Constructing a Nomogram for ICSI-ET Assisted Reproductive Outcomes in Infertile Men Based on Lifestyle and Sperm Quality: A Multicenter Retrospective Survey

Ke Wang¹Shihui Wang¹Yan Xu²Jie Bai³Mengmeng Ma³Yuping Fan³Xin Huang^4*You Zhang^3*Ningxin Qin^3*

• ¹Fudan University Shanghai Cancer Center, Shanghai, China
• ²Sanda University, Shanghai, China
• ³Shanghai First Maternity and Infant Hospital, Shanghai, China
• ⁴Shanghai Jiaotong University School of Medicine Xinhua Hospital, Shanghai, China

[Abstract] Objective: To identify risk factors influencing clinical pregnancy outcomes in infertile men undergoing intracytoplasmic sperm injection (lCSl) and to establish and validate a nomogram prediction model. Methods: A total of 1,037 infertile men who underwent lCSl-fresh embryo transfer (lCSl-ET) at the Reproductive Medicine Center of Tongji University Affiliated Obstetrics and Gynecology Hospital from February 2023 to February 2024 were included. Differences in demographic and laboratory indicators between 403 pregnancy cycles (study group) and 634 non-pregnancy cycles (control group) were analyzed. Lasso regression was applied to select predictive variables, which were further used in multivariate logistic regression to construct the nomogram. External validation was conducted using data from 290 infertile men who underwent ICSl treatment at Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine (March-June 2024). Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), confusion matrix and Decision Curve Analysis (DCA). Results: Age, body mass index, smoking, drinking, daily sleep time, daily exercise time, stress, progressive sperm motility rate, and sperm DNA fragmentation index were identified as predictors. In the training set, the model achieved an AUC of 0.919 (95% Cl: 0.900-0.938), the accuracy was 85.3% (95% CI:82.7%~87.5%), the sensitivity was 85.4% (95% CI: 81.1%~89.1%), the specificity was 85.2% (95% CI: 82.0%~88.1%). The validation set show that AUC of 0.930 (95% C1: 0.892-0.968), the accuracy was 80.7% (95% CI:74.8%~85.5%), the sensitivity was 78.8% (95% CI: 68.5%~86.9%), and the specificity was 81.9% (95% CI: 73.8%~88.4%). The external validation set results show that the AUC was 0.918 (95%CI: 0.876-0.959), the accuracy was 93.8% (95% CI: 90.4-96.0%), the sensitivity was 94.4% (95% CI: 88.4%~97.4%), and the specificity was 93.4% (95% CI: 88.8%~96.2%), indicating strong discrimination and calibration. Conclusion: Both sperm quality and lifestyle factors significantly affect clinical pregnancy outcomes in ICSl-ET cycles among infertile men. The developed nomogram demonstrates excellent predictive accuracy and reliability, providing a useful tool for clinicians to deliver individualized counseling and targeted interventions.