Value of high-sensitive cardiac troponin in predicting mortality in the emergency department

Abstract

Background: The present study aimed to determine the validity of hs-cTnT in predicting the mortality among patients presented to the emergency department (ED) likely to be acute coronary syndrome (ACS). Method: Patients in whom hs-cTnT was tested in the ED for a possible ACS composed the study cohort. Hs-cTnT levels of the study patients were obtained from the hospital computerized database. The outcome and mortality of the study patients was detected using the digitalized national mortality registry. All-cause mortality and cardiac mortality at the end of one month was the primary outcome. Results: 11795 patients were eligible for mortality detection and included into the final analysis. 1246 of 11795 patients were established to be dead during the study period and 358 of them supposed to be cardiac in origin. The Area Under the Curve (AUC) value of initial hs-cTnT for predicting one-month cardiac mortality was 0.869 (95% CI: 0.863-0.875) and 0.861 (95% CI: 0.855-0.867) for one-month all-cause mortality. The upper reference limit, 14 ng/L, has a sensitivity and specificity of 87% (95% CI: 77% to 94%) and 69% (95% CI: 68% to 70%), respectively, for predicting one-month cardiac mortality. Conclusions: The reference value of initial hs-cTnT does not have the ability to predict the cardiac mortality in a sufficient manner. However, reductions or increases in absolute or relative hs-cTnT levels are in concordance with mortality rates.