A biomechanical comparison of composite femurs and cadaver femurs used in experiments on operated hip fractures

Abstract Fourth generation composite femurs (4GCFs, models #3406 and #3403) simulate femurs of males 60 years. Proximal femur anthropometric measurements were noted. Strain gauge rosettes were attached and femurs were mounted in a hip simulator applying a combined subject-specific axial load and torque. Baseline measurements of resistance to deformation were recorded. Standardized femoral neck fractures were surgically stabilized before the constructs were subjected to 20,000 load-cycles. An optical motion tracking system measured relative movements. Median (95% CI) head fragment migration was 0.8 mm (0.4 to 1.1) in the 4GCF group versus 2.2 mm (1.5 to 4.6) in the cadaver group ( p =0.001). This difference in fracture stability could not be explained by observed differences in femoral anthropometry or potential overloading of 4GCFs. 4GCFs failed with fracture-patterns different from those observed in cadavers. To conclude, standard 4GCFs provide unrealistically stable bone-implant constructs and fail with fractures not observed in cadavers. Until a validated osteopenic or osteoporotic composite femur model is provided, standard 4GCFs should only be used when representing the biomechanical properties of young healthy femurs.