The Interaction Between Embryo Developmental Stage and Endometrial Preparation Protocol on Live Birth Outcomes in Frozen-Thawed Embryo Transfer: A Retrospective Cohort Study.

OBJECTIVES: To determine whether an interaction exists between embryo developmental stage and endometrial preparation protocol in frozen-thawed embryo transfer (FET) cycles, and to evaluate their independent associations with live birth after adjusting for key confounders, particularly female age. DESIGN: A retrospective cohort study of 1,634 first FET cycles, stratified by embryo type (D3 high-quality cleavage-stage vs. D5 non-top-quality blastocyst) and endometrial protocol (Natural Cycle [NC] vs. Artificial Cycle [AC]). PARTICIPANTS/MATERIALS, SETTING, METHODS: All included patients were undergoing their first FET cycle. The primary exposure variables were embryo type (D3 high-quality cleavage-stage vs. D5 non-top-quality blastocyst) and endometrial preparation protocol (NC vs. AC). The primary outcome was live birth per transfer cycle. A multivariable logistic regression model was employed to test for an interaction between embryo type and endometrial preparation protocol and to calculate adjusted odds ratios (aORs) after controlling for female age, anti-Müllerian hormone level, duration of Infertility, and endometrial thickness. Subgroup analysis was performed based on age (<35 years vs. ≥35 years). Model performance was evaluated using the Hosmer-Lemeshow test and the area under the receiver operating characteristic curve (AUC). Results No statistically significant interaction was found between embryo type and endometrial preparation protocol (adjusted P=0.453). In the adjusted model, both factors were independent predictors of live birth rate. The live birth rate was significantly higher following the transfer of a D5 non-top-quality blastocyst compared to a D3 high-quality cleavage-stage embryo (aOR=1.48, 95% CI: 1.14-1.93). The natural cycle protocol was associated with a significantly higher live birth rate than the artificial cycle protocol (aOR=1.78, 95% CI: 1.28-2.49). Age-stratified analysis revealed that the advantage of the natural cycle was particularly pronounced in patients aged <35 years, while a clear trend favoring blastocyst transfer was observed in patients aged ≥35 years. Limitations The retrospective design may introduce selection bias, and findings from a single center may not be fully generalizable. Conclusions Embryo developmental stage and endometrial preparation protocol are independent, non-interactive determinants of live birth outcomes in FET. This supports a simplified clinical decision-making model where these two factors can be optimized separately. Treatment strategies should be individualized: for younger patients (<35 years), priority should be given to the more physiological natural cycle protocol to optimize endometrial receptivity; for older patients (≥35 years), more active consideration can be given to utilizing developmentally competent D5 non-top-quality blastocysts in embryo selection.