: Dislocation Injury Data Collection Sheet For An Specimen .............................. 136 Appendix J: Ethics Board Certificates Of Approval ............................................................... 137 vi LIST OF TABLES Table 1 Decision analysis for the Posterior-screw fixation design ............................................... 24 Table 2 Accuracy of the measuring system .................................................................................. 32 Table 3 Experiment design indicating the groups and the number of specimens ......................... 36 Table 4 Strain measurements in five trials for a known load at a specific region ........................ 47 Table 5 Fixation device design decision analysis ......................................................................... 52 Table 6 Dislocation status of specimens post-injury .................................................................... 59 Table 7 Statistical summary of injury parameters of each group. ................................................ 59 Table 8 Intact group statistical summary ...................................................................................... 60 Table 9 Clamping force values of high, medium, and low for each specimen ............................. 61 Table 10 Intervertebral motion at C5/C6 joint for specimens with the three separate clamping force applications .......................................................................................................................... 61 Table 11 Repeated measurements ANOVA for ROM of the injured/fixed group in high, medium, and low clamping force conditions ............................................................................................... 62 Table 12 Repeated measurements ANOVA for NZ of the injured/fixed group in high, medium, and low clamping force conditions ............................................................................................... 63 Table 13 Injured/fixed group statistical summary ........................................................................ 64 Table 14 Injured/3-week group statistical summary ..................................................................... 64 Table 15 Injured/8-week group statistical summary ..................................................................... 65 Table 16 ANOVA for the raw ROM data ..................................................................................... 66 Table 17 ANOVA for the raw NZ data ........................................................................................ 66 Table 18 Results of the post hoc analysis of the ROM raw data using SNK test ......................... 66 Table 19 Results of the post hoc analysis of the NZ raw data using SNK test.. ........................... 67 Table 20 ANOVA for transformed data of ROM ......................................................................... 68 Table 21 ANOVA for transformed data of NZ ............................................................................. 69 vii LIST OF FIGURES Figure 1 Lateral xray image of the cervical spine: (A) a burst fracture injury type associated with a contusion type injury, (B) a fracture dislocation injury type, (C) a flexion distraction or facet subluxation injury type associated with distraction type injury [17] .............................................. 2 Figure 2 Anatomical planes with respect to the rat [27] ................................................................. 4 Figure 3 Rat spine and nerve roots [25] .......................................................................................... 5 Figure 4 Cranial/caudal view of the cervical and thoracic (T1 and T2) vertebrae [26] .................. 6 Figure 5 Anterior aspect of fifth cervical vertebra highlighting two anatomical features [25] ...... 7 Figure 6 Excised rat spine showing C3-T1 segments. .................................................................... 7 Figure 7 Transverse section of the rat spinal cord at C5 level [40] ................................................ 8 Figure 8 From left to right, an x-ray of an in vitro spine construct using C1-C2 transarticular screws [89], and lateral radiograph of spine construct in a patient treated with C3-C6 anterior cervical plate fusion [91]. ............................................................................................................. 12 Figure 9 Three mechanisms of spinal fixation studied in regeneration of nerve fibres, from left to right, rod and tranpedicular screws, compressive wiring of posterior processes, and retraction wiring of trasverse processes [24]................................................................................................. 13 Figure 10 Lateral radiographs showing progressive scoliosis in the unfixed rat in A, B ,and C, and fixed rat D [23] ....................................................................................................................... 14 Figure 11 Comparison between unfixed and fixed group showing the degree of scoliosis on the left and degree of scaffold misalignment on the right [23] ........................................................... 14 Figure 12 Conceptual graph of load sharing by a typical implant system [9] .............................. 16 Figure 13 A schematic of the experimental setup of the biomechanical assessment of the spine specimen [67] ................................................................................................................................ 17 Figure 14 Characteristic behavior of human cervical spine in an experimental study [100] ....... 18 Figure 15 Transverse schematic view showing the general outline of a spinal fixation device design holding the rat cervical vertebrae at the dent located between the transverse processes and the ridges dorsal to the transverse processes................................................................................. 22 viii Figure 16 From left to right, designs of A) Lateral-screw, B) Cam-screw, and C) Posterior-screw fixation devices ............................................................................................................................. 23 Figure 17 Experiment timeline showing the study groups ........................................................... 25 Figure 18 Custom-made experimental apparatus for rodent spine load applications.. ................. 27 Figure 19 Free body diagram of a single vertebra of the excised spine mounted on the experimental rig ............................................................................................................................ 27 Figure 20 FBD of the general outline of the proposed prototypes implanted on the spine subjected to bending moment (M) and shear loads (F). ................................................................ 28 Figure 21 Visual markers for insertion into the vertebrae ............................................................ 28 Figure 22 Schematic diagram of loading the spine mounted on the experimental apparatus....... 30 Figure 23 Step-wised loading cycles of flexion and extension versus time ................................. 31 Figure 24 The setup of the marker attachments on the precision rotation mount for the accuracy study .............................................................................................................................................. 32 Figure 25 A schematic diagram of a loading case representing the excised spine in the experimental apparatus in, A) neutral position, and B) subjected to an extension load ............... 34 Figure 26 A) Three step-wised loading cycles of flexion versus time , B) rotation-moment graph generated for the corresponding loading cycles............................................................................ 35 Figure 27 A) A model of the Sprauge-Dawley rat spine (C4-T3) highlighting the dents between the transverse processes and the ridges dorsal to the transverse processes in the cervical region, where the vertebral clamps sit, B) cervical vertebral clamps for dislocation injury, C) vertebral clamps attached to the spine.. ....................................................................................................... 37 Figure 28 UBC’s multimechanism SCI system [20] .................................................................... 38 Figure 29 Typical force-displacement curve of a specimen with a dislocation injury. ................ 39 Figure 30 Biomechanical fixation process .................................................................................... 41 Figure 31 A) A model of Sprauge-Dawley rat spine C4-T2 region, B) fixation device implanted at C5/C6 joint. ............................................................................................................................... 41 Figure 32 A schematic diagram of the clamping force measurement system .............................. 42 ix Figure 33 Custom strain gauged instrumented surgical forceps ................................................... 42 Figure 34 Schematic diagram of calibration of the instrumented forceps .................................... 43 Figure 35 Application of load FG by dynamight programmed in three cycles with a relative ramp waveform ...................................................................................................................................... 43 Figure 36 A) The overall calibration set-up, B) the highlighted region in a with the instrumented forceps subjected to a force applied by dynamight. ...................................................................... 44 Figure 37 Calibration graph of the instrumented forceps depicting the relationship between measured strain and applied load .................................................................................................. 44 Figure 38 Sample output graph of the clamping strain applied during the device implantation .. 46 Figure 39 Schematic diagram of sensitivity test of the instrumented forceps .............................. 47 Figure 40 Applied force-measured strain plot for different regions ............................................. 47 Figure 41 Histological slides of two specimens at the site of injury with A) 1.3, and B) 1.5 mm displacement severity .................................................................................................................... 51 Figure 42 Schematic diagram estimating the cross sectional area of the device for the second moment of area calculations ......................................................................................................... 53 Figure 43 FBD of the device acting as a beam ............................................................................. 55 Figure 44 Schematic diagram showing the cross sectional area of a rectangular bar for the second moment of area calculations ......................................................................................................... 55 Figure 45 Custom-designed MRI-compatible cervical spine fixation device for rats .................. 58 Figure 46 Intervertebral motion-applied moment graphs at C5/C6 joint for intact group specimens ...................................................................................................................................... 60 Figure 47 Intervertebral motion (ROM) at C5/C6 joint for each specimen of the injured/fixed group in the testing conditions of high, medium, and low clamping force applications.. ............ 62 Figure 48 Intervertebral motion (NZ) at C5/C6 joint for each specimen of the injured/fixed group in the testing conditions of high, medium, and low clamping force applications ........................ 63 Figure 49 Statistical comparison of degrees of intervertebral motion between groups. ............... 66 Figure 50 Results of the post hoc analysis of the ROM raw data using SNK test........................ 67 x Figure 51 Results of the post hoc analysis of the NZ raw data using SNK test ........................... 67 Figure 52 Plot of ROM transformed data of the three groups. ..................................................... 68 Figure 53 Results of the post hoc analysis of the transformed ROM data using SNK test .......... 68 Figure 54 Plot of NZ transformed data of the three groups. ......................................................... 69 Figure 55 Results of the post hoc analysis of the NZ transformed data using SNK test .............. 69 Figure 56 Histological slides of two specimens at the site of injury with A) 1.3, and B) 1.5 mm displacement severity .................................................................................................................... 70 Figure 57 Fixation device step highlighted with a defined height of H ........................................ 71 Figure B1 Spinal cord section of specimen 1 at 1.3 mm dislocation injury ................................. 94 Figure B2 Spinal cord section of specimen 2 at 1.3 mm dislocation injury ................................. 94 Figure B3 Spinal cord section of specimen 3 at 1.3 mm dislocation injury ................................. 94 Figure B4 Spinal cord section of specimen 1 at 1.5 mm dislocation injury ................................. 95 Figure B5 Spinal cord section of specimen 2 at 1.5 mm dislocation injury ................................. 95 Figure B6 Spinal cord section of specimen 3 at 1.5 mm dislocation injury ................................. 95 Figure B7 Spinal cord section of specimen 4 at 1.5 mm dislocation injury ................................. 96 Figure B8 Spinal cord section of specimen 5 at 1.5 mm dislocation injury ................................. 96 Figure C1 Force-displacement curve of dislocation injury for specimen (1) of injured/8-week group ............................................................................................................................................. 97 Figure C2 Force-displacement curve of dislocation injury for specimen (2) of injured/8-week group ............................................................................................................................................. 97 Figure C3 Force-displacement curve of dislocation injury for specimen (3) of injured/8-week group ............................................................................................................................................. 98 Figure C4 Force-displacement curve of dislocation injury for specimen (4) of injured/8-week group ............................................................................................................................................. 98 Figure C5 Force-displacement curve of dislocation injury for specimen (5) of injured/8-week group ............................................................................................................................................. 99 xi Figure C6 Force-displacement curve of dislocation injury for specimen (6) of injured/8-week group ............................................................................................................................................. 99 Figure C7 Force-displacement curve of dislocation injury for specimen (7) of injured/8-week group ........................................................................................................................................... 100 Figure C8 Force-displacement curve of dislocation injury for specimen (1) of injured/fixed group ..................................................................................................................................................... 100 Figure C9 Force-displacement curve of dislocation injury for specimen (2) of injured/fixed group ..................................................................................................................................................... 101 Figure C10 Force-displacement curve of dislocation injury for specimen (3) of injured/fixed group ........................................................................................................................................... 101 Figure C11 Force-displacement curve of dislocation injury for specimen (4) of injured/fixed group ........................................................................................................................................... 102 Figure C12 Force-displacement curve of dislocation injury for specimen (5) of injured/fixed group ........................................................................................................................................... 102 Figure C13 Force-displacement curve of dislocation injury for specimen (6) of injured/fixed group ........................................................................................................................................... 103 Figure D1 Rotation-moment curve at C5/C6 joint for intact specimen (1) ................................ 104 Figure D2 Rotation-moment curve at C5/C6 joint for intact specimen (2) ................................ 104 Figure D3 Rotation-moment curve at C5/C6 joint for intact specimen (3) ................................ 105 Figure D4 Rotation-moment curve at C5/C6 joint for intact specimen (4) ................................ 105 Figure D5 Rotation-moment curve at C5/C6 joint for intact specimen (5) ................................ 106 Figure D6 Rotation-moment curve at C5/C6 joint for intact specimen (6) ................................ 106 Figure D7 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (1) with device removed....................................................................................................................................... 107 Figure D8 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (2) with device removed....................................................................................................................................... 107 xii Figure D9 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (3) with device removed....................................................................................................................................... 108 Figure D10 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (4) with device removed....................................................................................................................................... 108 Figure D11 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (5) with device removed....................................................................................................................................... 109 Figure D12 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (6) with device removed....................................................................................................................................... 109 Figure D13 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (7) with device removed....................................................................................................................................... 110 Figure D14 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (1) with device ..................................................................................................................................................... 110 Figure D15 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (2) with device ..................................................................................................................................................... 111 Figure D16 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (3) with device ..................................................................................................................................................... 111 Figure D17 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (4) with device ..................................................................................................................................................... 112 Figure D18 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (5) with device ..................................................................................................................................................... 112 Figure D19 Rotation-moment curve at C5/C6joint for injured/8-week specimen (6) with device ..................................................................................................................................................... 113 Figure D20 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (7) with device ..................................................................................................................................................... 113 Figure D21 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (1) with high, medium, and low clamping forces .............................................................................................. 114 Figure D22 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (2) with high, medium, and low clamping forces .............................................................................................. 114 xiii Figure D23 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (3) with high, medium, and low clamping forces .............................................................................................. 115 Figure D24 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (4) with high, medium, and low clamping forces .............................................................................................. 115 Figure D25 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (5) with high, medium, and low clamping forces .............................................................................................. 116 Figure D26 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (6) with high, medium, and low clamping forces .............................................................................................. 116 Figure D27 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (1) with device ..................................................................................................................................................... 117 Figure D28 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (2) with device ..................................................................................................................................................... 117 Figure D29 rotation-moment curve at C5/C6 joint for injured/3-week specimen (3) with device ..................................................................................................................................................... 118 Figure D30 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (1) with device removed....................................................................................................................................... 118 Figure D31 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (2) with device removed....................................................................................................................................... 119 Figure D32 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (3) with device removed....................................................................................................................................... 119 Figure J1 Animal care certificate representing ethics approval for this study. ........................... 