Thesis (Ph. D.)--University of Hong Kong, 2007. / Title proper from title frame. Also available in printed format.

Thesis (M.S. )--University of Tennessee Health Science Center, 2006. / Title from title page screen (viewed on April 16, 2008). Research advisor: Denis J. DiAngelo, Ph.D. Document formatted into pages (xi, 89 p. : ill.) Vita. Abstract. Includes bibliographical references (p. 82-89).

Thesis (doctor of medicine)--Universitetet i Lund.

Thesis (doctor of medicine)--Universitetet i Lund.

Pitting and delamination wear of ultra high molecular weight polyethylene (UHMW-PE) tibial plateaux for total joint replacements have habitually been attributed to a fatigue crack growth mechanism associated with a combination of high sub-surface cyclic shear stresses and degradation of the chemistry and structure of UHMW-PE caused by the gamma-irradiation sterilisation standard procedure. However, the exact mechanisms by which cracks initiate and grow in UHMWPE are not known and the relationships between these mechanisms and pitting and delamination are only assumptions based on qualitative observations. A fracture mechanics approach based on the J-integral concept of plane strain crack initiation toughness was therefore applied in order to firstly obtain the fracture toughness and crack growth stability of UHMW-PE and secondly to determine the mechanisms by which pitting and delamination occur in vivo. It was necessary to modify the existing standard ASTM E813-89 for the treatment of experimental J data in order to accommodate for the large crack tip plasticity and pronounced ductile tearing. This modified method was then applied to a detailed investigation of the influence of sterilisation and ageing on the chemical, physical and mechanical properties of UHMW-PE. Simulated shelf and in vivo environments enabling a rapid ageing of UHMW-PE corresponding to 10 years of natural ageing were developed. Sterilisation was either conducted by gamma-irradiation in air or nitrogen, or by gas plasma. In virgin UHMW-PE, cracks propagated by a succession of plastic deformation and craze nucleation over thin layers of material, yielding a very high value of J (90 kJ/m 2 ) at 37 : C. Gammairradiation in air followed by 10 years ageing resulted in a highly brittle material with a crack initiation fracture toughness reduced by 78% and a mechanical behaviour approaching that of a linear elastic material i.e. creation of a "cup-and-cone" in tension and formation of 45" shear lips in threepoint bending. On the other hand, gas plasma sterilised UHMW-PE could not be differentiated from unsterile UHMW-PE in either its physical nor mechanical properties. Qualitative correlations existed between the presence and location of highly oxidised regions and the crack initiation fracture toughness of the material. Quantitatively, the J-initiation toughness exhibited a hyperbolic decrease with increasing density and oxidation index while the tensile secant modulus linearly increased with density. From these empirical relationships, a model was created which described the variation of the fracture toughness with depth within a UHMW-PE sample. This model indicated that the zones of high density, oxidation and crystallinity correspond to the areas of maximum shear stress and minimum fracture toughness and that the propensity of UHMW-PE to suffer pitting, delamination and high wear rates through a fracture mechanism is significantly increased by extensive oxidative degradation.

As the number of total knee arthroplasty (TKA) procedures increases annually, the patient demographic is shifting to include younger patients with higher expectations for post-operative function. The aim of this study was to compare movement patterns during activities of daily living among TKA patients and a healthy, age-matched group using 3D motion analysis. Specifically, this analysis looked at walking on level and inclined surfaces, as well as sitting up and down from a chair. It was predicted that (1) TKA patients would exhibit reduced knee extension moments at the operated limb and increased adduction moments at the contralateral limb during gait, (2) walking downhill would result in greater differences between TKA and control groups, compared to level walking, and (3) TKA participants would have greater flexion angles, moments and power values at the hip, compared to controls, during the sit-stand tasks. Seventeen participants (age=62±6 years, BMI=30±3 kg/m2, time after surgery=11±5 months) were recruited from the Ottawa Hospital, having undergone unilateral TKA by the same surgeon. An age-matched control group was composed of 17 individuals (age=63±8 years, BMI=27±4 kg/m2) who were recruited from the local community. Three dimensional (3D) biomechanical assessment was conducted with all participants performing five trials of walking on level and inclined surfaces, stair ascent and descent as well as sit-stand tasks. Results from this study were focused on gait and sit-stand transitions, showing that TKA participants exhibited altered gait patterns on both walking surfaces, with significantly smaller knee flexion angles and moments, as well as reduced peak power at the knee. The TKA group also experienced reduced knee extension moments; however, this was only significant for downhill walking. Consistent with our hypothesis, downhill walking resulted in greater discrepancies between the groups compared to level walking. Contrary to our third hypothesis, TKA participants exhibited significantly smaller peak hip flexion angles and moments during the sit-stand task, along with reduced hip abduction angles and knee abduction moments. The reduced knee flexion kinematics and kinetics observed during gait tasks, combined with the differences in frontal plane mechanics observed during the sit-stand task suggest that altered loading patterns persist six to twelve months after surgery. This may be a result of continued pre-operative movement patterns as well as the surgery itself, and should be kept in mind when developing rehabilitation programs for this patient population.

