A BIOMECHANICAL ANALYSIS

Background: This biomechanical study was designed to evaluate the predictive ability of dual-energy x-ray absorpti-ometry, cortical bone geometry as determined with computed tomography, and radiography in the assessment of tor-sional load to failure in femora with and without notching. Methods: Thirteen matched pairs of cadaveric femora were randomized into two groups: a notched group, which con-sisted of femora with a 3-mm anterior cortical defect, and an unnotched group of controls. Each pair then underwent torsional load to failure. The ability of a number of measures to predict femoral torsional load to failure was assessed with use of regression analysis. These measures included dual-energy x-ray absorptiometry scans of the proximal and the distal part of the femur, geometric measures of both anterior and posterior cortical thickness as well as the polar moment of inertia of the distal part of the femur as calculated on computed tomography scans, and the Singh os-teoporosis index as determined on radiographs. Results: The torsional load to failure averaged 98.9 N-m for the notched femora and 143.9 N-m for the controls; the dif-ference was significant (p < 0.01). Although several variables correlated with torsional load to failure, distal femoral bone-mineral density demonstrated the highest significant correlation (r = 0.85; p < 0.001). Moreover, multiple regres-sion analysis showed that a combination of distal femoral bone-mineral density and polar moment of inertia calculated with the posterior cortical thickness (adjusted r2 = 0.79; p < 0.001) had the strongest prediction of torsional load to fail