RESUMO
OBJECTIVE: The aim of this study was to determine the effectiveness of hematological parameters measured at the moment of admission to the emergency room in predicting in-hospital mortality and to determine cut-off values of strongly predictive values. SUBJECTS AND METHODS: The study began with approval of the ethics committee. In total, 1,929 patients over 18 years of age, whose date could be obtained, were included in the study. From the hemogram parameters, white blood cells (WBC), red cell distribution width, mean platelet volume, and ratio of neutrophils to lymphocytes (NLR) values were determined and recorded. CK-MB and high-sensitive Troponin T values were recorded as cardiac markers. For statistical analysis, "SPSS for Windows Version 21" package program was used. FINDINGS: About 71.7% (n = 1384) of the patients were male and 28.3% (n = 545) of the patients were female. About 92.5% of the patients (n = 1785) were discharged from the hospital, whereas the remaining 144 patients (7.5%) were exitus in the hospital. When the efficacy of hematological parameters and cardiac markers in predicting mortality was examined by receiver operating charecteristics analysis, NLR was found to be the strongest predictor (area under the curve [AUC], 0.772, standard deviation [SD] = 0.022, 95% confidence interval [CI]). It was found that the WBC value came in second place after NLR as a strong predictor of mortality (AUC, 0.749, SD = 0.024, % 95 CI). CONCLUSION: The use of predictors for the prediction of mortality for ST elevation myocardial infarction patients is of great importance for faster implementation of treatment modalities. We found that WBC and especially NLR values obtained with a simple method can be used as powerful predictors.
