Comparison of vaginal versus intramuscular progesterone in programmed cycles for frozen-thawed blastocyst transfer in patients with endometriosis

BackgroundPrevious studies have shown that due to the presence of endometrium progesterone resistance in patients with endometriosis, it is considered that higher levels of progesterone may be required to achieve live birth during programmed frozen-thawed embryo transfer (FET) cycles. Currently, the optimal progesterone support in FET cycles remains a contentious issue, and it mainly focused on the general infertile population, without specific attention to infertile patients with endometriosis. This study aimed to compare the pregnancy outcomes between vaginal or intramuscular progesterone administration in patients with endometriosis, and to determine whether the stage of endometriosis moderates the differences.MethodsThis retrospective cohort study included patients with endometriosis who underwent their first single frozen-thawed blastocyst transfer in a programmed cycle from January 2018 to April 2024 at a university-affiliated reproductive medical center. According to the routes of luteal support, patients were divided into vaginal progesterone and intramuscular progesterone groups. Analyses were conducted using multivariate regression models and subgroup analysis. Interaction tests were employed to determine whether the revised American Society for Reproductive Medicine (r-ASRM) stages of endometriosis moderated the differences between the routes of progesterone administration and pregnancy outcomes.ResultsA total of 825 programmed frozen-thawed blastocyst transfer cycles were included in the analysis, with 362 cases using vaginal progesterone and 463 cases using intramuscular progesterone. In the overall cohort, clinical pregnancy rate of the vaginal progesterone group was 49.17%, comparable to 44.06% of the intramuscular progesterone group (aOR 0.82, 95% CI 0.61-1.11). Similarly, there was no statistically significant difference in miscarriage rates between the two groups (16.85% versus 24.51%; aOR 1.57, 95% CI 0.90-2.75). In the subgroup analysis in patients classified as r-ASRM stages I-II, clinical pregnancy rate of vaginal progesterone group was significantly higher than that of intramuscular group (aOR 0.74, 95% CI 0.58-0.93, P = 0.011). Whereas, in patients with stages III-IV, no significant differences in pregnancy outcomes between the two groups were detected. Interaction tests between the routes of progesterone administration and r-ASRM stages were significant (P = 0.036).ConclusionsIn the first single frozen-thawed blastocyst transfer cycles for endometriosis patients with r-ASRM stages I-II, vaginal progesterone favours a higher clinical pregnancy rate compared to the intramuscular progesterone.