P-669 The impact of luteal phase support on pregnancy outcomes in relation to serum progesterone levels on the day of frozen embryo transfer in natural cycles

Abstract Study question Is luteal phase support (LPS) beneficial in low and normal serum progesterone (P) levels the day of frozen embryo transfer (FET) in natural cycles (NC)? Summary answer LPS in case of low serum P on the day of NC-FET results in similar pregnancy rates as compared to patients with normal P levels. What is known already The use of FET cycles has increased exponentially in recent years. Endometrial preparation in FET cycles can be achieved either in a natural cycle, a stimulated cycle or in an artificial cycle (AC). There is increasing evidence that proper serum P levels around the time of FET are important in AC and NC. Recent studies in AC have shown that enhanced progesterone supplementation in case of low serum P levels could reduce early pregnancy loss rates and normalize live birth rates. Whether this rescue strategy could also be beneficial in NC-FET has not been investigated in detail. Study design, size, duration A retrospective study was conducted at a tertiary university-based referral hospital encompassing 459 unique patients who had a NC-FET with or without hCG trigger from November 2020 onwards. Only patients who had an in-house validated serum P measurement (Cobas 6000®,Roche) on the day of FET were included. Blastocyst transfer was performed 6 days after a spontaneous luteinizing hormone (LH) peak or 7 days after hCG administration. Ongoing pregnancy rate (OPR) was the primary outcome. Participants/materials, setting, methods Outcomes were compared in patients with low serum P (<10ng/dl) levels, who received LPS from the day of FET onwards (group 1) and patients with normal serum P (≥10ng/dl) levels (group 2). Group 2 was subdivided into patients who did not received LPS (group 2a) and patients who did (group 2b), as by practitioner’s choice. Multivariable regression analysis was performed adjusting for age, BMI, endometrial thickness, rank of FET, embryo quality and indication for treatment. Main results and the role of chance Mean ages were 31.8 years (group 1) and 33.3 years (group 2, p = 0.004). BMI was significantly lower in group 2 (23.7 kg/m2) compared to group 1 (26.7 kg/m2, p < 0.001). Indication for treatment, menstrual cycle length, anti-Mullerian hormone (AMH) levels and parity were comparable between both groups. An hCG ovulation trigger was administered in 26.0% of the cycles in group 1 and 20.4% in group 2 (p = 0.28). Endometrial thickness was equal in both groups (8.3mm, p = 0.84). In group 1, 40.3% of the patients had a transfer of a top quality embryo as compared to 46.3% in group 2 (p = 0.33). Positive hCG rate (54.6%, 64.0% and 65.0% for group 1, 2a and 2b respectively, p = 0.26) and miscarriage rate per positive hCG (2.4%, 9.2% and 6.7% for group 1,2a and 2b respectively, p = 0.37) were comparable between groups. OPR per started cycle was similar between groups (46.8%, 55.4% and 55.6% for group 1, 2a and 2b respectively, p = 0.37). Multivariable logistic regression analysis showed no relationship between OPR and the three different groups (p = 0.13). A significant association was observed between OPR and patient age at embryo cryopreservation. Limitations, reasons for caution Although only patients were included who had serum P levels on the day of FET measured by a standardized in-house assay, the results of the study are limited by the retrospective nature and its associated bias. Confirmation of the results should be obtained in prospective studies. Wider implications of the findings Luteal phase deficiency was observed in only 16.8% of patients in our study population. Administration of LPS in patients with low serum P levels on the day of FET could rescue pregnancy outcomes. However, starting LPS when serum P levels are normal appears to have no added value. Trial registration number not applicable