Endometrial preparation: effect of estrogen dose and administration route on reproductive outcomes in oocyte donation cycles with fresh embryo transfer.

STUDY QUESTION: Is there a difference in live birth rates following endometrial preparation with either a constant or increasing estrogen dose in fresh embryo transfer from oocyte donation cycles? SUMMARY ANSWER: There is no difference in live birth rates between a constant dose versus an increasing dose of estrogen after fresh embryo transfer in oocyte donation cycles with oral or transdermal supplementation. WHAT IS KNOWN ALREADY: Endometrial preparation (EP) with estrogen and progesterone, and embryo-endometrial synchronicity are determinant for adequate embryo implantation. Estrogen is crucial and different exogenous administration patterns could imply variations on EP. Moreover, estrogen undergoes metabolization by the intestines and liver when administered orally, an effect that is bypassed by transdermal administration. Information on the effect of replacement patterns and route of administration of E on reproductive outcomes of women undergoing fresh embryo transfer from oocyte donation cycles is scarce. STUDY DESIGN, SIZE, DURATION: Retrospective cohort study including 8362 embryo transfers following ICSI, corresponding to 8254 patients, between October 2010 and March 2015. A total of 5593 (66.9%) patients received an increasing E dose (ID) (oral: 2 mg/day day(d)1-7, 4 mg days d8-12, 6 mg d13-embryo transfer; transdermal: 75 µg/3 days on d1-6, 150 µg/3 days d7-embryo transfer) while 2769 (33.1%) received a constant dose (CD) of estrogen (oral: 6 mg/day 1-embryo transfer; transdermal: 150 µg/3 days d1-embryo transfer). Embryos were generated by ICSI with fresh or vitrified donor oocytes fertilized with either fresh or frozen sperm from either the couple partner or donor. PARTICIPANTS/MATERIALS, SETTING,
METHODS: Cohort allocation was not related to patient characteristics; instead it reflected an internal policy change in E administration. Effect of estrogen dose (ID versus CD) on biochemical, clinical, ongoing and live birth rates, stratified by administration route, was analyzed by univariate and multivariate analysis adjusted by donor and recipient demographic and cycle characteristics. MAIN RESULTS AND THE ROLE OF CHANCE: No difference in live birth rate was found between CD and ID for oral (33.0 versus 32.5%, P = 0.81) and transdermal (35.3 versus 33.5%, P = 0.33) supplementation. Biochemical pregnancy rate was higher in CD than ID (53.7 versus 47.5%, P < 0.001) when patients received oral supplementation. Adjusted analysis confirmed that oral administration had a greater impact on biochemical pregnancy rates than transdermal (odds ratio (OR) 1.28; 95% confidence interval (CI) 1.11-1.48, P = 0.001 versus OR 1.13; 95% CI 1.00-1.30, P = 0.055). Sub-analysis of transfers between day 12 and 15 of estrogen supplementation showed no difference between CD and ID in pregnancy outcomes. Demographic variables and cycle characteristics were comparable between both groups. Moreover, the use of the oocyte donation model reduces confounding factors related to oocyte age, embryo aneuploidy, and embryo quality. LIMITATIONS, REASONS FOR CAUTION: The greatest limitation of this study is its retrospective nature. On the other hand, this study was performed using donated oocytes; although this is unlikely to affect the results, we cannot exclude the possibility that a high quality female gamete responds differently to endometrial state in comparison to a patient's own oocytes. WIDER IMPLICATIONS OF THE
FINDINGS: In fresh embryo transfer from oocyte donation cycles, changes in the protocol of E replacement do not seem to have an impact on clinical outcomes and performance; for this reason estrogen replacement protocols can be adjusted to the patient's characteristics and preferences as well as to the most cost effective strategy. STUDY FUNDING/COMPETING INTERESTS: None.