RESUMO
PURPOSE: Biomarkers are substances measured at the systemic level to evaluate organic responses in certain situations, establishing diagnoses, disease staging, and prognosis. Blood glucose is a biomarker recognized as a predictor of prognosis in children victims of traumatic brain injury (TBI). The scope of this study was to identify the accuracy of blood glucose as a biomarker of severe brain injury. METHODS: A retrospective analytical study was conducted through the consecutive review of medical records of children and teenage victims of TBI who underwent neurological surgery between 2016 and 2023 in a level 1 trauma center. Two groups were compared: children with Glasgow Coma Scale (GCS) score ≤ 8 and children with GCS > 8. We calculated the predictive values to define the accuracy of blood glucose as a biomarker of brain injury. RESULTS: Ninety-two medical records were included for analysis. Hyperglycemia predominated in cases with GCS ≤ 8 (48% vs 3%; p < 0.0001; OR, 30; 95% CI, 5.9902-150.2448). The glycemic measurement considering the cutoff point of 200 mg/dL or 11.1 mmol/L showed a specificity of 97%, a positive predictive value of 86%, an accuracy of 84%, and a likelihood ratio for a positive test of 16. CONCLUSION: Victims with GCS ≤ 8 are 16 times more likely to develop acute hyperglycemia after TBI when compared to those with GCS > 8. Blood glucose is a biomarker with an accuracy of 84% to predict severe brain injury, considering the cutoff point of 200 mg/dL or 11.1 mmol/L.