Predict Severe Traumatic Brain Injury

Severe traumatic brain injury (TBI) is associated with a 20-30% mortality rate and significant disability among most survivors. The Centers for Disease Control and Prevention (CDC) estimate that 2% of the U.S. population lives with disabilities directly attributable to TBI, with annual costs exceeding $76.5 billion. Current treatments are largely ineffective because they are instituted after irreversible damage has already occurred. By the time intracranial pressure (ICP) increases or brain tissue oxygen tension (PbtO2) decreases to harmful levels, it is often too late to reverse or repair the damage. A computerized method has been developed that can predict these injurious events ahead of time, allowing clinicians to intervene before further damage occurs. The goal of this proposal is to test these predictions in real time. The first phase of the project (Year 1) involves setting up the informatics infrastructure, with no patient interaction. In the second phase (Year 2), subjects, through surrogate decision-makers, will be enrolled in an observational study where data on intracranial pressure and brain tissue oxygen tension will be collected, and the prediction algorithm will be tested for accuracy. Clinical management will follow standard care protocols, and no additional interventions will be performed. Approximately 120 individuals will participate in this study at the University of Chicago and Ben Taub General Hospital in Houston. Data collected will include both the electronic medical record and data from bedside intensive care unit monitors. The electronic medical record includes demographic information, injury characteristics, laboratory values, and imaging data, while the intensive care unit monitor provides real-time vital signs such as intracranial pressure, brain tissue oxygen tension, and mean arterial pressure. These data will be securely stored in a research computer database. Efforts will be made to contact subjects or their caretakers at 6 mont