Comparative Analysis of Cervical Spine Management in a Subset of Severe Traumatic Brain Injury Cases Using Computer Simulation

BACKGROUND: No randomized control trial to date has studied the use of cervical spine management strategies in cases of severe traumatic brain injury (TBI) at risk for cervical spine instability solely due to damaged ligaments. A computer algorithm is used to decide between four cervical spine management strategies. A model assumption is that the emergency room evaluation shows no spinal deficit and a computerized tomogram of the cervical spine excludes the possibility of fracture of cervical vertebrae. The study's goal is to determine cervical spine management strategies that maximize brain injury functional survival while minimizing quadriplegia. METHODS/FINDINGS: The severity of TBI is categorized as unstable, high risk and stable based on intracranial hypertension, hypoxemia, hypotension, early ventilator associated pneumonia, admission Glasgow Coma Scale (GCS) and age. Complications resulting from cervical spine management are simulated using three decision trees. Each case starts with an amount of primary and secondary brain injury and ends as a functional survivor, severely brain injured, quadriplegic or dead. Cervical spine instability is studied with one-way and two-way sensitivity analyses providing rankings of cervical spine management strategies for probabilities of management complications based on QALYs. Early collar removal received more QALYs than the alternative strategies in most arrangements of these comparisons. A limitation of the model is the absence of testing against an independent data set. CONCLUSIONS: When clinical logic and components of cervical spine management are systematically altered, changes that improve health outcomes are identified. In the absence of controlled clinical studies, the results of this comparative computer assessment show that early collar removal is preferred over a wide range of realistic inputs for this subset of traumatic brain injury. Future research is needed on identifying factors in projecting awakening from coma and the role of delirium in these cases.