Serum anti-Mullerian hormone as a surrogate for antral follicle count for definition of the polycystic ovary syndrome

Context: Despite its frequency, the polycystic ovary syndrome (PCOS) is still a difficult diagnosis in endocrinology, gynecology, and reproductive medicine. To help solve this issue, the Rotterdam consensus conference proposed to include the ultrasonographic follicle count as a new diagnostic criterion, in addition to hyperandrogenism and oligo-anovulation. Unfortunately, its assessment does not offer sufficient reliability worldwide. Objective: The aim of our study was to check whether anti-Mullerian hormone (AMH) measurement in the serum could be a surrogate for antral follicle count in the diagnostic criteria of PCOS. Design, Setting, and Patients: Serum AMH was measured with a second-generation immunoassay in a cohort of 73 PCOS patients and 96 controls, and its diagnostic power was evaluated by receiver operating characteristic curves. PCOS was diagnosed according to the Rotterdam definition. Results: Serum AMH levels were 3-fold higher in PCOS patients than in controls (81.6 vs. 33.5 pmol/liter; P < 0.001) and were significantly related to the follicle number in the two groups. The area under the receiver operating characteristic curve for the AMH assay was 0.851, indicating a good diagnostic potency. Setting the threshold at 60 pmol/liter offered the best compromise between specificity (92%) and sensitivity (67%). Conclusions: The serum AMH level is an accurate marker of the ovarian early antral follicle number and offers a good diagnostic potency. In situations where accurate ultrasonographic data are not available, AMH could thus be used instead of the follicle count as a diagnostic criterion and incorporated as such in the Rotterdam definition of PCOS.