Anti-Müllerian hormone in guiding the selection of a freeze-all versus a fresh embryo transfer strategy: a cohort study

Purpose: To explore an interaction effect between serum anti-Müllerian hormone (AMH) levels and the relative treatment effect of a freeze-all versus a fresh embryo transfer strategy on live birth. Methods: This was a retrospective cohort study investigating couples with infertility and eligible for both freeze-all and fresh embryo transfer between 2017 and 2019. Women with an absolute indication for a freeze-all strategy were excluded. Multivariable fractional polynomial interaction analysis within a logistic regression model was used to evaluate whether the treatment effect of a freeze-all versus a fresh transfer strategy varied at different AMH levels. Non-linear interactions were also considered. The primary outcome was the live birth after the first transfer. Results: A total of 13,503 women underwent a fresh embryo transfer and 2247 women underwent a freeze-all strategy. Live birth rates were slightly higher in the freeze-all group compared to those in the fresh embryo transfer group (35% vs 33%). There was a non-linear interaction between baseline serum AMH levels and the relative treatment effect of a freeze-all strategy versus a fresh transfer strategy on live birth (P = 0.0161). The benefit on live birth from a freeze-all embryo transfer strategy was greatest in women with a high serum level (> 7 ng/ml). The interaction remained valid when different imputation methods were used. Conclusion: As serum AMH level increased, there was a nonlinear increase in relative treatment effect of a freeze-only transfer versus a fresh transfer strategy on live birth, and such an effect reaches its maximum in women with high AMH levels.