Background The paradoxical role of the adipose tissue on bone homeostasis is increasingly recognized, resulting partly from the mechanical osteogenic stimulus imposed by fat mass, but also from the possible deleterious effect of the adipose tissue as a secretor of inflammatory cytokines and adipokines. Adipokines have been previously associated with bone strength, mainly in adult samples. However, particularly before peak bone mass attainment, the mediators of this process remain to be clarified.

Objectives To identify biological links between adiposity and bone mineral density in 13-year-old girls.

Methods We used data from 132 girls born in 1990, recruited in public and private schools in Porto, Portugal, and evaluated at 13 years old as part of a large cohort study. Anthropometric assessment included fat mass estimation by bioelectrical impedance. Bone mineral density was measured at the non-dominant forearm by dual-energy X-ray absorptiometry. Pubertal development status was estimated using menarche age. Serum concentrations of leptin, adiponectin, osteoprotegerin (OPG), receptor activator of nuclear factor kappa-B ligand (RANKL), collagen type 1 cross-linked C-telopeptide (CTX), and procollagen I N-terminal propeptide (PINP) were determined using commercially available enzyme-linked immunosorbent assays. We used principal components analysis with Varimax rotation to estimate correlations between the six serum parameters (log-transformed). The final number of components retained was defined according to eigenvalues. Associations between the resulting components and forearm bone mineral density (in mg/cm²) and fat mass (in kg) were estimated using menarche age-adjusted linear regression coefficients and 95% confidence intervals.

Results We identified three independent components (Comp1 to Comp3) which explained 65% of the variance in the data. Comp1 was defined by a direct correlation with leptin (0.67) and inverse with CTX (-0.63). Comp1 was positively associated with BMD (17.04; 95%CI: 9.33, 24.74) and fat mass (3.62; 95%CI: 2.58,4.65), translating possible coexistence of increased leptin with decreased resorption. Comp2 was defined by a direct correlation with PINP (0.70) and inverse with OPG (-0.67). Comp2 was directly associated with BMD (7.45; 95%CI: 0.00,14.9) but not significantly with fat mass (0.68; 95%CI: -0.40,1.78), probably reflecting a local environment favoring bone formation. Comp3 was defined by a direct correlation with adiponectin (0.64) and inverse with RANKL (-0.72). Comp3 was negatively associated with BMD (-10.71; 95%CI: -19.22,-2.20) and fat mass (-1.55; 95%CI: -2.81,-0.29), favoring a negative effect of adiponectin on bone turnover.

Conclusions Our results suggest that leptin and adiponectin are important mediators of independent and opposing fat-bone associations early in adolescence and that the different mechanisms underlying these associations may be observable at the bone turnover level.

Disclosure of Interest None Declared