A Prediction Model for Pregnancy Outcomes in Frozen-Thawed High-Quality Single Blastocyst Transfer Cycles

Background: Frozen‑thawed high‑quality single blastocyst transfer has become a mainstream strategy to reduce multiple pregnancy risks in assisted reproductive technology, yet personalized prediction for pregnancy outcomes remains limited in clinical practice. Few validated models have incorporated key clinical and embryological factors for this specific population. The aim of this study was to identify predictors of clinical pregnancy following frozen-thawed, high-quality single blastocyst transfer, and to develop a nomogram for predicting pregnancy outcomes. Methods: We retrospectively analyzed 3398 cycles of frozen-thawed, high-quality single blastocyst transfer performed at the Third Affiliated Hospital of Sun Yat-sen University from July 2016 to December 2023. Cycles were randomly allocated to training (n = 2378) and validation (n = 1020) datasets at a 7:3 ratio. Predictors of clinical pregnancy were identified using univariate and multivariate logistic regression and were subsequently used to construct a nomogram. Model performance was assessed by area under the curve (AUC), calibration curves, and decision curve analysis (DCA). Results: Multivariate analysis identified four independent predictors of clinical pregnancy (all p < 0.05): maternal age at oocyte retrieval, endometrial thickness, fertilization method (conventional in vitro fertilization [IVF] vs. intracytoplasmic sperm injection [ICSI]), and blastocyst developmental stage (day-5 vs. day-6). Stratified analysis revealed that, compared with patients younger than 35 years, significantly lower odds of clinical pregnancy were first observed in the 35–37 years old subgroup (odds ratio [OR] = 0.753, 95% confidence interval [CI]: 0.612–0.928). Optimal pregnancy rates occurred at an endometrial thickness of 10.0–11.9 mm (OR = 1.973, 95% CI: 1.435–2.713). The model demonstrated moderate discriminative capacity, with AUC values of 0.63 (95% CI: 0.60–0.65) and 0.62 (95% CI: 0.59–0.66) in the training and validation sets, respectively. Calibration curves showed good agreement between predicted and observed probabilities, while DCA confirmed clinical utility across relevant threshold probabilities. Conclusions: Four clinical parameters independently predicted the clinical pregnancy rate: maternal age <35 years, endometrial thickness of 10.0–11.9 mm, ICSI fertilization, and day-5 blastocyst transfer. This validated nomogram shows moderate predictive capacity for the clinical pregnancy rate in frozen-thawed, high-quality single blastocyst transfer cycles and may serve as a useful tool for clinical application.