Comparative assessment of five serum antimüllerian hormone assays for the diagnosis of polycystic ovary syndrome.

OBJECTIVE

To determine whether the different antimüllerian hormone (AMH) immunoassays on the market offer the same performance for the diagnosis of polycystic ovary syndrome (PCOS).

DESIGN

A total of 95 serum AMH samples were retrospectively evaluated for a period of 3 months in the same laboratory.

SETTING

Academic center laboratory.

PATIENT(S): Forty-eight control women with regular menses and no hyperandrogenism and 47 patients with classic PCOS (i.e., hyperandrogenism plus oligoanovulation) attending our department for infertility.

INTERVENTION(S): None.

MAIN OUTCOME MEASURE(S): AMH measurement using five commercial assays. Method comparison and evaluation of the diagnostic performance by receiver operating characteristic analysis.

RESULT(S): Values obtained with Gen II and AL-105i ELISAs were similar to those provided by EAI AMH/MIS, whereas automatic assays generated lower values. A significant mean difference was observed between Access Dxi (1.35 ng/mL) or Cobas (1.73 ng/mL) and EIA AMH/MIS ELISA. By ROC analysis each assay displayed similar efficiency for PCOS diagnosis. Sensitivities varied from 49% to 74% when setting the specificity at 92%. Cluster analysis run in the control group identified a subgroup of asymptomatic women with polycystic ovary morphology (PCOM). After exclusion of PCOM, the 95th percentile of controls was 4.2 ng/mL (30 pmol/L) with the automatic assays and 5.6 ng/mL (40 pmol/L) with the manual assays.

CONCLUSION(S): Performance of the different AMH assays for PCOS diagnosis is comparable, providing that different threshold values are used for manual and automatic assays. Measurement of serum AMH level appears as a robust tool for the definition of PCOM.