Relationships of Bone Mineral Variables with Body Composition, Blood Hormones and Training Volume in Adolescent Female Athletes with Different Loading Patterns

The aim of this investigation was to determine the relationships of areal bone mineral density (aBMD) and content (BMC) with body composition, blood hormone and training load variables in adolescent female athletes with different loading patterns. The participants were 73 healthy adolescent females (14–18 years), who were divided into three groups: rhythmic gymnasts (RG; n = 33), swimmers (SW; n = 20) and untrained controls (UC; n = 20). Bone mineral and body compositional variables were measured by dual-energy X-ray absorptiometry, and insulin-like growth factor-1 (IGF-1), estradiol and leptin were analyzed from blood samples. In addition, aerobic performance was assessed by a peak oxygen consumption test. No differences (p > 0.05) in weekly training volume were observed between rhythmic gymnasts (17.6 ± 5.3 h/week) and swimmers (16.1 ± 6.9 h/week). Measured areal bone mineral density and bone mineral content values were higher in rhythmic gymnasts compared with other groups (p p > 0.05) in measured bone mineral values were seen between swimmers and untrained control groups. Multiple regression models indicated that IGF-1 alone explained 14% of the total variance (R2 × 100) in lumbar spine aBMD, while appendicular muscle mass and training volume together explained 37% of the total variance in femoral neck BMC in the rhythmic gymnast group only. In swimmers, age at menarche, estradiol and appendicular muscle mass together explained 68% of the total variance in lumbar spine BMC, while appendicular muscle mass was the only predictor and explained 19 to 53% of the total variance in measured bone mineral values in untrained controls. In conclusion, adolescent rhythmic gymnasts with specific weight-bearing athletic activity present higher areal bone mineral values in comparison with swimmers and untrained controls. Specific training volume together with appendicular muscle mass influenced cortical bone development at the femoral neck site of the skeleton in rhythmic gymnasts, while hormonal values influenced trabecular bone development at the lumbar spine site in both athletic groups with different loading patterns