Total knee arthroplasty (TKA) places a large burden on the Canadian Health care system with over 700 million dollars spent on this procedure annually. Exercise has been shown to be effective in improving pain, physical function, mobility and quality of life post-operatively. There is very limited research on adherence to rehabilitation and exercise after TKA. The aim of this thesis was to increase understanding of exercise adherence in this population. The first manuscript in this thesis was a scoping review that evaluated exercise adherence and the quality of reporting of exercise interventions within post-operative TKA rehabilitation trials. A systematic search of scientific databases was conducted for randomized controlled trials (RCT) with an exercise intervention for post-operative TKA. In total, 112 articles were included in this review. This study found that the vast majority of articles (85%) were of poor quality having either high/unclear risk of bias. The majority of RCTs (63%, N=71) on post-operative TKA rehabilitation did not adequately report exercise adherence (e.g., definition, outcome measure used and results), while only 23% (N=15) provided a definition of adherence in the context of their study. Overall reporting of the exercise intervention was poor, with 15 items (of 19) of the Consensus on Exercise Reporting Template (CERT) reported less than 60% of the time. Inadequate reporting of exercise interventions and adherence to exercises leads to decreased reproducibility and translation into clinical practice. Proper reporting of rehabilitation exercises after TKA will ensure standardization for future studies and clinical replication. The second manuscript in this thesis was a qualitative study that aimed to understand the patient- related barriers and facilitators to exercise adherence in patients immediately after undergoing TKA. Using an interpretive description approach, semi-structured qualitative interviews were conducted. Seven participants were interviewed at 8-weeks post-operatively in order to better capture physical, psychological, social and contextual factors linked to exercise adherence. Interview questions explored participants’ experience with physical activity and exercise, motivation to perform physical activity, beliefs that exercise will reduce pain, the factors that limit their ability to engage in exercise, and the importance of using self-regulation to improve exercise adherence. Emergent themes were mapped onto the domains of the WHO adherence framework. This study identified 4 themes that fit within the WHO adherence framework: patient-related factors, condition-related factors, health care system, and social support. In particular, self-regulation, previous knowledge of exercise, post-operative complications, comorbidities, social support, and lack of guidance from health care providers were identified as personal and environmental characteristics that affect exercise adherence. The overall findings of this study suggest exercise adherence is a multifaceted construct with interconnected concepts. / Thesis / Master of Health Sciences (MSc) / Total knee replacement (TKR) surgery costs the Canadian Health care system millions of dollars per year. Exercise is an effective method for improving function and decreasing pain after surgery. However, it is unclear how much patients perform exercise as recommended by health care providers. The purpose of this study was to understand exercise behaviour after knee surgery. The first study was a scoping review that looked at exercise adherence and the characteristics of exercise treatments delivered after knee surgery in previously published clinical trials. The study found that most studies did not report enough information for replication of exercise treatment protocols (e.g. description of provided exercises/progression, who implemented the intervention, etc.). Similarly, exercise adherence was also poorly reported (e.g., definition, how adherence was measured). Poor reporting of exercise treatment methods, and how well participants complete exercises decreases the ability of researchers and therapists to apply the results of these trials. In the second study, patients were interviewed after knee surgery to understand why patients do or do not exercise after surgery. There were several factors, both positive and negative, that patients identified that affected their ability to exercise (e.g. self-regulation (referring to the ability to control thoughts, emotions and behaviour to pursue long-term goals), prior knowledge of exercise, having social support from family and friends, as well as lack of support from health care providers). The overall findings suggest that how well participants exercise after surgery is a complex issue. In order to improve exercise adherence, we need to have a better understanding of the individual factors that may influence adherence.

Joint mobility is dependent on comfort, thus pain is associated with mobility. This study examined the relationship between pain perception and joint mobility in arthritis patients before and after total knee arthroplasty. Pain perception was indicated by pain intensity, pain distress and pain expectation; joint mobility was represented by the degree of knee joint flexion. Visual analogue scales were used for pain intensity, pain distress and pain expectation; knee joint flexion as measured using a goniometer. Data were collected on 24 subjects preoperatively (T1), immediately postoperatively (T2) and forty-eight to seventy-two hours postoperatively (T3). Twenty-four subjects participated in the study. Significant paired t-test resulted for joint flexion at T1 and T3 and pain expectation at T2 and T3. Pearson product-moment correlation coefficients were significant for pain intensity and pain distress at all three times, pain intensity and joint flexion at T1 and pain intensity at T1 and at T2.

