We may conclude that after adjustment, leptin is the only adipokine of those measured that correlates to femoral neck and lumbar spine BMD and femoral neck BMC in the rhythmic gymnast group.

The aim of this study was to determine the relationships of bone mineral density (BMD) and content (BMC) with insulin-like growth factor-1 (IGF-1), IGF-binding protein-3 (IGFBP-3) and estradiol in pubertal female athletes. The participants were 170 healthy adolescent girls (13-15 years) who participated in competitive extramural athletic programs, i.e., sports games (n = 49), track sprinting (n = 24), rhythmic gymnastics (n = 23), swimming (n = 24) and cross-country skiing (n = 17). The control group (n = 33) consisted of girls who took part only in compulsory physical education classes at school. The whole-body BMD and femoral neck and lumbar spine BMD and BMC were measured using DXA, and the volumetric BMD was calculated. Venous blood samples to determine the concentration of IGF-1, IGFBP-3 and estradiol were drawn after an overnight fasting. After adjusting for age, body height and body mass, the relationships among BMD variables, IGF-1 and the IGF-1/IGFBP-3 molar ratio remained significant only in the rhythmic gymnast group. BMDs at the femoral neck and lumbar spine were also related to estradiol levels (r = 0.45-0.60; p < 0.05) only in the rhythmic gymnast group. No relationships were found among the measured BMD, IGF axis and estradiol in other athletic groups. Only BMC at the femoral neck remained associated with the IGF-1/IGFBP-3 molar ratio in the rhythmic gymnast group after adjusting for age, body height and body mass. Stepwise multiple regression analysis indicated that IGF-1 and estradiol together explained 42.6% (R(2) x 100) of total variance in the femoral neck BMD and IGF-1 alone 35.4% (R(2) x 100) of the total variance in the femoral neck BMC only in the rhythmic gymnast group. We conclude that femoral neck and lumbar spine BMD correlated with IGF-1, IGF-1/IGFBP-3 molar ratio and estradiol in rhythmic gymnasts. No relationships were found between bone parameters and the hormones used in other athletic groups.

Results suggest that the reported increase in the risk of diabetes in subjects that were G allele carriers at site 45 in obese populations compared with normal-weight populations can be linked instead to a change in muscle mass or the muscle itself present in this genotype group.