Are routine repeat imaging and intensive care unit admission necessary in mild traumatic brain injury?

Object More than 1.5 million Americans suffer a traumatic brain injury (TBI) each year. Seventy-five percent of these patients have a mild TBI, with Glasgow Coma Scale (GCS) Score 13–15. At the authors' institution, the usual practice has been to admit those patients with an associated intracranial hemorrhage (ICH) to an ICU and to obtain repeat head CT scans 12–24 hours after admission. The purpose of this study was to determine if there exists a subpopulation of mild TBI patients with an abnormal head CT scan that requires neither repeat brain imaging nor admission to an ICU. This group of patients was further classified based on initial clinical factors and imaging characteristics. Methods A retrospective review of all patients admitted to a Level I trauma center from January 2007 through December 2008 was performed using the hospital Trauma Registry Database, medical records, and imaging data. The inclusion criteria were as follows: 1) an admission GCS score ≥ 13; 2) an isolated head injury with no other injury requiring ICU admission; 3) an initial head CT scan positive for ICH; and 4) an initial management plan that was nonoperative. Collected data included age, etiology, initial GCS score, time of injury, duration of ICU stay, duration of hospital stay, and anticoagulation status. Primary outcomes measured were the occurrence of neurological or medical decline and the need for neurosurgical intervention. Imaging data were analyzed and classified based on the predominant blood distribution found on admission imaging. Data were further categorized based on the Marshall CT classification, Rotterdam score, and volume of intraparenchymal hemorrhage (IPH). Progression was defined as an increase in the Marshall classification, an increase in the Rotterdam score, or a 30% increase in IPH volume. Results Three hundred twenty-one of 1101 reviewed cases met inclusion criteria for the study. Only 4 patients (1%) suffered a neurological decline and 4 (1%) required nonemergent neurosurgical intervention. There was a medical decline in 18 of the patients (6%) as a result of a combination of events such as respiratory distress, myocardial infarction, and sepsis. Both patient age and the transfusion of blood products were significant predictors of medical decline. Overall patient mortality was 1%. Based on imaging data, the rate of injury progression was 6%. The only type of ICH found to have a significant rate of progression (53%) was a subfrontal/temporal intraparenchymal contusion. Other variables found to be significant predictors of progression on head CT scans were the use of anticoagulation, an age over 65 years, and a volume of ICH > 10 ml. Conclusions Most patients with mild TBI have a good outcome without the necessity of neurosurgical intervention. Mild TBI patients with a convexity SAH, small convexity contusion, small IPH (≤ 10 ml), and/or small subdural hematoma do not require admission to an ICU or repeat imaging in the absence of a neurological decline.