Clinical pregnancy after ICSI: an exploratory transfer-day analysis of female age, AMH, sperm source, and transfer stage

Purpose This study aimed to explore transfer-level factors associated with clinical pregnancy after intracytoplasmic sperm injection (ICSI) using a deliberately lower-claim analytical strategy. Methods This retrospective single-centre study screened 332 ICSI-linked records from 2020 to 2025 and analysed 300 embryo transfers from 300 couples. Eligibility criteria, complete-case handling, one-transfer-per-couple preparation, variable coding, logistic regression specification, bootstrap optimism correction, calibration assessment, and software details were documented to support reproducibility. The primary outcome was clinical pregnancy per embryo transfer, defined as an intrauterine gestational sac on transvaginal ultrasonography. Since only 37 events were available, the primary multivariable analysis was restricted to four prespecified candidate variables: female age, anti-Müllerian hormone (AMH), sperm source, and embryo-transfer stage. This restriction was driven primarily by statistical proportionality and the limited event count, rather than an assumption that the omitted variables were not clinically significant. The analysis was intended to examine exploratory associations that could frame cautious transfer-day counselling discussions once all four variables were known; however, it was not designed as a clinical decision support mechanism. Exploratory baseline comparisons, a univariable logistic regression analysis, a four-variable multivariable association analysis, bootstrap-corrected discrimination, descriptive calibration assessment, and a fuller eight-variable sensitivity analysis were used. Results Clinical pregnancy occurred in 37 of 300 transfers (12.3%). In the simplified multivariable analysis, older maternal age was associated with lower odds of clinical pregnancy (adjusted odds ratio aOR 0.91 per year, 95% confidence interval CI 0.83–0.99; p = 0.032), whereas higher AMH was associated with higher odds (aOR 2.02 per unit, 95% CI 1.39–2.94; p 0.001). A more invasive sperm-source category was associated with lower odds (aOR 0.39, 95% CI 0.17–0.92; p = 0.032), and blastocyst transfer was associated with higher odds than cleavage-stage transfer (aOR 3.90, 95% CI 1.29–11.74; p = 0.016). The apparent area under the curve was 0.779, and the bootstrap-corrected area under the curve was 0.750. Conclusion In this dataset, female age, AMH, sperm source, and transfer stage showed the clearest exploratory associations with clinical pregnancy per embryo transfer. Since the event count was limited and calibration, coefficient stability, and transportability remain uncertain, the findings should be interpreted as hypothesis-generating. They may support cautious transfer-day counselling discussions but are not sufficient for routine predictive implementation without larger external validation studies.