Subject specific finite element analysis of stress shielding around a cementless femoral stem.

Abstract

Stress shielding around a femoral stem is usually assessed experimentally using composite or human cadaver femurs. In the present study we have explored the feasibility of using subject specific finite element models to determine stress shielding in operated femurs. Cortical strain was measured experimentally on seven human cadaver femurs, intact and implanted with a straight cementless prosthesis. Two load configurations were considered: single leg stance and stair climbing. Subject specific finite element models derived from computed tomography of the same femurs were analysed intact and with an implant. Principal cortical strain was used to validate the finite element models. Stress shielding was defined as the change in equivalent (von Mises) strain between pre- and post-operative femurs. Cortical strain predicted by the finite element analyses showed to be close to unity with the experimental observations for both intact (R2=0.94, slope=0.99), operated femurs (R2=0.86, slope=0.86) and stress shielding (R2=0.70, slope=0.90). In the proximal calcar area, the region most prone to periprosthetic remodelling, the finite element models were found to successfully reproduce the stress shielding observed experimentally. The study shows that subject specific finite element models manage to describe the stress shielding pattern measured in vitro in the different femurs. Finite element models based on actual human femurs (cadaver and/or patient) could thus be a useful tool in the pre-clinical evaluation of new implants.