Predictors of progression of intracranial injury in minimal traumatic brain injury

Background. Minimal traumatic brain injury (MTBI) is a major public health problem. 15% of patients with MTBI will have an intracranial injury visualized on computed tomography scan (CT), and a small percentage of these patients will have progression of injury (POI) necessitating a neurosurgical intervention. Identifying risk factors for those most at danger of POI allows for the identification of POI before irreversible damage ensues while minimizing unnecessary CT and monitoring in those not at risk.;Objective. The purpose of this research was to identify CT and clinical variables that significantly correlate with POI and determine their relative risk, while describing the timeframe of POI. The ultimate goal was to propose a predictive model for POI that will assist with efficient patient management.;Methods. Clinical and cranial CT characteristic data were retrospectively collected from medical records of patients with a MTBI and a positive CT admitted to UMDNJ Level-1 Trauma Center over a 3-year period who underwent a minimum of 2 cranial CT within 24 hours prior to any neurosurgical intervention. A linear regression analysis was used to identify independent risk factors for POI. Based on the independent variables identified, a mathematical predictive model for POI was proposed.;Results. Of the 504 patients admitted with a MTBI in the study period, 342 patients met inclusion criteria. 110 of those patients (32%) showed POI on follow-up CT. Multivariate logistic regression identified a fall from height (OR=2.24, 95% CI: 1.21-4.15), male gender (OR=2.05, 95% CI: 1.12-3.77), intraparenchymal hemorrhage (OR=1.94, 95% CI: 1.14-3.29) and temporal (OR=1.96, 95% CI: 1.16; 3.29) or frontal lobe injuries (OR=1.77, 95% CI: 1.02-3.10) to be predictive of POI. These findings did not confirm the study hypothesis that age >55 years, multiple hemorrhages, epidural hemorrhage and the use of anticoagulants would be predictive of POI. The predictive function of the model had a sensitivity of 54% and a specificity of 81%. Over 99% of POI stopped by 48 hours from presentation to the emergency department.;Conclusion. A number of factors were identified to be predictive of POI. A predictive model that may be useful in identifying patients at low risk of injury progression was developed. A 48-hour period of neurologic monitoring will identify almost all POI. Prospective data is needed to evaluate this proposed predictive model.