SAN FRANCISCO – An anti-Müllerian hormone level of 3.4 ng/mL or greater identified polycystic ovary syndrome in a study of 31 nonobese adolescents.
That cutoff was determined to best discriminate between PCOS and controls, with a positive predictive value of 75% and a negative predictive value of 61% in a study of 15 nonobese adolescents with PCOS aged 13-21 years, and 16 controls, reported Dr. Aviva Sopher at the Endocrine Society’s Annual Meeting.
The goal of the study wasn’t to define a definitive anti-Müllerian hormone (AMH) cutoff; that may come later as Dr. Sopher’s group and others continue to investigate the matter. Instead, the project was a preliminary proof-of-concept effort to gauge the utility of AMH in adolescent PCOS diagnosis.
Dr. Aviva Sopher
For now, because there was "overlap in AMH values between PCOS and controls" and "a normal adolescent girl with polyfollicular ovaries and no other symptoms can have an AMH in the range that we think of [as signifying] PCOS, I wouldn’t use AMH on its own. I am suggesting the use of AMH in conjunction with clinical presentation and lab work," said Dr. Sopher, a pediatric endocrinologist at Columbia University in New York.
The hope, however, is that AMH will eventually replace the need for ultrasound; the transabdominal approach is "suboptimal" in adolescents, and transvaginal ultrasound is "overly invasive in this age group," she said.
The hormone is produced by growing follicles and is a marker of their number. Blood levels were assessed in the study by enzyme-linked immunosorbent assay (ELISA). PCOS was diagnosed by National Institutes of Health criteria (Fertil. Steril. 2010; 93:1938-41).
AMH was significantly higher in subjects with PCOS (4.4 ng/mL) than in controls (2.4 ng/mL), and correlated significantly with average ovarian size, the appearance of polycystic ovaries, free testosterone, and androstenedione.
The PCOS participants were 1.5-fold more likely to have an AMH level of more that 3.4 ng/mL than were the healthy controls, and that cutoff had a positive predictive value for PCOS of 75% and a negative predictive value of non-PCOS of 61%.
Mean ovarian size was similar in both groups (7.1 cc in subjects with PCOS versus 6.7 cc in controls), as were body mass index z-scores (0.45 vs. 0.19) and percent body fat (36.6% vs. 34.2%). The differences were not significant.
The subjects were at least 2 years post menarche. Exclusion criteria included premature birth; other potential causes of hirsutism and irregular menses; and birth control pill use within 3 months of enrollment. Normal-weight girls were selected "to exclude the confounding factor of obesity," Dr. Sopher said.
AMH is a useful adjunct to diagnose adolescent PCOS, and it has the potential to replace ultrasound as a marker of follicle count, she concluded.
The authors said they had nothing to disclose. The study was funded by the National Institutes of Health.