Objective: The aim of this study was to assess the capability of a single measurement of antimullerian hormone (AMH) to predict menopause status in late reproductive-aged women.
Methods: A group of 147 women, naturally fertile, aged 40 to 50 years with regular menstrual cycles were selected from the Tehran Lipid and Glucose Study cohort. Participants were assessed three times at 3-year intervals (T1-T3), and their blood levels of AMH were measured. The World Health Organization classification was used to define menopause status. The area under the receiver operating characteristics curve was calculated to assess the ability of AMH at T1 to discriminate between women who have reached menopause status and those who have not. Cutoff points and their relevant sensitivity, specificity, and positive and negative predictive values were calculated.
Results: Of 147 women, menopause occurred in 60. With use of the AMH level at T1, the probability of an accurate prediction of not reaching menopause status within the next 6 years for women aged 40 to 50 years was 88% (area under the receiver operating characteristics curve, 0.88; 95% CI, 0.83-0.94; P < 0.001). A threshold of 0.39 ng/mL for AMH had the optimal combined sensitivity and specificity for prediction with a positive predictive value of 0.90 (95% CI, 0.81-0.96) and negative predictive value of 0.76 (95% CI, 0.65-0.86). Results for a slightly lower (0.365 ng/mL) and higher (0.49 ng/mL) AMH threshold had negligible effect. Stratified analysis for women aged 40 to 44 and 45 to 49 years produced similar results.
Conclusions: Of every 10 women in late reproductive age with AMH levels greater than 0.39 ng/mL, only one will reach menopause status within the next 6 years. A single AMH measurement is a good predictor for the onset of menopause.