O-129 Characteristics of serum anti-mullerian hormone levels in women with polycystic ovarian syndrome (PCOS) compared to those without PCOS: a propensity score matched study

Abstract Study question What is the optimal serum anti-Mullerian hormone (AMH) cut-off for diagnosing PCOS, and what factors influence AMH value variations in women with PCOS? Summary answer AMH cut-off for PCOS diagnosis was 5.055 ng/mL (sensitivity=85.4%,specificity=75.8%). In PCOS women, AMH levels declined more slowly with age and correlated with serum testosterone concentrations. What is known already Elevated serum AMH levels are a hallmark of PCOS, yet a widely accepted diagnostic cut-off remains controversial. PCOS is characterized by diverse endocrine and metabolic disorders, which can influence AMH concentrations. Study design, size, duration This case-control study included comprehensive sampling from medical records of women who visited our fertility clinics over the past five years (2019–2024) and underwent AMH testing (n = 8233). Participants were allocated into PCOS (n = 1200) and non-PCOS groups (n = 7233). PCOS was diagnosed following the Rotterdam criteria. Age and BMI were recorded for all patients. For women with PCOS, additional data on clinical characteristics, serum hormonal profiles, metabolic parameters, and polycystic ovarian morphology on ultrasound were routinely collected. Participants/materials, setting, methods Propensity score matching (PSM) in a 1:1 ratio was performed using age and BMI as covariates. Receiver operating characteristic (ROC) curve analysis and the Youden Index were used to determine the optimal cut-off point of AMH for the diagnosis of PCOS. Age-specific AMH values were compared between the two groups by analyzing difference in slope changes. Multiple linear regression analysis was conducted to identify factors associated with AMH distribution among PCOS women. Main results and the role of chance After PSM, 1162 PCOS patients and 1162 controls were included in the final analysis. Serum AMH levels demonstrated diagnostic relevance for PCOS (AUC=0.88, p < 0.05), with 5.055 ng/dL identified as the optimal cut-off (sensitivity=85.4%, specificity=75.8%). AMH levels decreased progressively with age in the study cohort, with a slope coefficient of -0.053 (95%CI: -0.103 to -0.003, p < 0.001). There were differences in age-specific AMH distribution between the two groups. The age-related decline in AMH levels was less pronounced in PCOS women compared to controls, as evidenced by a significant difference in slope coefficients (0.125, 95%CI: 0.044 to 0.205, p = 0.003). Multiple linear regression analysis was performed to identify factors influencing AMH levels in women with PCOS, using clinical characteristics, laboratory tests, and ultrasound findings as covariates. Serum AMH levels were positively associated with total serum testosterone levels (B-value=0.46, 95%CI: 0.07–0.85, p = 0.02). Prolactin concentrations (B-value=0.04, 95%CI: -0.07–0.85, p = 0.09) and the presence of polycystic ovarian morphology on ultrasound (B-value=1.16, 95%CI: -0.09–2.42, p = 0.07) showed a trend toward influencing AMH levels, albeit not statistically significant. Limitations, reasons for caution Older women constituted a small proportion of the PCOS population (n = 103 aged ≥35 years; n = 21 aged ≥38 years), potentially limiting the analysis of AMH-age correlations in this study. Wider implications of the findings These findings provide an AMH cut-off for PCOS diagnosis and suggest that endocrine disorders play a critical role in regulating AMH activity in women suffering from this condition. With advancing age, serum AMH levels appear to be better preserved in women with PCOS compared to their non-PCOS counterparts. Trial registration number No