Outcomes of frozen embryo transfers from a large monocentric cohort (2982 cycles): towards a preferential use of cycles with a corpus luteum for endometrial preparation.

INTRODUCTION

To determine if endometrial preparation protocols for frozen embryo transfers (FET) impact clinical and neonatal outcomes.

METHODS

A monocentric cohort study including all first autologous FET cycles of a single blastocyst from January 1st, 2020 to September 30th, 2023. The primary endpoint was live birth (LBR). Secondary endpoints included hCG-positive pregnancy (> 100 IU/l), ongoing pregnancy (OPR), miscarriage, birthweight and term. Outcomes were also analyzed by subgroup: Day 5/Day 6 embryos. Among the 2982 FET cycles, 1507 (50.5%) were artificial cycles (AC) and 1475 (49.5%) were cycles with a corpus luteum (CLC). Univariate and multivariate analyses were performed using logistic regression models adjusting on age at transfer, age at oocyte retrieval, body mass index (BMI), indication and duration of infertility, IVF/ICSI, AMH, embryo quality, ovulatory status and year of transfer.

RESULTS

Age at the time of transfer (35.3 vs. 36.0 years old) and age at oocyte retrieval (34.8 vs. 35.4 years old) were significantly lower in AC compared to CLC. After adjustment, CLC significantly increased LBR (aOR = 1.39 [1.12;1.73], P = 0.003) and OPR (aOR = 1.39 [1.12;1.74], P = 0.003) and significantly decreased miscarriage (aOR=0.63 [0.49;0.79], P = 0.0001) compared to AC. In subgroup analyses, Day 5 embryos had significantly higher LBR, OPR, hCG-positive rates and lower miscarriage rates compared to Day 6 embryos. Term of birth did not differ between the two groups.

CONCLUSION

Our study pleads for a preferential use of CLC protocols for endometrial preparation for FET in ovulatory women.