Abstract
BACKGROUND: High complication rates continue to be reported for femoral neck fractures, where moderate to severe shortening has been associated with pain and inferior functional outcomes. The objective was to evaluate the biomechanical performance of a novel femoral neck fracture fixation device by comparing to a clinically relevant construct using commercially available cannulated screws. METHODS: Twelve Sawbones femurs were equally divided into two groups where Group I was instrumented with a new fixation system consisting of two partially threaded cannulated screws and a cross-screw and Group II was instrumented with three partially threaded cannulated screws. The fracture pattern resembled a Garden Type II fracture with superior fracture comminution with a 3° wedge. Mechanical testing was set to simulate heel strike. All constructs were preconditioned under a series of non-destructive tests (torsional, bending and compression). Thereafter, a step-wise increasing cyclic loading protocol to a maximum of 30,000 cycles or until failure. Fracture gap area collapse, rotation of the femoral head, screw displacement, failure modes and number of cycles to failure were compared between groups. RESULTS: At 10,000 cycles, Group I had significantly larger fracture gap area when compared to Group II (p = 0.020). The fracture gap was also assessed after the additional fatigue protocol, and Group I continued to demonstrate enhanced ability to resist femoral neck shortening compared to Group II (p = 0.001). For femoral head rotation, Group I constructs demonstrated an average of 0.19° less rotation of the femoral head about the neck after 10,000 loading cycles (p = 0.88). From the zero position (0 N) to pre-load (50 N), Group I screws demonstrated less backout when compared to Group II. While this initial decrease in backout was not statistically significant (p = 0.47), with each subsequent loading step, Group I screws demonstrated significantly less displacement than Group II (p < 0.05). During the fatigue testing up to 30,000 cycles, samples in Group I failed at 20,286 cycles and Group II failed at 14,001 on average, however, this was not statistically significant (p = 0.19). CONCLUSION: Constructs in Group I outperformed Group II by demonstrating the ability to retain an increased fracture gap area during cyclical testing and withstand screw displacement. LEVEL OF EVIDENCE: Basic Science.