Abstract
STUDY QUESTION: What is the impact of endometrial fluid (EF) on single, euploid frozen embryo transfer (FET) cycles on live birth rate (LBR) and is cycle cancellation for EF a worthwhile intervention?
SUMMARY ANSWER: The LBR of single euploid FETs was significantly lower by 20.2 percentage points when EF was persistent on the day of decision for progesterone start/trigger, but despite the lower LBR, cycle cancellation may not confer an improved chance at live birth.
WHAT IS KNOWN ALREADY: The incidence of EF in cycles ranges from 3% to 8%, thus, only a few small studies have been performed to evaluate its impact. Existing literature generally concludes that the presence of EF leading up to an embryo transfer is detrimental to successful implantation, however, these studies examined untested embryos.
STUDY DESIGN, SIZE, DURATION: A retrospective cohort study was performed at a single, academically affiliated infertility center in the USA from January 2014 to December 2022. Inclusion criteria comprised patients who underwent their first IVF cycle, had pre-implantation genetic testing for aneuploidy performed by trophectoderm biopsy, had at least one euploid embryo, and were undergoing their first FET. Cycles were subdivided into three groups: no EF present in the cycle (no EF group), EF present but resolved prior to the day of decision for progesterone start/trigger (EF resolved group), and lastly, EF persistent on the day of decision for progesterone start/trigger (EF persistent group). Clinical outcomes were compared between the groups. In a secondary analysis, all single, euploid FET cycles that were cancelled due to EF were identified. The first subsequent completed single, euploid FET during the same study period was identified and analyzed.
PARTICIPANTS/MATERIALS, SETTING, METHODS: Four thousand three hundred eight FET cycles met inclusion criteria. Four thousand one hundred forty documented no EF, 108 documented EF that resolved, and 60 documented EF that persisted. The primary outcome was LBR per ET. A logistic regression analysis was performed adjusting for baseline characteristics (age, BMI, gravidity, parity, reason for infertility) and cycle characteristics (method of fertilization, protocol, endometrial thickness achieved during FET, embryo grade/day cryopreserved). In our secondary analysis, 90 single, euploid FET cycles were identified as cancelled specifically due to the presence of EF. Following these 90 cancelled cycles, there were 58 cycles that were identified as the first subsequent completed FET after cancellation. For the first subsequent completed single, euploid FET after index cycle cancellation, the presence of EF and overall LBR were recorded.
MAIN RESULTS AND THE ROLE OF CHANCE: When EF was present, but resolved prior to decision for progesterone start/trigger, the LBR was 9.4 percentage points lower compared to the no EF group, however, this did not reach statistical significance (49.1% vs 58.5%, aOR 0.71 (95% CI 0.47, 1.05)). When EF was persistent on ultrasound on day of decision for progesterone start/trigger, the LBR was significantly lower by 20.2 percentage points when compared to the no EF group (38.3% vs 58.5%, aOR 0.50 (95% CI 0.28, 0.88)). Of 58 subsequent FET cycles identified after initial FET cancellation, 23 demonstrated EF recurrence (but were not cancelled). The overall LBR for all 58 subsequent cycles was 39.7% which was not significantly different from the EF persistent group (aOR 0.99 (95% CI 0.38, 2.64)).
LIMITATIONS, REASONS FOR CAUTION: The retrospective, single center design may limit generalizability. Analysis of only euploid embryos introduces selection bias. Challenges of studying EF include the variable and unknown etiology of EF and non-uniform documentation of quantity/measurement of EF. Given the low incidence of EF, the sample sizes are small in the EF groups but are similar in size to previous studies. Power analysis was conducted and to achieve an 80% chance of finding a 10% difference in LBR, 385 patients would be needed in each group. In light of the actual LBR difference of 20.2 percentage points (58.5% vs 38.3%), our sample size of 60 patients in the EF group provided an 88% chance of finding a statistically significant difference.
WIDER IMPLICATIONS OF THE FINDINGS: The LBR was 9.4 percentage points lower when EF resolved and 20.2 percentage points lower when EF was persistent. We identified a small cohort of subsequent cycles after initial cycle cancellation and the LBR achieved in these cycles was similar to the LBR in the index transfer cycle if the transfer was performed even in the setting of persistent EF at time of progesterone start/trigger; perhaps if EF is persistent, cycle cancellation may not confer an improved LBR. Overall, many cases of EF likely result from underlying endometrial pathology. Patients should be counseled accordingly.
STUDY FUNDING/COMPETING INTEREST(S): No funding was used for this study. The authors have no conflicts of interest.
TRIAL REGISTRATION NUMBER: N/A.