OPTIMIZING FROZEN EMBRYO TRANSFER (FET) SUCCESS: IMPROVED SUCCESS IN NATURAL CYCLES WITH THE USE OF HCG TRIGGER DESPITE SIMILAR OVERALL OUTCOMES WITH GONADOTROPIN STIMULATED, NATURAL, AND PROGRAMMED CYCLES

Abstract

To evaluate reproductive outcomes in FET cycles following hormonally programmed cycles vs alternate regimens of gonadotropin stimulated cycles or natural or letrozole-induced cycles. Within natural and letrozole cycles, pregnancy rates with and without hCG trigger to induce ovulation were compared. Retrospective cohort study. All programmed, natural cycle, letrozole, and gonadotropin stimulated FETs from a single academic center from 2017 – 2020 were evaluated. Programmed cycles consisted of oral micronized estradiol 2 mg TID starting day 1 of menses followed by intramuscular progesterone. Gonadotropin stimulated cycles consisted of ovulation induction with FSH and/or hMG starting on Day 1-5 of menses followed by hCG injection (“trigger”), with vaginal or intramuscular progesterone in the luteal phase. Natural cycles involved no medication in the follicular phase, whereas letrozole cycles consisted of letrozole 5 mg cycle day 3-7. For both natural and letrozole cycles, transfer was timed either to the patient’s own ovulation as measured by serum LH and progesterone or by ovulation induced by hCG trigger. Vaginal progesterone was used for luteal support. For the primary analysis, natural and letrozole cycles were included together in the natural group. One way- ANOVA was performed to compare baseline variables between cohorts. Chi-square was used to compare pregnancy outcome variables. Fisher’s Exact test was used to compare two groups when appropriate due to sample size. A p value <0.05 was considered significant. A total of 784 FETs were included: 728 programmed FETs, 24 natural cycle FETs, and 32 gonadotropin stimulated FETs. No statistical difference was found between all 3 cohorts based on age, BMI, endometrial thickness, or number of embryos transferred. Clinical pregnancy rates (CPR) or implantation rates (IPR) did not differ between programmed, natural/letrozole cycles, or gonadotropin stimulated FETs (54.3% vs 41.7% vs 59.4%, p=0.4) (46.3% vs 42.3% vs 48.8%, p= 0.9) respectively. Live birth/ongoing pregnancy rates were also not different (43.3% vs 41.7% vs 56.3%, p=0.3). Additionally, miscarriage rates did not differ between groups (10.3% vs 0% vs 3.1% p = 0.1). A subset analysis of natural cycle FETs showed no difference in letrozole induction versus spontaneous ovulation CPR (40% vs 42.9%, p=0.99), however natural/letrozole cycles with hCG trigger were associated with an increase in CPR (75% vs 25%, p=0.032). Our center utilizes stimulated FET cycles as an alternative for patients who do not achieve appropriate endometrial growth with traditional programmed FET. Our findings suggest stimulated FET cycles can be a successful alternative for patients whose lining is not appropriate in a programmed FET cycle. We use letrozole or natural cycle FETs in patients for whom a programmed or stimulated FET is not appropriate, with similar success rates. However, our study suggests use of an hCG trigger may improve outcomes in natural or letrozole cycles.