A risk score based on admission characteristics to predict progressive hemorrhagic injury from traumatic brain injury in children.

UNLABELLED

Traumatic brain injury (TBI) is one of the leading causes of death and disability in children, and progressive hemorrhagic injury (PHI) post TBI is associated with poor outcomes. Therefore, the objective of this study was to develop and validate a prognostic model that uses the information available at admission to determine the likelihood of PHI occurrence after TBI in children. The identified demographic data, cause of injury, clinical predictors on admission, computed tomography scan characteristics, and routine laboratory parameters were collected and used to develop a PHI prognostic model with logistic regression analysis, and the prediction model was validated in 68 children. Eight independent prognostic factors were identified: lower Glasgow coma scale score (3 ~ 8) (6 points), intra-axial bleeding/brain contusion (4 points), midline shift ≥5 mm (9 points), platelets <100 × 10/L (11 points), prothrombin time >14 s (6 points), international normalized ratio >1.25 (7 points), D-dimer ≥5 mg/L (14 points), and glucose ≧10 mmol/L (11 points). We calculated risk scores for each child and defined three risk groups: low risk (0-16 points), intermediate risk (17-36 points), and high risk (37-68 points). In the development cohort, the PHI rates after TBI for the low-, intermediate-, and high-risk groups were 10.1, 47.9, and 84.2%, respectively. In the validation cohort, the corresponding PHI rates were 10.9, 47.5, and 85.4%, respectively. The C-statistic for the point system was 0.873 (p = 0.586 by the Hosmer-Lemeshow test) in the development cohort and 0.877 (p = 0.524 by the Hosmer-Lemeshow test) in the validation cohort.

CONCLUSION

Using admission predictors, we developed a relatively simple risk score that accurately predicted the risk of PHI after TBI in children. What is Known: • TBI is one of the leading causes of death and disability in children, and PHI post TBI is associated with poor outcomes. •Prediction of patients at low risk of PHI could help reduce treatment costs, whereas identification of patients at high risk of PHI could direct early medical intervention to improve outcomes. What is New: • This study firstly developed a risk score system by assessing the admission information that could provide an earlier prediction of the occurrence of PHI after acute TBI in children.