KNEE BIOMECHANICS DURING IMPACT LANDING: UNDERSTANDING INJURY MECHANISMS AND DEVELOPING PREVENTION STRATEGIES YEOW CHEN HUA NATIONAL UNIVERSITY OF SINGAPORE 2009 KNEE BIOMECHANICS DURING IMPACT LANDING: UNDERSTANDING INJURY MECHANISMS AND DEVELOPING PREVENTION STRATEGIES YEOW CHEN HUA B.Eng.(Hons.), NUS A THESIS SUBMITTED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN BIOENGINEERING DIVISION OF BIOENGINEERING NATIONAL UNIVERSITY OF SINGAPORE 2009 TABLE OF CONTENTS CONTENTS PAGE Acknowledgments ix 1. SUMMARY x List of Tables xiii List of Figures xiv List of Abbreviations and Symbols xx 2. BACKGROUND 3. RATIONALE 24 3.1 Hypothesis 25 3.2 Objectives 25 3.3 Overview 29 4. METHODOLOGY 30 4.1 STAGE A - Understanding Biomechanics of Landing 4.1.1 Analysis of landing maneuvers performed by human subjects a. Subject recruitment, anthropometric measurement and marker placement 31 31 b. Motion capture and force plate systems 31 c. Types of landing protocols 32 d. Processing of kinematics, kinetics and energetics data 33 e. Statistical analysis 33 4.2 STAGE B – Investigation of Knee Injury Mechanisms 4.2.1 Developing a test platform for application of simulated landing impact on intact knee specimens 35 a. Specimen procurement and preparation 35 b. Experimental setup 36 c. Motion capture system and marker placement 39 d. Impact protocol 39 ii e. Data processing 4.2.2 Anterior cruciate ligament failure 40 41 a. Compressive force response 41 b. Magnetic resonance imaging 41 c. Dissection 41 4.2.3 Analysis of osteochondral damage in knee specimens 41 a. Explant extraction 41 b. Preparatory steps for general histology 42 c. Histological staining methods 42 d. Mankin scoring 43 e. Cartilage thickness measurement 43 f. Cartilage volume quantification 44 g. Statistical analysis 45 4.2.4 Finite element model of the tibiofemoral joint 45 a. Geometry reconstruction 45 b. Model mesh and material properties 46 c. Loading and boundary conditions 47 d. Validation of finite element model 48 4.2.5 Simulated landing impact to osteochondral explants with and without sustained compression 48 a. Explant extraction and preparation 48 b. Impact protocol (to assess degenerative changes) 48 c. Impact protocol (to assess effects of non-sustained and sustained compressions) 4.2.6 Analysis of osteochondral explant damage and degeneration 50 51 a. Cell viability assessment 51 b. Collagen and glycoaminoglycan quantification 52 c. Histology and Mankin scoring 52 d. Immunohistochemistry 53 e. Micro-computed tomography and cartilage volume quantification 53 iii f. Cartilage thickness measurement 54 g. Statistical analysis 54 4.3 STAGE C - Development of Injury Prevention Strategies 4.3.1 Restraints to prevent anterior cruciate ligament failure following impact 55 a. Specimen Procurement and Preparation 55 b. Type of restraints 55 c. Impact protocol 56 d. Processing of force and kinematics data 58 e. Damage assessment 58 f. Statistical analysis 59 4.3.2 Development of a knee protection device 60 a. Knee brace with anterior-sloped hinge 60 b. Shoe with carbon-fiber sole 62 4.3.3 Evaluation of the knee protection device under normal landing conditions a. Analysis of landing in unbraced and braced knee conditions b. Determination of anterior tibial translation and axial tibial rotation 63 63 64 c. Measurement of contact joint force of knee brace 65 d. Processing of kinematics and kinetics data 65 e. Statistical Analysis 65 4.3.4 Evaluation of the knee protection device under large simulated landing impact a. Analysis of cadaveric knee specimens in unbraced and braced knee conditions 66 66 b. Processing of compressive force and kinematics data 67 c. Damage assessment 68 d. Statistical analysis 68 5. RESULTS 69 5.1 STAGE A - Understanding Biomechanics of Landing iv 5.1.1 Regression relationships of knee kinematics and kinetics with landing height a. Time profiles of knee kinematics, kinetics and energetics 70 70 b. Peak ground reaction force 71 c. Knee flexion angle 71 d. Knee flexion angular velocity 72 e. Knee joint power 73 f. Regression relationships 73 5.1.2 Effect of landing height and technique on sagittal knee joint kinematics and energetics 75 a. Time profiles 75 b. Peak ground reaction force 75 c. Knee flexion angle 76 d. Knee flexion angular velocity 77 e. Knee joint power 78 f. Eccentric work 79 5.1.3 Differences in energy-dissipating strategies between landing techniques in both sagittal and frontal planes 80 a. Joint angles at initial contact 80 b. Peak joint angle 81 c. Joint angular velocity 82 d. Joint moments 83 e. Joint power 84 f. Eccentric work 85 5.2 STAGE B – Investigation of Knee Injury Mechanisms 5.2.1 Porcine specimens during simulated landing impact 86 a. Peak compressive force 86 b. Anterior tibial translation and axial