SUN-021 Age-Specific Multiples of the Median Level of Serum Anti-Mullerian Hormone Is a Potential Marker for Diagnosis of Polycystic Ovary Syndrome

Abstract Serum anti-Mullerian hormone (AMH) levels are significantly higher in women with polycystic ovary syndrome (PCOS) than in normal ovulatory women. Different diagnostic cut-off values of AMH for discriminating women with PCOS from normal controls have been proposed. This is attributed partly to the different assay methods used with different calibration, as well as the age-related changes in serum AMH levels. We propose that it may be more appropriate to use age-specific multiples of the median (MoM) of AMH value instead of a “one for all ages” cut-off as a diagnostic threshold. Hence, we conducted a retrospective study to validate the performance of age-specific MoM of AMH value in the diagnosis of PCOS. We studied on a cohort of 751 women presented to the clinic for menstrual disorders or fertility treatment, including 473 women diagnosed with polycystic ovary syndrome by the Rotterdam criteria and 278 normal ovulatory controls. Their archived serum samples, collected at the early follicular phase, were retrieved and assayed for AMH by the automated Access AMH assay. The MOM AMH of each subject was calculated based on the age-specific reference ranges recently established by our group. Our results showed that MOM AMH was significantly higher in women with PCOS compared to controls (p<0.0001). When stratified into five-yearly age groups, there was no significant difference in MOM AMH (p>0.05) among women with PCOS aged 21-25, 26-30 and 31-35 years, but those aged 36-40 years had significantly higher MOM AMH (p<0.05) compared to the other younger age groups. Among the ovulatory controls, no significant difference was observed in MOM AMH among all the age groups (p>0.05). The area under the receiver-operator characteristic curve was 0.852 (95% CI 0.825-0.877) (p<0.0001) for discriminating women with PCOS from ovulatory controls by MOM AMH. The best cut-off value of MOM AMH was 1.44, and the corresponding sensitivity and specificity were 76% and 79% respectively. At the fixed specificity of 80% and the corresponding sensitivity of 73% (with positive and negative likelihood ratios of 3.8 and 0.33 respectively), the cut-off value of MOM AMH was 1.5. In conclusion, age-specific MOM AMH is a promising surrogate of antral follicle count in the diagnosis of PCOS.