The use of programmed frozen embryo transfer (FET) cycles has increased substantially in recent years, so the investigation of methods to optimize outcomes are warranted. Luteinizing hormone (LH) level is not always part of routine testing during FET cycles, but prior investigations on the impact of midcycle LH levels on pregnancy outcomes in programmed FET cycles are mixed. We aim to study the impact of elevated midcycle LH levels on pregnancy outcomes in programmed FET cycles that utilize intramuscular progesterone.

Single center retrospective cohort study to compare pregnancy outcomes of programmed FET cycles between January 2020 and July 2023 based on midcycle LH level. Midcycle LH levels were not obtained at the time of the cycle, but rather retrospectively from saved serum that was drawn within 2 days of initiation of progesterone administration. An LH level > 20IU/mL was designated “elevated” and LH level ≤ 20IU/mL was designated “normal” based on reference standards. Endometrial preparation was achieved by administration of estradiol supplementation for 14-21 days with the goal endometrial thickness (EMT) of ≥7mm prior to IM progesterone administration. The number of embryos transferred was consistent with ASRM guidelines.

Unadjusted and adjusted risk ratios (RR) and 95% confidence intervals were estimated by GEE estimation to calculate modified Poisson regression models with robust standard errors to account for lack of independence among observations. Variables assessed for confounding included: age of female at time of cryopreservation, endometrial thickness, estradiol level, cycle classification and number, and embryo quality and PGT status. Risk ratios were adjusted for the variables listed.

781 cycles were reviewed and those that utilized pituitary suppression, natural protocol or were part of other studies were excluded. Any cancelled cycles or those missing crucial values were also excluded. 395 programmed FET cycles comprised of 306 patients were identified, and 89 of these patients underwent multiple FETs during the study period. Elevated midcycle LH was observed in 4% (17/395) of programmed FETs. Mean age of female at time of embryo cryopreservation (32.38 versus 31.24yo) were similar between normal LH and elevated LH groups, respectively. In adjusted models, elevated LH was significantly negatively associated with live birth rate (RR 0.36, 95% CI 0.14-0.93), and clinical pregnancy rate (RR 0.30, 95% 0.11-0.81) and positively associated with a negative pregnancy test (RR 1.48, 95% 1.00-2.20).

Elevated LH >20IU/mL prior to starting progesterone is associated with lower live birth rate and clinical pregnancy rate compared with normal LH levels in programmed FET cycles.