137 xiv ACKNOWLEDGEMENTS I would like to first thank my supervisor Dr. Tom Oxland for his excellent mentorship and guidance. Without his patience and encouragement at times of difficulty, I would not have been able to succeed. His persistence taught me to continue work no matter what I believe might happen and that every single small step is a learning experience and not a failure. His on-going support has made my graduate experience a rewarding one. Many thanks to all my fellow students, the research engineers in particular Qingan Zhu, faculty, and other members of the Orthopaedic and Injury Biomechanics Group for all of their support, ideas, advice, and technical assistance. Special thanks to Colin Russell for his support and help in every step of this work, and Tim Bhatnagar for his guidance and assistance throughout this work and especially on the testing days. Many thanks to Dr. Jie Liu for his ideas, immense experience, and help with the specimens and surgical procedures, Jason Chak for the extra set of hands during testing even in the wee hours, and for their support in the form of brainstorming, chocolate and chit-chat: Erin Lucas, Katharine Wilson, and Angela Melnyk. Last but not least, I would like to thank Dr. Wolfram Tetzlaff and his team for all their time, support and training with the animal care of this work, in particular Peggy Assinck, Femke Streijger, and Clarrie Lam. Research grant from the Natural Sciences and Engineering Research Council of Canada (NSERC) is gratefully acknowledged. A very special thank you to my family and friends who have been with me for all, or even only part of my journey. Your support and encouragement has been greatly appreciated. Mom and Dad, thank you for having so much confidence in me, loving me, and being there to listen and to offer advice at all times. Mo, even in just the last bit of this journey you have been a source of so much strength, motivation, and help, thank you for everything, big and small, that made my life easier. Without your help especially on the computer and technical side of work, I would not have been able to carry on. Farid, thank you for your serenity and unwavering belief in me. And lastly, to all my family members from Iran, thank you for being there for me when I needed you, for your support and words of wisdom, and for all your love that I feel from across the ocean. xv DEDICATION I would like to dedicate this thesis to my family, who have been there for me in every step of my life. 1 1 INTRODUCTION 1.1 Background Spinal cord injury (SCI) and its treatment methods have been a topic of interest among researchers and clinicians for decades. Approximately 12,000 cases of acute SCI occur annually in North America [1-4]. The annual incidence rate of SCI in Canada is 35/year per million population [5], which corresponds to about 1000 injuries per year. The overwhelming economic cost of SCI cases affects individuals, families, and the health care system [1, 2]. Over 30 years of SCI data collection shows that the most common causes of SCI cases include primarily motor vehicle accidents in 35% and secondly, low energy falls in about 31% of all cases. Motor vehicle accidents and low energy falls were seen mostly in young adults and the elderly, respectively [1, 4]. Sekhon and Fehlings (2001) showed that the incidence of SCI has not decreased but that survival rates have continued to improve [2]. A study by Pickett et al. (2006) was conducted on the SCI cases in Canada between 1997 and 2001 to understand the current epidemiology of acute traumatic SCI; complete SCI and central cord syndrome accounted for the most prevalent clinical cases [4]. Central cord syndrome is a form of incomplete acute SCI associated with damage to the large nerve fibers carrying information from the cerebral cortex to the spinal cord [6]. It was reported that cervical SCI was the direct cause of death in majority of mortality cases accounting for 75% of all cases. In a study by Tator (1983), it was concluded that cervical injuries are approximately three times more frequent than thoracic, thoracolumbar, and lumbosacral injuries [7]. Sekhon and Fehlings (2001) also reported that approximately 55% of acute SCI cases occur in the cervical region [2]. This is likely due to a number of factors such as shape and size of the vertebrae (for example, vertebral canal diameter, Torg ratio - ratio of the diameter of vertebral canal to the width of vertebral body, Pavlov ratio - the ratio of the canal width to the vertebral body width, etc…), curvature and stiffness of the cervical spine, as well as the loading conditions (point of contact, rate, direction, magnitude of loads, etc…) [8-11]. The spinal cord is protected by the vertebrae, numerous connecting tendons and ligaments, as well as fibrous, elastic intervertebral discs. SCIs occur as a result of a variety of loading ranges and conditions on the vertebral column. They are usually initiated by a mechanical insult 2 followed by secondary biological injury patterns. The additional damaging processes of the secondary injury can be categorized into vascular, electrolyte, and biochemical deficits. Vascular damages such as hemorrhage and ischemia, electrolyte imbalances caused by plasma membrane compromise leading to an overload with Ca ++ , and protease activation are usually followed by necrosis of the cells [12-14]. An injury mechanism is defined by the type of force applied to a system, its direction, and motions associated with the injury. Some examples of primary injury mechanisms include acute compression, distraction or shear. Secondary injury could consist of vascular, electrolyte, biochemical changes [12]. Vertebral burst fracture and vertebral dislocation are the two most common injury patterns of SCIs. SCIs associated with burst fractures account for about 30-48% of cases, while 29-45% of cases are caused by dislocation injuries [2, 4]. Complete cord transection rarely occurs even in cases of complete acute SCI [2, 15, 16]. A burst fracture injury is usually followed by contusion and compression of the spinal cord through impingement of the cord by bone fragments of the fractured vertebrae, whereas in a fracture dislocation injury two adjacent vertebrae are displaced with respect to each other. Fracture dislocation injury involves shearing of the spinal cord due to the dislocation of the vertebrae. The third most common mechanism of injury is the flexion-distraction or facet subluxation, which involves stretching of the spinal cord (Figure 1) [17] . Figure 1 Lateral xray image of the cervical spine: (a) a burst fracture injury type associated with a contusion type injury. (b) a fracture dislocation injury type. (c) a flexion distraction or facet subluxation injury type associated with distraction type injury [Illustrations and caption text adopted from Choo A.M. thesis dissertation Figure 1.14 (2007)] [17] 3 The type of primary injury has an effect on the spinal cord pathology and its neurological deficit [7]. To find out if the secondary pathology process is a result of interactions between physical and biological processes at the time of primary injury, an appropriate experimental model is required. This model should mimic the injury process closely. After creating the injury, investigations at the cellular level are initiated, which usually involves a comprehensive histology analysis. Histological results consist of signs of hemorrhage in the spinal cord in early hours after the injury, which could be followed by cellular necrosis or apoptosis [13]. Previous studies have focused on treatment methods for all types of SCI, regardless of mechanism of injury and severity levels, hoping to target a specific treatment responsive to all forms; however, disregarding the heterogeneity of SCIs might be a reason for the current failure in finding a proper therapy. It has been proposed that future studies target specific types of SCI [18, 19]. Earlier work by Choo et al. (2007 and 2008) focused on the modeling of SCI mechanisms at 5 minute and 3 hour time-points relating the secondary pathology to the primary mechanisms of injury in rats [20, 21]. Results showed differences in cellular and vascular damage between different SCI mechanisms. More specifically, they demonstrated that among three different types of primary injury mechanisms, fracture-dislocation produced the greatest axonal degeneration. Furthermore, since the secondary degeneration patterns differed depending on the type of primary injury inflicted, it was concluded that different neuroprotective strategies for treatment may be required in cases of distinctly different SCIs. However, the effect of longer survival times post-injury, on the relationship between primary mechanisms of SCI and secondary pathology patterns has not been studied yet. 1.2 Motivation To investigate the effect of longer survival times post-injury on the relationship between primary mechanisms of SCI and secondary pathology patterns, it is required to first design a longer-term survival study. In particular, a long-term experimental model in rats subjected to an acute SCI is needed to follow-up the previous studies. In studies involving survival of SCI for longer periods of time, it is important to stabilize the spinal column to prevent further damage to the cord after the initial mechanical insult. Rigid fixation of the rat vertebrae during the post- injury period is vital to the success of these investigations, where the spinal column is substantially destabilized [22]. Therefore, the need for a spinal stabilization device is established. 4 Several studies have used different methods to achieve stabilization of the spinal column post- injury [22-24]. However, to date, no fixation device has been designed specifically for rats. 1.3 Cervical Spine Anatomy Of The Rat The rat cervical spine has many features in common with that of the human. This is likely due to the fact that both are vertebrate mammals. However, the rat spine is not simply a scaled down version of the human spine. Literature on the anatomy of the human spine is useful in understanding that of the rat, for which there is less published material. To further the understanding of the rat anatomy, several books and papers were consulted for assistance in identifying the cervical spine vertebral features [25-30]. Studying the anatomy of the spine requires the knowledge of the spatial terminology. The anatomical planes, describing various aspects of the anatomy of rodents, are illustrated in Figure 2 [27]. Sagittal plane refers to the plane, which divides the body into left and right portions. Transverse plane divides the body into cranial/anterior and caudal/posterior (head and tail) portions. Finally, frontal plane divides the body into dorsal and ventral (back and front) portions. Figure 2 Anatomical planes with respect to the rat. Sagittal plane refers to the plane, which divides the body into left and right portions. Transverse plane divides the body into cranial/anterior and caudal/posterior portions. Finally, frontal plane divides the body into dorsal and ventral portions [Illustration adopted from Wingerd (1988)] [27] 5 The rat spine consists of seven cervical, thirteen thoracic, six lumbar vertebrae, four sacral, and from twenty-seven to thirty-one coccygeal vertebrae (Figure 3) [31]. In the human, a total of thirty-three vertebral segments exist with seven cervical, twelve thoracic, five lumbar, five sacral, and four coccygeal vertebrae. Figure 3 Rat spine and nerve roots [Illustration adopted from Greene (1935)] [25] 1.3.1 Cervical Vertebrae Images of the individual rat vertebrae from C1 to T2 are depicted in Figure 4 [26] . The first two cervical vertebrae are distinct in appearance, and have particular names of Atlas and Axis respectively. Atlas or C1 articulates with the base of the skull and it does not have a vertebral body, whereas the Axis has the identifiable peg called the odontoid process about 6 which the Atlas rotates to allow turning of the head. The special characteristics of these two vertebrae allow a larger range of motion in the neck region. Common to all cervical vertebrae are the transverse processes. In particular for C3 to C6, the existence of transverse foramina is distinct, which give passage to local arteries and veins and plexes of sympathetic nerves. Furthermore, C3 to C6 all have similar features in terms of smaller vertebral bodies than the first two vertebrae, narrower laminae, and short spinous processes. The sixth cervical vertebra has two extra ventral processes as shown in Figure 4f. The thoracic vertebrae are characterized by long spinous processes, except for T1. Figure 4 Cranial/caudal view of the cervical and thoracic (T1 and T2) vertebrae [Illustrations adopted from Johnson et al. (1999)] [26] A unique feature of the cervical vertebrae from C3 to C7 is the ridge dorsal to the transverse process (Figure 5) [25]. This feature can be beneficial when designing devices to clamp and hold the vertebrae of this region in place. 7 Figure 5 Anterior aspect of fifth cervical vertebra highlighting two anatomical features [Illustration adopted from Greene (1935)] [25] 1.3.2 Ligaments And The Intervertebral Disks In reviewing the anatomy literature of the rat, little information on the structure of ligaments and intervertebral disks (IVDs) of the rat was found [30-35]. Some references, however, suggest that the microstructure of these tissues are similar in the vertebrates [36-38]. In the rat, adjacent vertebrae are connected via a network of ligaments and, more distinctly, an intervertebral disc (Figure 6). The intervertebral disc of the rat similar to the human is a mixture of water and fibrous cartilage, with a central nucleus pulposus and the stiffer annulus fibrosus, which surrounds the pulposus [37, 38]. The annulus fibrosus is composed of a lamellae structure of concentric sheets of collagen fibers connected to the vertebral end plates, which are the interface between the bone and the disc. The sheets are orientated at various angles. Although both the annulus fibrosus and nucleus pulposus are composed of water, collagen, and proteoglycans (PGs), the amount of fluid (water and PGs) is greatest in the nucleus pulposus. PG molecules are important because they attract and retain water. Stiffness of the disk is maintained by the hydrophilic properties of the proteoglycans. An important characteristic of water contained in the nucleus pulposus is to resist compression. The major role of the intervertebral disk is to distribute the load between the vertebrae and facilitate intersegmental motion [39]. Figure 6 Excised rat spine showing C3-T1 segments. The intervertebral disk between C5 and C6 is highlighted. Caudal Rostral C5-C6 IVD 8 1.3.3 The Spinal Cord The spinal cord is protected within the walls of the vertebral foramen (Figures 5 and 7). The cord itself consists of both grey and white matter, the former taking on a butterfly shape in the transverse plane of the cord. The white matter surrounds the grey matter [40]. A transverse section of the rat spinal cord is shown in Figure 7, highlighting the heterogeneous structure. The grey matter consists of neuron somata and white matter contains the surrounding axons. White matter also contains the supporting cells, called the glia, which are responsible for myelinating the axons. A network of vasculature permeates neural tissue and these capillaries are more numerous in the grey matter [41]. Figure 7 Transverse section of the rat spinal cord at C5 level. [Illustration adopted from Watson et al. (2008)][40] 1.4 Cervical Spine Injury The injuries occurring at the cervical region can be divided in two categories: injuries to the spinal column and those to the spinal cord. In this section, a brief summary of important characteristics of each will be discussed. 1.4.1 Spinal Column Injury Injuries to the vertebral column occur as a result of a mechanical impact to the spine. A variety of loading conditions such as compression, distraction, or shear could produce injuries 9 seen clinically such as burst fractures, fracture-dislocations, or flexion-distractions [20, 21, 42- 47]. Overall, these three vertebral column injuries account for about 80% of the column injuries observed in clinical cases [2, 4]. 1.4.2 Spinal Cord Injury Spinal cord injury is initiated by a primary injury, which is a mechanical insult to the cord. The primary injury is thought to be irreversible. Following the initial injury a series of interrelated mechanisms called the secondary injury occurs in the cord [2, 12]. This concept was first introduced by Allen in 1911, where he was using drop weights on dog spinal cords. The known weights in grams were dropped from a specific height measured in cm creating a given impact [48]. Secondary injuries initiated by the primary injury are a series of vascular, electrolyte, biochemical, and cellular processes. Hemorrhage and ischemia, influx of ions due to plasma membrane damage, and neurotransmitter accumulation are some of the examples of secondary injury in the spinal cord [7, 13, 14, 49]. Secondary degeneration and scar tissue formation are the two major issues that prevent repair of the spinal cord following an SCI [12]. Experimental SCI has been studied using a variety of models including in vivo (animal subjects), in vitro (animal and human cadavers), and mathematical simulations. Most of these models aim to reproduce the real life injury event. In vivo animal models can be classified by injury mechanisms: contusion, transection, dislocation, and distraction. Contusion and transection are the most common models [2, 20, 50]. Spinal cord contusion models were introduced in 1911 with Allen’s weight drop model in dogs which was explained earlier [48]. This model was also practiced in rabbits, cats, and primates [51-53]; but towards 1980’s, these models mostly used small animals such as rats [54, 55]. Kwo and Young (1989) developed the NYU impactor which was more advanced in providing detailed accurate measurements (±1%) of tissue displacement and more flexible in creating different severities of injury [56]. The MACSIS impactor was designed based on the NYU impactor and it was validated by creating consistent SCIs [50]. The Ohio State University Electromagnetic SCI Device [57] and the Infinite Horizon [44] not only measure displacement in tissues but also the force applied to the cord [58]. Other than weight drop, the contusion injury has been modeled by 10 compressing the cord using forceps [59], clips [60, 61], or balloons placed in the epidural space [62]. Transection is another widely used model in spinal cord regeneration studies. The advantages to this model are its reproducibility, simplicity, and the standardized methodology; however, it is not a clinically relevant mechanism of injury in humans [16, 63, 64]. SCI dislocation models attempt to mimic the clinical cases in which the cord is sheared due to vertebral motion [2]. 45% of all cases of human SCI are as a result of dislocation or fracture dislocation injuries [2, 43, 47, 65]. In a study by Fiford et al. (2004), lateral dislocation of the vertebral column was induced in a rat model [43]. Dorsal dislocation injury was modeled by Choo et al. in a rat model (2007, 2008, 2009) [20, 21, 63]. In a study by Clarke et al. (2008), dislocation mechanisms in rats were compared in terms of severity and biological damage. Lateral displacement was modeled after Fiford et al. (2004) and dorsal dislocation was reproduced using the same SCI system used by Choo et al. (2007). It was found that dorsal dislocation was more severe than lateral dislocation in rats [47]. Axial tension to the spinal column leading to spinal cord distraction is another clinically relevant mechanism of injury, which has been studied previously [45, 46, 66]. The main advantage of in vivo animal models mentioned above is that they simulate injuries and produce associated neurologic and physiologic responses. Additionally, they include the musculature, skin, and other structures that surround the cervical spine and that may influence its response to impact. However, animal models are not identical to human in anatomy, size, and structure. In addition, the ethical implications of using animals in such tests puts restrictions on the use of this type of model in today’s research environment. In vitro animal and human spine models are also widely used to study the biomechanics of spinal injuries [67-71]. These cadaveric models offer similar anatomy and bone material properties to live tissue but do not generally include the natural musculature and they lack the physiological response upon injury. These models allow for measurement of the individual vertebral 11 kinematics and the general kinetic response of the column. However, by not incorporating musculature in the model, the model results should be carefully interpreted. Mathematical models of the cervical column have been developed as a tool to assess injury [72- 74]. These models have been developed for specific injuries and range from simple two- dimensional to detailed three-dimensional finite element models that can include musculature of the spinal cord or the brain. These models allow for the determination of site specific loads and strains as well as a more general response. Although these mathematical models are found useful for further investigation of SCI, without the experimental data and a complete knowledge of material properties of soft tissues, it is fairly complicated to validate these models. 1.5 Spine Stabilization 1.5.1 Current State of Research To date, no fixation device in survival studies of SCI has been designed specifically for rodents. Currently, in most of the survival injury models, the muscles and the skin incisions are sutured in layers at the site of injury. Afterwards, the skin is closed with wound clips [75-82]. It seems that in most, if not all of the animal spinal cord experiments the importance of vertebral column fixation has been ignored. During head and neck movements, the cervical vertebrae flex in different directions, accompanied by possible motion of the spinal cord. Following spinal cord lesions due to injury, abnormal motion of the cord may cause increased bleeding, inflammation, increase in edema, and other pathological processes at the lesion site. Some of the rats that have received spinal cord injury operation may even develop scoliosis which is a form of misalignment of the spinal columns [22]. It is evident that excessive spinal cord strain after injury will seriously interfere with its healing process [22]. In clinical cases of spinal surgery, however, vertebral column fixation is a routine procedure [83]. It is believed that the degree of mechanical fixation, type, and length of fixation are some of the factors that influence the success of bony fusion in these cases [84, 85]. Fusion of the construct is achieved to restore spinal stiffness, maintain column alignment, prevent deformity formation, and finally reduce pain [9]. In one study, it was has been reported that surgical decompression and fusion followed by stabilization of the thoracolumbar spine resulted in several advantages including some level of improved neurologic function in the 12 patients [86]. However, this effect has to be further investigated in future studies and cannot be generalized based on these results. Currently, in clinical cases, fusion of the spinal segments post injury is achieved using several fixation methods to restabilize the spine. These range from implanting stabilizing cages [87], plate fixation [88] , different types of rod/ screws [89], or finally any combination of the methods mentioned [90]. Figure 8A shows a transpedicular technique between C1 and C2 in a cadaveric model [89] and Figure 8B illustrates a lateral radiograph of the spine construct of a patient treated with C3-C6 anterior cervical plate fusion [91]. These methods can be compared with each other using experimental in vitro methods [87, 89, 92, 93] or mathematical modeling [94]. Range of motion is the parameter of choice for comparison, establishing the degree of stabilization. Load sharing by the implant and the spine is also another question that can be addressed during these experiments [92]. Figure 8 From left to right, an X-ray of an in vitro spine construct using C1-C2 transarticular screws [89], and lateral radiograph of spine construct in a patient treated with C3-C6 anterior cervical plate fusion [91]. [Illustrations taken from Nassos et al. (2009), and Kelly et al. (2010)] In a study by Liu et al. (2003), the importance of fixation of the spine post-injury was investigated. The injury model was a transection of the rat spinal cord in the thoracic region. Post-injury, strong stainless steel wires were used for internal fixation of the vertebral column. The wires were tightened around three vertebral spinous processes through the base of the interspinous ligaments[22]. Fourteen days post-injury, the histological slides of the spinal cord 13 segments were compared between the fixed (N = 6) and unfixed group (N = 6). Results showed that in the fixed group less severe secondary degeneration was found; a reduction in the volume and thickness of the scar tissue was observed, and finally the regeneration of the nerve fibres was facilitated. It was concluded that the fixation of the vertebral column is a crucial procedure for SCI experiments [22]. Another study by Cheng and Olson (1995) also showed the importance of fixation in allowing nerve fibre regeneration. The SCI model was again a cord transection injury in rats studying the effect of three different mechanisms of spinal fixation along with application of fibrin glue in fibre regeneration. The three mechanisms consisted of wiring of transverse processes under a distraction load, wiring of posterior spinal processes in a compressive mode and finally using transpedicular screws and rods (Figure 9). Data was obtained at different time-points: 7, 14, 21, and 30 days post-injury. Results showed that compressive wiring with the application of fibrin glue was the most promising in promoting fibre regeneration. This study concluded that using microscopic instrumentation with regards to the small size of rat skeleton would decrease the efficiency of using rod and transpedicular screws, although in theory and in clinical cases this method is known to be successful [24]. Using wires around the processes in rats is extremely time-consuming during surgery and it is difficult to achieve a consistent fixation due to the unknown tensile load applied to the wires. Figure 9 Three mechanisms of spinal fixation studied in regeneration of nerve fibres, from left to right, rod and tranpedicular screws, compressive wiring of posterior processes, and retraction wiring of trasverse processes [Illustrations adopted from Cheng and Olson (1995)] [24] In a study by Rooney et al. (2008), the alignment of implanted tissue-engineered scaffolds was evaluated by comparing two groups of rats one fixed and the other unfixed. Complete cord transection was used to create the injuries in adult rats [23]. The scaffolds were implanted at the 14 site of injury. A metal rod fixed to the spinous processes above and below the injury site was used as the method of fixation. Radiography (two and four weeks post-op) and magnetic resonance microscopy (MRM) (4 weeks post-op) were the two techniques performed to measure the degree of scoliosis progression in the spinal column. Results showed that spinal fixation had proved to lessen or prevent scoliosis and reduce scaffold misalignment in the transection injury model of rat spinal cord (Figures 10 and 11). Figure 10 Lateral radiographs showing progressive scoliosis in the unfixed rat in A B and C, and fixed rat D. Radiographs were taken immediately (A) 2 weeks (B) and 4 weeks (C) post implantation of the scaffold into the transected cord. Arrows denote location of the scaffold at each time-point processes, (D) shows the aligned spine in the fixed rat [Illustration and text caption taken from Rooney et al. (2008)] [23] Figure 11 Comparison between unfixed and fixed group showing the degree of scoliosis on the left and degree of scaffold misalignment on the right [Illustrations taken from Rooney et al. (2008)] [23] The above studies demonstrate the importance of fixation of the spine following an SCI. However, the methods used for fixation were surgically difficult or time-consuming, and some involved very small rod and transpedicular screws that would not be practical in the small rodent D 15 skeleton. This clearly shows the need for a spinal fixation device for rodents in experimental SCI studies, which both restabilizes the injured spine and is practical when it comes to implantation. The question might arise that weather or not fusion can occur in animal models post-injury without the use of a fixation device. This depends on variety of factors beginning with the severity of the induced injury as an important one. Previous studies on animal fusion models, however, only have focused on the evaluation of fusion after minor decortication of the vertebral processes such as in modeling spinal orthrodesis or laminectomy procedures [95-97]. In these procedures, the mechanical stability of the spine was not compromised compared to the case when a severe dislocation injury would be induced. It is anticipated that achieving fusion after a severe destabilizing spinal injury fusion would be very difficult or rare [97]. Success of fusion can be assessed using manual palpation, radiographic, biomechanical, and histological evaluations [95-97]. In other studies, it has been shown that using fixation post spinal injury in rodent survival models results in a reduction in the volume and thickness of the scar tissue, reduction in the degree of scoliosis progression, as well as an increase in the regeneration of nerve fibres [22-24]. These results might suggest that a decrease in the severity of spinal cord secondary injury or scoliosis progression might be as a result of the development of spinal fusion due to a decrease in the intervertebral motion. 