Hip fracture remains a major public health concern due to the significant number of occurrences and mortality rates. Intertrochanteric fractures are the most common type of fracture and are typically caused by a fall. Intramedullary osteosynthesis is a common surgical practice to repair intertrochanteric fractures, but revisions are often required. The work presented in this thesis aims to improve the realism and fidelity of computational models of hip fracture, which can be an effective alternative to costly and labor-intensive clinical in-vivo and experimental in-vitro testing. Intersubject variability is inherently present in anatomy and material relations. Statistical shape and intensity models can be used to characterize anatomic and material property variability in a training set population and can be used to evaluate subject-specific fracture behavior. While prior computational models of hip fracture have evaluated bone strains to assess fracture risk, the current study advances the state of the art by utilizing a novel technique in the extended finite element method (XFEM) to assess hip fracture and develops a computational approach to evaluate fracture repair.

Natural and implanted finite element models were used to predict fracture patterns and evaluate bone-implant load share in subjects with varying geometry and bone quality. A model validation study demonstrated the capability of XFEM in generating unique subject-specific fracture patterns. Femurs generated from a previously published statistical model were fractured to capture the range of patient variability in fracture pattern and load at the onset of fracture. Overall femur size, bone thickness, and bone quality had large effects on load at the onset of fracture. Using one of the average subject models, a study was performed to investigate the effects of surgical alignment, implant material, and loading variability on hip fracture repair. Although, surgical alignment had little effect on load share in the bone-implant construct, mal-alignment caused an increase in peak implant stress and bone strain, which could result in implant failure and delayed fracture healing. Muscles were added to a fracture repair model to capture loading condition variability, which resulted in a more balanced load share between the bone and the implant, indicating the importance of musculoskeletal modeling. These studies included novel techniques for evaluating hip fracture and repair that could be used to aid the surgical community by providing guidance on how implant alignment can promote ideal bone healing conditions.