P-534 Serum progesterone levels on the day of embryo transfer do not affect pregnancy outcomes after euploid blastocyst transfer: data from artificial cycles with intramuscular progesterone

Abstract

Abstract Study question Do serum progesterone levels on the day of embryo transfer (ET) have an impact on implantation and pregnancy outcomes? Summary answer Serum luteal progesterone levels were not associated with pregnancy outcomes under the condition of utilizing 60mg intramuscular progesterone for luteal support in artificial cycles. What is known already It was recognized that circulating progesterone level might be associated with frozen-thawed embryo transfer (FET) success, and a serum progesterone level above a certain threshold at the time of blastocyst transfer improved live birth rates (LBR) and reduced risk of miscarriage in artificial FET cycles. However, this hypothesis is controversial as the previous findings were based on data using vaginal progesterone as luteal phase support with various doses and timing of administration. Studies evaluating euploid-only FET cycles with intramuscular progesterone were lacking. Study design, size, duration The study incorporated infertile couples undergoing (preimplantation genetic testing, PGT) from 1st January 2018 to 31st July 2021. It is a retrospective cohort study including 771 patients who underwent single frozen-thawed euploid blastocyst transfer after an artificial endometrial preparation cycle with 60mg intramuscular progesterone daily for luteal support. Each patient only contributed one cycle per cohort. Participants/materials, setting, methods Patients with recurrent implantation failure, endometrium 7.0 mm and cycles with gonadotrophin releasing hormone analogue down regulation were excluded. Serum progesterone measurements were taken on the day of ET, approximately 20±2 hours after the last injection of progesterone. The primary outcome was LBR based on serum progesterone levels. Secondary outcomes were the relationship between serum progesterone levels and clinical pregnancy rate, miscarriage rate as well as obstetrical and neonatal outcomes. Main results and the role of chance The median of serum progesterone levels was 11.80 ng/ml [9.95, 14.60] (median/IQR) and the overall LBR, clinical pregnancy rate and miscarriage rate were 55.6%, 64.33% and 10.69% respectively. Patients with the lowest centile of serum progesterone level (≤P10, ≤8.2ng/ml) had a similar clinical pregnancy rate (62.0% vs 64.6%), live birth rate (50.6% vs 56.2%) and miscarriage rate (12.2% vs 10.5%) compared with the rest of patients. After dividing patients in deciles according to serum progesterone levels, no differences in LBR were observed among groups and no correlations were found between progesterone levels and the other pregnancy outcomes. Multivariate regression analysis confirmed that serum progesterone levels did not affect the LBR after adjusting for possible confounders (age, body mass index, PGT indication, primary infertility, history of miscarriage, endometrium thickness, day of blastocyst and embryo quality) with adjusted (OR) 0.99, 95%CI 0.96-1.03), while the day of blastocyst (D6 vs. D5: aOR 0.47, 95%CI 0.33-0.65) as well embryo quality (good quality embryo vs. available embryo: aOR 1.88, 95%CI 1.38-2.58) were associated with LBR independently. Similarly, serum progesterone in their lowest levels did not negatively impact other pregnancy outcomes (ex. clinical pregnancy rates, miscarriage rates) and the perinatal outcomes. Limitations, reasons for caution This is a retrospective cohort study and the results only apply for patients under artificial cycles with intramuscular progesterone. Moreover, the time interval between the last administration of progesterone and the blood test was not controlled in the present study. Wider implications of the findings Serum progesterone level may not influence the FET outcomes independently in an artificial cycle with intramuscular progesterone. A daily dose of 60mg progesterone intramuscularly for luteal support is sufficient and the measurement of serum progesterone does not need to be performed in advance for most of the patients. Trial registration number not applicable