1.5.2 Concept Of Load Sharing The concept of load sharing by an ideal implant system is shown in Figure 12. This conceptual illustration shows the load percentage versus time where load is shared by the implant, and fusion mass. Initially, the implant takes the load, then gradually the fusion mass gets stiffer and takes the load to a greater extent, causing the implant to unload [9].Although rigid spine instrumentation is excellent for stabilizing the spine, there are disadvantages associated with using these devices. Sometimes, the implant continues to carry much of the load, rather than sharing it with the bone graft. This could result in an increased incidence of construct failures, bone resorption, or both [98]. It has been shown that fusion in the spine is facilitated with an increase in the mechanical stability of the fusion construct [83, 99]. 16 Figure 12 Conceptual graph of load sharing by a typical implant system. Load is shared by the implant and fusion mass. Initially, the implant takes the load then gradually the fusion mass takes the load to a greater extent, causing the implant to unload. [Illustration taken from White and Panjabi (1990)][9] 1.5.3 Definition Of Spinal Stabilization And Biomechanical Assessment An intact spine has a certain range of motion (ROMi). Once an injury has occurred, this range of motion increases. The goal of implanting a spinal fixation device is to reduce the range of motion of an injured spine to facilitate fusion. The degree of change in biomechanical fixation (BF) can be measured by comparing the ratios of ranges of motion of a fixed specimen (ROMf) to an intact specimen (ROMi) between different cases, Equation 1: BF = ROMf / ROMi : Equation (1) A typical experimental set-up for in vitro testing of the spine is one in which the excised spine is mounted and secured rigidly at two ends and the intervertebral motion is tracked while the spine is being loaded in the desired direction from the neutral state to a loaded state (Figure 13). 17 Figure 13 A schematic of the experimental setup of the biomechanical assessment of the spine specimen: (b) a human cadaveric lumbar spine specimen was fixed to the testing surface, with coordinate axes as shown. (c) A force transducer was used to measure the applied load. The loading apparatus consisted of (d) a displacement controlled linear actuator with (e) an optical position encoder for the determination of actuator position. [Illustration adopted from Little (2005)] [67] Several studies demonstrated the concept of biomechanical fixation experimentally [88, 100- 102]. Figure 14 shows the characteristic behaviour of a human cervical spine specimen comparing the displacement due to an applied moment of an intact specimen, to one with an induced discectomy [a discectomy is a surgical procedure in which the central portion of an intervertebral disc, the nucleus pulposus is removed [103]], and another which has been stabilized using a fixation device [100]. These curves were created by tracking the intervertebral motion between C5 and C6 vertebrae of a spine specimen subjected to different flexion and extension moments. Comparing the three curves gives valuable information regarding the degree of range of motion and neutral zone of a spine specimen. Neutral zone (NZ) is defined as the region of high flexibility around the neutral position, whereas the range of motion (ROM) refers to the total motion on the load-deformation curve [104]. It was determined that ROMf is smaller than the ROMi, while ROMinjured is larger than the intact one. This demonstrates that a certain degree of biomechanical fixation has been achieved by means of a fixation device, which can be measured according to Equation 1. For clarification, NZ and ROM of the intact specimen are indicated by arrows corresponding to the grey graph of Figure 14. 18 Figure 14 Characteristic behavior of human cervical spine in an experimental study; graph illustrates the relative motion of vertebrae at C5/C6 level subjected to known flexion and extension applied moments. Grey curve shows an intact spine specimen, black shows the specimen with a discectomy, and the red illustrates the fixed specimen. Arrows indicate the ROM and NZ of the specimens [Illustration adopted from Pitzen et al. (2003)] [100] 1.6 Objectives And Hypothesis The overall objective of this project is to design a custom rat cervical spinal stabilization device and to evaluate the biomechanical fixation provided by this device over eight weeks post- injury. In this study, we ask the question; Does this fixation device change the stability of the rat spinal column over time? The two specific objectives of this project are: 1) To design a custom MRI-compatible spinal stabilization device for the unstable cervical spine in a rat model to prevent additional mechanical injury following a dislocation injury; 2) To evaluate the biomechanical fixation provided by this device, immediately and eight weeks post-injury. ROMi NZi Moment (Nmm) D is p la ce m en t (d eg r ee s) Intact Disc ACFP ROMf ROMinjured 19 This device will prevent further damage to the spinal cord by stabilizing the column. Potential benefits of this device include reduction in pain and secondary damage, as well as facilitation in fusion between the vertebrae [105]. Hypothesis: It is hypothesized that following the biomechanical evaluation of the spinal stabilization system, the device-spinal column stiffness is more than the injured spinal column immediately after the injury. It is anticipated that the overall stiffness of the spine increases after 8 weeks due to healing. Healing refers to the creation of scar tissue resulting in bony fusion at the site of injury, which occurs due to the reduction in mobilization of the dislocated segments and load sharing between the implant and the spine. 1.7 Project Scope This project aims to design a custom MRI-compatible stabilization device for the cervical spine of rats. It also aims to evaluate the device performance by measuring the degree of biomechanical fixation provided by this device in one loading plane. A combination of an in vitro and in vivo animal model was used to evaluate the fixation device performance. This project examines the biologic and mechanical response of the spine to mechanical injury. The assessment of the system is merely biomechanical and the development of fusion is not assessed histologically. Assessments of cord tissue damage and animal function are not performed. The injury model is a dislocation type in the rat cervical spine. 20 2 METHODS 2.1 Implant Design In animal studies of spinal cord injury with accompanying column damage (i.e. vertebrae, and discs), it is crucial to restabilize the spine. After creating an acute dislocation injury to the spine, the fixation device is implanted on the corresponding injured vertebrae. This device aims to stabilize the spine by reducing the range of motion, facilitating bony fusion, as well as to prevent further mechanical damage to the spinal cord. In this way, the possibility of fusion between the vertebral segments at the site of injury is enhanced. 2.1.1 Design Criteria The fixation device is required to achieve a specific set of design requirements, so that it can be implanted in live animals and provide its function. Together with a relative ease of surgical implantation and MRI-compatibility, it should enhance the stability of the injured spine. In this section, the design requirements considered in the device design process are introduced:  Structural stiffness: The device is required to provide mechanical fixation to the injured spine. Structural stiffness of the device depends on its design and its elastic modulus. The device must be designed such that it rigidly stabilizes the injured spine. Rigid stabilization of the spine reduces the intervertebral motion of the injured spinal segment to a level that could facilitate fusion. Structural stiffness of different designs can be experimentally measured by applying the expected loads to the device and measuring the produced deflection. This can also be done theoretically by estimating the loads applied to the device when implanted on the spine and calculating the resulting deflection. The ratio of the applied load to the resulting deflection would represent the stiffness of the device.  Biocompatibility: Biocompatibility refers to the ability of a material to perform its desired function with respect to a medical therapy, without eliciting undesirable local or systemic effects in the recipient or beneficiary of that therapy [106, 107]. Similar to other medical devices that are categorized 21 in Medical Devices Regulations as class III by FOOD AND DRUGS ACT in Canada or other countries or regions, the spine fixation device is required to be biocompatible with the live tissues into which it is implanted [108].  MRI-compatibility: Advantages associated with the use of MRI demand that researchers utilize this tool to visualize and investigate the structural details of internal tissues. Future survival studies to be conducted by our research group involve extensive amount of MRI to assess the patterns of spinal cord tissue damage following injury. Therefore, it is required for this device to be made of an MRI-compatible material.  Ease of surgical implantation: The device needs to be relatively simple for the clinician to handle and also would need minimal tissue exposure to prevent endangering the survival of animals. The design of the device should account for the unique morphological characteristics of the rat cervical vertebrae.  Versatility: This criterion is for the implant to accommodate different sizes and/or strains of animals. A certain design feature can enable the device to be implanted on animals with some variability in size, geometrical ratios, and shape of the vertebrae.  Cost: The manufacturing cost is another factor, which needs to be addressed in the design process. The goal of the device design process is to minimize its manufacturing cost.  Size: The small animal size creates several design limitations. The design must avoid a bulky implant for the small rat spine. After the device implantation, the muscles and the skin would be sutured over the incision region. The size of the implant must be small enough to sit under the soft tissues and would not create further pain and discomfort for the animals endangering their survivability. 22 2.1.2 Design Prototypes Several prototypes were conceptualized and considered to compare against the design criteria. It was decided that a design would be selected, which would generally fit on the spine at the dent located between the transverse processes and the ridges dorsal to the transverse processes of adjacent vertebrae (Figure 15). This vertebral feature allows the fixation device to clamp the vertebrae in place providing it with additional rigidity. Other fixation methods such as wiring of the vertebral processes or insertion of small rods and pedicular screws were not considered due to the fact that these implantation techniques were time-consuming and impractical for operation and it was not possible to ensure consistency of the fixation provided by these methods. The three most relevant prototypes demonstrating different methods of clamping the spine are shown in Figure 16. The proposed material for these prototypes is Polyetheretherketone (PEEK) due to its excellent biocompatibility and MRI-compatibility characteristics [109, 110]. Other important properties of this material are its weight reduction and chemical resistance characteristics. PEEK is a rigid, bio-inert polymer that is relatively light and cost effective, and it is widely used in medical and clinical applications [109, 110]. Figure 15 Transverse schematic view showing the general outline of a spinal fixation device design holding the rat cervical vertebrae at the dent located between the transverse processes and the ridges dorsal to the transverse processes Lateral-screw fixation design: This prototype consists of two parts that clamp the vertebrae from both lateral sides and are fastened together with two screws. The screws enter the device from a lateral direction. This design has two slots to accommodate different sizes, providing versatility in addition to rigidity for the subjects (Figure 16A). 23 Cam-screw fixation design: This design is similar to the previous design and consists of two parts; however, the parts come together with the use of two cam-type screws holding the vertebrae tightly close to each other. The purpose of using cams in this design was the mechanical simplicity provided by them in clamping the two parts compared to the use of screws (Figure 16B). Posterior-screw fixation design: In this design, two parts that clamp the vertebrae from lateral sides are fastened together with two screws located at the posterior side of the spine. Similar to the lateral-screw design, two slots accommodate different sizes of spinal columns (Figure 16C). The use of an instrumented forceps was decided during the implantation of this device, in order to monitor and measure the clamping force before tightening the screws. Figure 16 From left to right, designs of A) Lateral-screw, B) Cam-screw, and C) Posterior-screw fixation devices These designs were compared with each other based on all the established design criteria, except biocompatibility and MRI-compatibility, which are related to the material selected for the device. The scores were given to each prototype based on a subjective assessment. During this assessment, quantitative analyses such as Finite Element Analysis (FEA), full cost analysis, and prototype fabrication were not performed, due to the lack of resources to test or construct each design prototype. In the decision analysis process, each criterion was given a weight, where the total weight of all the design criteria would add up to a 100. Also, each prototype was given a certain score for that criterion, where the total score of the three prototypes must add up to a 100. Finally, the total weighted score of each prototype was calculated according to Equation 2. 24 i: a design criterion Weighted Scorei = Weighti × Score Total Weighted Score = ∑i Weighted Scorei : Equation (2) As an example, decision analysis procedure for the posterior-screw fixation design is shown below (Table 1). Table 1 Decision analysis for the Posterior-screw fixation design Design Criteria Structural stiffness Ease of surgical implantation Versatility Cost Size weight 30 40 10 10 10 score 28 50 35 40 40 Total Weighted Score = ∑i Weighted Scorei = 30×28 + 40×50 + 10×35+ 10×40 +10×40 =3990 The prototype with the highest total weighted score was selected as the winner. Detailed engineering drawings of the final design was prepared and sent to the Mechanical engineering department machine shop at the University of British Columbia. Note that there might be a variation in the geometrical dimensions of different subjects vertebrae at the ridges dorsal to the transverse processes and the transverse processes. It is required to determine if these variations might cause some limitation in the use of the device between subjects. 2.2 Biomechanical Evaluation Of Implant Fixation provided by the device was biomechanically evaluated in a rat cervical spine model. The induced injury was a dorsal dislocation of C6 with respect to C5. The evaluation was conducted in two stages; first in vitro and secondly in vivo. Two in vitro groups of intact (control) and injured/fixed immediately post-injury were evaluated, as was one group for the in vivo experiment-injured with a survival time of eight weeks. The biomechanical evaluation 25 procedure was similar in both experiments. The general steps included creating a dislocation injury at the C5/C6 joint, harvesting the spinal column from the third cervical vertebra to the first thoracic vertebra. Next, the excised spine was subjected to known step-wise loads and the two- dimensional (2D) intervertebral motion was measured at the injury level. 2.2.1 Experimental Protocol All experiments were conducted in accordance with the University of British Columbia Animal Care Committee. The biomechanical evaluation of the device was tested in two groups of injured animals: first, immediately post-injury and second, at eight weeks post-injury. The results were compared to a control group of animals with intact spines. The timeline of this experiment is illustrated in Figure 17. Note that initially, an injured group with no fixation at time zero was considered for biomechanical testing. In the pilot work, however, it was found that specimens of this group post-injury were fairly unstable at the injured intervertebral joint. Upon loading, these specimens were further damaged and separated at the dislocated joint. Figure 17 Experiment timeline showing the study groups 2.2.1.1 Custom Loading Apparatus The spine machine at UBC is a device capable of applying loads in different directions to the spine and is widely used for human cadaveric specimens and large animal models in UBC’s biomechanics lab [111]. Due to the small size of the excised spinal column of rodents, a custom- made experimental apparatus was designed to biomechanically evaluate the spinal column. The custom apparatus applied a known shear force to the top of the spine, creating a bending moment at C5/C6 joint. The rat spinal column (C3-T1) was mounted on a screw with a similar 26 size to the diameter of the spinal canal. The screw would protrude into the spinal canal from T1 to the lower half of C7 in order to not interfere with the motion at the C5/C6 junction. It was found that further protrusion of the screw would have caused permanent destruction of the inner wall of the vertebral canal and so it was prevented. Another screw with a shorter length was inserted into the cranial two vertebrae (C3 and upper half of C4). The top screw was attached to a balancing weight connected via a string guided on a pulley to hold the column vertically at all times. A preliminary experiment was performed to find out the required balancing weight for an excised spinal specimen. This measured weight (9.68 grams) was required to hold the spinal column vertical and to not pull the specimen through the screws attaching it to the apparatus. The vertical orientation of the spine was visually examined by comparing its alignment to a ruler. If the alignment was not satisfactory, the pulley guiding the balancing weight was slightly moved to eliminate this issue. This was made possible by mounting the pulley on a sliding bar. Another string was attached to the top screw to apply a shear force at distance d (moment arm) from the C5/C6 junction. This force created a bending moment of flexion or extension at the joint, depending upon the direction of the applied force. When the spine was loaded, distance d slightly decreased; however, for consistency, it was always measured when the spine was initially oriented in the neutral position. A schematic diagram of this apparatus is illustrated in Figure 18. For simplicity, the pulley attached to the balancing weight is not shown. This pulley was mounted on the sliding bar which was placed at a distance of 45 cm from the excised spine. This height was selected in a preliminary test to minimize the source of error associated with the alignment of the balancing weight in the apparatus. Similarly, the two pulleys guiding the strings for loading were also mounted at a large distance from the spine specimen. The pulleys were made of plastic and were lubricated to reduce friction between the strings and the pulleys. 27 Figure 18 Custom-made experimental apparatus for rodent spine load applications. The excised spine containing six vertebrae from C3 to T1 is attached to the apparatus via two screws. The top screw is attached to a balancing weight mounted via pulleys. Pulleys are designed to be located high and they are not shown in this picture. Load is applied through a string also attached to the top screw. This load is applied at a distance d from the C5/C6 joint. The load shown is in the ventral direction, which would create a flexion moment. A dorsally directed force would create an extension moment. Visual markers shown by circular “+” signs are inserted into the vertebrae at C5 and C6 levels to track the intervertebral motion upon loading. The free body diagram of a single vertebra of the excised specimen is shown in Figure 19. The specimen is mounted in the experimental apparatus subjected to load F in the ventral direction, which produces a bending moment M in the specimen and a reaction shear force and an opposite force at the bottom of the vertebral segment. Figure 19 Free body diagram of a single vertebra of the excised spine mounted on the experimental rig. Letter F represent the external shear load applied at the vertebral endplate, M represents the resultant bending moment, W represents the force of gravity, and e the height of the spinal segment. F F M e W 28 The general outline of the device free body diagram subjected to external loads that are applied through the bony interface when the device is implanted is shown in Figure 20. The spine would be subjected to shear loads at the top segment (Figure 18), which would produce forces (F) and bending moments (M = shear force “F” × moment arm “d”) at the C5/C6 joint (Figure 19). These loads would be partially transferred through the vertebral interface to the device subjecting it to a bending moment M and shear forces F due to the load sharing between the spine and the implant (note that in the worst case scenario, loads F and M are entirely transferred to the device as shown in Figure 20). Figure 20 FBD of the general outline of the proposed prototypes implanted on the spine subjected to bending moment (M) and shear loads (F). Shear loads are assumed to act as concentrated loads. Loading procedure of the excised spine: Due to the small size of the rat spine, small and wireless visual markers were selected, in order to not obstruct or interfere with the spinal motion. The markers consisted of two small circular “+” signs glued to a pin at a specific distance from each other (Figure 21). The spinal column with the inserted markers was mounted on the custom-made experimental rig for the biomechanical evaluation (Figure 22). Figure 21 Visual markers for insertion into the vertebrae F M F 29 The stiffness of the device-spine system was measured by applying known loads and recording the unknown displacement between two spinal segments of C5 and C6. This way load- deformation plots could be generated to compare against the control or intact group. The experimental rig was designed to allow the device-spine system be subjected to flexion and extension loads. The applied loads were standard weights of 0.24, 0.50, 0.74, and 0.98 N, which were determined in the pilot work. Stepwise loading sequence was from lowest to highest. Note that for the injured/fixed group, the maximum load applied was selected to be the 0.74 N. The reason for this selection was that the injured specimens immediately post-injury were fairly unstable, and as soon as the maximum load of 0.98 N was applied under consecutive loading cycles additional tissue damage was obvious despite the fixation device was implanted. The directed shear loads were applied at a distance “d” from the C5/C6 junction through the use of strings guided by pulleys. This produced a flexion or extension bending moment about the C5/C6 joint. The moment arm “d” was (Mean ± S.D.) 13 ± 1 mm producing bending moments of 3.2, 6.4, 9.6 and 12.8 Nmm. Between each loading step a time window of 30 seconds was given to allow the specimen to creep accounting for viscoelastic effects [68, 102]. This was verified in the pilot work to ensure that most of the creep in the specimen would occur as a result of loading in the first 30 seconds. To reduce the viscoelastic effects, loading in flexion and extension was repeated three times and the last cycle was selected for the motion measurement purposes (Figure 23). In biomechanical testing, adequate preconditioning of the specimen is necessary, allowing for viscoelastic behavior of the spine [102, 112]. 