Discriminatory Value of Steroid Hormones on Polycystic Ovary Syndrome and Clustering of Hyperandrogenism and Metabolic Factors

We determined (1) if 11-oxygenated androgens better identify polycystic ovary syndrome (PCOS) diagnosis in women with obesity compared to total or free testosterone (T) and free androgen index; (2) how biochemical hyperandrogenism and metabolic factors cluster in a cohort of women with infertility and obesity. Women with obesity and PCOS comprised the study group (N = 132). Ovulatory women with obesity and idiopathic, tubal or male factor infertility were the control group (N = 83). Steroid hormones were measured by means of liquid chromatography tandem mass spectrometry. Receiver operating characteristic curves and principal component analysis were used. Women with obesity and PCOS had higher 11-ketotestosterone (11 KT) (1.22 nmol/L [0.84; 1.65] vs 1.05 [0.78; 1.35], P = .04) compared to controls, but not 11β-hydroxyandrostenedione 4.30 [2.87; 5.92] vs 4.06 [3.22; 5.73], P = .44). 11-ketotestosterone (area under the curve: 0.59) did not better discriminate PCOS in women with obesity compared to: total T (0.84), free T (0.91), and free androgen index (0.85). We identified 4 principal components (PCs) in the PCOS group (72.1% explained variance): (1) insulin resistance status; (2) blood pressure; (3) obesity; (4) androgen status and 4 PCs in the control group (68.7% explained variance) with variables representing metabolism being dispersed in component 2, 3, and 4. Eleven-oxygenated androgens do not aid in the diagnosis of PCOS in women with obesity. Insulin resistance is the strongest PC in the PCOS group. There is no major dominant characteristic that defines obese non-PCOS women.