A biomechanical comparison of various methods of stabilization of subtrochanteric fractures of the femur

Subtrochanteric femoral fractures with and without bony contact were simulated in cadaver specimens, fixed with one of seven different types of intramedullary or plate implants, and tested biomechanically. The implants used were Enders pins, Zickel nail, compression hip screw, AO angled blade plate, and intramedullary locked nail systems of the Klemm-Schellman, Brooker-Wills, and Grosse-Kempf types. Femur-implant constructs using intramedullary devices were a maximum of 5% as stiff in torsion as intact cadaveric femora tested in the same manner, while plate-fixed fractures were nearly 50% as stiff. In bending, all devices except the Enders pins were approximately 80% as stiff as intact femora. Loss of bony contact at the fracture site had little effect on stiffness except in the case of the keyless compression hip screw, where the screw rotated freely in the barrel. In combined bending and compression to failure, a test to simulate forces due to body weight, the intramedullary locked rods were found to support between 300 and 400% of body weight while the plate systems failed at loads between 100 and 200% of body weight.