30 Figure 22 Schematic diagram of loading the spine mounted on the experimental apparatus. A) in neutral position B) in flexion after applying a known load. The excised spine containing six vertebrae from C3 to T1 is attached to the apparatus via two screws. The top screw is attached to a balancing weight mounted via pulleys. Pulleys are designed to be located high and they are not shown in this picture. Load is applied through a string also attached to the top screw. This load is applied at a distance d from the C5/C6 joint. Visual markers shown by circular “+” signs are inserted into the vertebrae at C4, C5, and C6 levels to track the intervertebral motion upon loading. A B 31 Figure 23 Step-wised loading cycles of flexion and extension versus time Motion Capture System: The position of markers was tracked using camera snapshots (Phantom V9 camera with 1632×1200 resolution - Vision Research Inc. Wayne, NJ, USA). The angle between the marker carriers was found before and after loading. The change in angle between the markers placed into the two vertebrae represented their relative motion due to different loads. Accuracy Study of the Motion Capture System: The accuracy of measuring 2D intervertebral rotations using the motion capture system explained above was evaluated. For this purpose, a high precision rotation mount (Thorlabs - Newton, NJ, USA) was used. It had an accuracy of 0.04 degrees. One marker was attached to a rigid block placed in the middle section of the rotation mount. Another marker was attached on the rotation mount. The rotation mount was turned in small increments and snapshots were taken after each rotation (Figure 24). The change in angle between markers was measured by analysing the images and this value was compared to the actual rotation taken place. The details of this mathematical calculation are discussed in the “Kinematic Analysis”. An accuracy of 0.5 degrees (the average accuracy of five trials with different rotational angles) and precision of 0.2 degrees (standard deviation of the average accuracy of the five trials with different rotational angles) were found for this method (Table 2). The level of accuracy and precision was clearly sufficient, since in the pilot work the range of motions of the spine specimens were found to be about 20 degrees. If this total motion be divided by the number of segments (six), the intervertebral motion at each level would be more than three degrees. 32 Figure 24 The setup of the marker attachments on the precision rotation mount for the accuracy study. Top and bottom figures depict the top and the side views of the setup. One marker is attached to a stationary block in the middle and another to the rotary mount. Turning the rotary mount enables the attached marker to rotate with respect to the stationary one. This results in a change of angle, which is detectable by taking snapshots before and after each turn. Table 2 Accuracy of the measuring system Trial Range of Rotation (degree) Incremental steps (degree) Accuracy (degree) 1 0 - 0.20 0.04 0.50 2 2.40 - 8.00 0.2 0.58 3 6.00 - 6.20 0.04 0.59 4 7.80 - 8.00 0.04 0.59 5 20.00 - 27.00 1.00 0.19 Average ± S.D. 0.5 ± 0.2 2.2.1.2 Kinematic Analysis A schematic diagram of a loading case representing an excised spine specimen in, A) neutral position, and B) subjected to a 0.24 N extension load is shown in Figure 25. Marker positions were obtained using the coordinates of the two indicated points. The change in angle between the markers before and after each loading case can be measured, as outlined below. 33 Neutral case:  Equation of marker inserted in C5 (1) Y=a1X+b1  Equation of marker inserted in C6 (2) Y=a2X+b2 Angle between the markers: α1 = arctan a1- arctan a2 Subjected to an extension load of 0.25 N:  Equation of marker inserted in C5 (3) Y=a3X+b3  Equation of marker inserted in C6 (4) Y=a4X+b4 Angle between the markers: α2 = arctan a3- arctan a4 Therefore, ∆ Angle = α2- α1 34 Figure 25 A schematic diagram of a loading case representing the excised spine in the experimental apparatus in, A) neutral position, and B) subjected to a 0.24 N extension load. Marker line equations are also shown for each case. After generating rotation-moment graphs for individual specimens, NZ and ROM were calculated. NZ was defined as the region within which the spinal motion was produced with no external load. The entire motion measured from the neutral position was defined as ROM. These values were obtained from both flexion and extension motion of the third cycle of loading to obtain the total NZ and ROM of the specimen (Figure 26). (1) Y= a1X+b1 (4) Y= a4X+b4 (3) Y= a3X+b3 (2) Y= a2X+b2 B A 35 Figure 26 A) Three step-wised loading cycles of flexion versus time , B) rotation-moment graph generated for the corresponding loading cycles. Points of cycle one, two, and three are shown with cross, triangular, and circular signs, respectively. NZ (the spinal motion in the neutral position) and ROM (the total spinal motion) were defined as parameters corresponding to the third cycle. Same calculation was done for the motion in extension. These values were added to obtain the total NZ and ROM of the specimen. 2.2.1.3 Biomechanical Evaluation For the in vivo experiment, nine male Sprague-Dawley rats (Mean ± S.D. weight = 291 ± 11 g) underwent the injury process and were sacrificed after eight weeks. There were more animals selected for this group to be certain of the number of animals surviving at the experimental end-point. For the in vitro experiment, twelve Sprague-Dawley rats (Mean ± S.D. weight = 293 ± 12 g) were divided into two groups of six animals each (Table 3). Another group of animals, intended to have an eight week survival time (N = 3), were injured with the dislocation level of 1.7 mm. The increase in the displacement level was implemented to find out if the animals would survive the new dislocation level. Due to the consequent animal care complications, the animals were sacrificed three weeks post-injury. The biomechanical evaluation was performed on these specimens; however, due to the small number of subjects and the change in the dislocation level, only a qualitative analysis of the results was carried out. x x x x o o o o     Rotation Moment Time M o m e n t x x x x x      o o o o o ROM NZ x  o 36 The animals were ordered from the Center for Disease Modeling (CDM) and Charles River (CR) through the UBC’s Animal Care Centre. Table 3 Experiment design indicating the groups and the number of specimens Groups Number of Specimens Intact (Control) N=6 Injured/Fixed N=6 Injured/8-Week N=9 Surgical procedures: Animal weights were monitored and recorded. Animals were then anesthetized with isofluorane and kept within stereotactic-surgical frame and prepared for surgery by shaving the area over the cervical spine. The cervical spine of the animal was exposed between C2 and C7. Dorsal ligaments between C5 and C6 were transected and the C5/6 facet joints were removed. Ligament and facet removal was performed to mimic the type of posterior element fracture and ligament injury seen in bilateral facet fracture-dislocation [63], and to produce consistent injuries where the likelihood of residual dislocation following injury is reduced [17]. The surgical procedures were conducted by Dr. Jie Liu at International Collaboration on Repair Discoveries (ICORD) who is a microsurgery specialist with over 20 years of experience in this field. Injury Procedures: During the injury, custom-designed vertebral clamps were used to rigidly hold the vertebrae [20, 63]. These vertebral clamps made of PEEK were designed to sit inside the dent located between the transverse processes and the ridges dorsal to the transverse processes against the lateral masses same as the current implant (Figure 27A, and B). For dislocation (Figure 27C), a rostral clamp held C4 and C5 while a caudal clamp held C6 and C7. 37 Figure 27 A) A model of the Sprauge-Dawley rat spine (C4-T3) highlighting the dents between the transverse processes and the ridges dorsal to the transverse processes in the cervical region, where the vertebral clamps sit, B) cervical vertebral clamps for dislocation injury, C) vertebral clamps attached to the spine. Letters C and R represent caudal and rostral clamps. Block arrow denote direction of injury vector. The test was performed with the vertebral clamps attached to the vertebrae. The injury model consisted of a dislocation mechanism injury between C5 and C6 vertebrae. The injury was induced using a multimechanism SCI system (Figure 28, [20]) at 364 mm/s velocity and 1.5 mm displacement. The displacement level of 1.5 mm was established based on the results of a pilot study (“Establishing a Dislocation Magnitude in a Pilot Study”). C R A B C 38 Figure 28 UBC’s multimechanism SCI system. Injuries were produced with an electromagnetic linear actuator under displacement feedback control. The apparatus had seven degrees of freedom for positioning the actuator and animal at any orientation relative to each other. The actuator was mounted to a rotary axis, on a translating radial arm (R, θ), which in turn was mounted to a motorized z-axis. The z-axis allowed for continuous and incremental (50μm, 0.002in) positioning. A stereotactic frame, mounted to an x-y table, mounted to a turntable (ω), served as the specimen platform. A customized damped-vibration table acted as the system’s base. Sensors included a linear variable differential transformer to measure displacement, interchangeable load cells (22, 225, and 444 N), and interchangeable accelerometers (50 and 500 g). [Illustration adopted from Choo et al. (2007a)] [20] A pre-load magnitude of 2 N (previous preliminary work by Tim Bhatnagar, Jie Liu, and Maryam Shahrokni) was applied before the test was initiated. When the test was ready to commence, the load actuator was used to move the caudal clamps up slightly, until the load cell read ~2N (about 10% of maximum load – this value was selected based on previous preliminary experiments). Application of this pre-load magnitude ensures that the injury would be created consistently across different animals. The cervical vertebrae in the spine could be sitting in different positions with respect to each other, when the animal is placed in the stereotactic frame with the dislocation vertebral holders attached. Essentially, this could cause that the displacement value applied to the segments create different injuries in different animals. When a pre-load is applied to the spine prior to the test, it is assured that the spinal segments of all specimens are subjected to a similar initial pre-load ensuring that a maximum consistency in injury creation is obtained across the subjects. 39 The dislocation injury between the C5 and C6 segments was created with an electromagnetic linear actuator (Test Bench ELF LM-1, Bose, Eden Prairie, MN) integrated in the multimechanism SCI system shown in Figure 28. The injuries were created at a peak velocity of approximately 364 mm/s and displacement severity of 1.5 mm. The caudal vertebral clamps holding C6 and C7 were coupled to the actuator thereby dislocating the C5/C6 joint dorsally, then they were held for 0.5 seconds, and returned to the initial position [20, 63]. Parameters of injury: Each injury was characterized by several parameters that were obtained during the course of the injury:  Displacement (mm) refers to the magnitude of vertebral displacement due to the induced dislocation injury.  Force (N) is the load magnitude applied to the vertebral column during the injury.  Velocity (mm/s) refers to the stroke velocity of the induced injury. For illustration, a typical force-displacement plot of a specimen is shown in Figure 29. The failure point, indicated by a sharp drop in the force magnitude, is recognized as the biomechanical tissue failure. Upon examining the site of injury for each specimen, disruption of intervertebral disk was observed. The injury plots of all the specimens are found in the Appendices. Figure 29 Typical force-displacement curve of a specimen with a dislocation injury. The biomechanical tissue failure point is indicated by point A. 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) A 40 Forces were monitored during coupling between the caudal clamp and the actuator to ensure no damage to the cord occurred prior to the test stroke. This was done by monitoring the force magnitudes plotted in real time. After creating the injury, there were two methods to confirm the consistency of the dislocation. First, the force-displacement magnitudes were monitored at the point of the sharp drop in the curve (Point A shown in Figure 29). The second method was to verify this by the surgeon’s expertise; the surgeon observed the site of injury using a microscope looking for characteristic hemorrhage as well as the degree of intervertebral motion physically at the site of injury. These two methods were independently performed and the results were compared. A typical data collection sheet was filled out during the injury to monitor injury parameters, for instance maximum displacement, and maximum force, and the animal’s physiological condition such as heart rate- HR, and blood Oxygen saturation level- SPO2 (Appendix I). Fixation Procedures: The two injured groups of specimens received spinal column stabilization provided by the fixation device post-injury. After the initial mechanical insult, the fixation device was implanted across the C5 and C6 segments using custom instrumented surgical forceps which were used to ensure that the appropriate clamping force was applied providing consistency. A stretching load was applied on the rodent spinal column to straighten it while implanting the device. While clamping the device and tightening the screws to hold the vertebrae, strain measurements from the outfitted forceps were taken and recorded for further analysis of the clamping force. Once the appropriate range of force magnitude was achieved, the screws were tightened and the fixation device was secured to the spine. These procedures are shown in Figure 30. 41 Figure 30 Biomechanical fixation process, A) the overall setup, where the strain signal from the instrumented forceps is read during the device implantation, B) the desired level of force magnitude is achieved and the screws are tightened, C) an image of the device implanted is shown. A lateral view of the dents located between the transverse processes and the ridges dorsal to the transverse processes in the cervical region of the rat is shown in Figure 31A. This is the place where the fixation device held the vertebrae (same place where the vertebral clamps are attached). The fixation device implanted on the spine is shown in Figure 31B. Figure 31 A) A Model of Sprauge-Dawley rat spine C4-T2 region highlighting the dents between the transverse processes and the ridges dorsal to the transverse processes in the cervical region, where the device is holding on to, B) fixation device implanted at C5/C6 Joint, holding the adjacent vertebrae as well. Clamping Force Measurement System: The clamping force measurement system was used to measure forces applied to the fixation device while implanting it on the vertebrae. This system consisted of a custom set of instrumented surgical forceps, a signal conditioning system (NI USB-9237, National Instrument, Mississauga, Canada), and LabVIEW software for building data acquisition and instrument control applications. A schematic diagram of this system is depicted in Figure 32. The strain B A A B C 42 gauges were instrumented manually with one active gauge and one dummy gauge to remove the temperature effects (Figure 33). Figure 32 A schematic diagram of the Clamping Force Measurement System. Arrows indicate the direction of clamping force F applied from the instrumented forceps to the implant. Figure 33 Custom strain gauged instrumented surgical forceps Calibration Method: A schematic diagram of the instrumented forceps subjected to a known load FG using a servohydraulic testing machine (Instron DynaMight, Instron, Norwood, MA) is depicted in Figure 34. A rigid block was placed between the forceps teeth to simulate the fixation device. The location of the strain gauge (SG) is labeled on the diagram. Load (FG) was applied at a distance (a) from the pivot point. Load (FG) with a relative ramp waveform under force control was applied in three cycles (Load cell (200 lbs. (1kN)) with the sample rate of 50 Hz) (Figure 35). Strain values were measured synchronously with the applied force data. The applied force and the strain measurements of the last cycle were compared and plotted to determine if a linear 43 relationship exists between the strain and the load. The overall calibration set-up is shown in Figure 36. A rigid aluminum block was placed and secured between the forceps teeth, while a custom-made indenter applied the force to the position where experimentally, the fingers would hold the forceps. A thin layer of foam was attached between this indenter and the forceps to simulate the finger touch. Figure 34 Schematic diagram of calibration of the instrumented forceps. The instrumented forceps is subjected to a known load FG which represents the grip force at a distance a from the pivot point. Strain gauge location is labeled as SG. A rigid block is placed between the forceps teeth to simulate the fixation device. Figure 35 Application of load FG by DynaMight programmed in three cycles with a relative ramp waveform. Force applied started from 0N ramping to -50N and going back to 0N in each cycle (sample rate of 50 Hz). 44 Figure 36 A) The overall calibration set-up, B) the highlighted region in A with the instrumented forceps subjected to a force applied by DynaMight. To analyze the collected measurements a custom MATLAB script was written to match the frequencies of load and strain values and synchronize the data (Appendix E). The graph obtained represents a linear relationship between the measured strain and the applied load based on Equation 3 (Figure 37). Note that only the third cycle of the collected data is shown. Force (N) = -41803 × strain - 50.754 : Equation (3) Figure 37 Calibration graph of the instrumented forceps depicting the relationship between measured strain and applied load. Equation of the line is also shown on the graph. y = -41803x - 50.754 R² = 0.9995 -60 -50 -40 -30 -20 -10 0 -1.40E-03 -1.20E-03 -1.00E-03 -8.00E-04 -6.00E-04 -4.00E-04 -2.00E-04 0.00E+00 2.00E-04 F o rc e ( N ) Strain (m/m) Load Cell Forceps Block Instron Electronic Controller strain signal load signal 45 Force-Strain Conversion and data acquisition: The injury group of specimens with the immediate post-injury end-point was implanted with the fixation device. The strain magnitudes recorded were converted to force values based on the calibration curve. To find the effect of applying different clamping forces on the spine stiffness three different levels of force magnitudes were determined for implementing during the implantation process. This would also show that if there were an inconsistency in surgeon’s load application during the device implantation how it would influence the spinal stiffness of the subjects. This categorization was defined based on a pilot test where the surgeon applied different levels of load on a cadaveric rat spine using the instrumented forceps and data was analyzed afterwards based on Equation 3 to establish the limits:  High Force: 15-25 N  Medium Force: 10-15 N  Low Force: 3-10 N The biomechanical evaluation was conducted on the spine after the fixation device had been applied with a known clamping force. The order of the clamping force application was selected as high, medium, and low. The order was selected to minimize the error associated with the destructive testing at the clamping site that might have occurred due to low a clamping force. The strain data collected during the device placement on the spine was noisy due to the physical vibrations of the surgeon’s hand. Therefore, to remove the noise, a MATLAB script was written (MATLAB - The MathWorks. Natick, MA, USA) – Appendix F. This code script receives the raw data and plots it while taking the average of the peak region of the graph where it was corresponding to the time when the force was applied on the device. It also shows the average strain value in an output box. A sample graph is shown in Figure 38. 46 Figure 38 Sample output graph of the clamping strain applied during the device implantation Strain measurement system sensitivity analysis: To find out if the measured strain is sensitive to the location of the grip force FG, value of “a” can be changed and strain can be measured accordingly (Figure 34). An experiment was conducted; where known forces (up to 30N) were applied in a set-up similar to the calibration procedure (Figure 36) at different locations on the forceps handles such that distance (a) was changed. In this set-up, a block was placed between the forceps teeth, and the location of grip force application was changed from a most proximal region to the strain gauge (L1) to a most distal region (L4) (Figure 39). The total region that FG could be applied at was 6 cm long. The applied force-strain measurement relationship for each region from L1 to L4 was plotted (Figure 40). To obtain the precision associated with this method, a known load was applied at a certain location in several trials. The precision was defined as the standard deviation of the strain measurements in five trials with a known applied load (20N) at a specific region (L1) (Table 4). Average Peak Strain 47 Figure 39 Schematic diagram of sensitivity test of the instrumented forceps. Strain gauge location is labeled as SG. A rigid block was placed between the forceps teeth to simulate the fixation device. The instrumented forceps was subjected to a known load FG, representing the grip force. FG was applied at four regions, labeled L(1)-L(4), which were located at known distances from SG. Figure 40 Applied force-measured strain plot for different regions (L1, L2, L3, and L4) Table 4 Strain measurements in five trials for a known load (20N) at a specific region (L1) Trials Applied Load Measured Strain 1 -20.05 -6.95E-04 2 -20.09 -6.90E-04 3 -20.06 -6.93E-04 4 -20.08 -6.91E-04 5 -20.05 -6.94E-04 Mean ± S.D. -20.07 ± 0.02 -6.93E-04 ± 2.07E-06 -40 -35 -30 -25 -20 -15 -10 -5 0 -1.40E-03 -1.20E-03 -1.00E-03 -8.00E-04 -6.00E-04 -4.00E-04 -2.00E-04 0.00E+00 F o rc e ( N ) Strain (m/m) L(4) L(3) L(2) L(1) L4 L3 L2 L1 48 The results show that when the grip force was applied closer to the strain gauge, the strain was measured more accurately for the same value of applied load (FG). The differences associated with the strain readings of regions L2, L3, and L4 with respect to L1 were 25%, 59%, and 80%, respectively. From an experimental point of view, the surgeon would not hold the forceps handles at regions L3 or L4 due to the difficulty of grasp. The results above illustrated that the difference in the strain measurements due to a change in the load application location was small within the first two proximal regions (maximum of 25%). However, it is important to ensure that the surgeon applies the grip force consistently at a point close to the strain gauge. The precision of the strain measurements was defined as the standard deviation of each strain measurement in five trials with respect to the known load applied. Precision using this method was found 2.07×10 -6 (m/m) (Table 4). Recovery and Survival – Sacrifice and Tissue Harvest: Following the injury and fixation, in vitro group of animals were euthanized while deeply anesthetized by an over-dose of isofluorane at the corresponding time-point. The spinal column was harvested from C3 to T1 (C3, C4, C5, C6, C7, and T1). In the animals designated to the in vivo survival study, the muscle and the skin was sutured over top of the fixation clamp. The animals were allowed to recover normally and appropriate post- operative procedures were taken: 48 hours post-procedure: Animals received 3 times daily bladder expression until bladder reflexes have re-established (for approximately 7 days). Animals also received injections of Baytril (10mg/kg) during this period, if required, due to having bladder infections. Mashed-up food, treats, and transgel/water/Ensure were provided well within reach. Buprenorphine (0.03 mg/kg sc) and fluid supplement (Ringers Solution) (5cc) were given subcutaneously twice daily. Special monitoring sheets were used to follow up the animal well-being post surgery (Appendix H). This ensured that different physiological signs, such as weight, condition of wounds, physical appearance, clinical signs, behavior, and possible tumors, would be given a score and appropriate precautions would be taken place. The humane end-point was achieved if an animal reached an overall score of 20 by filling out the monitoring sheet. Based on the definition of the Canadian Council on Animal 49 Care (CCAC), humane end-point refers to the moment in the experiment at which pain and/or distress experienced by an investigational animal are ended, minimized or reduced by either killing it humanely, or by discontinuing the painful procedure [113]. 