tibial rotation 88 5.2.2 Human cadaveric specimens during simulated landing impact a. High-speed video capture of impact compression 89 89 v b. Compressive force, posterior femoral displacement and axial tibial rotation 5.2.3 Osteochondral damage in porcine specimens 91 93 a. Histological observations 93 b. Mankin score distribution 94 c. Mankin score comparison between specimens 96 5.2.4 Osteochondral damage in human cadaveric specimens 97 a. Cartilage volume 97 b. Cartilage thickness 97 c. Histological observations 100 d. Mankin score profiles 101 e. Mankin score comparison between regions 104 5.2.5 Finite element analysis of a tibiofemoral joint during simulated landing impact a. Validation with experimental results b. Prediction of compressive force, relative anterior tibial translation and axial tibial rotation c. Prediction of peak anterior cruciate ligament and tibial cartilage stresses 5.2.6 Damage and degenerative changes in osteochondral explants 105 105 106 108 111 a. Compressive force response 111 b. Cell viability 112 c. Collagen and glycoaminoglycan quantification 115 d. Histological observations 116 e. Immunohistochemistry 118 f. Mankin scores 120 g. Cartilage volume 121 5.2.7 Comparisons between non-sustained and sustained impact compression conditions 122 a. Compressive force response 122 b. Histological observations 122 c. Mankin scores 123 vi d. Overall cartilage thickness 124 e. Zonal cartilage thickness 125 5.3 STAGE C - Development of Injury Prevention Strategies 5.3.1 Comparisons between unrestrained and restrained impact compression conditions 127 a. ACL failure 127 b. Specimen responses 127 c. Compressive force 130 d. Relative anterior tibial translation 131 e. Axial tibial rotation 132 f. Histological observations 133 g. Mankin scores 135 h. Cartilage thickness 136 i. Cartilage surface morphology and volume 136 j. Subchondral bone plate thicknesss 138 5.3.2 Comparisons between unbraced and braced knee conditions during normal landing 139 a. Peak ground reaction force and contact force 139 b. Joint flexion angles 141 c. Anterior tibial translation 142 d. Axial tibial rotation 142 5.3.3 Comparisons between unbraced and braced knee conditions during large simulated landing impact 143 a. ACL failure 143 b. Peak compressive force and contact force 143 c. Relative anterior tibial translation and axial tibial rotation 144 d. Histological observations 145 e. Mankin scores 145 f. Cartilage thickness 146 6. DISCUSSIONS 6.1 Understanding biomechanics of landing 147 148 vii 6.2 Investigation of knee injury mechanisms 161 6.3 Development of injury prevention strategies 192 7. CONCLUSION 207 8. RECOMMENDATIONS 209 9. BIBLIOGRAPHY 211 10. APPENDICES 233 11. PUBLICATION LIST 250 viii ACKNOWLEDGMENTS This work was funded by the grant entitled, ‘R175-000-062-112: Construction and validation of a dynamic 3D Finite Element (FE) model of the tibio-femoral joint’ from the Ministry of Education-Academic Research Fund, and co-supported by the Defence Medical and Environmental Research Institute. The work would not have been possible without the careful guidance from my supervisors, Prof James Goh and Dr Peter Lee. I wish to thank them for their selfless support and mentoring throughout the course of my doctoral studies. I would like to express my grateful thanks to the staff from the Division of Bioengineering, namely Assoc/Prof Toh, Annie, Millie, Dorothy, Ernest, Matthew, Yen Ping and Jasmine, who have bestowed help upon me in one way or another. Many thanks to the friendly team from Department of Orthopaedic Surgery - Grace, Hazlan, Dominic, Siew Leng, Irene, Soon Chiong, Jamaliah and Amit. Also to the DMERI staff, Chee Hoong, Serene, Jianzhong, Kok Yong, Kaizhen, Douglas and Lee Tong, it was fun knowing you all, thanks! And the NUSTEP colleagues, Elaine, Hock Hee, Wendy, Evi, Wan Ping, Eriza, Julee, Haifeng, Hongbin, Eugene See and Eugene Wong, I appreciate all your ‘combat service’ support! And to Joe and Alvin from the Impact Lab for lending me their high-speed video cameras, and Chris Au from Functional Imaging Centre (NUH) for conducting the MRI scans, you guys are crucial in my project, thanks! Not to forget my precious FYP students, Kian Siang, Puay Yong, Ngee Hung, Swee Ting, Shaun, Rubab, Wei Lung, Chin Yang and Yihao, this thesis is the fruit of our relentless teamwork over the past years! ~ Smooth seas not make skilful sailors ~ ix photomicrographs using the Mankin scoring system (Mankin et al., 1971; Xie et al., 2006); they were blinded against the test conditions to eliminate bias. d. Immunohistochemistry Additional histological sections were obtained from the embedded control and impacted explants for IHC analysis (Ultravision Detection System, LabVision Corp, USA). The sections were labeled