Muscle indices do not fully account for enhanced upper extremity bone mass and strength in gymnasts.

OBJECTIVES Muscular forces are an important determinant of bone strength, but bone may also adapt to non-muscular loading. We tested the hypothesis that loads associated with childhood gymnastics yield high arm bone mass (BMC), bone size and bone strength, independent of arm lean mass (FFM) and muscle cross-sectional area (CSA). METHODS Total body DXA and distal radius pQCT scans were performed on 33 post-menarcheal girls (19 ex/gymnasts, 14 non-gymnasts). Physical activity and calcium intake were assessed by questionnaire. For the non-dominant arm, pQCT measured bone strength indices and bone CSA (total, cortical) (4%, 33% sites); DXA measured arm FFM, arm BMC and skull BMC. Multiple regression analyses assessed gymnastic exposure, arm FFM, gynecological age and stature as predictors of bone parameters. RESULTS Bone outcomes at loaded upper extremity sites were 10-42% greater in ex/gymnasts than non-gymnasts. Gymnastic exposure remained a consistent, significant predictor of upper extremity skeletal parameters after accounting for the effects of muscle parameters, gynecological age and height. CONCLUSIONS Considering the effects of either arm FFM or muscle CSA, indices of bone mass, geometry and theoretical strength are disproportionately elevated after gymnastic exposure. Thus, non-muscular loading may be a distinct and important determinant of human skeletal structure.