Outcome comparisons of controlled ovarian hyperstimulation protocols in adenomyosis patients: a real-world retrospective cohort study

To compare the efficacy of different COH protocols on pregnancy outcomes in adenomyosis patients. A real-world retrospective cohort study analyzed 1486 IVF-ET cycles in adenomyosis patients who received COH regimens between 2018 and 2021. Pregnancy outcomes were compared among patients under different COH protocols. The short-acting long protocol achieved the highest live birth (47.92%) and cumulative clinical pregnancy (68.84%) rates. The antagonist protocol showed lower fresh-cycle pregnancy (36.63% vs. 52.10%, p = 0.036), live birth (21.78% vs. 36.55%, p = 0.009), and cumulative clinical pregnancy rates (39.31% vs. 53.29%, p < 0.001) compared to the long/ultra-long protocol. Multivariable logistic regression confirmed that the COH protocol was an independent predictor of both clinical pregnancy (Wald χ² = 8.127, p = 0.043) and cumulative clinical pregnancy (Wald χ² = 40.344, p < 0.001). In patients <35 years with a normal ovarian reserve (anti-Müllerian hormone ≥ 1.2 ng/mL), live birth rates and cumulative clinical pregnancy rates were similar between the antagonist protocol and long/ultra-long protocols (27.59% vs. 44.94%, p = 0.098; 63.83% vs. 63.87%, p = 0.995), with significantly less gonadotropin used in the antagonist protocol. In adenomyosis patients, the long/ultra-long provided better fresh-cycle outcomes. While the antagonist protocol had lower overall pregnancy rates, it preserved cumulative pregnancy rates in young patients with normal ovarian reserves and reduced gonadotropin exposure.