OR37-5 Anti-Mullerian Hormone (AMH) may be a useful adjunct in the diagnosis of PCOS in adolescents

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: OR37-Pediatric Endocrinology: HPG Axis Disorders
Clinical
Monday, June 17, 2013: 11:15 AM-12:45 PM
Presentation Start Time: 12:15 PM
Room 102 (Moscone Center)
Aviva B Sopher*1, Galina Grigoriev1, Tamara Cameo1, Diana Laura1, Jodi P Lerner1, R Jeffrey Chang2, Donald J McMahon1 and Sharon E Oberfield1
1Columbia University Medical Center, New York, NY, 2University of California - San Diego, La Jolla, CA
Introduction:The diagnosis of PCOS in adolescents is challenging as its diagnostic features, menstrual irregularity, hirsutism and acne, are common in this group. Early diagnosis of PCOS may prevent development of insulin resistance, dyslipidemia and ultimately type 2 diabetes mellitus, cardiovascular disease and the metabolic syndrome. AMH, a product of ovarian granulosa cells of growing follicles, may be a marker of follicle number, specifically small antral follicles, and is increased in PCOS. Imaging of the ovaries by ultrasound in adolescents is suboptimal transabdominally and invasive transvaginally. In contrast, AMH is measured in a routine blood sample, thus may be useful in the diagnosis of PCOS in this group.

Objective:To determine if nonobese adolescents with PCOS have higher AMH than controls, to examine hormones and features of PCOS that correlate with AMH, and to calculate an AMH value that discriminates between PCOS and controls.

Design/Methods: 31 adolescent girls (13-21 yr):15 PCOS(BMIz:0.45±0.79,%BF:36.6±8.4) and 16 controls (BMIz:0.19±.60, %BF 34.2±6.5) at least two years post-menarche had  Ferriman-Gallwey scoring (FG) for hirsutism, scale for acne (0-3), fasting blood for AMH and reproductive hormones, transabdominal pelvic ultrasound for ovarian size and polycystic appearance (PCO), and dual-energy x-ray absorptiometry (DXA, GE Lunar DPX(L) & Prodigy) for percentage body fat (%BF). PCOS was diagnosed by the NIH criteria. Group differences were evaluated by Student's t-test, hormonal value differences were adjusted for menstrual age by ANCOVA and relationships between AMH and PCOS markers in the entire group were evaluated by Pearson correlations. A discriminant analysis was performed to determine a cutoff for AMH between PCOS and controls

Results: AMH was higher in PCOS (4.4±3.4 ng/mL) than controls (2.4±1.3 ng/mL) (P<0.05) and correlated with average ovarian size (r=0.42), PCO (r=0.57), free testosterone (r=0.46) and androstenedione (r=0.42)(P<0.03). Ovarian size was similar in PCOS (7.1±2.6 cc) and controls (6.7±1.8 cc)(p=0.64). An AMH cutoff of 3.4 ng/mL had a positive predictive value for PCOS of 75% and a negative predictive value of non-PCOS of 61%. Those with PCOS  were 1.49 more likely to have AMH >3.4 ng/mL (confidence interval 0.98–2.26 ng/mL). However, a stepwise logistic regression analysis, which included AMH, DHEAS and FG showed FG to be the only variable to significantly predict PCOS (P=0.0128).

Conclusions: In this small group of nonobese adolescents, AMH was higher in PCOS than controls and correlated with androgens and PCO. Degree of hirsutism was the best predictor of PCOS. Our AMH cutoff value of 3.4 ng/mL is slightly higher than a prior proposed cutoff in adults of 2.8 ng/mL, reflecting the physiologic decline of AMH that begins at about age 25 years. Our data suggest that AMH may be a useful adjunct in the diagnosis of PCOS in adolescents.

Nothing to Disclose: ABS, GG, TC, DL, JPL, RJC, DJM, SEO

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm