This thesis is missing pages 98-107, all of which are not in the other copies of the thesis. -Digitization Centre / In total knee arthroplasty with bone defect of the tibia, it was believed that with older designs of tibial tray, both block and stem augments must be used with the tibial tray to improve the knee stability. Obviously, the extended stem causes more difficulties to the surgery as well as more suffering to the patients. Getting rid of the extended stem and still maintaining enough stability is therefore very desirable. The newest tray design, Deltafit Keel tray, which provides much more contact with the human bone structure, may provide enough stability without the extra long stem. The objective in this project is to answer the questions - Is the stem augmentation definitely required alongside the block implant for the cases of bone defect in TKA (Total Knee Arthroplasty) when using the Deltafit Keel tibial tray design? In other words, does the configuration of Deltafit Keel tray with a block provide enough stability in the cases of bone defect? In order to give a reliable answer, three configurations have been studied by conducting both experiments and FEA simulation. The three cases are Deltafit Keel tibial tray only (case 1-no bone defect defect), tray with block augment (case 2-with bone defect assumed) and tray with block and extended stem (case 3-with bone defect assumed). In this study, three commercially available composite bones with isotropic material properties are utilized. For each configuration, the bones are clamped in a testing apparatus and 3000 N static compressive load is imposed on the top surface of the tibia tray at central, medial and lateral locations. In experiment, the strains and displacements at strategically selected locations were measured by strain rosettes (strain gages) and DVRT (Differential Variable Reluctance Transducer) displacement transducers, respectively. In order to simulate the three cases, FE model is established by employing several advanced software including CATIA, True Grid Mesh generator and Abaqus. In order to compare with the experimental results, nine cases (three implant configurations with three different loading positions for each) have been simulated using Abaqus/Standard 6.4. In addition to the nine-case studies, the influence of load offsetting is also investigated by shifting the nodal load along medial-lateral and anterior-posterior directions. It is found that load shifting one node in either direction does not cause significant change in either strain or displacement. Furthermore, FE results of adjacent elements are checked as well and no sudden changes are observed. Since the discrepancy of the output from adjacent elements is negligible, an average value of the elements can be used to represent the output in a small region to compare the experimental strain measured by strain rosettes. Both the experimental data and FEA simulation results lead to the conclusion that comparable stability can be achieved with the configuration of Deltafit Keel tibial tray and a block as compared to the case of Deltafit Keel tray only without bone defect. Moderate improvement of stability, but with significant stress shielding, is found when the extended stem is implanted. For the amount of bone defect and the bone material properties used in this study, the Deltafit Keel tray with a block is the best choice because it is able to provide adequate stability and avoid excessive stress shielding. The loss of a substantial amount of bone to implant an extended stem to trade for the excessive stability may not be worthwhile. Besides, stress shielding is a potential problem which may exist if the extended stem is used. / Thesis / Master of Applied Science (MASc)