Predictors of Mortality among Head Trauma Patients Reaching the ICU Alive

Background: Traumatic brain injury (TBI) is a leading cause of morbidity and mortality worldwide. Ninety percent occurs in low and middle income countries.Aim of Work: To construct a predictive model for mortality in head injury patients on the basis of easily available parameters.Methods: Prospective randomized study that included 100 head trauma patients admitted to the ICU in a period of 15 months. Demographic data, diabetes, hypertension and cardiac history were recorded. Admission blood samples were obtained for CBC, coagulation profile, kidney and liver function tests, lactate level and random glucose level. Receiver operating curve (ROC) analysis including the area under the ROC and multivariable logistic regression were used to identify independent mortality predictors of admission parameters to create a prognostic model.Results: A total of 58 patients died (58%) out of 100 patients included in this study. Multivariate analysis revealed that age 75 years (HR=25.49, 95% CI=2.36 to 275.66), GCS 9 (HR=4.18, 95% CI=1.58 to 11. 08), serum creatinine higher than 1.5 mg/ml (HR= 8.75 (95% CI=3.10 to 24.71), PaCO2 45 mmHg (HR=14.86 (95% CI=4.85 to 45.49) and history of cardiac disease (HR=0.37, 95% CI=0.15 to 0.91) were associated with high mortality rate.Conclusion: Age 75 years, GCS 9, PaCO2 45 mmHg, serum creatinine exceeding 1.5 mg/ml and cardiac diseases are independent significant predictors of mortality in head trauma patients.