A Compound Risk Indicator for Subject-Specific Prediction of Hip Fracture in Sideways Falls

Low-trauma hip fracture, usually caused by fall from standing height, has become a main source of morbidity and mortality among the elderly. However, the hip fracture risk is subject-dependent as it is related to the subject’s anthropometric parameters, the kinematic/kinetic variables in fall, and the hip anatomy. A compound fracture risk indicator is proposed in this study to define the correlation between the hip fracture risk and subject’s parameters such as anthropometric parameters, bone mineral density (BMD), height of body mass center (HMC), and hip soft tissue thickness (STT). In this study, a two-level, subject-specific biomechanical model composed of a whole-body dynamics model and a proximal femur finite element model was used to predict the hip fracture risk for 80 subjects. The required information was obtained from the subject’s whole body and hip DXA (dual energy X-ray absorptiometry) images. Then, multivariate nonlinear least-square fits of power functions were used to investigate the effect of parameters on hip fracture risk. Results indicated that although BMD is the dominant parameter affecting the hip fracture risk, other parameters such as hip soft tissue thickness and anthropometric parameters also have considerable effects. This finding suggests that the hip fracture risk cannot be accurately predicted by the currently available single factor predictors as they do not consider all multifactorial parameters. However, the proposed function can properly predict the risk of hip fracture in the absence of any computational simulation and biomechanical model. It also quantifies the effect of musculoskeletal-, and organ-level parameters on hip fracture risk and demonstrates which individual is more likely to experience hip fracture.