48 hours to two weeks post-procedure: Animals were monitored continuously post-injury twice a day for a period of two weeks and then once a day if fully recovered. If the physiological conditions were back to normal based on the monitoring sheets, the animals were considered normal. Special considerations due to severity of injury and its effect on the forelimbs: Acutely post-injury, animals were not able to move or use their forelimbs. Therefore, they were monitored more regularly, turned over in the cage, and their bedding would be cleaned, if required, to avoid infection. The animals were also hand-fed a liquid nutritional supplements (Ensure®) due to not being able to feed themselves during this phase. At eight weeks post injury, in vivo group of animals were euthanized and the spinal columns were harvested similar to the in vitro group. 2.2.1.4 Statistical Methods This study investigated the hypothesis that the device-spinal column stiffness over time is greater than the injured spinal column immediately after the injury. This hypothesis was tested using a one-way analysis of variance at a 95% level of significance (parameters: Range of Motion and Neutral Zone). The Kolmogorov-Smirnov test was performed to determine the normality of the range of motion and neutral zone data sets. A post hoc test, Student Newman- Keuls test (SNK) was used [114]. All the statistical analyses were analyzed using Statistica 7 (StatSoft, Tulsa, Oakland/USA). Note that the injured/fixed specimens had three separate values for ROM and NZ data for each test condition of high, medium, and low clamping force. If there was a significant difference between the motion associated with the three clamping force conditions, the medium condition would be selected to obtain the final ROM and NZ values for that individual specimen. If there 50 was no significant difference, these values were averaged and considered as the specimen ROM and NZ. 2.2.2 Establishing A Dislocation Magnitude In A Pilot Study Motivation: Prior to this study, experiments conducted with the UBC’s multimechanism SCI system were with immediate post-injury end-points or with survival time of three hours. Therefore, establishing an appropriate dislocation magnitude for a longer survival time was an important step that needed to be taken before starting the main study. This displacement must not only produce a distinct histological spinal cord injury, but also must be tolerated by the animals such that they can survive. Pilot Study Design: A group of male Sprague-Dawley rats were randomly selected (Mean ± S.D. weight = 356 ± 6 g) and divided in two groups with different levels of injury severity. Light and severe dislocation injury levels were defined at 1.3 (N = 3) and 1.5 mm displacement (N = 5), respectively and allowed to survive for 3 weeks. Close monitoring and gross behavioral assessments were performed during lifespan. One week post-injury, Basso-Beattie-Bresnahan (BBB) open field locomotor rating scale (Appendix A) for assessing locomotion following spinal cord injuries was performed. This rating was completed by two experienced individuals at ICORD. At the experimental end-point, all animals were euthanized while deeply anesthetized (overdosed - respiratory arrest) via transcardial perfusion of 250 mL of phosphate buffered saline (0.12M PBS) followed by 500mL of 4% parafomaldehyde in PBS. Spinal cords were post-fixed overnight in 4% paraformaldehyde and cryoprotected in graded sucrose in PBS (12, 18, and 24%) before being frozen in isopentane cooled with dry ice. Cords were transversely cryosectioned at 20μm [17]. Appropriate histochemical techniques were employed to characterize tissue damage qualitatively. The stain used in this process was Eriochrome Cyanine stain for myelin which illustrates the damage in the white matter. Sections were imaged (5× objective lens) using an AxioPlan2 microscope (Carl Zeiss, Thornwood, NY) equipped with monochrome camera with RGB unit (Retiga-Exi, QImaging, Burnaby, BC) using Northern Eclipse software (Empix Imaging, Mississauga, ON) [17] (Figure 41). 51 Figure 41 Histological slides of two specimens at the site of injury with A) 1.3, and B) 1.5 mm displacement severity 52 3 RESULTS 3.1 Implant Final Design Decision analysis is one of the critical phases in the design process. Based on the established design criteria (Table 5), the posterior-screw fixation model was selected as the final design with the score of 3990. The final design of the spinal fixation device made of PEEK is illustrated in Figure 45. Engineering drawings of the device can be found in Appendix G. Table 5 Fixation device design decision analysis Lateral-screw Cam-screw Posterior-screw Criteria Analyzed Qualitatively Weight Score Weighted Score Score Weighted Score Score Weighted score Structural stiffness 30 36 1080 36 1080 28 840 Ease of surgical implantation 40 25 1000 25 1000 50 2000 Versatility 10 35 350 30 300 35 350 Cost 10 40 400 20 200 40 400 Size 10 40 400 20 200 40 400 Total Weighted Score 100 3230 2780 3990 All the design criteria are listed in Table 5 except biocompatibility and MRI-compatibility, which are related to the material selected for the device (PEEK). The structural stiffness of the device was an important criterion, which was given the second highest weight (30). This criterion, basically, was the determining factor in terms of the mechanical functionality of the device. Ease of surgical implantation, however, was the most important criterion and was given the highest weight (40). This was due to the fact that it determines how dangerous an operation could be for the animal and how it would effect its survivability during and after the operation, as well as how difficult and time-consuming it is for the surgeon to implant the device. The other three criteria, versatility, cost, and size were given equal weights (10). Below, the rationale behind the scoring of different designs is explained: 53 Structural stiffness: The three proposed prototypes were similar in the proposed manufacturing material and the design in consisting of two parts holding the vertebrae from the lateral direction, and having a C- shaped cross-sectional area, as well as similar geometrical dimensions. The stiffness for the three individual designs, however, is expected to slightly differ depending on the use of screws or cams and the direction of the applied force. Based on this qualitative analysis, given that the designs have similar shapes (consisting of two parts holding the vertebrae, C-shaped cross-sectional area, and similar geometrical dimensions) and would be made of PEEK, the applied clamping force (force F in Figure 32) would be the determining factor in the selection process. The clamping force is applied from the device to the vertebrae in order to stabilize the injured spine. The Lateral-screw and the Cam-screw designs were given a higher score for this criterion. In the Lateral-screw design, the screws are coincident with the forces that need to be applied to the vertebrae for fixation (36). The Cam- screw design holds the two parts of the device from lateral side using a cam-shape linkage. The force applied from the linkage also coincides with the force required to clamp the vertebrae (36). The Posterior-screw design achieves the lowest score in this criterion, since the clamping force is applied through the threads of the screws located posteriorly. This force is perpendicular to the direction of the force that needs to clamp the vertebrae (28). For the Posterior-screw design, a theoretical analysis was performed to estimate the structural stiffness of the device (Figure 42): Calculations of second moment of area of the proposed design: Figure 42 Schematic diagram estimating the cross sectional area of the device for the second moment of area calculations. Letters h and b refer to the geometrical properties defining the cross sectional area, and I refers to the second moment of area of the corresponding cross-section. 54 Cross-sectional area geometrical properties: b1 = 7.6 mm h1 = 14 mm b2 = 6.4 mm h2 = 12 mm 𝐼𝑂 = 7.6 × 143 − (6.4 × 123) 12 = 816 𝑚𝑚4 Estimated loads applied to the device: Based on the data previously published by our group, the applied load to a Sprague-Dawley rat spinal segment in the dislocation injury is about 13 N to 31 N [17]. Therefore, when the device is implanted a portion of this force would be transferred to the implant due to the load sharing between the spine and the implant. For the worst case scenario, it is assumed that all the applied load is transferred to the device (maximum load of 31 N). Therefore: Moment arm for the force Fmax: d Applied moment: M’= Fmax.d = 31 N×13 mm= 403 Nmm Maximum moment transferred through the bony interface to the device: M ≈ M’ Estimated deflection of the device under the specified loads (Figure 43): Assumptions:  Device acts as a beam with a length of l (10mm).  Shear loads F are acting at distance “a” from both sides. Magnitude of “a” is assumed (3mm).  The x axis is positive to the right and the y axis is positive upward.  The initial condition of the beam is shown in black and the beam deflected under the applied load is shown in blue. 55 Figure 43 FBD of the device acting as a beam. The initial condition of the beam is shown in black and the beam deflected under the applied load is shown in blue. Using superimposition the device maximum deflection in the y direction can be found as follows [115]: Deflection, 𝑦𝑚𝑎𝑥 = − 𝑀𝑙2 2𝐸𝐼 + 𝐹𝑎2 6𝐸𝐼 (𝑎 − 3𝑙) 𝑦𝑚𝑎𝑥 = − 403 𝑁𝑚𝑚 10 𝑚𝑚 2 2 3.6 𝐺𝑃𝑎 816 𝑚𝑚4 + 31𝑁 × 3 𝑚𝑚 2 6 3.6 𝐺𝑃𝑎 816 𝑚𝑚4 (3 𝑚𝑚 − 3 × 10 𝑚𝑚) 𝑦𝑚𝑎𝑥 = −0.0069 𝑚𝑚 + −0.00043 𝑚𝑚 = −0.0073 𝑚𝑚 Structural stiffness: Stiffness = load/deflection Stiffness = 31 N /0.0073 mm = 4247 N/mm This stiffness value could be compared to the stiffness of a simple rectangular cross-sectional bar made of PEEK that would be attached to the posterior side of the vertebrae. Cross-sectional area geometrical properties (Figure 44): Figure 44 Schematic diagram showing the cross sectional area of a rectangular bar for the second moment of area calculations. Letters h and b refer to the geometrical properties defining the cross sectional area. M F F l-2a a a x y ymax 56 b = 1.9 mm h = 14 mm 𝐼𝑥 = 𝑏ℎ3 12 = 1.9 × 143 12 = 434 𝑚𝑚4 Assuming similar applied loads as shown previously, maximum deflection can be found as follows: Deflection, 𝑦𝑚𝑎𝑥 = − 𝑀𝑙2 2𝐸𝐼 + 𝐹𝑎2 6𝐸𝐼 (𝑎 − 3𝑙) 𝑦𝑚𝑎𝑥 = − 403 𝑁𝑚𝑚 10 𝑚𝑚 2 2 3.6 𝐺𝑃𝑎 434 𝑚𝑚4 + 31𝑁 × 3 𝑚𝑚 2 6 3.6 𝐺𝑃𝑎 434 𝑚𝑚4 (3𝑚𝑚 − 3 × 10𝑚𝑚) 𝑦𝑚𝑎𝑥 = −0.013 𝑚𝑚 + −0.00080 𝑚𝑚 = −0.014 𝑚𝑚 Stiffness = load/deflection Stiffness = 31 N /0.014 mm = 2214 N/mm In an in vivo study in sheeps, fusion was achieved 12 weeks post-injury with a resultant flexion- extension intervertebral ROM of 4 degrees using a spinal fixation device [116]. In another sheep in-vitro study with a similar biomechanical evaluation set-up, the ROM of an intact cervical spine joint was found to be about 13 degrees [117]. These two studies show that fusion was obtained while a 70% reduction in ROM was achieved during a period of 12 weeks post-fixation. Using these values and estimating for the rat, a 70% reduction in ROM of the intact cervical spine results in about 5 degrees of intervertebral ROM (the intact rat spine ROM at C5-C6 is found about 16 degrees. These results can be found in section 3.2.2.1). At about 12 weeks post- injury in a rat model, if this or a smaller intervertebral ROM is observed, it is anticipated that a bony fusion most probably has been formed. Comparing this degree of motion to 0.014 mm deflection (equivalent to about 0.014 degree) achieved by using a simple rectangular bar, it is anticipated that a simple bar made of PEEK would provide sufficient mechanical stability, which would lead to fusion formation at the dislocated joint. This also ensures that the proposed design with a deflection magnitude half of the rectangular bar would provide more sufficient mechanical stability, ensuring that fusion would be achieved at the dislocated joint. This 57 conclusion is made by assuming that the device holds the spine in a way that there is minimal motion between them. This assumption can be validated at the experimental end-point. Ease of surgical implantation: Although the Lateral-screw design is relatively compact, it is very invasive to the tissues surrounding the implant. The surgeon is required to expose a wider area of the lateral-posterior side of the animal spine by removing the skin and muscles. Implanting this device endangers the survival of the animal during the operation, and increases the recovery time post-injury (25). The Cam-screw design is bulky, and in order to engage the locking mechanism, the surgeon is required to expose a wide area on the lateral-posterior side. Due to this surgically invasive procedure, again, a low score was given (25). The Posterior-screw design is compact and easy for the clinician to work within the microscopic environment (50). Versatility: All three designs obtained similar scores in this criterion. A slightly higher score was given to Lateral-screw (35), and Posterior-screw designs (35), which are similar in consisting of two slots to accommodate different sizes of animals. Cost: The Lateral-screw, and Posterior-screw designs are similar in design and contain sharp angles which is easier to machine compared to curved surfaces. However, the Cam-screw design consists of a cam shape linkage, which is more difficult to machine, and would be given a lower score compared the other designs (20). Size: The Lateral-screw and Posterior-screw designs are compact in size occupying similar space when implanted. However, the Cam-screw design is bulky and was given a lower score compared the other designs (20). 58 Figure 45 Custom-designed MRI-compatible cervical spine fixation device for rats 3.2 Biomechanical Evaluation 3.2.1 Dislocation Injury Parameter Repeatability The dislocation statuses of individual specimens are shown in Table 6. Plots of force- displacements and surgeon’s clinical examinations were verified to ensure that the dislocation injury was created. The point at which the specimen failed for the injured/8-week and injured/fixed groups were 1.07 ± 0.17 mm and 1.01 ± 0.13 mm, respectively (Mean ± S.D). A summary of injury parameters along with their statistical measures are shown in Table 7. Maximum displacement, force, and velocity are listed. Additional load-displacement curves for individual specimens of each group are shown in Appendix D. 59 Table 6 Dislocation status of specimens post-injury obtained by graph analysis and clinical examination Group Specimen Specimen failure [mm] Clinical Examination In ju re d /8 -w e e k 1 0.85 Dislocated 2 0.95 Dislocated 3 0.90 Dislocated 4 1.25 Dislocated 5 1.30 Dislocated 6 1.15 Dislocated 7 1.10 Dislocated Mean ± S.D. 1.07 ± 0.17 In ju re d /F ix e d 1 0.90 Dislocated 2 1.10 Dislocated 3 1.25 Dislocated 4 0.95 Dislocated 5 0.95 Dislocated 6 0.91 Dislocated Mean ± S.D. 1.01 ± 0.13 Table 7 Statistical summary of injury parameters of each group, * denotes that two specimens were excluded from analysis. Group Max Displacement (mm) Max Force (N) Max Velocity (mm/s) Mean S.D. Of Displacement Mean S.D. Of Max Force Mean S.D. Of Max Velocity Injured/Fixed N=6 1.55 0.03 11.8 2.2 364 15 Injured/8-week N=7* 1.55 0.49 11.3 4.0 363 9 3.2.2 Kinematic Analysis Of In Vitro Specimens In this section, the results of the kinematic analysis of in vitro specimens are shown in the form of rotation-moment plots and summary tables of NZ and ROM values. To remove the viscoelastic effects, only the last cycle of loading has been shown for each specimen. The results of individual specimens of each group are shown in Appendix D. 60 3.2.2.1 Intact Specimens The spinal characteristic graphs of the intact specimens are presented in Figure 46 showing the intervertebral motion at C5/6 joint due to the applied moments. The result of statistical measurement of this data is summarized in Table 8. NZ and ROM for the intact group are 3.4 ± 2.8 and 18.1 ± 3.3 degrees, respectively (Mean ± S.D.). Figure 46 Intervertebral motion-applied moment graphs at C5/C6 joint for intact group specimens Table 8 Intact group statistical summary Intact Group Specimens NZ (deg.) ROM(deg.) 1 2.5 15.0 2 0.4 14.9 3 4.2 17.7 4 7.4 22.6 5 5.5 21.7 6 0.4 16.5 Mean ± S.D. 3.4 ± 2.8 18.1 ± 3.3 -15 -10 -5 0 5 10 15 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) Rotation-Moment curves at C5/C6 Joint for Intact Group specimen1 specimen2 specimen3 specimen4 specimen5 specimen6 61 3.2.2.2 Injured/Fixed Specimens In this section, the clamping force values of high, medium, and low force for each specimen are shown in Table 9 along with their means, and standard deviations. The NZ and ROM values obtained from the three clamping force cases (high, medium, and low) are shown in Table 10. The results of individual specimens of each group are depicted in Appendix D. Table 9 Clamping force values of high, medium, and low for each specimen Clamping Force (N) Specimen High Med Low 1 24.2 14.3 6.5 2 24.4 11.1 4.2 3 24.8 13.3 6.6 4 22.6 11.6 3.3 5 17.5 12.4 7.5 6 23.7 11.2 7.3 Mean ± S.D. 22.9 ± 2.7 12.3 ± 1.3 5.9 ± 1.8 Table 10 Intervertebral motion at C5/6 joint for specimens with the three separate clamping force applications Clamping Force Level High Med Low Specimen NZ (deg.) ROM (deg.) NZ (deg.) ROM (deg.) NZ (deg.) ROM (deg.) 1 4.3 14.8 3.6 7.2 3.5 27.3 2 6.7 26.9 5.9 26.1 4.5 31.2 3 3.9 14.4 4.7 13.0 5.2 15.2 4 3.3 20.5 1.2 23.9 0.7 17.8 5 8.0 24.9 3.9 33.4 15.2 29.2 6 4.8 8.7 1.5 12.4 9.8 11.0 Mean ± S.D. 5.2 ± 1.8 18.3 ± 7.0 3.5 ± 1.8 19.3 ± 10.0 6.5 ± 5.2 22.0 ± 8.4 The data listed in Table 10 for ROM is also shown in the box and whisker plot form (Figure 47). 62 Figure 47 Intervertebral motion (ROM in degrees) at C5/6 joint for each specimen of the injured/fixed group in the testing conditions of high, medium, and low clamping force applications. Black dots and whiskers represent the mean and standard deviations from the mean values. The results of the repeated measurements ANOVA for the specimens of injured/fixed group is shown in Table 11. The obtained p-value for the group (0.45) is much greater than 0.05, meaning that there is no significant difference across the high, medium and low groups in their average ROM value. Table 11 Repeated measurements ANOVA for ROM of the injured/fixed group in high, medium, and low clamping force conditions Source of Variation Df Sum of Squares Mean Square F p-value Group 2 42.10 21.05 0.862 0.451478 Subject 5 846.32 169.26 6.931 0.004849 Residuals 10 244.21 24.42 The data listed in Table 10 for NZ is also shown in plot form in Figure 48. 63 Figure 48 Intervertebral motion (NZ in degrees) at C5/6 joint for each specimen of the injured/fixed group in the testing conditions of high, medium, and low clamping force applications. Black dots and whiskers represent the mean and standard deviations from the mean values. The results of the repeated measurements ANOVA for the specimens of injured/fixed group are shown in Table 12. The obtained p-value for the group (0.23) is much greater than 0.05, meaning that there is no significant difference across the high, medium and low groups in their average NZ value. Table 12 Repeated measurements ANOVA for NZ of the injured/fixed group in high, medium, and low clamping force conditions Source of Variation Df Sum of Squares Mean Square F p-value Group 2 27.23 13.62 1.686 0.2339 Subject 5 86.94 17.39 2.153 0.1414 Residuals 10 80.76 8.08 Since there was no significant difference between the results of different clamping force conditions, the overall NZ and ROM values for each specimen were obtained by averaging these values. NZ and ROM for the injured/fixed specimens are 5.0 ± 2.4 and 19.9 ± 7.5 degrees, respectively (Mean ± S.D.). The statistical summary of this data is listed in Table 13. 64 Table 13 Injured/fixed group statistical summary Injured/Fixed Specimen No. NZ (deg.) ROM(deg.) 1 3.8 16.4 2 5.7 28.0 3 4.6 14.2 4 1.8 20.7 5 9.0 29.2 6 5.4 10.7 Mean ± S.D. 5.0 ± 2.4 19.9 ± 7.5 3.2.3 Kinematic Analysis Of In Vivo Specimens In this section, the results of kinematic analysis of in vivo specimens are depicted in the form of rotation (degrees) versus applied moment (Nmm) graphs and summary tables of NZ and ROM values. Again, only the last cycle of loading has been shown for each specimen. In Appendix D, the results of individual specimens are depicted. Note that the original number of specimens in this group was originally 9; however, two specimens were euthanized three days post-injury due to not regaining the expected recovery. 3.2.3.1 Injured/3-Week Specimens Qualitative Analysis NZ and ROM for the injured/3-week specimens with the fixation device are 0.3 ± 0.2 and 1.4 ± 0.9 degrees, respectively (Mean ± S.D.). For the specimens with the device removed, the NZ and ROM values are 1.8 ± 1.5 and 4.1 ± 3.1, respectively (Mean ± S.D.) (Table 14). The individual graphs for these specimens are depicted in the Appendix D. Table 14 Injured/3-week group statistical summary Injured/3-week With Device With Device Removed Specimen No. NZ (deg.) ROM (deg.) NZ (deg.) ROM (deg.) 1 0.3 0.9 2.8 2.8 2 0.1 0.9 0.2 1.8 3 0.6 2.5 2.5 7.6 Mean ± S.D. 0.3 ± 0.2 1.4 ± 0.9 1.8 ± 1.5 4.1 ± 3.1 65 3.2.3.2 Injured/8-Week Specimens NZ and ROM for the injured/8week specimens with the fixation device are 0.3 ± 0.3 and 0.7 ± 0.4 degrees, respectively (Mean ± S.D.). For the specimens with the device removed, the NZ and ROM values are 0.7 ± 0.5 and 1.5 ± 0.7, respectively (Mean ± S.D.) (Table 15). The rotation-moment graphs of individual specimens can be found in Appendix D. Table 15 Injured/8-week group statistical summary Injured/8-week With Device With Device Removed Specimen No. NZ (deg.) ROM (deg.) NZ (deg.) ROM (deg.) 1 0.0 0.4 0.9 1.1 2 0.1 0.6 0.1 0.6 3 0.8 0.8 0.4 2.6 4 0.4 0.4 0.7 1.2 5 0.4 1.4 0.3 1.2 6 0.1 0.8 0.7 2.0 7 0.1 0.4 1.7 1.9 Mean ± S.D. 0.3 ± 0.3 0.7 ± 0.4 0.7 ± 0.5 1.5 ± 0.7 3.2.4 Statistical Summary Result Of All Groups 3.2.4.1 Degree Of Intervertebral Motion Across Groups The degrees of intervertebral motion for the three study groups is given in Figure 49. Means and confidence intervals of ROM and NZ values in degrees are compared across the groups. (*), (**), and (***) represent p-values of 0.0002, 0.006, and 0.04, respectively. The ANOVA results for ROM and NZ are shown in Tables 16 and 17, which indicate that there is strong evidence against the null hypothesis that the mean response is the same across groups. The results of the SNK test are shown in Tables 18 and Figure 50 for ROM and in Table 19 and Figure 51 for NZ. The p-values marked in red represent a significant difference between the two specified groups. 