Comparison of the GRACE score, TIMI score and a New Laboratorial Score to predict adverse outcomes in acute coronary syndrome

Abstract Introduction Acute coronary syndrome (ACS) is a clinical entity which includes a heterogeneous group of patients with different outcomes. Risk scores are in this setting a resourceful tool to identify the subset of patients with a worse prognosis, in order to plan therapeutic and surveillance strategies. Aim To create a risk score – Laboratory Risk Score (LRS) – which exclusively includes analytical and echocardiographic parameters, as a predictor of adverse outcomes (in-hospital mortality and 1-year mortality), and compare it with other well-known scores: GRACE Score (GS) and TIMI-score (TS). Methods A retrospective cohort study was conducted, which included patients admitted in the Cardiology Department with the diagnosis of ACS. In order to calculate the new LRS, the authors attributed the value of 1 to each of the satisfied condition from the following: leucocytes >11,7g/L, hemoglobin <13.3g/dL, red cell distribution width >14%, prothrombinemia <90%, glycaemia at admission >143mg/dL, urea >53.5mg/dL, creatinine >1.16mg/dL, reactive C-protein >1.0mg/dL, maximum troponin >35.0ng/dL, natriuretic brain peptide >416 pg/dL and left ventricular ejection fraction <40%. LRS resulted from the sum of the satisfied conditions. ROC curves for LRS, GS and TS to predict in-hospital mortality and to predict 1-year mortality were constructed. The statistical analysis was performed in SPSS and Medcalc. p value <0.05 was considered statistically significant. Results 1714 patients (70.4% male, average age 69±13 years-old) were included in this study. Intra-hospital mortality rate was 6.8% and 1-year mortality rate after de discharge was 4.8%. The areas under the ROC curves for predicting in-hospital mortality were the following: 0,790 (LRS, p<0,001), 0,793 (GS, p<0.01), 0.817 (TS, p<0.001). For predicting 1-year mortality, the areas under the ROC curves were: 0,715 (LRS, p<0,001), 0,761 (GS, p<0,001), 0.742 (TS, p<0.001). Pairwise comparison of ROC curves showed no significant differences between the scores. Conclusion The above-mentioned risk scores, including the new LRS, are obtained with non-invasive and widely available parameters and displayed a good performance in predicting in-hospital and 1-year mortality. Pairwise comparison of ROC curves demonstrated that the new laboratorial score was not inferior predicting adverse outcomes. The SRL is an easily obtained score, that shows a statistical significance in predicting mortality, especially the prediction of in-hospital mortality. Funding Acknowledgement Type of funding sources: None.