Comminuted subtrochanteric fractures pose a clinical challenge; locking plate technology has been theorized to offer treatment advantages. A comminuted subtrochanteric femoral fracture model was created with a 2 cm gap below the lesser trochanter in fifteen matched pairs of human cadaveric femora confirmed to be non-osteoporotic. The femora were randomized to treatment with a trochanteric femoral nail (TFN), proximal femoral locking plate (PFLP), or 95° angled blade plate (ABP). Each was tested under incrementally increasing cyclic load up to 90,000 cycles to simulate progressive weight bearing during three months.

The TFN was the strongest implant: it withstood significantly more cycles, failed at a significantly higher force, and withstood a significantly greater load than either plate (p< 0.001). Varus collapse was significantly lower in the TFN construct (p< 0.0001). Mode of failure differed among implants, with damage to the femoral head through implant cut-out in five of ten blade plate specimens and two of ten nail specimens, whereas no damage to the femoral head bone was observed in any of the PFLP specimens.

The TFN was biomechanically stronger than the PFLP and this may have clinical relevance during the slow healing of subtrochanteric femoral fractures. The PFLP was biomechanically equivalent to the ABP but failure occurred without significant damage to the femoral head, suggesting that although biomechanically equivalent, the PFLP might have clinically relevant advantages in its mode of failure over the ABP.