Sex-specific Associations Between Muscle Performance And Bone Mineral Density During Adolescence And Young Adulthood

Abstract

Measures of muscle size are considered the best predictors of areal bone mineral density (aBMD). However, the role of muscle function may be particularly important for public health since muscle function can be improved via the modifiable factor of physical activity; whereas, muscle size is heavily influenced by the non-modifiable factor of genetics. PURPOSE: The aim of this study was to analyze the contribution of muscle mass in the relationship between muscle function and aBMD (g/cm2) of the whole body, hip, and right arm. METHODS: The sample included 462 participants from 9 to 20 yrs (212 girls:13.7±2.3yrs; 250 boys:13.8±2.4yrs) without clinical conditions affecting muscle or bone. Bone variables were assessed by DXA. Muscle function was operationalized as the maximum power per body mass (W/kg) measured during a vertical counter movement jump on a force platform and the maximum handgrip strength (kg) assessed with a dynamometer. Muscle size (kg) was operationalized as appendicular lean soft tissue (aLST) from the total DXA body scan. Mediation model effects were estimated using the PROCESS macro, and 95% bootstrap confidence intervals were constructed to examine the indirect (mediated by muscle size) effects of muscle function on bone outcomes. Body height and maturity offset were used as covariates. RESULTS: The mediation analyses indicated an indirect effect of vertical jump power on aBMD of WBLH and total hip through aLST in both boys and girls (p<0.001). In boys, aLST explained 59.0% (β=0.001 95%CI [0.001, 0.002]) and 33.7% (β=0.002 95%CI [0.001, 0.003]) of the total effects of vertical jump power on aBMD of WBLH and total hip, respectively. In girls, the percentages of the total effects of vertical jump power on aBMD explained by aLST were 69.3% (β=0.001 95% CI [0.001, 0.002]) at the WBLH and 51.9% (β=0.002 95%CI [0.001, 0.003]) at the total hip. Associations between handgrip strength and aBMD of the right arm were also partially explained by aLST, specifically 31.6% (β=0.001 95%CI[0.001, 0.002]) in boys (p<0.001) and 55.2% (β=0.002 95%CI[0.001, 0.003]) in girls. CONCLUSION: The higher proportion of explained variability due to muscle size that we report in girls (when compared to boys) suggests that girls may be missing out of the “value added” osteogenic effect of muscle function associated with physical activity.