66 Figure 49 Statistical comparison of degrees of intervertebral motion between groups, (*), (**), and (***) represent p-values of 0.0002, 0.006, and 0.04 respectively. Table 16 ANOVA for the ROM raw data Source of Variation Df Sum of Squares Mean Square F p-value Between groups 2 1362.31 681.160 31.997 2.56E-06 Error (within a group) 16 340.61 21.290 Table 17 ANOVA for the raw NZ data Source of Variation Df Sum of Squares Mean Square F p-value Between groups 2 62.903 31.451 7.100 0.006206 Error (within a group) 16 70.873 4.430 Table 18 Results of the post hoc analysis of the ROM raw data using SNK test. The p-values marked in red represent a significant difference between the two specified groups. Injured/8-week Intact Injured/Fixed Injured/8-week 0.000164 0.000171 Intact 0.000164 0.501466 Injured/Fixed 0.000171 0.501466 0 5 10 15 20 25 30 35 Intact Injured (t=0 w) Injured (t=8 w) D e g re e s o f In te rv e rt e b ra l M o ti o n a t C 5 /C 6 Plots of Means and 95% Confidence Intervals ROM NZ * ** * *** Inta Injured/Fixed Injured/8-week 67 Figure 50 Results of the post hoc analysis of the ROM raw data using SNK test Table 19 Results of the post hoc analysis of the NZ raw data using SNK test. The p-values marked in red represent a significant difference between the two specified groups. Injured/8-week Intact Injured/Fixed Injured/8-week 0.036650 0.005770 Intact 0.036650 0.187077 Injured/Fixed 0.005770 0.187077 Figure 51 Results of the post hoc analysis of the NZ raw data using SNK test 3.2.4.2 ROM From the data in Figure 49, note that the variance in data-sets is not the same across groups. This does not meet the assumption needed for ANOVA. To overcome this, the data were transformed to allow for equal variation across the subject groups. This is called applying a variance-stabilizing transformation [118]. However, the conclusions of the analysis apply to the transformed populations. The analysis was performed on both the regular and transformed scales and the same results were achieved. Also, the sample size across the groups are not equal; but, we can still use ANOVA with a slight adjustment for unbalanced data [119]. The plots of the transformed data are shown in Figure 52. Based on these plots it appears that the variation is approximately equal across the groups. The one-way ANOVA analysis is used for the transformed data. 68 Figure 52 Plot of ROM transformed data of the three groups The ANOVA result of the transformed ROM data is given in Table 20. This analysis supports the hypothesis that the means between some of the groups are different (P-value < 0.05). To see where the difference among groups lies, a Student Newman-Keuls test (SNK) (Figure 53), indicates that injured/fixed and intact were not significantly different, but both were different from the injured/8-week. The results from the transformed data are comparable to those using the regular scale (Table 16). Table 20 ANOVA for transformed data of ROM Source of Variation Df Sum of Squares Mean Square F p-value Between groups 2 30.034 15.017 96.249 1.20E-09 Error (within a group) 16 2.496 0.156 Figure 53 Results of the post hoc analysis of the transformed ROM data using SNK test 69 3.2.4.3 NZ From the data shown in Figure 49, again, there appears to be some problems with the assumption of constant variance across the groups. Similar to the previous section, the log transformation is performed on the raw data (Figure 54). Figure 54 Plot of NZ transformed data of the three groups The ANOVA result of the transformed NZ data is given in Table 21. This analysis indicates that there is strong evidence against the hypothesis that the means of the transformed NZ for all groups is the same (P-value < 0.05). For the post hoc analysis, a SNK test was used again (Figure 55). This analysis indicates that the injured/fixed and intact groups were not significantly different, but both were different from the injured/8-week. The results from the transformed data are comparable to those using the regular scale (Table 17). Table 21 ANOVA for transformed data of NZ Source of Variation Df Sum of Squares Mean Square F p-value Between groups 2 15.251 7.625 8.174 0.003583 Error (within a group) 16 14.927 0.933 Figure 55 Results of the post hoc analysis of the NZ transformed data using SNK test 70 3.3 Pilot Study Results In Establishing A Dislocation Magnitude Both of the injury severities were tolerated by the animals in terms of survivability. The neurological deficiencies were more pronounced in the 1.5 mm dislocation injury group. The average of behavioral analysis score of the 1.3 mm dislocation injury group was 20 in the BBB locomotor scale, whereas this score had an average of 15 for the 1.5 mm dislocation injury group. The tissue samples acquired from the 1.5 mm displacement injured animals were distinctly different from the 1.3mm displacement injury. This severe injury was characterized qualitatively by identifying shrunken neuronal cell bodies in the grey matter and smaller area of spared myelin in the white matter observed in the spinal cord sections (Figure 56). A larger lesion size in the 1.5 mm dislocation injury was grossly visible. The histological slides of individual specimens are shown in Appendix B. Based on the above points, it was concluded that the 1.5 mm dislocation injury is both survivable for the animals and produces a histologically distinct injury in the spinal cord tissue. Therefore, this severity was selected for the displacement parameter of injury in the design of the actual study. Figure 56 Histological slides of two specimens at the site of injury with A) 1.3, and B) 1.5 mm displacement severity 71 4 DISCUSSION 4.1 Implant Performance The two overall objectives of this project were to design a custom MRI-compatible spinal stabilization device for the unstable cervical joint and to evaluate the biomechanical fixation provided by this device immediately after injury and at eight weeks post-injury. The performance of the implant was evaluated by measuring the degree of biomechanical fixation that it provides to the injured spine. The results of this study demonstrated that the custom-designed fixation device was effective in stabilizing the rat cervical spine up to eight weeks post-injury. 4.2 Limitations This study highlights the design and biomechanical evaluation of a novel fixation device for rodents. The results yield the desired information; however, this study, as with many biomechanical studies, has several inherent limitations raised that need to be addressed in any related future work. These limitations are listed here along with the suggested solutions. 4.2.1 Design Limitations The results show that magnitude of the clamping force applied to the device does not influence the stabilization of the spine. This could be explained by a limitation in the device design. In Figure 57, the region highlighted in the image refers to a step with a defined height of H. Decreasing this height ensures that application of an increased clamping force value would enhance the stiffness of the corresponding spinal joint. The device with the current design, however, did provide sufficient biomechanical fixation of the spine leading to restoring its stability post-injury. Figure 57 Fixation device step highlighted with a defined height of H 72 Another limitation arises when this device is intended for use in the distraction model of spine injury. As mentioned previously, a stretching load is applied on the spinal column during the implantation of the device. Since in a distraction injury the column is subjected to a stretching force and the muscles along the column are partially injured at the site of injury, implanting the device on the spine using this method would further distract the spinal segments. This was investigated in a pilot study (April 2010) conducted by Chen K. et al. from the Orthopaedic and Injury Biomechanics Group (unpublished study). Following solutions could help when using this device in the distraction model:  Modifications in the method of insertion of the device, which would not require a stretching load  Modification in the design of the device. For instance, incorporating larger teeth that could hold the vertebrae with a greater force. 4.2.2 Biomechanical Evaluation One limitation concerning the current study is that the physiologic loads in the rat spine are not known. One Newton forces in four incremental steps were applied based on the prior pilot test. The corresponding moment applied to the spine at C5/C6 joint is about 14 Nmm. It is still possible that higher or lower loads arise in vivo. A two-dimensional kinematic analysis was implemented. To fully illustrate spine characteristic behaviour a three-dimensional assessment is required. The introduction of another camera in the evaluation setup to capture motion in an additional plane would provide a three dimensional representation of the event. Placement of markers in direction that would enter a different plane of motion could cause a resulting out of plane motion. Using a second camera and improving the alignment of specimens in the rig would eliminate this problem to a great extent. It is worth mentioning that the Optotrak Motion Capture System is capable of measuring 3D motion between the C5 and C6 vertebrae and it was considered for this purpose. However, there were many challenges concerning the use of Optotrak, for instance, the wired IRLED 73 connection, as well as the relatively high mass of infrared markers compared to the mass of excised spine. As a result, this method was not feasible. The experimental apparatus designed for the biomechanical evaluation was custom designed with relatively high dimensional tolerances. Perhaps same apparatus design made by a manufacturer or experienced machinist would eliminate the errors regarding the machining process and manual assembly of the parts which would lead to an enhanced specimen alignment in the rig. The alignment of the specimens in the apparatus was also challenging due to the possible destructive testing method currently implemented. To mount the specimens, the bottom and top vertebrae of the excised spines were attached using two screws. If this attachment was not secure enough, the excised spine would turn about its axis interfering with the specimen alignment or twist subjecting it to additional unknown loads. Placement of markers was implemented by pressing the pins into the bony vertebral bodies. This was done manually. Therefore, locating the markers in a specific point with a certain angle repeatedly was difficult to achieve. The pin insertion angle was important, since it should only be inserted into the vertebral body and not the intervertebral disk which might interfere with the resultant intervertebral of motion. Investigating additional methods would be beneficial in future studies. For example, a hole can be bored into the vertebrae using a high precision surgical drill with the desired angle. Then the markers can be inserted inside the hole. Another method that can be investigated is removing the tissues off the vertebrae at the location where the marker is attached. Then the markers can be glued to the bony surface. 4.2.3 Statistical Limitations When comparing groups with small number of subjects, it is difficult to say with certainty that the average for the sample is truly representative of the population average. A larger number of subjects would be better, if possible, to ensure the average from the sample is more representative of the population average. 74 It would have been more appropriate if the clamping force levels of high, medium and low were applied in a random order to each specimen, instead of always using the same order. This is because there could be a carryover effect when going from one condition to the other. However, this randomization was implemented in a preliminary pilot work. The excised spines immediately after injury were quite unstable and having a low or medium clamping force to start with would have damaged them. Therefore, due to the destructive nature of the test, the order of clamping force application was selected as high, medium and low. It would be optimal to have a repeated measures analysis in which the same subject is tested before treatment (the control), immediately after injury, and at 8-weeks. In this way, the possibility that the differences between groups were due to random variation caused by a small sample, is reduced, and the chance that the differences were truly due to treatment is increased. However, this requires an in vivo method of kinematic analysis, which goes beyond the scope of this project. 4.3 Biomechanical Fixation There was no significant difference in spinal motion between the intact and injured/fixed specimens, meaning that the device restores the stiffness of the injured spine to its original condition. Therefore, the biomechanical fixation provided by the device immediately post-injury results in the stabilization of the spine similar to the intact spine. Currently, there is no study in the literature reporting the change of intervertebral motion in a rat model. The current study shows that immediate reduction in intervertebral motion using the fixation device restores the motion of the injured spine to the intact specimen. Previous human and animal in vitro studies report that post-surgery, the spinal motion is considerably decreased to intact or lower than the intact level [120-123], depending on the type of fixation device used. In a study by Xu et al. (2006), thoracolumbar human cadaver specimens were subjected to a variety of loading conditions. The specimens were tested in four stages of intact, destabilized in the middle segment, fixed with a dynamic pedicle screw, and fixed with a rigid type of pedicle screw. Results showed that the group with the rigid fixation device was significantly stiffer than the dynamic one, which was in the range of the intact spine (ROM of fixed specimens with the rigid fixation and the dynamic fixation were 68% and 30% less than the intact ROM, 75 respectively). It was also shown that the dynamic device provides a comparable stability with the rigid device in other loading directions. Some potential advantages of using the dynamic devices are reduction in stress shielding and implant failure rates, as well as the protection of adjacent segment degeneration by allowing a controlled motion at the unstable joint [122]. In another study by Gunzburg et al. (2009), in an in vivo ovine model, a destabilizing procedure was performed in the lumbar spine region. Radiographs were taken in full extension and flexion of subjects, in intact, destabilized and fixed with a fixation device, and with the same fixation device using a tension band to stabilize the spine in flexion as well as extension. Results showed that 16% and 43% reduction in ROM were achieved with the fixation device and with fixation along with the band compared to the intact [123]. In a study by Maciejczak et al. (2001), 45% immediate reduction in ROM of C5/C6 in flexion/extension was achieved after a discectomy followed by fusion. Fusion was performed using the Cloward technique in the application of an interbody dowel bone graft [120]. The Cloward procedure is a method in which cervical herniated discs, bony spurs and osteophytes are removed through an anterior approach and the two adjacent bodies are fused with a bony or artificial graft [124]. In another study by Finn et al. (2009), the flexion/extension range of motion of a fusion construct under a ±1.5 Nm moment was compared with those of an intact and one with a cervical arthroplasty device. The cervical arthroplasty device in this case was an artificial disk implant. C4/C5 intervertebral ROM of the intact specimens and those with the arthroplasty device were similar and about 75% more than the specimens with the fusion construct [121]. These studies showed that a wide range of reduction in intervertebral motion is achieved immediately after fusion operation, from no change to up to 75% reduction compared to the intact motion. The difference in reduction of intervertebral motion by fixation in the current study compared to those of the literature could be due to the severity of the current injury model. The device design also could be another determining factor in restabilising the injured spine to the intact level. However, this degree of immediate fixation provided by the device was clearly sufficient as the spinal motion after eight weeks was significantly decreased, indicating the possibility of fusion at the injured site. This study shows that the reductions in NZ and ROM at 8 weeks post-injury were 79% and 92%. These values are comparable to the change in intervertebral motion in other animal models [125]. In a study by Erulkar et al. (2001), a single level fusion was performed in a rabbit model with a survival time of 5 weeks post-operation. The NZ and ROM of the fused specimens were 76 significantly decreased from that of intact specimens in flexion and extension (60% to 80%). It was found that fusion substantially reduced the intervertebral motion, but did not eliminate it [125]. In a study by Huang et al. (2006), a comparison was made between an allograft bone and a titanium implant, combined with a fixation plate, following a corpectomy in a bovine model 16 weeks post-operation (corpectomy refers to the removal of a vertebral body [126]). Results showed that after removing the fixation plate, the allograft bone subjects were about 30% stiffer than the subjects with the titanium cage implant upon biomechanical evaluation at the study end- point. This could perhaps indicate fusion at the construct, but has to be further histologically and radiographically determined [127]. In a study by Toth et al. (2002), function of an interbody fusion device was evaluated in 25 sheep at 3, 6, 12, 18, and 24 months. Radiographic, biomechanical, and histological measures were taken to evaluate the fusion. Results demonstrated a trend of increased fusion stiffness, radiographic fusion, and histological fusion at 3 to 24 months [125, 127, 128]. In this study, at eight weeks post-injury, the stability of the spinal column was more than the intact condition resulting in a stiffer joint at the site of injury. This condition is also observed in clinical cases [129, 130]. In these cases, the aim is to restore the original stiffness of the spine along with the rehabilitation practice. Several studies focused on the question of whether or not the stiffness of the spine would be recovered to its initial measures. In a follow-up study by Lin HL et al. (2008), cervical range of motion in patients with anterior cervical discectomy and interbody fusion were compared using radiographic imaging. The results showed that there was an overall decrease in the ROM and flexion movement in the follow-up evaluations of 14.5- months; however, a longer term follow-up study was suggested [130]. 4.3.1 Effect Of Implant Clamping Force On Specimen Kinematics It was shown that there is no significant difference between the three clamping force applications while implanting the device (p-values of 0.45 for ROM and 0.23 for NZ). This means that the magnitude of the clamping force applied to the device does not influence the stabilization of the spine. This could be explained by the device design. This design characteristic eliminates inconsistencies associated with the surgeon’s hand in the device implantation and application of the clamping force. 77 There is no concern regarding the rate of growth of rodents and how it could affect the device clamping force on the vertebrae during the course of implant existence. A recent study on aging and its effects on morphology of the cervical spine in a rat model has determined that the there is approximately a significant difference of 15% in the vertebral canal widths between the young (3 months) and aged groups (12-18 months) [131]. In the current study, the male SD rats had a maximum age of three months and three weeks indicating that they would be categorized in the young group. This leads to the fact that the vertebral growth in this group would not have an effect on the clamping force applied from the device at the bony interface. In a study by Cramer et al. (2004), changes in the lumbar vertebrae of rats were evaluated following fixation in a subluxation model of the spine. In the fixed group, the device was removed at different time-points up to twelve weeks post-op. It was found that the fixed group had more facet joint degenerative changes compared to the control group (P < 0.0001), however, not much degenerative change in the vertebral bodies and intervertebral disks were identified. A direct relation was found between the fixation duration and development of degenerative changes in facet joints [132]. These comparisons were made using a dissecting microscope (6X and 12X magnifications). To provide a higher detailed analysis in this study, it is suggested that a more reliable method such as CT imaging would be used in future studies. Also, the fixation device was made of stainless steel with a high elastic modulus compared to PEEK providing a relatively stiff vertebral fixation. As mentioned, the duration of fixation has a relationship with the degree of degenerative changes. It would be expected that following fusion, if the device is removed, these effects would be reduced. The current study has evaluated a spinal fixation device following an SCI after eight weeks, which was proven to provide adequate fixation of the vertebrae after a dislocation injury-type. The device can be safely removed at this end-point based on the results and therefore the effects of degenerative changes if present could be minimized. Another concern would be the effect of biological loosening over time. According to Wolff’s law the bone resorption could occur at the bone-implant interface over time. During the course of the current study, no detrimental effect of loosening was observed at the end-point of eight weeks indicated by the kinematic analysis. Longer term studies might be appropriate to investigate the effect of time on the device loosening. 78 5 CONCLUSIONS AND FUTURE WORK 5.1 Conclusions A novel spinal fixation device for rats was designed and its performance in providing biomechanical fixation to an injured spine confirmed that the device provides stabilization of the cervical spine post-injury, both acutely and after eight weeks. No significant difference was observed in the spine stabilization when different levels of clamping force were applied during the device implantation. This shows that any inconsistency in the magnitude of the applied load within the evaluated range during the device implantation with this specific design would not affect the spine stabilization. 5.2 Contributions This device is a novel well-characterized fixation device for rodent studies of SCI. It is suitable for dorsal dislocation injury and allows long term survival studies. This device will both prevent further damage to the spinal cord and will stabilize the column. Potential benefits of this device could include a reduction in secondary damage and the volume of scar tissue in the spinal cord post-injury [22]. This study furthers fundamental understanding of SCI mechanics in survival studies and helps facilitate future studies establishing a link between injury specific observations and clinical treatment of human SCI. 5.3 Future Directions This spinal fixation device will be used in several survival studies to further investigate mechanisms of injury, characterize cord deformation using MRI, and study the effects of different treatment methods over time. The device design can be further modified for use in the distraction injury model. It can be also further customized for use in the thoracic or lumbar regions of the spine, as well as for other strains of rats or mice depending on their size and vertebral structure. 79 REFERENCES 1. Jackson, A. B., Dijkers, M., Devivo, M. J., Poczatek, R. B. , A demographic profile of new traumatic spinal cord injuries: Change and stability over 30 years. Archives of Physical Medicine and Rehabilitation, 2004. 85 (11): p. 1740-1748. 2. Sekhon, LH, Fehlings, MG, Epidemiology, demographics, and pathophysiology of acute spinal cord injury. Spine, 2001. 26(24 Suppl.): p. S2-12. 3. O'Connor, P.J., Trends in spinal cord injury. Accident; analysis and prevention, 2006. 38(1): p. 71-7. 4. Pickett, G E, Campos-Benitez, M, Keller, JL, Duggal, N, Epidemiology of traumatic spinal cord injury in canada. Spine, 2006. 31(7): p. 799-805. 5. Canadian Paraplegic Association. Canadian facts on spinal cord injury. 2010 5/13/2010]; Available from: http://www.canparaplegic.org/en/SCI_Facts_67.html?backfromtrail=true. 6. Schneider, RC, Cherry, G, Pantek, H. , The syndrome of acute central cervical spinal injury: with special reference to the mechanisms involved in hyperextension injuries of the cervical spine. Neurosurgery, 1954. 11: p. 546–577. 7. Tator, C. H., Spine-spinal cord relationships in spinal cord trauma. Clinical Neurosurgery, 1983. 30: p. 479-494. 8. Mayer, PR., Spinal Cord Injury, W.B. Saunders, in Diagnosis and Management of Disorders of the Spinal Cord., Young RR and Woolsey, Editor. 1995: Philadelphia. p. 103-104. 9. A.A. White, M.M. Panjabi, Clinical biomechanics of the spine. 2nd ed. 1990, Philadelphia: Lippincott Williams & Wilkins. 10. Frank P. Castro, Jr, James Ricciardi, Michael E. Brunet, Michael T. Busch, Thomas S. Whitecloud, Stingers, the Torg Ratio, and the Cervical Spine. The American Journal of Sports Medicine, 1997. 25(603). 11. Yue WM, Tan SB, Tan MH, Koh DC, Tan CT, The Torg-Pavlov ratio in cervical spondylotic myelopathy: a comparative study between patients with cervical spondylotic 80 myelopathy and a nonspondylotic, nonmyelopathic population. Spine (Phila Pa 1976), 2001. 26(16): p. 1760-4. 12. Tator, C. H., Fehlings, M. G., Review of the secondary injury theory of acute spinal cord trauma with emphasis on vascular mechanisms. Journal of Neurosurgery: Spine, 1991. 75(1): p. 15-26. 13. Stokes, BT, Jakeman, LB, Experimental modelling of human spinal cord injury: A model that crosses the species barrier and mimics the spectrum of human cytopathology. Spinal Cord : The Official Journal of the International Medical Society of Paraplegia, 2002. 40(3): p. 101-109. 14. Kwon, B., Tetzlaff, W., Grauer, J.N., Beiner, J., Vaccaro, A. R., Pathophysiology and pharmacologic treatment of acute spinal cord injury. The Spine Journal, 2004 4: p. 451– 464. 15. Norenberg, M.D., Smith, J., Marcillo, A. , The pathology of human spinal cord injury: Defining the problems. Journal of Neurotrauma, 2004. 21(4): p. 429-440. 16. Kakulas, B. A. , Pathology of spinal injuries. Central Nervous System Trauma : Journal of the American Paralysis Association, 1984. 1(2): p. 117-129. 17. Choo, A. M., Clinically Relevant Mechanisms of Spinal Cord Injury: Contusion, Dislocation, and Distraction, in Mechanical Engineering. 2007, The University Of British Columbia: Vancouver. p. 202. 18. Tator, C. H., Review of treatment trials in human spinal cord injury: Issues, difficulties, and recommendations. Neurosurgery, 2006. 59(5): p. 957-82; discussion 982-7. 19. Hawryluk, Gw, Rowland, Jw, Kwon, Bk, Fehlings, Mg, Protection and repair of the injured spinal cord: a review of completed, ongoing, and planned clinical trials for acute spinal cord injuries. Neurosurgical Focus, 2008. 25: p. E14. 20. Choo, A.M., Liu, J., Lam, C.K., Dvorak, M., Tetzlaff, W., Oxland, T.R., Contusion, Dislocation and Distraction – Primary Hemorrhage and Membrane Permeability in Distinct Mechanisms of Spinal Cord Injury. J.Neurosurg Spine 2007a. 6(3): p. 255-266. 81 21. Choo, A. M., Liu, J., Dvorak, M., Tetzlaff, W., Oxland, T. R. , Secondary pathology following contusion, dislocation, and distraction spinal cord injuries. Experimental Neurology, 2008. 212(2): p. 490-506. 22. Fei Liu, MD, Zhuo-Jin Luo, Si-Wei You, Xi-Ying Jiao, Xiao-Mei Meng, Ming Shi, Chun-Ting Wang, Gong Ju, Significance of Fixation of the Vertebral Column for Spinal Cord Injury Experiments. Spine, 2003. 28(15): p. 1666–1671. 