Can AMH levels predict the need to step up FSH dose for controlled ovarian stimulation following a long GnRH agonist protocol in PCOS women?

Abstract

Background: To explore the role of anti-Mullerian hormone (AMH) in predicting the need to step up recombinant FSH (rFSH) dose following long GnRH agonist protocol in IVF/ICSI cycles of polycystic ovarian syndrome (PCOS) women. Methods: This is a retrospective cohort study of 825 PCOS women undergoing long GnRH agonist protocol enrolled from Jan 2019 to Dec 2021. The daily rFSH dose at which the first response to rFSH were recorded. The dose at which the first response to rFSH was based on folliculometry during follow up in which two or more follicles reached ≥ 11 mm. A receiver operating characteristic (ROC) curve analysis was done to investigate the ability of AMH to predict the need to step up initial rFSH dose. Results: PCOS women who needed to step up initial rFSH dose had a significantly higher AMH compared with those didn’t step up initial rFSH dose (11.37 ± 3.25ng/ml vs. 8.69 ± 3.16ng/ml, p < 0.001). In multivariate logistic regression analysis, increased AMH level was an independent factor for the need to step up initial rFSH dose in PCOS patients after adjusted for confounding factors. ROC curve analysis showed AMH could predict the need to step up initial rFSH dose (AUC = 0.738, 95%CI: 0.704–0.773), having 75.4% specificity and 63% sensitivity when the threshold AMH concentration was 9.30ng/ml. 58.8% PCOS women with AMH > 9.30 ng/ml required increased rFSH dose compared to 18.8% of women with AMH ≤ 9.30ng/ml (p < 0.001). Although the clinical pregnancy rate and live birth rate were not significantly different, there was a higher incidence of OHSS among women with AMH > 9.30 ng/ml vs. AMH ≤ 9.30ng/ml (20.8% vs. 15.3%, p = 0.043). Conclusion: PCOS women with AMH > 9.30 ng/ml were resistant to rFSH stimulation and require increased dose for the cycle recruitment of ovarian follicles.