Bone mineral density and bone turnover in relation to serum leptin, alpha-ketoglutarate and sex steroids in overweight and obese postmenopausal women.

OBJECTIVE: Recent studies have shown that parallel changes in body weight and bone mass can be partially mediated via circulating leptin. Therefore, among the hormones involved in bone and mineral metabolism, such as oestrogens, testosterone and parathormone, leptin has recently become a subject of considerable interest. The aim of this study was to assess associations between leptin, E(2), testosterone, dehydroepiandrosterone sulphate (DHEA-S), SHBG, alpha-ketoglutaric acid (AKG) and bone mineral density (BMD) and bone turnover markers in overweight and obese postmenopausal women.
DESIGN: Eighty healthy, postmenopausal Caucasian women were studied. BMD of the lumbar spine (L(2)-L(4)) and femoral neck regions were examined using the dual X-ray absorptiometry (DXA) method. Associations were evaluated in stepwise multiple regression analysis, including information on the possible confounders and effect modifiers, for example, age, years since menopause, height and weight.
RESULTS: Femoral neck BMD was positively correlated with weight (r = 0.52, P < 0.000001), body mass index (BMI) (r = 0.48, P < 0.000006), hipline (r = 0.48, P < 0.00006), waistline (r = 0.45, P < 0.00002) and DHEA-S (r = 0.36, P < 0.0008). Correlations of E(2), SHBG, testosterone and leptin, as well as biochemical markers of bone turnover with L(2)-L(4) and femoral neck BMD were not found. In the whole study group, significant predictors of L(2)-L(4 )BMD were BMI (beta = 0.35, P < 0.01) testosterone (beta = 0.27, P < 0.05) and osteocalcin (OC) (beta = 0.22, P < 0.05) (R(2) = 0.23), while predictors of femoral neck BMD were BMI (beta = 0.42, P < 0.001), testosterone (beta = 0.24, P < 0.05), E(2) (beta = 0.19, P < 0.05), as well as osteocalcin (beta = 0.20, P < 0.05) (R(2) = 0.41). In the subgroup with BMI 30-39.9, the significant predictors of both L(2)-L(4 )and femoral neck BMD were testosterone (beta = 0.32, P < 0.05, R(2) = 0.19; beta = 0.33, P < 0.05, R(2) = 0.29) and osteocalcin (beta = 0.34, P < 0.05, R(2) = 0.19; beta = 0.45, P < 0.01, R(2) = 0.29). In the subgroup with waist : hip ratio (WHR > or = 0.85, the predictor of L(2)-L(4 )BMD was E(2) (beta = 0.38, P < 0.05) (R(2) = 0.21), whereas the predictors of femoral neck BMD were BMI (beta = 0.29, P < 0.05) and testosterone (beta = 0.35, P < 0.01) (R(2) = 0.36).
CONCLUSION: The main endocrine variable predicting lumbar spine BMD in overweight and obese postmenopausal females was testosterone, while the main determinants of femoral neck BMD were both testosterone and E(2). No effect was found of serum leptin on examined indicators of bone status.