Oral Presentation ANZBMS-MEPSA-ANZORS 2022

The use of osseointegrated implants for lower limb amputees has gained popularity over the past few decades, particularly as an alternative to conventional socket-based prostheses. Unfortunately, up to 5.5% of these implanted femurs fail within the first year of surgery [1].  The ability to predict these fractures and the external loading conditions that cause them may help design better implants and improve surgical procedure. Thus, the aim of this study was to investigate whether computational modelling can accurately predict fractures that occurred in osseointegrated femurs. A finite element (FE) model of a composite femur with an osseointegrated implant was generated using a set of QCT images. Over 15,000 combinations of unique external loads (force and moment directions) were applied individually to the implant while the nodes on the proximal femur were fixed. Each of the loads was gradually increased until the femur fractured, defined by the point in which the volume of contiguous failed elements reached 350 mm³ [2]. We compared the fracture sites predicted by the FE model to in vivo fracture data reported in literature. Our analysis predicted that 99.3% of the 15,056 fractures occurred at two distinct regions, the superior trochanteric region, and adjacent to the proximal tip of the implant (Fig. 1(b)). These findings are consistent with those of Hoellwarth et al [1] who reported that 20 out of 22 femoral fractures in osseointegrated patients occurred in the trochanteric region and that 19 out of 22 occurred within 2 cm of the proximal tip of the implant. This preliminary work demonstrates that FE analysis can predict the location of in vivo fractures in femurs with osseointegrated implants. Future work will focus on validating specimen specific finite element models and using them to study the effect of implant positioning on femoral fractures.