with primary monoclonal anticollagen type I or II antibodies (Chemicon International, United States), prepared at a dilution of 1:200, overnight at 4oC. Biotinylated goat anti-mouse secondary antibodies were then applied at 1:200 for 30min, followed by addition of strepavidin peroxidase for 45min and finally the chromagen-substrate solution for 3min at 37oC. e. Micro-computed tomography and cartilage volume quantification The explants were analysed using the MicroCT scanner (SMX-100CT X-ray CT Sys, Shimadzu, Japan) at Day (before impact), (after impact), and 14. The scan settings were: X-ray voltage (32kV), X-ray current (115µA), detector size (5”), scaling coefficient (100) and pixel spacing (0.0143mm/pixel). The scans were then reconstructed to create the 3D explant geometry using VGStudioMax (Version 1.2, Volume Graphics, Germany). The cartilage was segmented from the 3D volume using a consistent set of threshold gray-values (20086-24334), which was found to allow a reasonable demarcation of the cartilage region from the underlying bone. The segmented cartilage volume of each explant was subsequently calculated to determine the change in volume over the different time-points. 53 f. Cartilage thickness measurement To assess the extent of cartilage deformation, an image-processing software ImageJ (Version 1.4, National Institutes of Health, USA) was used to obtain the average thickness of the cartilage based on the photomicrographs in each condition. Thicknesses of the various cartilage zones were also measured based on the different cell arrangements (Loening et al., 2000) (Fig. 4.2.8). Fig. 4.2.8: Photomicrograph of a typical explant histological section, based on Hematoxylin & Eosin stains. Superficial, middle and deep cartilage zones were demarcated based on the different cell arrangements. g. Statistical analysis Student’s t-test (SigmaStat 3.1, SysTat Software Inc, USA) was used to compare between 1-mm and 2-mm displacement test groups in terms of load response, and between control and impacted test groups in terms of cell viability, GAG and collagen content, Mankin scores and cartilage volume (paired t-test). Normalization was based against the control menisci-covered explants at Day 0/1. Student t-test was also conducted to detect differences in peak impact stress between non-sustained and sustained compression conditions. One-way ANOVA was used to examine differences in peak impact stress and Mankin score between regions within each condition and compartment, and to compare the Mankin scores and 54 cartilage thicknesses between the three conditions. All significance levels were set at p=0.05. 4.3 STAGE C – Development of Injury Prevention Strategies 4.3.1 Restraints to prevent anterior cruciate ligament failure following impact a. Specimen Procurement and Preparation Twenty porcine hind legs (pig age ~2months, weight ~40kg) were procured from a local abattoir (Primary Industries, Singapore). They were prepared in the same manner as mentioned previously, prior to mounting onto the MTS. b. Type of restraints A 5-kg weight was added via a pulley system to the femoral stage to provide a slight posterior preload (Meyer et al., 2008) (Fig. 4.3.1A). In order to assess the effect of inhibiting anterior tibial translation or axial tibial rotation on the knee joint during impact compression, restraint fixtures were fabricated and added into the setup. To restrain anterior tibial translation (or relative posterior femoral translation), the femoral stage was connected to the base plate using wire rope, which was made taut using the adjustable rod with locking nut (Fig. 4.3.1B). To restrain axial tibial rotation, an extension rod was built into the rotator jig that holds the tibial potting cup and two blocking rods were placed on both sides of the extension rod to restrain the rotational motion of the jig, together with the tibia (Fig. 4.3.1C). Finally, these two sets of restraint fixtures were used together to determine the overall effect of inhibiting both factors on the knee joint response (Fig. 4.3.1D). 55 c. Impact protocol Four specimens were used as non-impact (NI) controls while the remaining sixteen porcine knee specimens were divided into four test groups with four specimens each: IC (Impact Compression without restraints, Fig. 4.3.1A), ICA (added with Anterior tibial translation restraint, Fig. 4.3.1B), ICR (added with axial tibial Rotation restraint, Fig. 4.3.1C) and ICC (Combination of both restraints, Fig. 4.3.1D and 4.3.1E). Fig. 4.3.1: A) Experimental setup for inducing ACL failure in a soft tissue-intact porcine knee joint via impact compression; posterior femoral translation and axial tibial rotation were unrestrained. B) the femoral stage was connected to the base plate using wire rope, which was made taut using the adjustable rod with locking nut so as to restrain anterior tibial translation or relative posterior femoral translation. C) an extension rod was built into the rotator jig that holds the tibial potting cup and two blocking rods were placed on both sides of the extension rod to restrain the rotational motion of the jig, together with the tibia. D) both sets of restraint fixtures are used together in the same setup. 