Predictors of mortality in neurosurgical patients after traumatic brain injury

Abstract

Introduction: Traumatic brain injury (TBI) is considered a relevant worldwide cause of mortality and morbidity. Objective: To evaluate independent factors at hospital admission and during hospitalization that predicts in-hospital mortality in patients with TBI submitted to neurosurgery. Methods: Retrospective cohort analyzed data of patients that underwent neurosurgery due to the TBI. Evaluated predictor factors were age, gender, Glasgow Coma Scale (GCS) at admission, sequels before and after neurosurgery, trauma mechanism, and neurosurgical procedure. The survival analyses were verified. Positive and negative predictive values were calculated. Results: A total of 312 patients were analyzed, majority male (81.4%) with mild TBI (55.8%), and mean age of 49.6 ± 23.5 years. GCS at admission was an independent predictor of mortality (HR = 0.87; 95% CI: 0.81 to 0.93; p < 0.001). The GCS ≤ 13 was the optimal cut-off value to predict survival with a significant difference between GCS groups (≤13 or >13; p<0.001). Patients with GCS ≤ 13 had a six times greater risk of mortality (HR 6.04; 95 CI 2.39 to 15.3; p<0.001). The positive and negative predictive values were 35.8% and 97%, respectively. Conclusions: The GCS in admission is an important predictor of in-hospital mortality in TBI patients that need a neurosurgical procedure.