This retrospective observational study included severe
trauma patients with available medical records from January 2017 to December 2018 but excluded those with IHI. Logistic regression analysis was conducted to identify the risk factors for mortality in non-IHI patients, and adjustments were performed for relevant covariates. An area under the receiver operating characteristics curve (AUROC) analysis was performed to examine the primary outcome of our study, which was mortality at hospital discharge in severe non-IHI
trauma patients.
RESULTS: Of the 483 severe non-IHI
trauma patients included in the study, 86 patients (17.8 %) died. The multiple logistic regression analysis demonstrated ISI (odds ratio [OR], 2.300; 95% CI, 1.183-4.470) was significantly associated with mortality in the non-IHI group. Additionally,
trauma and injury severity score (TRISS; OR, 0.538; 95% CI, 0.447-0.649),
lactate (OR, 1.410; 95% CI, 1.252-1.588),
creatinine (OR, 1.554; 95% CI, 1.221-1.979), and activated partial thromboplastin time (aPTT; OR, 1.050; 95% CI, 1.021-1.080) were independently associated with mortality at hospital discharge. The AUROC values for TRISS,
lactate, aPTT,
creatinine, and ISI were as follows: 0.892 (95% CI, 0.861-0.918), 0.838 (95% CI, 0.803-0.870), 0.754 (95% CI, 0.712-0.792), 0.650 (95% CI, 0.606-0.693), and 0.848 (95% CI, 0.813-0.879), respectively. The AUROC for the multiple logistic regression model with ISI was 0.942 (95% CI, 0.917-0.962). In a model in which TRISS was omitted, the addition of ISI to other predictors significantly improved the AUROC to 0.900 (95% CI, 0.869-0.925) (p=0.039).
CONCLUSION: The initial ISI in the ED after
trauma was associated with mortality in severe non-IHI
trauma patients. In conjunction with other prognostic indicators, it could be used as an early prognostic marker, particularly if TRISS is unavailable.