Predicting Outcomes of Penetrating Cardiovascular Injuries at a Rural Center by Different Scoring Systems

OBJECTIVE: To compare the anatomical and physiological scoring systems and the outcomes of surgical management of penetrating cardiovascular trauma at a rural center. METHODS: Seventy-seven patients underwent emergency surgery at our center between January/2012 and October/2018 due to penetrating cardiovascular trauma. Injury Severity Score (ISS), Revised Trauma Score (RTS), New Injury Severity Score (NISS), and Trauma and Injury Severity Score (TRISS) were calculated. The validation of these risk scores to predict mortality was assessed by the area under the receiver operating characteristic curve analysis. RESULTS: All trauma scores were correlated with mortality. As ISS, NISS, and TRISS values increased and RTS values decreased, the mortality rate increased. The area under the curve (AUC) in the receiver operating characteristic curve analysis was 0.943 for TRISS, 0.915 for RTS, 0.890 for ISS, and 0.896 for NISS (P<0.001 for each). Logistic regression analysis revealed that scores were correlated with mortality (P<0.001 for each). By investigating cardiac injuries alone, only TRISS and RTS results correlated with mortality for cardiac injuries (Mann-Whitney U test, P=0.003 and P=0.01, respectively). The AUC was only statistically significant for TRISS and RTS (AUC=0.929, P<0.05 for both). For vascular injuries, all the scores were significantly correlated with in-hospital mortality (Mann-Whitney U test, P<0.001 for each). TRISS had the highest AUC (AUC=0.946, P<0.001). CONCLUSION: TRISS has the highest predictivity for in-hospital mortality in patients with penetrating cardiovascular trauma.