Development and validation of a nomogram to predict the probability of death after surgical evacuation for traumatic intracranial hemorrhage

Here we describe the derivation and validation of a prognostic nomogram for patients with Traumatic Intracranial Hemorrhage (tICH) after surgical evacuation. This is a retrospective study based on 245 patients admitted to the Department of Neurosurgery of Huashan Hospital affiliated to Fudan University, between August 2005, and August 2023. We divided the dataset into primary and validation data by the ratio of 7:3. The LASSO regression model was used for predictor selection. The nomogram was developed using Cox regression models. The predictive performance of the nomogram was assessed by concordance index (C index) and calibration in the primary and validation cohorts. We also used decision curve analysis (DCA) to describe the clinical value. The main outcome was death related to tICH. The nomogram incorporated age, GCS-E, history of hypertension, and cerebellar hematoma, which was selected by the LASSO regression model. The nomogram showed good calibration and discrimination in the primary and validation data, with a 1-year C-index of 0.882 (95% CI, 0.777 to 0.987) and 0.818 (95% CI, 0.669 to 0.968), respectively. Decision curve analysis indicated that the nomogram is clinically useful when the patient or doctor's threshold probability ranges from 10 to 100%. In this study, we found that the tICH-related mortality rate was 11.42% (28/245). In the elderly cohort aged ≥ 65 years, the mortality rate increased to 28.13%(18/64). The nomogram we developed here can be conveniently used to predict the long-term prognosis of patients with tICH after surgical evacuation.Retrospectively registered: KY2024-860.