We designed this study to evaluate whether controlling nutritional status (CONUT) score was valuable in predicting outcome of patients with traumatic brain injury (TBI).
Clinical and laboratory data of patients admitted with TBI in our hospital between January 2015 and January 2018 was retrospectively collected. We divided patients into two groups according to in-hospital and 90-day outcome respectively and compared baseline characteristics of two groups. Multivariate analyses were conducted to evaluate whether CONUT score was associated with outcome of included patients. Receiver operating characteristics (ROC) curves were drawn to compare the predictive value of CONUT score and constructed models.
A total of 365 patients were included in our study. Multivariate logistic regression analyses showed that CONUT score was independently associated with in-hospital mortality (OR = 1.244, 95% CI: 1.099-1.408, p = 0.001) and 90-day outcome (OR = 1.441, 95% CI: 1.193-1.741, p < 0.001). ROC curve showed that area under the ROC curve (AUC) of CONUT score for predicting in-hospital mortality and 90-day outcome were 0.790 (95% CI: 0.745-0.836, p < 0.001) and 0.839 (95% CI: 0.792-0.886, p < 0.001), respectively.
CONUT score is independently associated with in-hospital mortality and 90-day outcome of patients with TBI.

Copyright © 2020 Elsevier B.V. All rights reserved.