Predicting osteoporotic hip fracture risk

Predicting osteoporotic hip fracture risk

Researchers from the MultiSim project, based at the University of Sheffield’s Insigneo Institute and Sheffield Teaching Hospitals NHS Foundation Trust, have developed a model that is able to stratify subjects closely matching their actual hip fracture status.

By 2020, over 100,000 new hip fractures will occur each year in the UK. The annual cost of treatment of these fractures will exceed £2 billion. Even after treatment (surgery), hip fractures remain associated with excess mortality that lasts up to several years post-surgery. Therefore, the prevention of osteoporotic hip fractures is a high-priority healthcare problem.

Currently, the development of effective fracture prevention strategies is impeded by the lack of an accurate understanding of how different risk factors determine hip fracture risk in a subject. To quantify the accuracy of a model (which encapsulates a certain understanding), researchers usually test the ability of that model to correctly stratify subjects by their known fracture status.

The MultiSim researchers started out by developing three different models, each accounting for a different hazard/protective factor influencing hip fracture. One model predicted the range of impact force magnitudes, as induced by potential falls a subject can suffer. Another model predicted the range of impact force attenuation magnitudes, such as due to falling on a soft floor, due to wearing a hip protector or thick clothing during fall, and due to the presence of fat and muscles around the hip region. Yet another model predicted, per individual, the maximum load their thigh bone can resist for different impact orientations, based on computed-tomography scans of their bones. An individual’s fracture risk was then determined by combining the results of the three models.

Fracture risk predicted by the combined model correctly identified the fracture status in 40 (out of 49) subjects with fracture and in 38 (out of 49) subjects without fracture. All subjects were recruited during a previous study and represented postmenopausal British women in the Sheffield region. This shows that the mechanistic understanding of the determinants of hip fracture risk captured in this model can be considered accurate.

A multiscale model to predict current absolute risk of femoral fracture in a postmenopausal population (Biomechanics and Modeling in Mechanobiology) P. Bhattacharya, Z. Altai, M. Qasim, M. Viceconti