Introduction: Gait biomechanics are associated with knee osteoarthritis (OA) structural progression, but no studies have included: i) all three lower extremity joints, ii) non-frontal plane factors, iii) temporal loading patterns, and iv) progression from structural and symptomatic perspectives. This dissertation addressed gaps in our understanding of lower limb biomechanics and their implication for determining whether we have identified and are targeting the most effective biomechanical variables in the development and evaluation of conservative interventions to slow knee OA structural and symptom progression (progression to TKA). Methods: 54 patients with knee OA underwent baseline gait analysis. Three-dimensional hip, knee, and ankle angles and moments were calculated. Waveform characteristics were determined using Principal Component Analysis (PCA), and knee adduction moment (KAM) peak and impulse were calculated. At follow-up 5-8 years later, 26 patients reported undergoing total knee arthroplasty (TKA). Unpaired Student’s t-tests detected differences in baseline demographic and gait characteristics between TKA and no-TKA groups. Receiver operating curve analysis determined discriminative abilities of these differences. Stepwise discrimination analysis determined which multivariate combination best classified the TKA group. Logistic regression analysis determined the predictive ability of the multivariate model. Results: There were no baseline differences in clinical and spatiotemporal gait characteristics, but the TKA group showed significant gait biomechanical differences, including higher KAM magnitude (KAMPC1), less difference between early and mid-stance KAM (KAMPC2), higher KAM peak and impulse, reduced early stance knee flexion and late stance knee extension moments (KFMPC2), and reduced stance dorsiflexion moments (AFMPC4). The multivariate discriminant function with the highest classification rate (74.1%) combined KAMPC1, KFMPC2, and AFMPC4, with sensitivity of 84.6 and specificity of 71.4. A one-unit increase in the model score increased risk of progression to TKA six-fold. Conclusion: Higher KAMPC1 scores suggest higher overall loading during gait. Lower KFMPC2 and AFMPC4 scores suggest inability to unload the knee and therefore sustained loading. Interventions reducing overall load and altering patterns of loading (i.e. increase unloading) may reduce risk of progression to TKA. Future research should determine how components of the discriminant model can be altered conservatively, and what impact alterations have on the risk of progression to TKA.

The Oxford Unicompartmental Knee Replacement (UKR) is a commonly performed procedure, with a good clinical outcome at 15 years, however, radiolucent lines are commonly found beneath the tibial tray. With the projected increase in knee arthroplasty, particularly in younger patients, implant longevity is of paramount importance. The aim of this thesis is to understand how fixation is achieved with the Oxford UKR and how it can be improved. A histological study demonstrated that in the presence of a radiolucent line the tibial bone-cement interface is made up of a combination of direct bony contact, fibrocartilage and fibrous tissue. The radiolucency is more marked when there is more soft tissue. However in all cases there is some direct bony contact. Cemented and cementless fixation was compared in a randomised controlled study using radiostereometric analysis and fluoroscopic imaging of the interfaces. In the first year the cementless tibial component subsided on average 0.28 mm and had an increased posterior slope of 0.40°, whereas the cemented component only subsided 0.09 mm, with a 0.10° increase in slope. In the second year both components had very little further subsidence (mean<0.05 mm) and no increase in posterior slope. In the second year a single cementless tibial component subsided greater than 0.15 mm, whereas four cemented components, all with radiolucencies, subsided more than 0.15 mm. At two years the cemented components had a significantly higher prevalence of radiolucency (62% v 29%), with 24% having a complete radiolucency, whereas no cementless components had a complete radiolucency. Two designs of lateral UKR were also compared. These had a flat tibial component that predominantly transmits compressive loading, and a domed component that also transmits shear. There was a lower prevalence of radiolucency in the domed tibia (13% v 60%), even though there was a similar amount of migration as the cemented medial tibial component. In conclusion radiolucent lines, both partial and complete, are common with cemented components, and may, in part, be a result of compressive loading. They are associated with good long-term results and direct bone cement contact indicating satisfactory fixation. However, they are also associated with increased migration and soft tissue at the interface suggesting that the fixation, although satisfactory, is suboptimal. The cementless components had no complete radiolucencies and low levels of migration in the second year. This suggests that bone ingrowth and secure fixation occurs reliably, and therefore that cementless fixation may be better than cemented for the Oxford UKR.