23. Rooney, GE, Vaishya, S, Ameenuddin, S, Currier, BL, Schiefer, TK, Knight, A, Chen, B, Mishra, PK, Spinner, RJ, Macura, SI, Yaszemski, MJ, Windebank, AJ., Rigid Fixation of the Spinal Column Improves Scaffold Alignment and Prevents Scoliosis in the Transected Rat Spinal Cord. Spine, 2008. 33(24): p. E914–E919 24. Cheng, Henrich, Olson, Lars, A new surgical technique that allows proximodistal regeneration of 5-HT fibers after complete transection of the rat spinal cord. Experimental neurology, 1995. 136: p. 140-161. 25. Greene, E., Anatomy of the Rat. The American Philosophical Society, Philadelphia, 1935: p. 19,149. 26. Johnson, DR, McAndrew, TJ, Oguz, O., Shape differences in the cervical and upper thoracic vertebrae in rats (Rattus norvegicus) and bats (Pteropus poiocephalus): can we see shape patterns derived from position in column and species membership? Journal of Anatomy, 1999. 194(02): p. 249-253 27. Wingerd, B.D., Rat Dissection Manual. 1988, Baltimore: Johns Hopkins. vii-viii,9-10. 28. Rowett, H. , The Rat as a Small Mammal. third ed. ed. 1974, London: John Murray Ltd. 11-13. 29. Wells, T. , The Rat: A Practical Guide. 1964, London: Heinemann Educational Books Ltd. 10-14. 30. Howell, A. , Anatomy of the Wood Rat. 1926, Baltimore: The Williams & Wilkins Company. 124-127. 31. Hebel, R, Stromberg, Melvin W., Anatomy and Embryology of the Laboratory Rat. 1986, Wörthsee: BioMed Verlag. 17-18. 82 32. Olds, R J Olds; Joan R, A color atlas of the rat - Dissection guide. 1979, New York: Wiley. 112. 33. Hebel, Rudolf, M. W. Stromberg, Anatomy of the Laboratory Rat. 1976 Baltimore: Williams & Wilkins. 34. Chiasson, Robert B., Laboratory Anatomy of The White Rat Fifth ed. 1987: McGraw-Hill Science/Engineering/Math. 35. David G. Smith, Michael P. Schenk, Michael Schenk, A Dissection Guide and Atlas to the Rat. 2001, Englewood: Morton Publishing Company. 120. 36. Kent, George C., Comparative Anatomy of the Vertebrates. seventh ed. 1992: William C Brown Publishers. 681. 37. Tran CM, Markova D, Smith HE, Susarla B, Ponnappan RK, Anderson DG, Symes A, Shapiro IM, Risbud MV, Regulation of CCN2/connective tissue growth factor expression in the nucleus pulposus of the intervertebral disc: role of Smad and activator protein 1 signaling. Arthritis and rheumatism, 2010. 62(7): p. 1983-92. 38. Agrawal A, Guttapalli A, Narayan S, Albert TJ, Shapiro IM, Risbud MV., Normoxic stabilization of HIF-1alpha drives glycolytic metabolism and regulates aggrecan gene expression in nucleus pulposus cells of the rat intervertebral disk. American journal of physiology. Cell physiology, 2007. 293(2): p. C621-31. 39. Margareta Nordin, Victor H. Frankel, Basic biomechanics of the musculoskeletal system. Third ed. 2001: Lippincott Williams & Wilki. 500. 40. Charles Watson, George Paxinos, Gulgun Kayalioglu, The spinal cord: A Christopher and Dana Reeve Foundation Text and Atlas. 2008: Elsevier Ltd. 408. 41. Banks, WA, Kastin, AJ, Arimura, A., Effect of spinal cord injury on the permeability of the blood-brain and blood-spinal cord barriers to the neurotropin PACAP. Experimental neurology, 1998. 151(1): p. 116-23. 42. Sparrey, C., Choo, A., Liu, J., Tetzlaff, W., Oxland, T.R., The distribution of tissue damage in the spinal cord is influenced by the contusion velocity. Spine, 2008. 33(22): p. E812-E819. 83 43. Fiford, RJ, Bilston, LE, Waite, P, Lu, J., A vertebral dislocation model of spinal cord injury in rats. Journal of neurotrauma, 2004. 21(4): p. 451-8. 44. Scheff, SW, Rabchevsky, AG, Fugaccia, I, Main, JA, Lumpp, JE Jr., Experimental Modeling of Spinal Cord Injury: Characterization of a Force-Defined Injury Device. Journal of neurotrauma, 2003. 20(2): p. 179-93. 45. Dabney Kw, Ehrenshteyn M, Agresta CA, Twiss JL, Stern G, Tice L, Salzman SK, A Model of Experimental Spinal Cord Trauma Based on Computer-Controlled Intervertebral Distraction: Characterization of Graded Injury. Spine, 2004. 29(21): p. 2357-64. 46. Maiman, DJ, Myklebust, JB, Ho, KC, Coats, J., Experimental spinal cord injury produced by axial tension. Journal of spinal disorders, 1989 2(1): p. 6-13. 47. Elizabeth C. Clarke, Anthony M. Choo, Jie Liu, Clarrie K. Lam, Lynne E. Bilston, Wolfram Tetzlaff, Thomas R. Oxland, Anterior Fracture-Dislocation Is More Severe than Lateral: A Biomechanical and Neuropathological Comparison in Rat Thoracolumbar Spine. Journal of Neurotrauma, 2008. 25(4): p. 371-383. 48. Allen, A.R., Surgery of experimental lesion of spinal cord equivalent to crush injury of fracture dislocation of spinal column. A preliminary report. Journal of the American Medical Association, 1911. 57: p. 878–880. 49. Maikos, Jason T., Shreiber, David I., Immediate Damage to The Blood-Spinal Cord Barrier Due to Mechanical Trauma. Journal of Neurotrauma, 2007. 24(3): p. 492-507. 50. Young, W, Spinal cord contusion models. Progress in Brain Research, 2002. 137: p. 231- 255. 51. Ferraro, Armando, Experimental medullary concussion of the spinal cord in rabbits: Histologic study of the early stages. Arch Neurol Psychiatry, 1927. 18(3): p. 357-373. 52. Albin, MS, White, RJ, Acosta-Rua, G, Yashon, D., Study of Functional Recovery Produced by Delayed Localized Cooling After Spinal Cord Injury in Primate. Journal of Neurosurgery, 1968a. 29: p. 113-120. 53. Goodkin, R., Campbell, J.B., Sequential pathologic changes in spinal cord injury: A preliminary report. Surgical Forum, 1969. 20: p. 430–432. 84 54. Noble, L.J., Wrathal, J.R., spinal cord contusion in the rat: morphometric analyses of alterations in the spinal cord. Experimental neurology, 1985. 88: p. 135-149. 55. Wrathal, J.R., Pettegrew, R.K., Harvey, F., Spinal cord contusion in the rat: production of graded, reproducible, injury groups. Experimental neurology, 1985. 88: p. 108-122. 56. Kwo, S, Young, W, Decrescito, V, Spinal cord sodium, potassium, calcium, and water concentration changes in rats after graded contusion injury. journal of Neurotrauma, 1989. 6: p. 13-24. 57. Stokes, BT, Noyes, DH, Behrmann, DL, An electromechanical spinal injury technique with dynamic sensitivity. Journal of Neurotrauma, 1992. 9(3): p. 187-95. 58. Grill, R.J., User-defined variables that affect outcome in spinal cord contusion/compression models. Experimental neurology, 2005. 196: p. 1-5. 59. Gruner, J.A., Yee, A.K., Blight, A.R. , Histological anf functional evaluation of experimental spinal cord injury: evidence of a step-wise response to graded compression. Brain research, 1996. 729: p. 90-101. 60. Joshi, M, Fehlings, MG., Development and characterization of a novel, graded model of clip compressive spinal cord injury in the mouse: Part 1. Clip design, behavioral outcomes, and histopathology. Journal of Neurotrauma, 2002a. 19(2): p. 175-90. 61. Joshi, M, Fehlings, MG., Development and Characterization of a Novel, Graded Model of Clip Compressive Spinal Cord Injury in the Mouse: Part 2. Quantitative Neuroanatomical Assessment and Analysis of the Relationships between Axonal Tracts, Residual Tissue, and Locomotor Recovery. 2002b. 19(2): p. 191-203. 62. Vanicky, I., Urdzikova, L., Saganova, K., Cizkova, D., Galik, J., A simple and reproducible model of spinal cord injury induced by epidural balloon inflation in the rat. journal of Neurotrauma, 2001. 18: p. 1399-1407. 63. Choo, A.M., Liu, J., Liu, Z., Dvorak, M., Tetzlaff, W., Modeling spinal cord contusion, dislocation, and distraction: characterization of vertebral clamps, injury severities, and node of Ranvier deformations. Journal of neuroscience methods, 2009. 181(1): p. 6-17. 64. Taoka, Yuji, Okajima, Kenji, Spinal cord injury in the rat. Progress in Neurobiology, 1998. 56: p. 341-358. 85 65. Kiwerski, J., The influence of the mechanism of cervical spine injury on the degree of the spinal cord lesion. Paraplegia, 1991. 29(8): p. 531-6. 66. Silberstein, M, McLean, K., Non-contiguous spinal injury: clinical and imaging features, and postulated mechanism. Paraplegia, 1994. 32(12): p. 817-23. 67. Little, Jesse S., Khalsa, Partap S., Human Lumbar Spine Creep during Cyclic and Static Flexion: Creep Rate, Biomechanics, and Facet Joint Capsule Strain. Annals of Biomedical Engineering, 2005. 33(3): p. 391–401. 68. Yamamoto, I, Panjabi, MM, Crisco, T, Oxland, T. R., Three-dimensional movements of the whole lumbar spine and lumbosacral joint. Spine, 1989. 14(11): p. 1256-60. 69. Espinoza, Orías AA, Malhotra, NR, Elliott, DM. , Rat disc torsional mechanics: effect of lumbar and caudal levels and axial compression load. The Spine Journal, 2009. 9(3): p. 204-209. 70. Panjabi, M, Abumi, K, Duranceau, J, Oxland, T, Spinal stability and intersegmental muscle forces. A biomechanical model. Spine, 1989 14(2): p. 194-200. 71. Thomas R. Oxland, Ruey-Mo Lin, and Manohar M. Panjabi, Three-Dimensional Mechanical Properties of the Thoracolumbar Junction. Journal of Orthopaedrc Resear< h, 1992. 10: p. 573-580. 72. Greaves, Carolyn Y., Gadala, Mohamed S., Oxland, Thomas R., A Three-Dimensional Finite Element Model of the Cervical Spine with Spinal Cord: An Investigation of Three Injury Mechanisms. Annals of Biomedical Engineering, 2008. 36 (3): p. 396–405. 73. Maikos, Jason T., Qian, Zhen, Metaxas, Dimitris, Shreibe, David I., Finite Element Analysis of Spinal Cord Injury in the Rat. JOURNAL OF NEUROTRAUMA 2008. 25: p. 795–816. 74. Kimpara, H, Nakahira, Y, Iwamoto, M, Mik,i K, Ichihara, K, Kawano, S, Taguchi, T., Investigation of anteroposterior head-neck responses during severe frontal impacts using a brain-spinal cord complex FE model. Stapp car crash journal, 2006. 50: p. 509-44. 75. Berrocal, Y, Pearse, DD, Singh, A, Andrade, CM, McBroom, JS, Puentes, R, Eaton, MJ, Social and Environmental Enrichment Improves Sensory and Motor Recovery after 86 Severe Contusive Spinal Cord Injury in the Rat. JOURNAL OF NEUROTRAUMA, 2007. 24: p. 1761–1772. 76. Stackhouse, Scott K., Murray, Marion, Shumsky, Jed S., Effect of Cervical Dorsolateral Funiculotomy on Reach-to-Grasp Function in the Rat. JOURNAL OF NEUROTRAUMA 2008. 25: p. 1039–1047 77. Blanco, Jennifer E., Anderson, Kim D., Steward, Oswald, Recovery of forepaw gripping ability and reorganization of cortical motor control following cervical spinal cord injuries in mice. Experimental neurology, 2007. 203: p. 333–348. 78. Anderson, Kim D., Gunawan Ardi, Steward Oswald, Quantitative assessment of forelimb motor function after cervical spinal cord injury in rats: Relationship to the corticospinal tract. Experimental neurology, 2005. 194: p. 161– 174. 79. Anderson, Kim D., Abdul, Marim, Steward, Oswald, Quantitative assessment of deficits and recovery of forelimb motor function after cervical spinal cord injury in mice. experimental Neurology, 2004. 190: p. 184– 191. 80. Fischer, Florence R., Peduzzi, Jean D., Functional Recovery in Rats With Chronic Spinal Cord Injuries After Exposure to an Enriched Environment. Journal of Spinal Cord Medicine, 2007. 30: p. 147–155. 81. Kloos, Anne D., Fisher,Lesley C., Detloff, Megan R., Hassenzahl,David L., Basso, D. Michele, Stepwise motor and all-or-none sensory recovery is associated with nonlinear sparing after incremental spinal cord injury in rats. Experimental neurology, 2005. 191: p. 251– 265. 82. Anderson, Kim D., Gunawan, Ardi, Steward, Oswald, Spinal pathways involved in the control of forelimb motor function in rats. Experimental neurology, 2007. 206: p. 318– 331. 83. Winegar, CD, Lawrence, JP, Friel, BC, Fernandez, C, Hong, J, Maltenfort, M, Anderson, PA, Vaccaro, AR., A systematic review of occipital cervical fusion: techniques and outcomes. Journal of neurosurgery. Spine, 2010. 13(1): p. 5-16. 87 84. Lee YP, Robertson C, Mahar A, Kuper M, Lee DS, Regev GJ, Garfin SR., Biomechanical Evaluation of Transfacet Screw Fixation for Stabilization of Multilevel Cervical Corpectomies. Journal of spinal disorders & techniques, 2010. 85. Samir Lapsiwala, Edward Benzel, Surgical management of cervical myelopathy dealing with the cervical–thoracic junction. The Spine Journal, 2006. 6(6 Suppl.): p. 268S-273S. 86. Tasdemiroglu, E, Tibbs, PA, Long-term follow-up results of thoracolumbar fractures after posterior instrumentation. Spine, 1995. 20: p. 1704–8. 87. Sengupta, Dilip K, Mehdian, SMH, Mulholland, Robert C, Webb, John K, Ohnmeiss, Donna D, Biomechanical evaluation of immediate stability with rectangular versus cylindrical interbody cages in stabilization of the lumbar spine. BMC Musculoskeletal Disorders, 2002. 3(23). 88. Kandziora, Frank; Kerschbaumer, Fridun; Starker, Michael; Mittlmeier, Thomas, Biomechanical Assessment of Transoral Plate Fixation for Atlantoaxial Instability. Spine, 2000 25 (12): p. 1555-1561. 89. Nassos, JT, Ghanayem, AJ, Sasso, RC, Tzermiadianos, MN, Voronov, LI, Havey, RM, Rinella, AS, Carandang, G, Patwardhan, AG., Biomechanical Evaluation of Segmental Occipitoatlantoaxial Stabilization Techniques. SPINE, 2009. 34(25): p. 2740–2744. 90. Rathonyi, GC, Oxland, TR, Gerich, U, Grassmann, S, Nolte, LP, The role of supplemental translaminar screws in anterior lumbar interbody fixation: a biomechanical study. European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society, 1998. 7: p. 400-407. 91. Kelly, MP, Mok, JM, Berven, S., Dynamic Constructs for Spinal Fusion: An Evidence- Based Review. Orthopedic Clinics of North America, 2010. 41: p. 203–215. 92. Cheng, Boyle C., Burns, Paul, Pirris,Stephen, Welch, William C., Load Sharing and Stabilization Effects of Anterior Cervical Devices. Journal of Spinal Disorders & Techniques, 2009. 22: p. 571–577. 88 93. O'Brien, JR, Dmitriev, AE, Yu, W, Gelb, D, Ludwig, S., Posterior-only Stabilization of 2-column and 3-column Injuries at the Cervicothoracic Junction A Biomechanical Study. Journal of Spinal Disorders & Techniques, 2009. 22: p. 340–346. 94. Galbusera, F, Bellini, CM, Costa, F, Assietti, R, Fornari, M., Anterior cervical fusion: a biomechanical comparison of 4 techniques. Journal of Neurosurgery: Spine, 2008. 9: p. 444–449. 95. S.D. Boden, J.H. Schimandle, W.C. Hutton, An experimental lumbar intertransverse process spinal fusion model. Radiographic, histologic, and biomechanical healing characteristics. Spine, 1995. 20(4): p. 412–420. 96. Ali Riza Gezici, Rüçhan Ergün, Kamil Gürel, Fahri Yilmaz, Önder Okay, Ömer Bozdoğan, The Effect of Risedronate on Posterior Lateral Spinal Fusion in a Rat Model. Journal of Korean Neurosurgical Society, 2009. 46(1): p. 45-51. 97. Feiertag, M.A., Boden, S.D., Schimandle, J.H., Norman, J.T., A rabbit model for nonunion of lumbar intertransverse process spine arthrodesis. Spine, 1996 21(1): p. 27- 31 98. Steinmetz, Michael P, Benzel, Edward C, Apfelbaum, Ronald I, Axially Dynamic Implants for Stabilization of the Cervical Spine. Neurosurgery, 2006. 59(4): p. 378-389. 99. Greene, DL, Crawford, NR, Chamberlain, RH, Park, SC, Crandall, D Biomechanical comparison of cervical interbody cage versus structural bone graft. Spine Journal 2003. 3: p. 262–269. 100. Pitzen, T, Lane, C, Goertzen, D, Dvorak, M, Fisher, C, Barbier, D, Steudel, WI, Oxland, TR, Anterior cervical plate fixation: biomechanical effectiveness as a function of posterior element injury. Journal of Neurosurgery, 2003. 1 Suppl.: p. 84-90. 101. Abumi K, Panjabi M, Duranceau J, Biomechanical evaluation of spinal fixation devices. Part III. Stability provided by six spinal fixation devices and interbody bone graft. Spine, 1989. 14(11): p. 1249-55. 102. Panjabi MM, Abumi K, Duranceau J, Crisco JJ., Biomechanical evaluation of spinal fixation devices: II. Stability provided by eight internal fixation devices. Spine, 1988. 13(10): p. 1135-40. 89 103. Reed Group. Discectomy: Definition. 2010 6/6/2010]; Available from: http://www.mdguidelines.com/discectomy/definition. 104. Panjabi, MM., The stabilizing system of the spine. Part II. Neutral zone and instability hypothesis. Journal of spinal disorders, 1992 5(4): p. 390-6; discussion 397. 105. Liu, F, Luo, ZJ, You, SW, Jiao, XY, Meng, XM, Shi, M, Wang, CT, Ju, G. , Significance of Fixation of the Vertebral Column for Spinal Cord Injury Experiments. Spine, 2003. 28(15): p. 1666–1671. 106. Williams, David, On the mechanisms of biocompatibility. Biomaterials, 2008. 29(20): p. 2941-2953. 107. Williams, David, Revisiting the Definition of Biocompatibility. Medical Device Technology 2003. 14(8). 108. Department of Justice Canada. Medical Devices Regulations SOR/98-282. 1998 February 8th, 2010 5/13/2010]; Available from: http://www.emergogroup.com/files/Canada-Medical-Devices-Regulations-v2.pdf. 109. Matweb. Overview of materials for Polyetheretherketone, Unreinforced. 2010 august 13 2008; Available from: http://www.matweb.com/search/DataSheet.aspx?MatGUID=2164cacabcde4391a596640d 553b2ebe. 110. Zeus Industrial Products Inc. PEEK. 2010 May 13 2010]; Available from: http://www.zeusinc.com/extrusionservices/materials/peek.aspx. 111. Goertzen, Darrell J., Lane, Chris, Oxland, Thomas R., Neutral zone and range of motion in the spine are greater with stepwise loadingthan with a continuous loadingprotocol. An in vitro porcine investigation. Journal of Biomechanics, 2004. 37: p. 257–261. 112. Panjabi, M, Biomechanical evaluation of spinal fixation devices: I. A conceptual framework. Spine. Spine, 1988. 13(10): p. 1129-1134. 113. The Canadian Council on Animal Care, (CCAC). Choosing an appropriate endpoint in experiments using animals for research, teaching and testing. 1999 9/24/2010]; Available from: http://www.ccac.ca/en/Publications/PUBLICAT/RESOURCE/VOL222/Supart8222.htm. 90 114. Glantz, Stanton a., Primer of Biostatistics. 6 ed. 2005: McGRAW-HILL. 106-118. 115. Joseph Shigley, Charles Mischke, Richard Budynas, Joseph Shigley, Charles Mischke, Richard Budynas, Mechanical Engineering Design. Seventh ed. 2004, New York: McGraw-Hill. 116. Kandziora F, Pflugmacher R, Scholz M, Eindorf T, Schnake KJ, Haas NP, Bioabsorbable interbody cages in a sheep cervical spine fusion model. Spine, 2004. 29(17): p. 1845-55; discussion 1856. 117. Pflugmacher R, Schleicher P, Gumnior S, Turan O, Scholz M, Eindorf T, Haas NP, Kandziora F, Biomechanical comparison of bioabsorbable cervical spine interbody fusion cages. Spine, 2004. 29(16): p. 1717-22. 118. Montgomery, Douglas C., Chapter 3, in Design and Analysis of Experiments. 2005, John Wiley & Sons Inc. p. 80. 119. Montgomery, Douglas C., Chapter 3, in Design and Analysis of Experiments. 2005, John Wiley & Sons Inc. p. 75. 120. Maciejczak, A, Ciach, M, Radek, M Jr, Radek, A, Awrejcewicz, J., Immediate Stiffness of the C5–C6 Segment after Discectomy with the Cloward Technique: An in Vitro Biomechanical Study on a Human Cadaveric Model. Neurosurgery, 2001. 49: p. 1399– 1408. 121. Finn, MA, Brodke, DS, Daubs, M, Patel, A, Bachus, KN., Local and global subaxial cervical spine biomechanics after single-level fusion or cervical arthroplasty. European spine journal 2009. 18: p. 1520–1527. 122. Xu, HZ, Wang, XY, Chi, YL, Zhu, QA, Lin, Y, Huang, QS, Dai, LY, Biomechanical evaluation of a dynamic pedicle screw fixation device. Clinical biomechanics, 2006. 21(4): p. 330-336. 123. Gunzburg, R, Szpalski, M, Callary, SA, Colloca, CJ, Kosmopoulos, V, Harrison, D, Moore, RJ, Effect of a novel interspinous implant on lumbar spinal range of motion. European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society, 2009. 18(5): p. 696-703. 91 124. General Electric Healthcare. Neuroradiology: Cloward's procedure. 2010 7/9/2010]; Available from: http://www.medcyclopaedia.com/library/topics/volume_vi_1/c/clowards_procedure.aspx. 125. Erulkar, Jonathan S., Grauer, Jonathan N., Patel, Tushar C, Panjabi, Manohar M, Flexibility Analysis of Posterolateral Fusions in a New Zealand White Rabbit Model. Spine, 2001. 26(10): p. 1125-1130. 126. Medical Dictionary. Corpectomy: Definition. 2010 8/14/2010]. 127. Huang, P, Gupta, MC, Sarigul-Klijn, N, Hazelwood, S, Two in vivo surgical approaches for lumbar corpectomy using allograft and a metallic implant: a controlled clinical and biomechanical study. The spine journal : official journal of the North American Spine Society, 2006. 6(6): p. 648-58. 128. Toth, JM, Estes, BT, Wang, M, Seim, HB 3rd, Scifert, JL, Turner, AS, Cornwall, GB, Evaluation of 70/30 poly (L-lactide-co-D,L-lactide) for use as a resorbable interbody fusion cag. Journal of neurosurgery, 2002. 97(4 Suppl): p. 423-32. 129. Nabhan, A, Ishak, B, Steimer, O, Zimmer, A, Pitzen, T, Steudel, WI, Pape, D., Comparison of Bioresorbable and Titanium Plates in Cervical Spinal Fusion: Early Radiologic and Clinical Results. Journal of Spinal Disorders & Techniques, 2009. 22(3): p. 155-161. 130. Lin, HL, Cho, DY, Liu, YF, Lee, WY, Lee, HC, Chen, CC, Change of cervical balance following single to multi-level interbody fusion with cage. British journal of neurosurgery, 2008. 22(6): p. 758-63. 131. Laing, A.C., Cox, R., OXLAND, T. R., Tetzlaff, W., The effects of age on the mophometry and dgenrative state of the crvical sine in a rat model. Submitted to The Anatomical Record, 2010. 132. Cramer, GD, Fournier, JT, Henderson, CN, Wolcott, CC., Degenerative Changes Following Spinal Fixation in a Small Animal Model. Journal of manipulative and physiological therapeutics, 2004. 27(3): p. 141–154. 133. Basso, DM, Beattie, MS, Bresnahan, JC., A sensitive and reliable locomotor rating scale for open field testing in rats. Journal of neurotrauma, 1995. 12(1): p. 1-21. 92 APPENDICES Appendix A: The 21-Point Basso, Beattie, Bresnahan Locomotor Rating Scale and operational definitions of categories and attributes [133] 93 94 Appendix B: Histological Slides Of Pilot Study Figure B1 Spinal cord section of specimen 1 at 1.3 mm dislocation injury Figure B2 Spinal cord section of specimen 2 at 1.3 mm dislocation injury Figure B3 Spinal cord section of specimen 3 at 1.3 mm dislocation injury 95 Figure B4 Spinal cord section of specimen 1 at 1.5 mm dislocation injury Figure B5 Spinal cord section of specimen 2 at 1.5 mm dislocation injury Figure B6 Spinal cord section of specimen 3 at 1.5 mm dislocation injury 96 Figure B7 Spinal cord section of specimen 4 at 1.5 mm dislocation injury Figure B8 Spinal cord section of specimen 5 at 1.5 mm dislocation injury 97 Appendix C: Force-Displacement Curves For Injured Specimens Figure C1 Force-displacement curve of dislocation injury for specimen (1) of injured/8-week group Figure C2 Force-displacement curve of dislocation injury for specimen (2) of injured/8-week group 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 98 Figure C3 Force-displacement curve of dislocation injury for specimen (3) of injured/8-week group Figure C4 Force-displacement curve of dislocation injury for specimen (4) of injured/8-week group 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 99 Figure C5 Force-displacement curve of dislocation injury for specimen (5) of injured/8-week group Figure C6 Force-displacement curve of dislocation injury for specimen (6) of injured/8-week group 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 100 Figure C7 Force-displacement curve of dislocation injury for specimen (7) of injured/8-week group Figure C8 Force-displacement curve of dislocation injury for specimen (1) of injured/fixed group 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 101 Figure C9 Force-displacement curve of dislocation injury for specimen (2) of injured/fixed group Figure C10 Force-displacement curve of dislocation injury for specimen (3) of injured/fixed group 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 102 Figure C11 Force-displacement curve of dislocation injury for specimen (4) of injured/fixed group Figure C12 Force-displacement curve of dislocation injury for specimen (5) of injured/fixed group 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 103 Figure C13 Force-displacement curve of dislocation injury for specimen (6) of injured/fixed group 0 2 4 6 8 10 12 14 16 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 F o rc e ( N ) Displacement (mm) 104 Appendix D: Rotation Versus Applied Moment Plots Of Third Cycle For Individual Specimens A) Intact Group Figure D1 Rotation-moment curve at C5/C6 joint for intact specimen (1) Figure D2 Rotation-moment curve at C5/C6 joint for intact specimen (2) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) 105 Figure D3 Rotation-moment curve at C5/C6 joint for intact Specimen (3) Figure D4 Rotation-moment curve at C5/C6 joint for intact specimen (4) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) 106 Figure D5 Rotation-moment curve at C5/C6 joint for intact specimen (5) Figure D6 Rotation-moment curve at C5/C6 joint for intact specimen (6) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) 107 B) Injured/8-week Group with Device Removed Figure D7 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (1) with device removed Figure D8 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (2) with device removed -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) 108 Figure D9 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (3) with device removed Figure D10 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (4) with device removed -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) 109 Figure D11 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (5) with device removed Figure D12 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (6) with device removed -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) 110 Figure D13 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (7) with device removed C) Injured/8-week Group with Device Figure D14 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (1) with device -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) 111 Figure D15 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (2) with device Figure D16 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (3) with device -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) 112 Figure D17 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (4) with device