Novel score for mortality risk prediction 6 months after acute coronary syndrome

Abstract

Aim. To create a prediction score for assessing the mortality risk 6 months after hospitalization with acute coronary syndrome (ACS).Material and methods. Based on the results of ACS RECORD-3 register (Russia), we determined independent mortality predictors 6 months after ACS by performing multivariate regression analysis in patients discharged alive from the hospital with known outcomes.Results. The following predictors were obtained during the analysis: non-prescription of aspirin at discharge (odds ratio (OR) 5,8; 95% confidence interval (CI) 2,315,0; p<0,0001), newly diagnosed heart failure, pulmonary edema or shock in a hospital (OR 5,7; 95% CI 2,6-12,7; p<0,0001), age ≥75 years (OR 5,3; 95% CI 2,710,6; p<0,0001), non-prescription of beta-blockers at discharge (OR 5,0; 95% CI 2,3-10,8; p<0,0001), in-hospital management without  immediate percutaneous coronary intervention (PCI) (primary PCI during ST-segment elevation ACS or PCI during the first 72 hours with non-ST-segment elevation ACS) (OR 3,9; 95% CI 1,69,8; p=0,004), the initial serum creatinine ≥100 µmol/L (OR 3,1; 95% CI 1,6-6,1; p=0,001), body mass index <30 kg/m2 (OR 2,8; 1,2-6,3; p=0,014). Each of them was evaluated at one point and was a component of the RECORD-6 score. Prediction sensitivity and specificity for the new score were 73,3% (95% CI 60,1-83,5) and 71,4% (95% CI 68,9-73,7), respectively; prediction accuracy, estimated as the area under the ROC curve was 0,931 (95% CI 0,897-0,964). The cut-off point was considered 3 points, which had the best ratio of prediction sensitivity and specificity. The mortality after 6 months with a value of <3 points was 1,6%, and with a value of ≥3 points — 10,1% (relative risk (RR) 0,16; 95% CI 0,09-0,28; p<0,0001), and the mortality after 12 months was 7,8% and 22,5%, respectively (RR 0,35; 95% CI 0,25-0,49; p<0,0001). Relative to the GRACE risk score for 6-month mortality showed that the prediction value of the RECORD-6 score was at least no worse.Conclusion. The novel RECORD-6 risk score is an accurate and simple prediction tool for assessing the mortality risk 6 months after discharge from the hospital. The prediction accuracy of the RECORD-6 risk score is not lower the GRACE risk score.