56 Fig. 4.3.1E: Impact testing platform for applying simulated landing impact to the porcine knee specimens (with intact joint capsules). In an unrestrained condition, the specimens were allowed anterior tibial translation (relative posterior femoral translation) and axial tibial rotation upon impact compression. In a semi- or fully-restrained condition, anterior tibial translation and axial tibial rotation restraint fixtures were added to the platform. All specimens were subjected to the same impact protocol, which was adopted from a previous report (Yeow et al., 2008). Briefly, the mounted specimens were adjusted via MTS to eliminate tensile/compressive preloading and to maintain a neutral joint position (frontal-plane alignment of longitudinal anatomical tibial axis with femoral axis). Impact compression was performed by displacement control at a single haversine of 10-Hz frequency to simulate a landing impact (Richards et al, 1996; Yeow et al., 2008). The compression trial was successively repeated with incremental actuator displacement of mm; after each trial, the specimens were returned to the initial position before the subsequent trial. The compressive force response was measured by the triaxial load-cell (9347B, Kistler, Switzerland). The end-point of the impact tests was either when ACL failure or a visible bone fracture was present. 57 d. Processing of force and kinematics data A significant drop in compressive force response of at least 70% was used to indicate a major ACL failure; the force drop was estimated from the difference between the peak compressive force during impact compression and the mean compressive force during the post-compression time period 300-500ms. The posterior femoral displacement and axial tibial rotation angle at peak compressive force were determined based on the trajectories of the tibial and femoral markers. Positive and negative axial tibial rotation angles indicated external- and internal-rotated positions relative to the initial position respectively. Increase and decrease in axial tibial rotation angles over time referred to external and internal rotations respectively. Positive posterior femoral displacement represented posterior translation of the femur. e. Damage assessment The presence or absence of ACL failure was further confirmed via dissection. Cylindrical explants (4-mm diameter, 7-mm height) were extracted from same tibial osteochondral regions (Fig. 4.3.2) mentioned previously. Similar procedures, consisting of histology, staining, Mankin scoring and microCT were performed. In addition, the subchondral bone plate thickness was quantified from the microCT slices for all explants using CTAnalyser (Version 1.9, Skyscan, Belgium). 58 Fig. 4.3.2: Locations of tibial osteochondral regions, where cylindrical explants were extracted. The selected regions were the anterior (A), exterior (E), interior (I) and posterior (P) sites in both medial (M) and lateral (L) compartments of the tibial plateau. The anterior, exterior and posterior tibial osteochondral sites were located beneath the menisci while the interior sites were exposed. f. Statistical analysis Paired t-test was used for all test specimens to compare the drop in compressive force response (Fdrop) between the final compression trial that resulted in a substantial Fdrop (major ACL failure) or a visible bone fracture and the prior trial that did not result in major ACL failure or bone fracture. One-way ANOVA was performed between the test groups to identify any differences in terms of peak compressive force, drop in compressive force, posterior femoral displacement and axial tibial rotation. One-way ANOVA was also conducted between the test groups to detect differences in terms of peak compressive load, Mankin scores, cartilage thickness, cartilage volume and subchondral bone plate thickness between test groups. All significance levels were set at p=0.05. 