Outcomes after frozen embryo transfer failure: changing the protocol does not improve live birth.

OBJECTIVE

To study whether changing the frozen embryo transfer (FET) protocol for endometrial preparation impacts pregnancy outcomes for women who did not achieve a live birth after their initial FET.

DESIGN

Retrospective cohort study.

SUBJECTS: The study included 17,989 FET cycles after a prior failed FET that occurred from 2012 to 2022. There were four groups of patients studied: those who underwent a programmed FET after a failed initial programmed FET, those who underwent a natural FET after a failed initial natural FET, those who underwent a natural FET after a failed initial programmed FET, and those who underwent a programmed FET after a failed initial natural FET.

EXPOSURE: The subsequent FET protocol, after a failed initial FET.

MAIN OUTCOME MEASURES: The primary outcome was live birth. Secondary outcomes included clinical pregnancy and pregnancy loss.

RESULTS

For patients who failed an initial programmed FET, there was no difference in live birth for those who switched to natural FET protocol vs. those who repeated a programmed FET (42.1% vs. 40.7%, adjusted risk ratio [RR] 1.02, 95% confidence interval [CI] [0.97, 1.08]). Secondary outcomes of clinical pregnancy and pregnancy loss were also not statistically different. For patients who failed an initial natural FET, there was also no difference in live birth for those who switched to a programmed FET protocol vs. those who repeated a natural FET (39.9% vs. 38.2%, adjusted RR 1.06 95% CI [1.00, 1.13]). These findings were consistent in a subgroup analysis restricted to euploid transfers. In a planned subgroup analysis examining various types of natural FET, for women who initially failed programmed FET, live birth was higher for those who changed protocols to a true natural FET, vs. those who repeated a programmed FET (50.3% vs. 40.7%, adjusted RR 1.2, 95% CI [1.03, 1.39]).

CONCLUSION

For women who do not achieve a live birth after an FET in the setting of adequate endometrial preparation, there is no significant association with live birth, clinical pregnancy or pregnancy loss after changing or repeating the transfer protocol. Further research is needed to clarify the effect of changing to a true natural FET protocol.