Performance of the BIG Score in predicting mortality in normotensive children with trauma

Abstract

Objectives: Children have a larger reserve for traumatic hemorrhagic shock, requiring a score that uses physiologic variables other than hypotension. Recently, the BIG score comprising admission base deficit, international normalized ratio, and the Glasgow Coma Scale has been reported to predict traumatic mortality. We aimed to validate the performance of the BIG score in mortality prediction of normotensive children with trauma. Methods: We reviewed 1046 injured children (<18 years) who visited a Korean academic hospital from 2010 to 2018, excluding those with age-adjusted hypotension. In-hospital mortality, the BIG score and its predicted mortality, Revised Trauma Score, and Pediatric Trauma Score were calculated. We compared areas under the curve (AUCs) for in-hospital mortality of the 3 scores and did in-hospital and BIG-predicted mortalities. Results: Of the 1046 children, 554 were enrolled with a 4.9% in-hospital mortality rate. The median BIG score was higher in the nonsurvivors (6.4 [interquartile range, 4.4–9.2] vs 20.1 [16.5–24.8]; P < 0.001). The AUC of the BIG score was 0.94 (95% confidence interval [CI], 0.92–0.96), which was higher than that of Pediatric Trauma Score (0.87 [95% CI, 0.84–0.90]; P < 0.001). The AUC of the BIG score tended to be higher than that of Revised Trauma Score without statistical significance (0.90 [95% CI, 0.87–0.92]; P = 0.130). We noted a parallel between in-hospital and BIG-predicted mortalities. The hemorrhage-related nonsurvivors showed higher median base deficit and BIG score than did the isolated traumatic brain injury-related ones. Conclusions: The BIG score can predict mortality with excellent accuracy in normotensive children with trauma.