Clinical tool developed to identify children with TBI at low-risk for ICU-level care

Researchers retrospectively reviewed 284 patients aged 16 years or younger with mild traumatic brain injury (TBI). They assessed covariates that included age, Glasgow Coma Scale score, pupillary response, CT characteristics and the Injury Severity Score. They determined significant predictors for model inclusion with stepwise logistic regression. They used binary recursive partitioning to derive the clinical decision rule.
The five final predictor values included in the clinical decision rule were a midline shift of less than 5 mm, intraventricular hemorrhage, non-isolated head injury, postresuscitation GCS score less than 15 and cisterns absent. Investigators noted the clinical decision rule could determine that 37 of 40 patients needed ICU-level care and 154 patients of 244 did not need ICU-level care. There was a 98.1% negative predictive value. – by Monica Jaramillo
Disclosures: The researchers report no relevant financial disclosures.