Comparative Evaluation of Computed Tomography-Based Scoring Systems For Predicting Mortality In Adults With Traumatic Brain Injury: A Retrospective Study

Objectives: To evaluate and compare the prognostic performance of five commonly used computed tomography (CT)-based scoring systems—Rotterdam, Helsinki, Stockholm, Marshall, and The neuroimaging radiological interpretation system (NIRIS)—in predicting in-hospital and 30-day mortality among patients with traumatic brain injury (TBI). Material and Methods: This retrospective observational study included 278 adult patients with TBI presenting to the emergency department of a tertiary care hospital in Ahmedabad, India, during 2024. All patients underwent non-contrast CT of the head within two hours of presentation. CT images were independently assessed by neuroradiologists and scored according to the five CT-based scoring systems. Mortality outcomes were evaluated at hospital discharge and at 30 days post-injury. Prognostic accuracy was determined using sensitivity, specificity, positive and negative predictive values, and the area under the receiver operating characteristic curve (AUC). Results: The overall 30-day mortality was 15.1% (n=42). Higher scores in all systems correlated with increased mortality ( p <0.001). The Stockholm score demonstrated the best performance (specificity 91%, AUC 0.89), followed by the Helsinki score (AUC 0.86). Rotterdam and Marshall scores also showed strong discriminatory ability, while NIRIS was moderately predictive. Diffuse axonal injury, cerebral oedema, and midline shift were significant imaging predictors of poor outcome. Conclusion: All five CT-based scoring systems are valid predictors of mortality in TBI patients. the Stockholm and Helsinki scores demonstrated superior Prognostic performance and may be incorporated into early emergency department assessment, particularly when combined with clinical variables such as Glasgow coma scale. these findings highlight the importance of structured imaging-based scoring for improving early risk stratification and guiding management.