Mechanical loading associated with exercise is known to benefit bone health;​ however, most studies explore exercise benefits on bone mass independent of bone structure and strength. The purpose of this dissertation is to explore the response of the skeleton to exercise across the translational divide between animal- and human-based studies, with a particular emphasis on exercise-induced changes in bone structure and estimated strength.

To explore the skeletal benefits of exercise, models were used wherein loading is introduced unilaterally to one extremity. Unilateral exercise enables the contralateral, non-exercised extremity to be used as an internal control site for the influences of systemic factors, such as genetics and circulating hormones.

In study 1, a dose response between load magnitude and tibial midshaft cortical bone adaptation was observed in mice that had their right tibia loaded in axial compression at one of three load magnitudes for 3 d/wk over 4 weeks. In study 2, the ability of peripheral quantitative computed tomography to provide very good prediction of midshaft humerus mechanical properties with good short-term precision in human subjects was demonstrated. In study 3, collegiate-level jumping (long and/or high jump) athletes were shown to have larger side-to-side differences in tibial midshaft structure and estimated strength between their jump and lead legs than observed in non-jumping athletes. In study 4, prepubertal baseball players followed for 12 months were shown to gain more bone mass, structure and estimated strength in their throwing arm relative to their nonthrowing arm over the course of 12 months.

These cumulative data using a combination of experimental models ranging from animal to cross-sectional and longitudinal human models demonstrate the ability of the skeleton to adapt its structure and estimated strength to the mechanical loading associated with exercise. Study of these models in future work may aid in optimizing skeletal responses to exercise.