Background: The study was performed to validate the threshold for anti-Müllerian hormone (AMH) that suggests the presence of polycystic ovaries in women with polycystic ovary syndrome (PCOS). Methods: This prospective observational study included 59 infertile women: (A) 22 PCOS with hyperandrogenism (HA) and oligo-anovulation, (B) 15 suspected PCOS with either HA or oligo-anovulation and (C) 22 asymptomatic control women. The diagnosis of PCOS was made according to the NIH classification. Results: For serum AMH and follicle number, respectively, the areas under the receiver operating characteristic curve (A versus C) were 0.97 and 0.93. The best compromise between specificity (95% and 91%) and sensitivity (95% and 82%) was obtained with threshold values of 33 pmol/l and 13 follicles, respectively. Using a serum AMH > 33 pmol/l as a surrogate for either HA or oligo-anovulation in group B would lead to the diagnosis of PCOS in seven (46.6%) patients from this group. Conclusions: Our data confirms that the AMH assay has a high diagnostic potential, providing that an appropriate threshold is used. AMH measurement may be included as an ultimate diagnostic criterion for the diagnosis of PCOS when either HA or anovulation is missing and/or when no reliable antral follicle count can be obtained.
This prospective study had the objective to compare anti-Müllerian hormone (AMH) serum levels among women with endometriomas and those with other benign ovarian cysts and verify a possible correlation between the AMH post-operative decline and the achievement of pregnancy. The study included 57 women undergoing laparoscopic surgery. Serum levels of AMH were measured preoperatively and at 6 and 12 months postoperatively. AMH serum levels decreased in both endometriomas and other benign cysts, but significant recovery was reported only in endometriomas (p = .029). Through an interview, information on reproductive outcome was obtained. The secondary endpoint was to identify other predictors of spontaneous pregnancy onset. After surgery 27 women attempted to conceive. At 18-months follow-up, 12 of them got pregnant, with a live birth of 37%. We did not record statistically significant differences in reproductive outcome between women with AMH serum levels lower and higher than 1.1 ng/ml (41.7 and 53.3% respectively). There was no statistically significant difference in AMH serum levels between pregnant and non-pregnant women, neither before nor after surgery. AMH did not appear to have a predictive role on reproductive outcome.