The effect of long-term periprosthetic bone loss on the process of aseptic loosening of tibial total knee arthroplasty (TKA) is subject to debate. Contradicting studies can be found in literature, reporting either bone resorption or bone formation before failure of the tibial tray. The aim of the current study was to investigate the effects of bone resorption on failure of tibial TKA, by simulating clinical postoperative bone density changes in finite element analysis (FEA) models and FEA models were created of two tibiae representing cases with good and poor initial bone quality which were subjected to a walking configuration and subsequently to a traumatic stumbling load. Bone failure was simulated using a crushable foam model incorporating progressive yielding. Repetitive loading under a level walking load did not result in failure of the periprosthetic bone in neither the good nor poor bone quality tibia at the baseline bone densities. When applying a stumble load, a collapse of the tibial reconstruction was noticed in the poor bone quality model. Incorporating postoperative bone loss led to a significant increase of the failure risk, particularly for the poor bone quality model in which subsidence of the tibial component was substantial. Our results suggest bone loss can lead to an increased risk of a collapse of the tibial component, particularly in case of poor bone quality at the time of surgery. The study also examined the probability of medial or lateral subsidence of the implant and aimed to improve clinical implications. The FEA model simulated plastic deformation of the bone and implant subsidence, with further validation required via mechanical experiments.