Beyond Glasgow Coma Scale: Prehospital prediction of traumatic brain injury

Introduction: Early identification of traumatic brain injury followed by timely, targeted treatment is essential. We aimed to establish the ability of prehospital Glasgow Coma Scale score alone and combined with vital signs to predict hospital-diagnosed traumatic brain injury.

Methods: This study included adults from the 2017-2020 Trauma Quality Improvement Program data set with blunt mechanism. We calculated test characteristics of prehospital Glasgow Coma Scale score ≤12 alone and Glasgow Coma Scale score combined with heart rate and systolic blood pressure for predicting (1) any traumatic brain injury and (2) moderate to severe traumatic brain injury. Diagnostic performances were calculated in all patients and older adults (≥55 years). We used decision curve analysis to determine the net diagnostic benefit of prehospital Glasgow Coma Scale score combined with heart rate + systolic blood pressure over Glasgow Coma Scale score alone.

Results: Of 1,687,336 patients, 39.1% had any traumatic brain injury, 3.7% had moderate to severe traumatic brain injury, and 9.1% had a prehospital Glasgow Coma Scale score ≤12. Prehospital Glasgow Coma Scale score ≤12 alone had a sensitivity 83.1%, specificity 93.7%, negative predictive value 99.3%, and positive predictive value 33.7% for predicting moderate to severe traumatic brain injury. Adding prehospital heart rate <65/min and systolic blood pressure >150 mm Hg to Glasgow Coma Scale score ≤12 improved the positive predictive value for moderate to severe traumatic brain injury (55.3%), with a preserved negative predictive value of 96.4%. Decision curve analysis showed the traumatic brain injury prediction model including prehospital heart rate and systolic blood pressure had the greatest net benefit across most threshold probabilities.

Conclusion: Less than a third of adult blunt trauma patients with a prehospital Glasgow Coma Scale score ≤12 have moderate to severe traumatic brain injury. Supplementing Glasgow Coma Scale score with prehospital vital signs improves diagnostic accuracy, potentially by filtering out patients with altered consciousness due to shock. Future work should better identify patients for traumatic brain injury-specific treatments in prehospital settings, including triage destination.