Prevalence and risk factors for previous cesarean scar defect formation: a retrospective cohort study

To determine the prevalence of previous cesarean scar defect (PCSD) in a referred cohort, identify associated risk factors, and develop a nomogram for individualized risk estimation. This retrospective single-center study analyzed 156 women who underwent cesarean delivery between 2021 and 2023 at a tertiary referral center and received routine postoperative transvaginal ultrasound 1–6 months after surgery. PCSD was diagnosed when a wedge-shaped or cystic anechoic defect at the cesarean scar communicated with the uterine cavity and disrupted myometrial continuity. Patients were classified into PCSD ( n = 43) and non-PCSD ( n = 113) groups. Independent predictors were identified by multivariate logistic regression, and a nomogram was constructed and internally validated with 500 bootstrap resamples and ROC analysis. PCSD prevalence was 27.56% (43/156). Four independent predictors were identified: a higher number of cesarean deliveries (OR = 3.19, 95% CI: 1.27–8.05), emergency cesarean section (OR = 3.25, 95% CI: 1.38–7.64), retroverted uterus (OR = 2.28, 95% CI: 1.04–5.01), and low-positioned incision near the internal cervical os (OR = 2.91, 95% CI: 1.16–7.70). The model showed an apparent AUC of 0.858 (95% CI: 0.793–0.923) and an optimism-corrected AUC of 0.821 after 500 bootstrap resamples. Calibration was acceptable, with a calibration slope of 0.874, intercept of − 0.032, Hosmer–Lemeshow P = 0.609, and Brier score of 0.163. Decision curve analysis suggested positive net benefit across threshold probabilities of 0.14–0.86 in the development cohort. The nomogram showed acceptable apparent and internally validated performance for estimating PCSD risk in this development cohort. It may support early postoperative risk stratification and counseling, but external validation is required before clinical implementation.