Extracranial injuries are important in determining mortality of neurotrauma*

Publisher Website
Google Scholar
Abstract
This study aimed to assess the significance of extracranial injuries, as measured by Injury Severity Score, on 6-mo and 9-yr mortality of neurotrauma. Retrospective linked data cohort study. A major neurotrauma center in Western Australia. Six hundred eighty-three adult neurotrauma patients. Data were first used to validate the largest published international neurotrauma “extended” mortality prognostic model, in which extracranial injuries are considered significant if the patient has hypoxemia or hypotension on admission. Logistic and Cox regression, incorporating bootstrap techniques to adjust for overfitting, were used to assess the significance of Injury Severity Score in determining 6-mo and 9-yr mortality, respectively. Among a total of 683 patients admitted between 1994 and 2002, 636 (93.1%) had extracranial injuries. The international neurotrauma “extended” mortality prognostic model was poorly calibrated and underestimated the observed mortality (slope and intercept of the calibration curve were 2.14 and 0.35, respectively) when applied to our patients. Incorporating Injury Severity Score into the model improved its calibration. Injury Severity Score accounted for 11% of the variability and was the third most important factor after Marshall computed tomographic grading (17.8%) and pupil reactivity (14.5%) in determining 6-mo mortality. There was a notable increase in mortality between 6-mo (19.2%) and 24-mo follow-up (25.8%). Injury Severity Score remained important and accounted for 9.2% of the variability in determining 9-yr mortality after the injury. Hypotension and hypoxemia on admission were inadequate markers of extracranial injuries; incorporating more comprehensive extracranial injury assessment by the Injury Severity Score to the standard neurologic prognostic factors improved the accuracy of predictions on mortality after neurotrauma.