LAUSR

Abstract Objective (1) To investigate anti-Mullerian-hormone (AMH) in adolescents with polycystic ovary syndrome (PCOS) compared to age-matched girls without PCOS; (2) to determine whether AMH is a valuable marker for diagnosing PCOS and to explore the best cutoff value in adolescents; (3) using a multivariate predictive model for the diagnosis of PCOS. Methods Ninety girls (aged 10–20 years) recruited during two years for the PCOS-group (n = 45) or control-group (n = 45). PCOS diagnostic criteria according guidelines of the Endocrinology Expert Group, Obstetrics/Gynecology Branch of the Chinese Medical Association. Serum AMH and other sex hormones were measured. Logistic regression analysis to estimate the odds ratio of AMH and other variables for the diagnosis of PCOS. Receiver operator characteristics (ROC) curve analysis was performed to reveal the diagnostic potential. Results (A) AMH was significantly higher in PCOS patients than in controls (10.21 ± 5.85 ng/ml vs. 4.31 ± 2.84 ng/ml, p < .001). In PCOS-group, total testosterone (TT), free testosterone (FT) and biologically active testosterone (BioT) were significantly higher than in controls (p < .001). (B) Logistic regression suggests that AMH and TT are correlated with the diagnosis of PCOS (p < .05). (C) ROC curve analyses demonstrated that the optimal value of AMH for predicting PCOS was 6.32 ng/mL, with 69.8% sensitivity and 80.5% specificity. Furthermore, AMH combined with TT can provide 83.7% sensitivity and 80.5% specificity for diagnosing PCOS in adolescents. Conclusions AMH may be a useful biomarker for the diagnosis of PCOS in Chinese adolescent girls. A cutoff value of 6.32 ng/mL best discriminated between PCOS patients and controls. Besides AMH a multivariate predictive model should include TT.