Risk factors for postoperative recurrence of chronic subdural hematoma and construction and validation of an online dynamic nomogram prediction model

Objective To investigate factors associated with postoperative recurrence in patients with chronic subdural hematoma (CSDH) and to construct and internally validate an online dynamic nomogram for individualized recurrence-risk estimation. Methods This retrospective single-center study included 284 surgically treated patients with CSDH between January 2018 and May 2025. Before model development, the cohort was divided at a 7:3 ratio into a training cohort ( n = 198) and a validation cohort ( n = 86). Univariate analysis, multivariable logistic regression, predictor selection, and nomogram construction were performed exclusively in the training cohort, whereas the validation cohort was reserved for model performance assessment. Recurrence was defined as reaccumulation or enlargement of the hematoma on postoperative CT within 6 months after surgery, accompanied by recurrent or worsening neurological symptoms, regardless of whether repeat surgery was subsequently performed. Results Overall, 54 of 284 patients (19.01%) experienced recurrence; 33 recurrence events occurred in the training cohort and 21 in the validation cohort. Smoking history, laterality, hematoma density, greatest hematoma thickness, cerebral midline shift ≥10 mm, and postoperative pneumocephalus were statistically significant in the multivariable analysis. The AUC was 0.89 (95% CI, 0.82–0.95, sensitivity 0.82, specificity 0.85) in the training cohort and 0.74 (95% CI, 0.59–0.88, sensitivity 0.52, specificity 0.85) in the validation cohort. Bootstrap optimism correction yielded an apparent AUC of 0.892, a mean optimism of 0.021, and an optimism-corrected AUC of 0.871. In the training cohort, the Hosmer-Lemeshow test showed no statistically significant evidence of lack of fit ( χ 2 = 9.96, df = 8, p = 0.268). Conclusion The online nomogram provides a convenient framework for individualized recurrence risk estimation. Further external multicenter validation is warranted to assess its generalizability and clinical applicability.