Mechanical fatigue of whole rabbit-tibiae under combined compression-torsional loading is better explained by strained volume than peak strain magnitude.

The mechanical fatigue behavior of whole bone is poorly defined, particularly for the combined loading modes that occur in vivo. The purpose of this study was to quantify the fatigue life of whole rabbit-tibiae under cyclic uniaxial compression and biaxial (compression and torsion) loading, and to explore the relationship between fatigue life and specimen-specific finite element (FE) predictions of stress/strain. Twelve tibiae were tested cyclically until failure across a range of uniaxial-compressive loads. Another twenty-two tibiae were separated into three groups and loaded biaxially; peak compressive load was constant in all three groups (50% ultimate force) but torsion was varied (0%, 25%, or 50% of ultimate torque). FE models with heterogeneous linear-elastic material properties were developed from computed tomography. We assessed peak stress/strain and stressed/strained volume based on principal stress/strain, as well as von Mises and pressure modified von Mises criteria. A logarithmic (r2 = 0.68; p < 0.001) relationship was observed between uniaxial-compressive load and fatigue life. Biaxial tests demonstrated that fatigue life decreased with superposed torsion (p = 0.034). Strained volume, based on a maximum principal strain or pressure modified von Mises strain criteria, were strong predictors of fatigue life under both uniaxial (r2 = 0.73-0.82) and biaxial (r2 = 0.59-0.60) loads, and these outperformed equivalent peak stress- and strain-based measures. Our findings highlight the importance of evaluating strain distributions, rather than peak stress or strain, to predict the fatigue behavior or whole bone, which has important implications for the study of stress fracture.