59 4.3.2 Development of a knee protection device a. Knee brace with anterior-sloped hinge The prototype comprised of a tibial component and femoral component, connected by elastic bands (Fig. 4.3.3; See Appendix A.2 for detailed Solidworks drawings). These components can be secured to the thigh and shank using the brace cuffs. The tibial component was layered with ultra high molecular weight poly ethylene (UHMWPE) to provide shock absorption upon contact with the femoral component. The key feature of the brace prototype was the use of a unique joint that consists of a femoral component resting on an anteriorly-sloped (10deg) tibial component, which can facilitate the generation of posterior shear force upon impact compression, so as to relieve the anterior tibial translation induced in the actual knee joint. While previous studies (Pflum et al., 2004; Metey et al., 2005; Yeow et al., 2008) have shown that the posterior tibial slope can act to translate part of the tibiofemoral joint compressive force into anterior tibial shear forces, this anteriorslope feature helps transmit part of the compressive force into posterior shear forces at the brace joint to counter anterior tibial translation. As the posterior shear forces act on the lateral side of the tibia, it can also reduce internal tibial rotation. 60 Fig. 4.3.3: The brace prototype, (A) exploded view, (B) front elevational view, (C) side elevational view, (D) worn on the leg with slight flexion, and (E) worn on the leg with greater flexion. - Velcro straps; - Thigh cuff support; - Shank cuff support; - Femoral component; - Elastic straps; - UHMWPE layer; - Tibial component; – Connecting studs 61 b. Shoe with carbon-fiber sole The shoes with carbon fiber soles were normal running shoes added with Ushaped carbon-fiber-based soles at the forefoot and heel to assist in shock absorption during landing (Fig. 4.3.4; See Appendix A.2 for detailed Solidworks drawings). The brace prototype can be attached to the shoes with carbon fiber soles so as to increase the extent of impact force transmission into the brace joint. Fig. 4.3.4: Shoe incorporated with the carbon-fiber sole, (A) isometric view, side elevational view of the shoe during (B) toe strike, (C) heel strike, (D) heel off, and (E) toe off, and (F) attached to brace prototype. – Shoe body; – Anterior sole component; – Posterior sole component; – Sole grip pads 62 4.3.3 Evaluation of knee protection device under normal landing conditions a. Analysis of landing in unbraced and braced knee conditions Subject preparation and instrumentation were similar to the prior motion analysis studies. Each subject was asked to undergo five different test conditions in random sequence. The test conditions were: (1) Brooks Shoes (BS; Maximus Model, Brooks Sports Inc., US) (Fig. 4.3.5A), (2) shoes with Carbon Fiber Soles (CFS), (3) Brooks Shoes and Ossur 3DX GII Ligament Brace (BS+O) (Fig. 4.3.5B), (4) Brooks Shoes and brace Prototype (BS+P), and (5) shoes with Carbon Fiber Soles and brace Prototype (CFS+P). The subjects were instructed to perform single-leg landing by stepping off a 0.6-m platform with the dominant limb and landing bare-foot onto the force-plate; they were asked to employ their natural landing style. They were given minutes of practice and minutes of rest before commencement of actual landing trials. A trial is considered successful when the subject steps off the platform (without an upward and/or forward jump action) and adopts a stable landing posture. Five trials were conducted per test condition and the results were averaged from each set of five trials. Fig. 4.3.5: (A) Brooks Shoes and (B) Ossur 3DX GII Ligament Brace (A) (B) 63 b. Determination of anterior tibial translation and axial tibial rotation Fifteen retroreflective markers (25-mm diameter) were attached to the subject’s lower body based on the Plug-in-Gait Marker Set, specifically on the sacrum and bilaterally on the anterior superior iliac spine, lateral thigh, lateral femoral epicondyle, lateral shank, calcaneus, lateral malleolus and second metatarsal head, to facilitate the capturing of the subjects’ landing motion. Additional markers were placed on the anterior and lateral thigh, tibial tuberosity, lateral tibial epicondyle, and medial malleolus. Using Vicon Bodylanguage (See Appendix A.3 for pictorial description of procedures and codes), the markers on the anterior and lateral thigh were used to create the femur segment while the markers on the lateral shank, medial and lateral malleolus, and tibial tuberosity were used to create the tibia segment. In the static trial for BS, reference relationships were established between the femur segment and the lateral femoral epicondyle, and between the tibia segment and the lateral tibial epicondyle. During the dynamic trials, the epicondyle markers were removed due to obstruction by the knee braces. The referencing relationships obtained in the static trials were used to ‘re-create’ the lateral femoral and tibial epicondyle markers throughout the landing phase. These markers, together with the measured knee width, were then used to generate the origin markers for the tibia and femur segments. The femur