ObjectiveTo investigate the effect of using hemoglobin-revised trauma score (HB-RTS) on predicting mortality risk in trunk injury patients.
MethodsAccording to case control study rules, medical records of patients with trunk injury who were treated between July 2011 and October 2013, were collected. Patients were divided into two groups according to their survival prognosis within 28 days. The HB-RTS, revised trauma score (RTS), and injury severity score (ISS) of the two groups were calculated. The receiver operating characteristic (ROC) curve was drawn, and the area under the curve (AUC) and Youden's index were calculated. And then, we selected the optimal cutoff value using HB-RTS and RTS to predict the trunk injury patients' mortality and calculated their sensitivity, specificity, and accuracy. In the end, the correlation of ISS with HB-RTS and RTS using linear correlation analysis method was analyzed.
ResultsBesides RTS, HB-RTS and ISS in the death group were statistically significant higher than that in the survival group. The AUC of HB-RTS and RTS of the two groups were 0.922 and 0.888, respectively. The sensitivity of HB-RTS and RTS were 91.5% and 83%; and the specificity of HB-RTS and RTS were 95.7% and 76.6%, respectively. The correlation of HB-RTS and ISS was -0.592, while the correlation of RTS and ISS was 0.585.
ConclusionCompared to RTS, ROC of HB-RTS is enlarged, the specificity is increased, and the sensitivity is decreased. HB-RTS has stronger correlation with ISS than with RTS. The predictive effect of HB-RTS is higher than RTS in predicting mortality of trunk injury patients.
