In low AMH level, which protocol is best on IVF?

Abstract

Anti-müllerian hormone (AMH) is now considered one of the best predictive markers of ovarian response for ART and many clinical applications of AMH measurement has been proposed. Low AMH levels have been associated with poor ovarian response and may correlated to unsuccessful IVF outcomes. But, there is little result which protocol shows better outcomes in poor responder. The aim of this study was to find better IVF protocol for low AMH patients. Retrospective analysis. The study included 1413 IVF cycles from June 2010 to December 2012, controlled ovarian hyperstimulations were performed with GnRH agonist protocol (n=886) or GnRH antagonist protocol (n=527). The patients were divided into two groups according to serum AMH level (AMH<2.0 or >2.0 ng/ml). In each group, we analysed pregnancy rates in GnRH agonist and antagonist cycles. Patients’ age was higher in low AMH group than normo-high AMH group (38.0 vs. 34.5) and other demographics (infertility causes, period, etc.) had no differences. A total pregnancy rate was 48.5% (685/1413). In analysis of pregnancy rates between protocol groups, agonist group was 48.4% (429/886) and antagonist group was 48.5% (256/527), there was no significant difference (P=0.45). As we expected, low AMH group (AMH<2.0) showed significantly lower pregnancy rate than normo-high AMH group (33.0% [146/442] vs. 55.5% [539/971], p<0.001) regardless of protocol. In low AMH group (AMH<2.0), antagonist protocol group showed higher pregnancy rate than agonist group (38.5% [57/148] vs. 30.2% [89/294], P<0.0001). In normo-high AMH group (AMH>2.0), opposite result was shown (52.5% [199/379] vs. 57.4% [340/592], P<0.005). Our results showed that low AMH level could be considered as a predictor of poor outcome in IVF cycles. But, using GnRH antagonist protocol showed enhancing pregnancy rate in low AMH patients. In conclusion, if you encounter patients with low AMH, we recommend using GnRH antagonist protocol as good counterplan.