343 A New Computed Tomography-based Frontal Contusion Score for Patients with Traumatic Brain Injury

Abstract

Abstract INTRODUCTION Frontal contusions are characterized by gradually progressing hematoma/edema and rapid deterioration owing to central herniation, even if the patient is conscious at the time of admission. This study 1) examined how to better characterize frontal contusion in a series of traumatic brain injury (TBI) patients, and 2) developed and validated a new frontal contusion score (FCS) based on the shape of the frontal contusion to facilitate rapid, accurate assessment of the computed tomography (CT) findings of frontal contusion. METHODS This study retrospectively analyzed data from 206 consecutive patients with isolated frontal contusions. The new score is based mainly on the shape of the frontal brain contusion. Forward stepwise logistic regression was used to identify independent predictors of acute neurological deterioration and refractory intracranial hypertension. A receiver-operating characteristic (ROC) curve was then drawn based on the FCS. RESULTS >The incidence of acute neurological deterioration increased significantly as the FCS increased. FCS, obliteration of the basal cistern, and a serum sodium decrease of more than 10 mm ol/L within 24 hours were independent predictors of acute neurological deterioration. Each one-unit increase in FCS led to a 57% increase in the odds of acute neurological deterioration [odds ratio (OR), 1.57; 95% confidence interval (CI), 1.25 1.95]. The area under the curve (AUC) of the FCS that predicted acute neurological deterioration was 0.727 (95% CI 0.656 0.797). The incidence of refractory intracranial hypertension increased significantly with an increase in the FCS. Only FCS and obliteration of the basal cistern remained predictors of refractory intracranial hypertension. Each one-unit increase in FCS led to a 49% increase in the odds of refractory intracranial hypertension (OR, 1.49; 95% CI, 1.06 2.10). The area under the curve (AUC) of the FCS for predicting refractory intracranial hypertension was 0.647 (95% CI 0.532 0.763). The FCS was not an independent predictor of the 6-month mortality (OR, 0.87; 95% CI, 0.59 1.28) or 6-month unfavorable outcome (OR, 1.32; 95% CI, 0.93 1.87). CONCLUSION Therefore, the FCS is a valid evaluator of the character of frontal contusion. The clinical utility and generalizability of this score need to be validated in a large sample.