origin marker, hip joint centre and lateral femoral epicondyle were utilized to form the femoral coordinate system, while the tibia origin marker, ankle joint centre and lateral tibial epicondyle were utilized to form the tibial coordinate system. Anterior tibial translation was determined by the difference between the tibial and femur origin marker coordinates transformed onto the tibial coordinate system in 64 the anterior-posterior direction. Axial tibial rotation was estimated by the Euler angle between tibial and femoral coordinate systems in the superior-inferior direction transformed onto the tibial coordinate system. c. Measurement of contact joint force of knee brace Five subjects were randomly selected, whereby a pressure film sensor (K-scan, Tekscan Inc, US) was attached onto the surface of the tibial component so as to determine the distribution of impact stress and magnitude of impact force that was transmitted through the brace joint during landing. The film sensor was calibrated from a series of known forces applied using the material testing system (810-MTS, MTS Systems Corporation, USA) and set to collect stress data at a sampling rate of 100Hz. d. Processing of kinematics and kinetics data The software, Vicon Workstation 5.1 and Polygon 3.1, were used for data collection and processing respectively. The kinematics data were smoothed using a Woltring filter with a mean squared error of 20. Peak GRF, contact force, knee flexion angles, anterior tibial translation and axial tibial rotation of the dominant limb during landing phase were obtained for analysis. The landing phase was taken as the time between initial contact and peak knee flexion angle. All parameters were normalized to control condition BS. e. Statistical Analysis One-way ANOVA was used to compare the parameters between the test conditions. All significance levels were set at p=0.05. 65 4.3.4 Evaluation of the knee protection device under large simulated landing impact a. Analysis of cadaveric knee specimens during unbraced and braced knee conditions Four isolated cadaveric leg specimens and four pairs of cadaveric leg specimens (total: 12 specimens) were obtained from male donors (age: 56-89years). For the paired specimens, they were prepared in the same manner as described previously. With each pair of specimens, one was randomly chosen for the unbraced group while the contralateral specimen was used for the braced group. Each pair of specimens was subjected to the same impact protocol, which was adopted from a previous report (Yeow et al., 2009c). As the brace prototype was designed to be attachable to the shoe; hence in this case, this attachment was simulated by linking the prototype directly to the tibial potting cup so that impact loads can be partly transmitted into the brace joint upon compression (Fig. 4.3.6). The presence of ACL failure determined the end-point of the impact trials for the unbraced specimen within each pair; this end-point was then used to dictate the completion of the impact trials for the braced contralateral specimen. 66 Fig. 4.3.6: Experimental setups for the unbraced and braced conditions. b. Processing of compressive force and kinematics data A significant drop in compressive force response of at least 70% was used to indicate a major ACL failure; the force drop was estimated from the difference between the peak compressive force during impact compression and the mean compressive force during the post-compression time period 300-500 ms (Yeow et al., 2009c). The significant force drop determined the end-point of the impact trials for the unbraced specimen within each pair. The presence or absence of ACL failure was further confirmed via magnetic resonance imaging and dissection. The posterior femoral displacement (relative anterior tibial translation) and axial tibial rotation angle at peak compressive force were determined based on the trajectories of the tibial and femoral markers. Peak compressive force, posterior femoral displacement and axial tibial rotation were normalized to the unbraced condition. 67 c. Damage assessment ACL failure was determined via magnetic resonance imaging and dissection. Osteochondral explants were extracted from the same regions of the tibial cartilage previously described, and underwent similar procedures of histology, staining, Mankin scoring and cartilage thickness measurement. d. Statistical analysis Paired t-test was used for all test specimens to compare the peak compressive force, posterior femoral displacement and axial tibial rotation during final compression trial. One-way ANOVA, together with Holm-Sidak post-hoc test, was performed between the test groups to identify any differences in terms of cartilage thicknesses and Mankin scores at different tibial plateau regions. All significance levels were set at p=0.05. 