GnRH-a use before programmed frozen embryo transfer cycles for women with PCOS: a retrospective cohort study.

Background: Programmed protocols are most commonly chosen as endometrial preparation for women with polycystic ovarian syndrome (PCOS) undergoing frozen-thawed embryo transfer (FET) cycles. However, the efficacy of gonadotropin-releasing hormone agonist (GnRH-a) pretreatment before programmed cycles is still up for debate. This study was to compare the pregnancy and perinatal outcomes of PCOS patients receiving programmed cycles with and without GnRH-a pretreatment as endometrial preparation in FET cycles.

Methods: This is a retrospective cohort study conducted in the Reproductive Medicine Centre of Tongji Hospital. The primary analysis included 2733 FET cycles (223 were programmed cycles combined with GnRH-a pretreatment; 2510 were programmed cycles) during Jan. 2016 and Sept. 2022 from 1934 women with PCOS. Patients who had undergone both endometrial preparation protocols were further analyzed as a subgroup. The primary outcomes were pregnancy outcomes including live birth rate, clinical pregnancy rate, biochemical pregnancy loss rate, ectopic pregnancy rate, and multiple pregnancy rate. The secondary outcomes were perinatal outcomes. Propensity score matching (PSM) and generalized estimating equation were employed to eliminate essential confounders and account for patients with multiple cycles. The subgroup analysis included patients who underwent both endometrial preparation regimens and utilized the Wilcoxon's matched pairs test to compare the adjusted pregnancy outcomes rate, calculated by dividing the number of pregnancy outcomes by the number of cycles.

Results: The essential baseline variables of the patients were balanced after conducting PSM. Pregnancy outcomes of the total PCOS population exhibited no variances (P > 0.05) between protocols after adjustments. When focusing on patients who had received both protocols, GnRH-a administration was associated with increased adjusted live birth rates (P < 0.001), singleton live birth rates (P < 0.001), multiple live birth rates (P = 0.049), clinical pregnancy rates (P < 0.001), and lower miscarriage rates (P = 0.028). Further analysis of these patients indicated that the pregnancy outcomes of therapy with GnRH-a were superior to those without only in the second transfer cycle. No significant difference was exhibited in singleton perinatal outcomes in terms of gestational age, birth weight, delivery mode, gender, obstetric complications, and adverse birth outcomes between the two groups (P > 0.05).

Conclusions: GnRH-a pretreatment before programmed cycles may not affect pregnancy or perinatal outcomes of general women with PCOS in FET cycles but may be beneficial for PCOS patients who did not achieve a live birth during the first cycle receiving a programmed protocol. The conclusion should be considered with caution. Further well-designed studies are required to validate our findings.