New prehospital scoring system for traumatic brain injury to predict mortality and severe disability using motor Glasgow Coma Scale, hypotension, and hypoxia: a nationwide observational study

Objective Assessing the severity of injury and predicting outcomes are essential in traumatic brain injury (TBI). However, the respiratory rate and Glasgow Coma Scale (GCS) of the Revised Trauma Score (RTS) are difficult to use in the prehospital setting. This investigation aimed to develop a new prehospital trauma score for TBI (NTS-TBI) to predict mortality and disability. Methods We used a nationwide trauma database on severe trauma cases transported by fire departments across Korea in 2013 and 2015. NTS-TBI model 1 used systolic blood pressure <90 mmHg, peripheral capillary oxygen saturation <90% measured via pulse oximeter, and motor component of GCS. Model 2 comprised variables of model 1 and age >65 years. We assessed discriminative power via area under the curve (AUC) value for in-hospital mortality and disability defined according to the Glasgow Outcome Scale with scores of 2 or 3. We then compared AUC values of NTS-TBI with those of RTS. Results In total, 3,642 patients were enrolled. AUC values of NTS-TBI models 1 and 2 for mortality were 0.833 (95% confidence interval [CI], 0.815 to 0.852) and 0.852 (95% CI, 0.835 to 0.869), respectively, while AUC values for disability were 0.772 (95% CI, 0.749 to 0.796) and 0.784 (95% CI, 0.761 to 0.807), respectively. AUC values of NTS-TBI model 2 for mortality and disability were higher than those of RTS (0.819 and 0.761, respectively) (P<0.01). Conclusion Our NTS-TBI model using systolic blood pressure, motor component of GCS, oxygen saturation, and age was feasible for prehospital care and showed outstanding discriminative power for mortality.