Comparison of two different methods for measuring anti-mullerian hormone in a clinical series

Background: Anti Mullerian hormone (AMH) has previously been measured using a manual method, but a fully automated assay from Roche Diagnostics was recently introduced. The aim of this study was to compare the results from the AMH gen II ELISA and Elecsys Cobas AMH methods in a clinical setting to evaluate whether the assays achieve the goals of analytical performance. A prospective observational study with 23 women seeking laparoscopic sterilization was conducted. Blood samples were collected preoperatively as well as 1 week and 1, 3 and 6 months postoperatively; they were evaluated with the AMH gen II ELISA and Elecsys Cobas AMH methods. The assays were validated according to the optimal performance of biochemical assays: CVAnalytical < 0.25* CV (Within Biological Variation). Findings: We found a good correlation between the two methods; there was a bias of approximately 32 %. The total within-person biological variability ranged from approximately 21 to 32 %. The analytical variability of the AMH gen II ELISA and Elecsys Cobas methods ranged from 5.5 to 10.3 % and 2.8 to 3.3 %, respectively. Applying the goals for optimal assay performance, the Elecsys Cobas method achieved optimal performance throughout the measuring range, whereas the AMH Gen II only achieved optimal performance in the high end of the measuring range. Furthermore, the Elecsys Cobas assay had a low limit of quantitation of 0.5 pmol/l compared to 3.0 pmol/l for the AMH gen II ELISA. Conclusions: In the clinical setting, the Elecsys Cobas AMH assay performs well according to the optimal standard for biochemical assays.