The likelihood of a front seat occupant sustaining a femoral shaft fracture in a frontal crash has traditionally been assessed by an injury criterion relying solely on the axial force in the femur. However, recently published analyses of real world data indicate that femoral shaft fracture occurs at axial loads levels below those found experimentally. One hypothesis attempting to explain this discrepancy suggests that femoral shaft fracture tends to occur as a result of combined axial compression and applied bending. The current study aims to evaluate this hypothesis by investigating how these two loading components interact. Femoral shafts harvested from human cadavers were loaded to failure in axial compression, sagittal plane bending, and combined axial compression and sagittal plane bending. All specimens subjected to bending and combined loading fractured midshaft, whereas the specimens loaded in axial compression demonstrated a variety of failure locations including midshaft and distal end. The interaction between the recorded levels of applied moment and axial compression force at fracture were evaluated using two different analysis methods:​ fitting of an analytical model to the experimental data and multiple regression analysis. The two analysis methods yielded very similar relationships between applied moment and axial compression force at midshaft fracture. The results indicate that posteroanterior bending reduces the tolerance of the femoral shaft to axial compression and that that this type of combined loading therefore may contribute to the high prevalence of femoral shaft fracture in frontal crashes.