Retrieval-to-transfer interval and autologous frozen embryo transfer outcomes: age-stratified and joint associations with oocyte retrieval age

Background The retrieval-to-transfer interval (RTI) is the elapsed time between oocyte retrieval and frozen embryo transfer (FET), which in an autologous freeze-all cohort equals the duration of embryo cryostorage. Whether RTI is associated with FET outcomes independently of oocyte retrieval age, whether this association differs across retrieval-age strata, and whether the two risks combine additively or synergistically have not been formally evaluated in a large autologous FET cohort. Methods In a single-center retrospective cohort of 57,282 women undergoing their first autologous vitrified FET (January 2018–August 2025), maternal ages at retrieval and transfer were each categorized (<30, 30–34, 35–39, and ≥40 years) and cross-classified into eight age-combination groups, two of which combined small age-discordant subgroups. RTI was modeled both continuously and categorically (≤1, 2–4, and ≥5 months) across four retrieval-age strata using stratified restricted cubic splines, and effect modification was assessed on multiplicative and additive scales. The primary outcome was live birth; secondary outcomes were clinical pregnancy and clinical pregnancy loss (analyzed only among clinical pregnancies). Results Adjusted live birth was lowest when both oocyte retrieval and embryo transfer occurred at ≥40 years (OR 0.24, 95% CI 0.20–0.28, versus both at 30–34 years). After adjustment for retrieval age, RTI ≥5 months (versus ≤1 month) remained associated with lower live birth (OR 0.88, 95% CI 0.83–0.93) and clinical pregnancy (OR 0.79, 95% CI 0.75–0.84), and with higher clinical pregnancy loss (OR 1.19, 95% CI 1.08–1.30; all P < 0.001), greatest in the 30–39-year retrieval strata; in the 35–39-year stratum, adjusted live birth fell from 35.8% at 3 months to 28.6% at 24 months. A multiplicative interaction was detected for clinical pregnancy; odds-ratio-scale additive interaction showed no departure from additivity for any outcome. Conclusions A longer RTI was associated, in a dose-dependent manner, with less favorable FET outcomes, most markedly when oocytes were retrieved at 30–39 years. When advanced retrieval age and a long RTI occurred together, the associated risks combined approximately additively rather than synergistically. Because this design cannot separate maternal aging, cryostorage duration, and reasons for delay, these findings represent associations rather than causal effects; RTI merits consideration alongside retrieval age when counseling on transfer timing.