Abstract
OBJECTIVE: One of the main features of polycystic ovary syndrome (PCOS) is increased adipose tissue, which can result in hormonal disturbances. In the present study, we aimed to investigate which indicator of obesity could better associate with hormonal disturbances in PCOS women. METHODS: In this cross-sectional analysis, women with PCOS were included according to the Rotterdam criteria. Fasting blood samples were analyzed for biochemical, metabolic, and hormonal parameters. Anthropometric measures comprised body composition indices (assessed by bioelectric impedance analysis [BIA]), waist circumference, body mass index (BMI), and waist-to-height ratio (WHtR). Linear regression modeling was used to assess the association between anthropometric indices and hormonal imbalance, adjusted for age, mensuration status, and the homeostasis model assessment-estimated insulin resistance (HOMA-IR). Receiver operating characteristics (ROC) curves were utilized to ascertain the sensitivity, specificity, and optimal cut-off points of various anthropometric indices in identifying hyperandrogenism. RESULTS: A total of 129 PCOS women with a median (interquartile range [IQR]) age of 32.0 (23.0-32.0) years and a median BMI of 26.3 (23.00-29.70) kg/m(2) were enrolled. In the adjusted linear regression model, BMI (β = 0.053, P < 0.001), waist circumference (β = 0.021, P = 0.001), WHtR (β = 3.325, P = 0.002), total fat mass (β = 0.021, P = 0.002), trunk fat mass (β = 0.038, P = 0.006), and leg fat mass (β = 0.045, P = 0.004) were positively associated with free androgen index (FAI). In addition, BMI (β=-0.017, P = 0.003), waist circumference (β=-0.008, P = 0.002), WHtR (β=-1.167, P = 0.004), total fat mass (β=-0.008, P=0.003), trunk fat mass (β=-0.017, P=0.001), and leg fat mass (β=-0.018, P=0.004) were negatively associated with the serum level of sex hormone binding globulin (SHBG). WHtR showed the greatest area under the curve (AUC) value (AUC = 0.676, P = 0.001) for identifying hyperandrogenism (FAI ≥ 4.97 or total testosterone ≥ 0.7 ng/mL) in PCOS women with corresponding sensitivity of 87.30% and specificity of 39.70%. CONCLUSIONS: WHtR is related to hyperandrogenism in PCOS better than other anthropometric measures.