Figure D18 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (5) with device -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) 113 Figure D19 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (6) with device Figure D20 Rotation-moment curve at C5/C6 joint for injured/8-week specimen (7) with device -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) 114 D) Injured/Fixed Group At High, Medium, and Low Clamping Forces Figure D21 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (1) with high, medium, and low clamping forces Figure D22 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (2) with high, medium, and low clamping forces -25 -20 -15 -10 -5 0 5 10 15 20 25 -10 -5 0 5 10 A n g le ( D e g re e s ) Moment (Nmm) High F Med F Low F -25 -20 -15 -10 -5 0 5 10 15 20 25 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) High F Med F Low F 115 Figure D23 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (3) with high, medium, and low clamping forces Figure D24 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (4) with high, medium, and low clamping forces -25 -20 -15 -10 -5 0 5 10 15 20 25 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) High F med F Low F -25 -20 -15 -10 -5 0 5 10 15 20 25 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) High F Med F Low F 116 Figure D25 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (5) with high, medium, and low clamping forces Figure D26 Rotation-moment curve at C5/C6 joint for injured/fixed specimen (6) with high, medium, and low clamping forces -25 -20 -15 -10 -5 0 5 10 15 20 25 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) High F Med F Low F -25 -20 -15 -10 -5 0 5 10 15 20 25 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) High F Med F Low F 117 E) Injured/3-week Group with Device Figure D27 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (1) with device Figure D28 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (2) with device -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) 118 Figure D29 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (3) with device F) Injured/3-week Group with Device Removed Figure D30 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (1) with device removed -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) 119 Figure D31 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (2) with device removed Figure D32 Rotation-moment curve at C5/C6 joint for injured/3-week specimen (3) with device removed -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -20 -15 -10 -5 0 5 10 15 20 A n g le ( D e g re e s ) Moment (Nmm) -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 -15 -10 -5 0 5 10 15 A n g le ( D e g re e s ) Moment (Nmm) 120 Appendix E: Matlab Script For Calibration Of The Instrumented Forceps – Data Acquisition To Obtain The Relationship Between The Clamping Force And The Strain %a=pwd; %msgbox('Please choose the time column from you load data file'); %time = xlsread('load.csv', -1); %msgbox('Please choose the load column from you load data file'); %loadval = xlsread('load.csv', -1); allVal = xlsread('load.csv'); time = allVal(:, 1); loadval = allVal(:, 6); strain = textread('strain.txt', '%f', -1); interval = 0.000620; %t = size(strain); %strain_time = zeros(t(1, 1)); %strain_time(1, 1) = 0; %for i=2:t(1, 1) % strain_time(i, 1) = strain_time(i - 1, 1) + interval; %end %plot(loadval); %figure; %plot(strain); load_index = 0; flag = 1; minDif = 0.4; while (flag) load_index = load_index + 1; dif1 = loadval(load_index, 1) - loadval(load_index + 1, 1); dif2 = loadval(load_index + 1, 1) - loadval(load_index + 2, 1); dif3 = loadval(load_index + 2, 1) - loadval(load_index + 3, 1); if (dif1 >= minDif && dif2 >= minDif && dif3 >= minDif) flag = 0; end end fprintf(1, '%f\n', loadval(load_index, 1)); a = size(strain); strain_count = a(1, 1); samplerate = 10; sum = 0; counter = 0; n = ceil(strain_count / samplerate); new_strain = zeros(n, 1); for i = 1:strain_count, sum = strain(i, 1) + sum; 121 if (mod(i, samplerate) == 0) counter = counter + 1; new_strain(counter, 1) = sum / samplerate; sum = 0; end end if (mod(strain_count, samplerate) ~= 0) new_strain(counter + 1, 1) = sum / mod(strain_counter, samplerate); end save('temp.txt', 'new_strain', '-ASCII'); t = size(new_strain); %strain_time = zeros(t(1, 1)); %strain_time(1, 1) = 0; %for i=2:t(1, 1) % strain_time(i, 1) = strain_time(i - 1, 1) + interval; %end plot(strain, 'k'); %xlim([0, t(1, 1) * interval]); figure; strain_index = 0; flag = 1; minDif = 1.00000e-7; while (flag) strain_index = strain_index + 1; dif1 = new_strain(strain_index + 1, 1) - new_strain(strain_index, 1); dif2 = new_strain(strain_index + 2, 1) - new_strain(strain_index + 1, 1); dif3 = new_strain(strain_index + 3, 1) - new_strain(strain_index + 2, 1); dif4 = new_strain(strain_index + 4, 1) - new_strain(strain_index + 3, 1); if (dif1 >= minDif && dif2 >= minDif && dif3 >= minDif && dif4 >= minDif) flag = 0; end end interval = interval * samplerate; min_index = min(load_index, strain_index); a = size(loadval); load_count = a(1, 1); strainval = zeros(load_count, 1); for i = 1:load_index-1, strainval(i, 1) = new_strain(mod(i, min_index) + 1, 1); end %a = load_index + 1; base_time = time(load_index, 1); for i = load_index:load_count, time_diff = time(i, 1) - base_time; offset = floor(time_diff/interval); if (strain_index + offset > n) 122 strainval(i, 1) = new_strain(n, 1); else strainval(i, 1) = new_strain(strain_index + offset, 1); end end plot(time, loadval, 'r'); figure; plot(time, strainval, 'b'); save('new_strain.txt', 'strainval', '-ASCII'); l fprintf(1, '%f\n', new_strain(strain_index, 1)); 123 Appendix F: Matlab Script For Clamping Force Data Acquisition function varargout = untitled(varargin) % UNTITLED M-file for untitled.fig % UNTITLED, by itself, creates a new UNTITLED or raises the existing % singleton*. % % H = UNTITLED returns the handle to a new UNTITLED or the handle to % the existing singleton*. % % UNTITLED('CALLBACK',hObject,eventData,handles,...) calls the local % function named CALLBACK in UNTITLED.M with the given input arguments. % % UNTITLED('Property','Value',...) creates a new UNTITLED or raises the % existing singleton*. Starting from the left, property value pairs are % applied to the GUI before untitled_OpeningFcn gets called. An % unrecognized property name or invalid value makes property application % stop. All inputs are passed to untitled_OpeningFcn via varargin. % % *See GUI Options on GUIDE's Tools menu. Choose "GUI allows only one % instance to run (singleton)". % % See also: GUIDE, GUIDATA, GUIHANDLES % Edit the above text to modify the response to help untitled % Last Modified by GUIDE v2.5 24-Jan-2010 19:38:22 % Begin initialization code - DO NOT EDIT gui_Singleton = 1; gui_State = struct('gui_Name', mfilename, ... 'gui_Singleton', gui_Singleton, ... 'gui_OpeningFcn', @untitled_OpeningFcn, ... 'gui_OutputFcn', @untitled_OutputFcn, ... 'gui_LayoutFcn', [] , ... 'gui_Callback', []); if nargin && ischar(varargin{1}) gui_State.gui_Callback = str2funcdislocation); end if nargout [varargout4] = gui_mainfcn[4]); else gui_mainfcn(gui_State, varargin{:}); end % End initialization code - DO NOT EDIT % --- Executes just before untitled is made visible. function untitled_OpeningFcn(hObject, eventdata, handles, varargin) % This function has no output args, see OutputFcn. % hObject handle to figure % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % varargin command line arguments to untitled (see VARARGIN) % Choose default command line output for untitled handles.output = hObject; 124 % Update handles structure guidata(hObject, handles); % UIWAIT makes untitled wait for user response (see UIRESUME) % uiwait(handles.figure1); clc; %Check if the paths to 'nilibddc.dll' and 'nilibddc_m.h' have been %selected. If not, prompt the user to browse to each of the files. global theFirstTime; theFirstTime = 1; loadData(handles); % --- Outputs from this function are returned to the command line. function varargout = untitled_OutputFcn(hObject, eventdata, handles) % varargout cell array for returning output args (see VARARGOUT); % hObject handle to figure % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Get default command line output from handles structure varargout{1} = handles.output; function edit1_Callback(hObject, eventdata, handles) % hObject handle to edit1 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Hints: get(hObject,'String') returns contents of edit1 as text % str2double(get(hObject,'String')) returns contents of edit1 as a double % --- Executes during object creation, after setting all properties. function edit1_CreateFcn(hObject, eventdata, handles) % hObject handle to edit1 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles empty - handles not created until after all CreateFcns called % Hint: edit controls usually have a white background on Windows. % See ISPC and COMPUTER. if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor')) set(hObject,'BackgroundColor','white'); end function edit2_Callback(hObject, eventdata, handles) % hObject handle to edit2 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Hints: get(hObject,'String') returns contents of edit2 as text % str2double(get(hObject,'String')) returns contents of edit2 as a double % --- Executes during object creation, after setting all properties. function edit2_CreateFcn(hObject, eventdata, handles) % hObject handle to edit2 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles empty - handles not created until after all CreateFcns called 125 % Hint: edit controls usually have a white background on Windows. % See ISPC and COMPUTER. if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor')) set(hObject,'BackgroundColor','white'); end function edit3_Callback(hObject, eventdata, handles) % hObject handle to edit3 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Hints: get(hObject,'String') returns contents of edit3 as text % str2double(get(hObject,'String')) returns contents of edit3 as a double % --- Executes during object creation, after setting all properties. function edit3_CreateFcn(hObject, eventdata, handles) % hObject handle to edit3 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles empty - handles not created until after all CreateFcns called % Hint: edit controls usually have a white background on Windows. % See ISPC and COMPUTER. if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor')) set(hObject,'BackgroundColor','white'); end % --- Executes on button press in pushbutton1. function pushbutton1_Callback(hObject, eventdata, handles) % hObject handle to pushbutton1 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) global theFirstTime; if (theFirstTime == 1) theFirstTime = 0; else loadData(handles); end global currentfolder; global mat; global convertedMat; global baseline; global factor1 ; global constFact; global minVal; global maxVal; minmax = get(handles.radiobutton1, 'Value'); if minmax == 1 upperVal = str2num(get(handles.edit1, 'String')); lowerVal = str2num(get(handles.edit2, 'String')); 126 else factor = str2num(get(handles.edit3, 'String')); upperVal = maxVal + 0.05 * abs(maxVal); lowerVal = maxVal - (maxVal - minVal) * factor; end sum = 0; count = 0; m = size(mat); firstIndex = -1; lastIndex = -1; for k=1:m(1,1) if mat(k,1) >= lowerVal && mat(k, 1) <= upperVal if firstIndex == -1 firstIndex = k; end if k > lastIndex lastIndex = k; end sum = sum + mat(k,1); count = count + 1; end end average = sum / count; set(handles.averageEdit, 'String', num2str(average)); set(handles.averageForceEdit, 'String', num2str(average * factor1 + constFact)); avgVec = zeros(m(1,1), 1); avgForceVec = zeros(m(1,1), 1); for k=1:m(1, 1) if k >= firstIndex && k <= lastIndex avgVec(k, 1) = average; avgForceVec(k, 1) = factor1 * average + constFact; else avgVec(k, 1) = baseline; avgForceVec(k, 1) = factor1 * baseline + constFact; end end xx = 0:m(1,1)-1; %if (ptype.Value == 10) %Plot Data from channels in this group %fid = fopen('result.txt', 'at'); %fprintf(fid, 'Group# %f\n\n', i); %fprintf(fid, '%f\n', mat); %fclose(fid); figure; hold on; xlabel('Time (ns)'); ylabel('Strain'); plot(xx, mat, 'b',xx, avgVec, 'r');clear mat; clear avgVec; 127 figure; hold on; xlabel('Time (ns)'); ylabel('Force (N)'); plot(xx, convertedMat, 'k', xx, avgForceVec, 'r'); clear convertedMat; clear avgForceVec; %end %legend(channames); % --- Executes on key press with focus on radiobutton1 and none of its controls. function radiobutton1_KeyPressFcn(hObject, eventdata, handles) % hObject handle to radiobutton1 (see GCBO) % eventdata structure with the following fields (see UICONTROL) % Key: name of the key that was pressed, in lower case % Character: character interpretation of the key(s) that was pressed % Modifier: name(s) of the modifier key(s) (i.e., control, shift) pressed % handles structure with handles and user data (see GUIDATA) % --- Executes during object creation, after setting all properties. function radiobutton2_CreateFcn(hObject, eventdata, handles) % hObject handle to radiobutton2 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles empty - handles not created until after all CreateFcns called % --- Executes on key press with focus on radiobutton2 and none of its controls. function radiobutton2_KeyPressFcn(hObject, eventdata, handles) % hObject handle to radiobutton2 (see GCBO) % eventdata structure with the following fields (see UICONTROL) % Key: name of the key that was pressed, in lower case % Character: character interpretation of the key(s) that was pressed % Modifier: name(s) of the modifier key(s) (i.e., control, shift) pressed % handles structure with handles and user data (see GUIDATA) % --- If Enable == 'on', executes on mouse press in 5 pixel border. % --- Otherwise, executes on mouse press in 5 pixel border or over radiobutton2. function radiobutton2_ButtonDownFcn(hObject, eventdata, handles) % hObject handle to radiobutton2 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % --- Executes on button press in radiobutton1. function radiobutton1_Callback(hObject, eventdata, handles) % hObject handle to radiobutton1 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Hint: get(hObject,'Value') returns toggle state of radiobutton1 set(handles.edit3, 'Enable', 'off'); set(handles.edit1, 'Enable', 'on'); set(handles.edit2, 'Enable', 'on'); % --- Executes on button press in radiobutton2. function radiobutton2_Callback(hObject, eventdata, handles) % hObject handle to radiobutton2 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) 128 % Hint: get(hObject,'Value') returns toggle state of radiobutton2 set(handles.edit1, 'Enable', 'off'); set(handles.edit2, 'Enable', 'off'); set(handles.edit3, 'Enable', 'on'); function averageEdit_Callback(hObject, eventdata, handles) % hObject handle to averageEdit (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Hints: get(hObject,'String') returns contents of averageEdit as text % str2double(get(hObject,'String')) returns contents of averageEdit as a double % --- Executes during object creation, after setting all properties. function averageEdit_CreateFcn(hObject, eventdata, handles) % hObject handle to averageEdit (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles empty - handles not created until after all CreateFcns called % Hint: edit controls usually have a white background on Windows. % See ISPC and COMPUTER. if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor')) set(hObject,'BackgroundColor','white'); end function loadData(handles) global currentfolder; pwd; currentfolder = pwd; %if exist('NI_TDM_DLL_Path','var')==0 % [dllfile,dllfolder]=uigetfile('*dll','Select nilibddc.dll'); % NI_TDM_DLL_Path=fullfile(dllfolder,dllfile); %end %if exist('NI_TDM_H_Path','var')==0 % [hfile,hfolder]=uigetfile('*h','Select nilibddc_m.h'); % NI_TDM_H_Path=fullfile(hfolder,hfile); %end %Prompt the user to browse to the path of the TDM or TDMS file to read. [filepath,filefolder]=uigetfile('*.tdms','Select TDM or TDMS File'); Data_Path=fullfile(filefolder,filepath); [pathstr, name, ext, versn] = fileparts(Data_Path); %Apply the appropriate file type to 'ftype', depending on whether a TDM or %TDMS file was selected. Here, 'ftype' is used in the 'DDC_OpenFileEx' function. if strcmp(ext,'.tdms') ftype='TDMS'; else ftype='TDM'; end %Load nilibddc.dll (Always call 'unloadlibrary('nilibddc')' after finished using %the library.) NI_TDM_DLL_Path = fullfile(currentfolder, 'dev', 'bin', 'nilibddc.dll'); NI_TDM_H_Path = fullfile(currentfolder, 'dev', 'include', 'nilibddc_m.h'); loadlibrary(NI_TDM_DLL_Path,NI_TDM_H_Path); 129 %Open the TDM or TDMS file (Read Only) (Always call 'DDC_CloseFile' after finished creating or %reading a file.) pfile = libpointer('int32Ptr', 0); calllib('nilibddc','DDC_OpenFileEx',Data_Path,ftype,1,pfile); %Read and Display File Name DDC_FILE_NAME=libpointer('stringPtr','name'); pfilenamelen=libpointer('uint32Ptr',0); %Get the length of the 'DDC_FILE_NAME' string property err=calllib('nilibddc','DDC_GetFileStringPropertyLength',pfile.Value,DDC_FILE_NAME,pfilenamelen); if err==0 %Only proceed if File Name is found %Initialize a string of the length of the file name pfilename=libpointer('stringPtr',blanks(pfilenamelen.Value)); calllib('nilibddc','DDC_GetFileProperty',pfile.Value,DDC_FILE_NAME,pfilename,pfilenamelen.Value+1); disp(['File Name: ' pfilename.Value]); end %Read and Display File Description DDC_FILE_DESCRIPTION=libpointer('stringPtr','description'); pfiledesclen=libpointer('uint32Ptr',0); %Get the length of the 'DDC_FILE_DESCRIPTION' string property err=calllib('nilibddc','DDC_GetFileStringPropertyLength',pfile.Value,DDC_FILE_DESCRIPTION,pfiledesclen); if err==0 %Only proceed if File Description is found %Initialize a string of the length of the file description pfiledesc=libpointer('stringPtr',blanks(pfiledesclen.Value)); calllib('nilibddc','DDC_GetFileProperty',pfile.Value,DDC_FILE_DESCRIPTION,pfiledesc,pfiledesclen.Value+1); disp(['File Description: ' pfiledesc.Value]); end %Read and Display File Title DDC_FILE_TITLE=libpointer('stringPtr','title'); pfiletitlelen=libpointer('uint32Ptr',0); %Get the length of the 'DDC_FILE_TITLE' string property err=calllib('nilibddc','DDC_GetFileStringPropertyLength',pfile.Value,DDC_FILE_TITLE',pfiletitlelen); if err==0 %Only proceed if File Title is found %Initialize a string of the length of the file title pfiletitle=libpointer('stringPtr',blanks(pfiletitlelen.Value)); calllib('nilibddc','DDC_GetFileProperty',pfile.Value,DDC_FILE_TITLE,pfiletitle,pfiletitlelen.Value+1); disp(['File Title: ' pfiletitle.Value]); end %Read and Display File Author DDC_FILE_AUTHOR=libpointer('stringPtr','author'); pfileauthlen=libpointer('uint32Ptr',0); %Get the length of the 'DDC_FILE_AUTHOR' string property err=calllib('nilibddc','DDC_GetFileStringPropertyLength',pfile.Value,DDC_FILE_AUTHOR,pfileauthlen); if err==0 %Only proceed if File Author is found %Initialize a string of the length of the file author pfileauth=libpointer('stringPtr',blanks(pfileauthlen.Value)); calllib('nilibddc','DDC_GetFileProperty',pfile.Value,DDC_FILE_AUTHOR,pfileauth,pfileauthlen.Value+1); disp(['File Author: ' pfileauth.Value]); end %Read and Display File Timestamp DDC_FILE_DATETIME=libpointer('stringPtr','datetime'); 130 pyear=libpointer('uint32Ptr',0); pmonth=libpointer('uint32Ptr',0); pday=libpointer('uint32Ptr',0); phour=libpointer('uint32Ptr',0); pminute=libpointer('uint32Ptr',0); psecond=libpointer('uint32Ptr',0); pmsecond=libpointer('doublePtr',0); pwkday=libpointer('uint32Ptr',0); err=calllib('nilibddc','DDC_GetFilePropertyTimestampComponents',pfile.Value,DDC_FILE_DATETIME,pyear,pm onth,pday,phour,pminute,psecond,pmsecond,pwkday); if err==0 %Only proceed if File Timestamp is found disp(['File Timestamp: ' num2str(pmonth.Value) '/' num2str(pday.Value) '/' num2str(pyear.Value) ', ' num2str(phour.Value) ':' num2str(pminute.Value) ':' num2str(psecond.Value) ':' num2str(pmsecond.Value)]); end %Get Channel Groups DDC_CHANNELGROUP_NAME=libpointer('stringPtr','name'); DDC_CHANNELGROUP_DESCRIPTION=libpointer('stringPtr','description'); DDC_CHANNEL_NAME=libpointer('stringPtr','name'); %Get the number of Channel Groups pnumgrps=libpointer('uint32Ptr',0); calllib('nilibddc','DDC_GetNumChannelGroups',pfile.Value,pnumgrps); %Get Channel Groups only if the number of Channel Groups is greater than %zero if pnumgrps.Value>0 pgrps=libpointer('int32Ptr',zeros(1,pnumgrps.Value)); calllib('nilibddc','DDC_GetChannelGroups',pfile.Value,pgrps,pnumgrps.Value); end %fid = fopen('result.txt', 'wt'); %fclose(fid); %for i=1:pnumgrps.Value %For each Channel Group for i=1:1 %Get Channel Group Name pgrpnamelen=libpointer('uint32Ptr',0); err=calllib('nilibddc','DDC_GetChannelGroupStringPropertyLength',pgrps.Value(i),DDC_CHANNELGROUP_NA ME,pgrpnamelen); if err==0 %Only proceed if Channel Group Name is found pgrpname=libpointer('stringPtr',blanks(pgrpnamelen.Value)); calllib('nilibddc','DDC_GetChannelGroupProperty',pgrps.Value(i),DDC_CHANNELGROUP_NAME,pgrpname,pg rpnamelen.Value+1); else pgrpname=libpointer('stringPtr',''); end %Get Channel Group Description pgrpdesclen=libpointer('uint32Ptr',0); err=calllib('nilibddc','DDC_GetChannelGroupStringPropertyLength',pgrps.Value(i),DDC_CHANNELGROUP_DE SCRIPTION,pgrpdesclen); if err==0 %Only proceed if Channel Group Description is found pgrpdesc=libpointer('stringPtr',blanks(pgrpdesclen.Value)); 131 calllib('nilibddc','DDC_GetChannelGroupProperty',pgrps.Value(i),DDC_CHANNELGROUP_DESCRIPTION,pgrp desc,pgrpdesclen.Value+1); end figure('Name',pgrpname.Value); hold on; %Get Channels pnumchans=libpointer('uint32Ptr',0); %Get the number of Channels in this Channel Group calllib('nilibddc','DDC_GetNumChannels',pgrps.Value(i),pnumchans); %Get Channels only if the number of Channels is greater than zero if pnumchans.Value>0 pchans=libpointer('int32Ptr',zeros(1,pnumchans.Value)); calllib('nilibddc','DDC_GetChannels',pgrps.Value(i),pchans,pnumchans.Value); end channames=cell(1,pnumchans.Value); %for j=1:pnumchans.Value %For each Channel in group for j=1:1 %Get Channel Name pchannamelen=libpointer('uint32Ptr',0); err=calllib('nilibddc','DDC_GetChannelStringPropertyLength',pchans.Value(j),DDC_CHANNEL_NAME,pchanna melen); if err==0 %Only proceed if Channel Name is found pchanname=libpointer('stringPtr',blanks(pchannamelen.Value)); calllib('nilibddc','DDC_GetChannelProperty',pchans.Value(j),DDC_CHANNEL_NAME,pchanname,pchannamelen. Value+1); channames{j}=pchanname.Value; else channames{j}=''; end %Get Channel Data Type ptype=libpointer('voidPtr',uint8(0)); calllib('nilibddc','DDC_GetDataType',pchans.Value(j),ptype); %Get Channel Value if Data Type is 'Double'(10) if ptype.Value==10 pnumvals=libpointer('uint64Ptr',0); calllib('nilibddc','DDC_GetNumDataValues',pchans.Value(j),pnumvals); pvals=libpointer('doublePtr',zeros(1,pnumvals.Value)); calllib('nilibddc','DDC_GetDataValues',pchans.Value(j),0,pnumvals.Value,pvals); chanvals(:,j)=(pvals.Value)'; end end useMinMax = 0; global maxVal; global minVal; maxVal = -10000000; minVal = 10000000; 132 m = size(chanvals); global mat; global convertedMat; global factor1 ; factor1 = 76.997; global constFact; constFact = 100.39; mat = zeros(m(1,1), 1); convertedMat = zeros(m(1,1), 1); mat(1,1) = chanvals(1,1); convertedMat(1, 1) = factor1 * chanvals(1, 1) + constFact; aa = 0; for k=2:m(1,1) temp = chanvals(k, 1); if (temp > maxVal) maxVal = temp; end if (temp < minVal) minVal = temp; end mat(k, 1) = chanvals(k,1); convertedMat(k, 1) = temp * factor1 + constFact; end global baseline; baseline = chanvals(1,1); set(handles.edit1, 'String', num2str(minVal)); set(handles.edit2, 'String', num2str(maxVal)); end %Close TDM or TDMS file calllib('nilibddc','DDC_CloseFile',pfile.Value); %Unload nilibddc.dll unloadlibrary('nilibddc'); function averageForceEdit_Callback(hObject, eventdata, handles) % hObject handle to averageForceEdit (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Hints: get(hObject,'String') returns contents of averageForceEdit as text % str2double(get(hObject,'String')) returns contents of averageForceEdit as a double % --- Executes during object creation, after setting all properties. function averageForceEdit_CreateFcn(hObject, eventdata, handles) % hObject handle to averageForceEdit (see GCBO) if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor')) set(hObject,'BackgroundColor','white'); end 133 Appendix G: Drawings Of The Fixation Device 134 Appendix H: Animal Monitoring Sheets 135 136 Appendix I: Data Collection Sheet For An Individual Specimen Subjected To A Dislocation Injury Type Biomechanics of SCI Data Collection: Dislocation Date: UBC Machine Process Details Any observed noise in system (with jack-drive on/off): ______________________ Performed Warm-up cycle: Y N Blank Shot filename: _____________________________ Normal? Y N Dislocation Magnitude: __________ Experiment file name: _____________________________ Animal Details ID#: _________________ Mass (g): ______________________ Surgery Details: __________________________________________________________ ______________________________________________________________________________ __________________________________________________________________ Outcomes Max Displacement (mm): _________________ Max Force (N): ____________________ Preload magnitude: __________________ Notes on outcomes: _______________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ Pre Post Time Immed. Immed. After preload Immed. 30 s 1 min 5 min 10 min 20 min 30 min HR (bpm) SPO2 (%) Notes: __________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ 137 Appendix J: Ethics Board Certificates Of Approval Figure J1 Animal care certificate representing ethics approval for this study