Mortality After Traumatic Brain Injury in Elderly Patients: A New Scoring System

Traumatic brain injury (TBI) remains a life-threatening condition characterized by growing incidence worldwide, particularly in the aging population, in which the primary goal of treatment appears to be avoidance of chronic institutionalization.

To identify independent predictors of 30-day mortality or vegetative state in a geriatric population and calculate an intuitive scoring system, we screened 480 patients after TBI treated at a single department of neurosurgery over a 2-year period. We analyzed data of 214 consecutive patients aged ≥65 years, including demographics, medical history, cause and time of injury, neurologic state, radiologic reports, and laboratory results. A predictive model was developed using logistic regression modeling with a backward stepwise feature selection.

The median Glasgow Coma Scale (GCS) score on admission was 14 (interquartile range, 12-15), whereas the 30-day mortality or vegetative state rate amounted to 23.4%. Starting with 20 predefined features, the final prediction model highlighted the importance of GCS motor score (odds ratio [OR], 0.17; 95% confidence interval [CI], 0.09-0.32); presence of comorbid cardiac, pulmonary, or renal dysfunction or malignancy (OR, 2.86; 9 5% CI, 1.08-7.61); platelets ≤100 × 10⁹ cells/L (OR, 13.60; 95% CI, 3.33-55.49); and red blood cell distribution width coefficient of variation ≥14.5% (OR, 2.91; 95% CI, 1.09-7.78). The discovered coefficients were used for nomogram development. It was further simplified to facilitate clinical use. The proposed scoring system, Elderly Traumatic Brain Injury Score (eTBI Score), yielded similar performance metrics.

The eTBI Score is the first scoring system designed specifically for older adults. It could constitute a framework for clinical decision-making and serve as an outcome predictor. Its capability to stratify risk provides reliable criteria for assessing efficacy of TBI management.