68 [...]... landing phase of single-leg and double-leg landing 5 .1. 9 Comparison of knee flexion angular velocities between landing height and between landing techniques 5 .1. 10 Comparison of knee joint powers between landing height and between landing techniques 5 .1. 11 Comparison of knee eccentric work between landing height and between landing techniques 5 .1. 12 Comparison of hip, knee and ankle joint angles at initial... to devise a knee protection device to minimize knee injury risk during landing The project comprises of three stages: (A) understanding biomechanics of landing, (B) investigation of knee injury mechanisms, and (C) development of injury prevention strategies The findings indicated substantial regression relationships of landing kinetics and kinematics with landing height; landing height and technique... double-leg and single-leg landing in both sagittal and frontal planes 5 .1. 13 Comparison of peak hip, knee and ankle joint angles between double-leg and single-leg landing in both sagittal and frontal planes 5 .1. 14 Comparison of peak hip, knee and ankle joint angular velocities between double-leg and single-leg landing in both sagittal and frontal planes 5 .1. 15 Comparison of peak hip, knee and ankle... Unbraced and braced modes of the impact setup 5 .1. 1 Profiles of ground reaction force, knee flexion angle, angular velocity and joint power during landing phase 5 .1. 2 Regression relationship of peak ground reaction force with landing height 5 .1. 3 Regression relationships of knee flexion angles with landing height xiv 5 .1. 4 Regression relationship of knee flexion angular velocity with landing height 5 .1. 5... of knee joint power with landing height 5 .1. 6 Profiles of ground reaction forces, knee flexion angles, angular velocities, joint powers between single-leg and double-leg landing tasks 5 .1. 7 Comparison of peak ground reaction forces between landing height and between landing techniques 5 .1. 8 Knee flexion angles at initial contact, at peak ground reaction force and at maximum knee flexion during landing. .. double-leg and single-leg landing in both sagittal and frontal planes 5 .1. 16 Comparison of peak hip, knee and ankle joint powers between double-leg and single-leg landing in both sagittal and frontal planes 5.2 .1 A) Dissection photograph and magnetic resonance imaging scans, pre-test and post-test, of porcine knee specimens B) Profiles of the drop in compressive force response during simulated landing impact. .. knee joint kinematics and energetics sustained during landing phase, in response to the effect of landing technique 6 Energy-dissipating Strategies during Landing Landing-related injuries are common in intensive sports, such as volleyball, basketball and gymnastics (Dufek and Bates, 19 91; Ferretti et al., 19 92; Harringe et al., 2007; McKay et al., 20 01) These injuries can be incurred from typical landing. .. double-leg landing study by DeVita and Skelly (19 92) found that female volleyball and basketball players exhibited a peak GRF of 2-3 bodyweights (BW) during stiff and soft landing tasks from a 0.59-m height; their results suggested that soft landing (high knee flexion) reduced the impact stress on body tissues compared with stiff landing (low knee flexion) McNitt-Gray (19 93) further tested landing heights... examined the effects of gender, landing height and landing stiffness on knee joint energetics, there is still a lack of understanding on how the knee joint may respond in terms of energy dissipation between single-leg landing and double-leg landing, yet this issue is important in explaining why single-leg landing leads to a higher injury risk relative to double-leg landing (Olsen et al., 2004) Therefore,... 2008) 2 Landing Height With the aim of identifying the factors involved in the mechanisms of landing impact injuries, previous studies have performed motion analysis work on double-leg landing from various landing heights, comparing between genders, between soft and stiff landing styles, and between gymnasts and recreational athletes (DeVita and Skelly, 19 92; McNitt-Gray, 19 93; Seegmiller and McCaw, . techniques 5 .1. 10 Comparison of knee joint powers between landing height and between landing techniques 5 .1. 11 Comparison of knee eccentric work between landing height and between landing techniques. observations 14 5 e. Mankin scores 14 5 f. Cartilage thickness 14 6 6. DISCUSSIONS 14 7 6 .1 Understanding biomechanics of landing 14 8 viii 6.2 Investigation of knee injury mechanisms 16 1 . force and at maximum knee flexion during landing phase of single-leg and double-leg landing 5 .1. 9 Comparison of knee flexion angular velocities between landing height and between landing