Femoral fractures cause excess morbidity, disability and mortality and are a major health problem, which is likely to be aggravated by the ageing of population. Areal bone mineral density (aBMD) measured with dual energy X-ray absorptiometry (DXA) is the gold standard clinical measure to evaluate fracture risk in vivo. Three-dimensional measurements can be performed with Quantitative Computed Tomography (QCT), but inducing higher radiation dose. DXA-based and QCT-based Finite Element (FE) analysis have been developed in recent years in order to improve the prediction of femoral strength and therefore of fracture risk. However, these models should be validated with accurate experiments in vitro before a clinical application. The aim of this study was to evaluate the ability of aBMD, DXA-based and QCT-based FE models in predicting the femoral strength in two loading conditions. Thirty-six pairs of human femora were dissected and scanned with DXA and QCT. aBMD was measured in different sites of the proximal femur. For each pair, one femur was tested in one-legged stance configuration (STANCE) and the contralateral one in a backward sideways fall configuration (FALL). Nonlinear QCT-based FE and linear DXA-based FE models were generated reproducing the same loading configurations imposed in the experiments. For experiments and models, the femur failure loads were computed and compared. For the FALL configuration the best predictors of bone strength was the QCT-based FE (R2=0.85), followed by femoral neck aBMD (R2=0.80) and DXA-based FE (R2=0.76). For the STANCE configuration the best predictor was QCT-based FE (R2=0.80), which was similar to the DXA-based FE (R2=0.79), both were better than the femoral neck aBMD (R2=0.66). While QCT-based FE models have been found to be superior to both aBMD and DXA-based FE models in both configurations, the DXA-based FE models have shown good improvement for the STANCE configuration compared with femoral neck aBMD.
Disclosure: The authors declared no competing interests. The study was partially supported by the FP7 European program (MAMBO: PIEF-GA-2012-327357).