AMH is a Better Predictor of Live Birth in IVF Cycles When AMH and FSH Values Are Discordant [40F

Abstract

INTRODUCTION: Follicle stimulating hormone (FSH) and anti-Mullerian hormone (AMH) assist physicians in assessing ovarian reserve during infertility evaluations and in vitro fertilization (IVF) cycles. Despite usual congruity between these biomarkers, physicians are often faced with discordant values. Our study investigates which biomarker better predicts live birth (LB) in IVF patients with discordant FSH and AMH results using the Society for Assisted Reproductive Technology Clinical Outcomes Reporting System (SART CORS) database. METHODS: This IRB-approved retrospective cohort study compared live birth rate (LBR) with FSH and AMH values in patients who completed at least one cycle of autologous IVF as reported to SART CORS in the years 2013 and 2015. Chi-square test was used to compare the incidence of LB among the study cohorts. RESULTS: 44,696 patient charts met inclusion criteria. LBR was higher in patients with elevated FSH (>10mIU/mL) and normal AMH (>1 ng/dL) opposed to patients with normal FSH (<10) and low AMH (<1) (39% versus 26%, P<.001). In a multivariate logistic regression model of patients with normal FSH, low AMH had a lower risk of LB than patients with normal AMH, with relative risk of .87 (95% CI .83-.91). IVF cycle cancellation was higher in patient with normal FSH and low AMH compared to elevated FSH and normal AMH (30% versus 8%, P<.001). CONCLUSION: AMH is a stronger predictor of live birth in patients undergoing IVF when FSH and AMH values are discordant. Lower AMH is also associated with higher IVF cycle cancellation rates than those seen with elevated FSH values.