Anti-Müllerian Hormone and Follicle-Stimulating Hormone Are Poor Independent Predictors of Live Birth After Assisted Reproductive Technology.

Abstract

To query if anti-Müllerian hormone (AMH) and/or follicle-stimulating hormone (FSH) predict live birth at the University of Colorado Advanced Reproductive Medicine (CU ARM). This was a retrospective analysis using the Society for Assisted Reproductive Technology (SART) Clinic Outcome Reporting System database at CU ARM from 2017 to 2019 to identify the pregnancy outcomes of the initial fresh or frozen embryo transfer (FET) and their corresponding AMH and FSH. Fisher's exact tests were used to identify differences in pregnancy outcome by age group, and area under the receiver operator characteristic curves was used to quantify live birth prediction. A total of 1083 records from 557 patients were reviewed. After only including the first autologous transfer, 270 cycles were analyzed. Overall live birth (L/B) rate was 58.15% (157/270), which declined with increasing age group (p ≤ 0.01). Although AMH significantly decreased with increasing age (p < 0.001), it was not associated with pregnancy outcome (3.54ng/mL vs. 3.41ng/mL, p = 0.56); this relationship was unchanged after controlling for age in logistic regression models (p = 0.52). FSH was also not significantly related to pregnancy outcome (7.00IU/L vs 6.00IU/L, p = 0.15), and this relationship did not change after controlling for age (p = 0.61). Using AUC, the only variable predictive of live birth was age (p = 0.002). AMH and FSH are not associated with the probability of live birth. Only age was significantly associated with live birth in this series. AMH and FSH should therefore be used cautiously when counseling patients about ART outcomes.