Predicting neurosurgical clearance in the polytrauma patient with concomitant traumatic brain injury

The goal of this study is to develop a model based on previously used prognostic predictors in traumatic brain injury (TBI) patients with polytrauma, which will facilitate the decision-making of whether to clear these patients for non-cranial surgery. Data of eligible patients was obtained from a trauma database at a Level I trauma and academic tertiary referral center in the United States. The number of days seen by the neurosurgical service prior to clearance, injury severity score (ISS), post-trauma day 0 (PTD 0) of Glasgow Coma Score (GCS), intracranial pressure (ICP) score and computed tomography (CT) score, as well as the changes in GCS, ICP score and CT score between PTD 0 and day of clearance were the variables used in developing the model. The Neurosurgical Clearance Model (NCM) was developed using data from 50 patients included in the study. Patients were cleared by neurosurgeons 1.6 days later than it would appear possible based on a retrospective review of the patients' clinical conditions. A single model equation was developed, the ultimate result of which is a clearance probability value. The best cutoff clearance probability value was found to be 0.584 (or 58.4%) using Receiver Operator Characteristic curve analysis. Our data suggests that neurosurgeons are risk-averse in clearing polytrauma patients for non-cranial surgery. This pilot NCM, if reproduced and validated by other groups and in larger prospective studies, may become a useful tool to assist clinicians in this often-difficult decision-making process.