MON-206 Analysis of Concomitant AMH and Testosterone Results in 3213 Female Samples: Contrary to Common Belief, AMH and Testosterone Levels Are Not Elevated in Tandem in Most Women

Abstract

The most recent guidelines for PCOS (polycystic ovary syndrome) characterize Anti-Müllerian Hormone (AMH) as a developing marker for diagnosis of PCOS. Therefore an observational study was undertaken to look for patterns that would clarify the utility of AMH as a diagnostic tool for PCOS. Data included review of AMH, and testosterone (T), along with ICD10 codes. The four most common in order were: irregular menstruation, unspecified, encounter for screening for other suspected endocrine disorder, other ovarian dysfunction, and polycystic ovarian syndrome. Results from 3213 samples with both total T and AMH ordered were analyzed along with available demographic data. When 2924 results were analyzed from women between 18 and 50 years, and 9.0% had AMH>11 ng/mL, 8.1% had T >55 ng/dL, (where these cutoffs represent the upper limits of normal) while only 2.5% of patients had both elevated T and AMH. In 194 women 18-40 years old with ICD10 code E28.2 for PCOS elevated results were found in 20%for AMH and 62% for T, but only 6.2% for both. The T assay was certified for accuracy by CDC. The results were concordant with several studies concluding that women with PCOS have higher levels of both AMH and T. However in this analysis, coincident elevated AMH and Testosterone levels were present in only a minority of patients. Because age is an important variable for AMH, results were also analyzed in 5- groups: segregated into five age groups 18-25, 26-30, 31-40, 41-45, 46-50. The data underscore the importance of the previous observation that AMH naturally declines with age in reproductive women. Interestingly, a bimodal or trimodal pattern was noted in the distribution of AMH results from all age groups of women with elevated testosterone, but the pattern was not observed in patients with normal testosterone values. Previous studies have shown a strong correlation of AMH with antral follicle count, suggesting groups of patients with very few antral follicles, and groups of women with a high number of antral follicles based on elevated AMH results. We conclude that because AMH and testosterone seem to provide distinct information the levels of both may be useful in characterizing patients. Any PCOS diagnostic cut-off for AMH will need to be age related. Patterns of AMH and T levels in women with elevated testosterone suggest there may be multiple etiologies which are differentiated by AMH levels.