To develop a practical, pre-trigger risk scoring system using routinely available clinical parameters to predict moderate-to-severe ovarian hyperstimulation syndrome (OHSS) in in vitro fertilization and intracytoplasmic sperm injection (IVF/ICSI) cycles. We retrospectively analyzed women undergoing their first fresh IVF/ICSI cycle between January 2018 and July 2025. Only pre-trigger variables were included. Univariable logistic regression identified candidate predictors, which were entered into multivariable logistic regression with backward selection. Regression coefficients were converted into an integer-based risk score. Model performance was assessed using receiver operating characteristic (ROC) analysis, calibration, and internal bootstrap validation. A total of 1,641 women were included, of whom 125 (7.6%) developed moderate-to-severe OHSS. Younger age, higher AMH, lower FSH and E2, elevated LH/FSH ratio, and polycystic ovary syndrome (PCOS) were independently associated with OHSS. These six predictors were incorporated into a 0–13 point scoring system. OHSS incidence increased progressively with higher scores, enabling clinically meaningful stratification into low-, moderate-, high-, and very high-risk categories. The model demonstrated good discrimination (AUC 0.84; 95% CI 0.81–0.87) and excellent calibration (Hosmer–Lemeshow p = 0.61). Bootstrap validation confirmed robust performance with minimal optimism. The score retained predictive utility in both PCOS and non-PCOS subgroups. We developed a simple, pre-trigger risk scoring system that reliably predicts moderate-to-severe OHSS prior to final oocyte maturation, supporting proactive prevention and individualized stimulation strategies. External and prospective multicenter validation is warranted.
He et al. (Sat,) studied this question.