Introduction: The stress distribution within the polyethylene component of a TKR is ultimately dependent on the kinematics of the replaced knee. In turn, the kinematics are dependent on the design of the implant, the relative alignment of the components and tensions of the surrounding soft tissues. Clinical fluoroscopy studies have shown that unicondylar loading occurs in a high proportion of replaced knees. However, the impact on the polyethylene stresses is unknown. The aim of this study was to examine the influence of unicondylar loading during a gait cycle on the predicted kinematics and stresses generated by a commercially available TKR. Methods: A 3D finite element model was developed using explicit FE. The femoral component the TKR was modelled as a rigid surface while the tibial insert was meshed using 4-noded solid elements with non-linear polyethylene properties. The flexion angle, axial force, A-P force and I-E torque were applied to the model as a function of the gait cycle and were similar to those used by a Stanmore knee simulator. The axial force is normally applied through the centre of the femoral component (load case 1). Unicondylar loading was simulated by displacing the action of the axial force medially, by 10 (load case 2)and 20mm (load case 3). The resulting kinematics (A-P translation and I-E rotation) and the internal stresses within the polyethylene liner are reported for each of the three load cases. . Results: When subjected to uniform bi-condylar loading (lc1), the peak anterior translation of the polyethylene component was approx. 3.5mm and the peak internal rotation was 6 degrees. The peak von Mises stresses was approx. 16 MPa. Shifting the axial load medially by 10mm only had a minor effect on the predicted kinematics, increasing the anterior translation by 1mm and the internal rotation by 1 degree. The peak polyethylene stresses increased to 18 MPa. Shifting the axial load medially by 20mm had a significant effect on both the kinematics and stresses. The maximum anterior translation of the polyethylene increased to 8mm and there was an associated increase in the peak internal rotation, from 6 degrees to 20 degrees. The peak polyethylene stresses increased to approx. 20 MPa and significant amounts of plastic deformation were induced during the stance phase of the gait cycle. Discussion and conclusions: Explicit finite element analysis has for the first time enabled us to simulate the abnormal kinematics caused by unicondylar loading and the associated changes in the polyethylene stresses. For the particular design examined, uneven bicondylar loading (lc2) had a minor affect on the kinematics but did increases the polyethylene stresses. Unicondylar loading (lc3) significantly affected both the kinematics and the polyethylene stresses.