P-438 Does artificial endometrial preparation for frozen embryo transfers have poorer outcomes compared to other endometrial preparation?

Abstract

Abstract Study question Do artificial thaw cycles increase miscarriage rates compared to Letrozole and natural thaw cycles? Summary answer After adjusting for potential confounders, artificial thaw cycles had twice the risk of miscarriage compared to both natural and Letrozole thaw cycles. What is known already In frozen embryo transfers, miscarriage rates have been shown to be higher with artificial cycle regimen than with natural cycles whereas there is inconsistent evidence regarding live birth rates between the two types of endometrial preparation. Studies comparing artificial to Letrozole thaw cycles also gave inconsistent data as to whether one had better outcomes than the other. There is lack of studies comparing these three types of endometrial preparation with regards to pregnancy, miscarriage and live birth rates. Study design, size, duration This is a retrospective study on 712 frozen embryo transfers from August 2019 to March 2021 for women between 23 to 45 years old. The endometrial preparation distribution is as follow: 288 Letrozole cycles, 200 artificial cycles and 224 natural cycles. Only single blastocyst transfers were included. Exclusion criteria included women with endometrial cancer/hyperplasia, recurrent pregnancy loss, Mullerian anomalies or sexual dysfunction, cycles with preimplantation genetic testing and use of donor gametes. Participants/materials, setting, methods Anonymised data of patients undergoing frozen embryo transfers were obtained from the database in a fertility centre in Singapore. Primary outcome is miscarriage rate and secondary outcomes are pregnancy and live birth rates. Data was analyzed with Stata. Continuous variables of the different groups were compared using Student’s T-test while chi-square test was used for dichotomous variables. Logistic regression analysis was performed to assess the effect of independent variables on the outcomes. Main results and the role of chance Artificial thaw cycles had the highest miscarriage rate at 43.8%, compared to 25.3% in natural thaw and 33.0% in Letrozole thaw cycles (p = 0.028). Clinical pregnancy rate was highest in artificial thaw cycles at 48.0%, compared to 40.6% in natural thaw and 34.7% in Letrozole thaw cycles (p = 0.013). Live birth rate was highest in natural thaw cycles at 30.3%, followed by 27.0% in artificial thaw and 23.2% in Letrozole thaw cycles (p = 0.028). After adjusting for confounding factors, artificial thaw cycles have approximately twice the risk of miscarriage compared to Letrozole thaw (p = 0.007) and natural thaw cycles (p = 0.032). The chance of pregnancy was 1.5 times higher in artificial compared to Letrozole thaw cycles (p = 0.027) but similar to that in natural thaw cycles. The chance of a live birth in natural thaw cycle was 1.6 times higher than that in Letrozole thaw cycle (p = 0.027) but similar in artificial thaw cycle. The mean age of participants in the artificial group was significantly lower at 33.9 years, compared to 35.3 years and 35.5 years in the Letrozole and natural group respectively (p < 0.001). However in multivariate analysis, increasing age did not have a significant impact on either miscarriage (p = 0.907) or live birth rate (p = 0.066). Limitations, reasons for caution The retrospective nature of this study may have resulted in selection biases and unrecognized confounders may influence the results of this study. Wider implications of the findings A possible hypothesis which can explain higher miscarriage rates in artificial thaw cycles may be insufficient progesterone support due to the lack of corpus lutea unlike in natural or Letrozole thaw cycles. Optimizing progesterone support in women with low progesterone levels undergoing artificial thaw can be further explored. Trial registration number not applicable