Citation

Yao S, Song J, Li S, Cao C, Fang L, Wang C, Xu G. World Neurosurg. 2017; 101: 528-533.

Affiliation

Department of Neurosurgery, Cognitive Neuroscience Laboratory, Wuhan General Hospital, 627 Wu Luo Road, Wuhan, 430070, Hubei Province, PR China; Department of Neurosurgery, Wuhan General Hospital, Southern Medical University, 1063 Sha Tai Road, Guangzhou, 510515, Guangdong Province, PR China. Electronic address: yaoshun2008@yeah.net.

Copyright

(Copyright © 2017, Elsevier Publishing)

DOI

10.1016/j.wneu.2017.02.072

PMID

28249827

Abstract

OBJECTIVE: The Helsinki computerized tomography (CT) scoring system was developed to predict long-term outcome in patients with traumatic brain injury (TBI) two years ago; however, it has not yet been external validated. This study aimed to determine whether this system could be used as an independent predictor for TBI.

METHODS: This retrospective cohort study was performed on 302 consecutive patients with TBI. Univariate and multivariate logistic regressions and receiver operating characteristic (ROC) curve analyses were employed to determine the relationship between initial Helsinki CT scores and mortality and unfavorable neurological outcome at 6 months post-injury. Outcomes were assessed using the Glasgow Outcome Scale (GOS, scores of 1-3 defined as unfavorable outcome).

RESULTS: Of all patients, mortality was 17.9% and unfavorable outcome was 41.4%. The Helsinki CT score was significantly associated with the 6-month outcome in univariate analyses (p<0.05). After adjustment for other factors in the multivariate regression analysis, the Helsinki CT score remained an independent predictor for mortality (OR 1.22, 95% CI 1.08-1.39, p = 0.002) and unfavorable outcome (OR 1.14, 95% CI 1.04 - 1.26, p = 0.007). ROC curve analyses showed that the Helsinki CT score possessed good discrimination ability for mortality (AUC 0.81, 95% CI 0.75-0.87, p<0.001) and moderate discrimination ability for unfavorable outcome (AUC 0.74, 95% CI 0.69-0.80, p<0.001). Moreover, at 1.9 hours following TBI, the Helsinki CT score was most accurate for predicting mortality (accuracy, 74.5%) and unfavorable outcome (accuracy, 71.5%).

CONCLUSION: The Helsinki CT score showed good prognostic discrimination and can be used as an independent predictor for long-term outcome prediction in TBI patients.

Copyright © 2017 Elsevier Inc. All rights reserved.

Language: en

Keywords

Computerized tomography; Helsinki CT score; Long-term outcome prediction; Traumatic brain injury