Background The optimal endometrial preparation protocol for frozen embryo transfer (FET) remains controversial, with different cycle regimens and luteal phase support strategies across studies yielding conflicting results. This study aimed to compare the pregnancy outcomes of modified natural cycles (mNC) versus hormone replacement therapy (HRT) cycles, both with intensive luteal support using vaginal micronized progesterone and oral dydrogesterone. Methods This retrospective cohort study included 2365 FET cycles (1892 HRT and 473 mNC) in ovulatory women. Both groups received vaginal progesterone (800mg/day) and oral dydrogesterone (30mg/day) from the day after ovulation trigger or upon progesterone initiation. Propensity score matching was used to balance baseline characteristics, resulting in 1419 HRT and 473 mNC cycles for analysis. Treatment effect estimates with 95% confidence intervals were estimated using appropriate regression models. Results The propensity score-matched population had similar live birth rate (34.7% in the mNC group and 34.8% in the HRT group; aRR 1.02, 95% CI 0.80–1.29), pregnancy rate (54.3% vs 51.3%), clinical pregnancy rate (42.9% vs 42.0%), ongoing pregnancy rate (35.5% vs 35.7%), and miscarriage rate (7.8% vs 7.1%). There were no significant differences in multiple pregnancy rates, gestational age at delivery, birthweight, preterm birth rates between the two protocols. Conclusions In ovulatory women undergoing FET with intensive luteal phase support, the use of HRT or mNC for endometrial preparation yields comparable pregnancy and live birth rates.
