Accuracy of the grace score in predicting mortality in patients with covid-19

Background: The GRACE and TIMI scores have been well validated for assessment of prognosis in non-ST-elevation acute coronary syndromes (NSTE-ACS). The purpose of this study was to elucidate the accuracy of predicting mortality in patients with NSTE-ACS and COVID-19 compared to a control population. Methods: The HCA Healthcare database yielded 22,243 patients at 152 hospitals across the US who meet the inclusion criteria of the study between March 2020 and November 2020. The GRACE score was applied to each patient retrospectively and a received operator characteristic (ROC), area under the curve (AUC) approach was used to compare the prediction of mortality in patients with COVID-19 to the control group. A logistic regression approach was used to elucidate the impact of COVID-19 to mortality after adjusting for other risk factors. Results: In the study population, 13.8% of the study patient had COVID-19. The mortality rate among the mortality rate among the COVID-19 group and the control population was 39.5% and 14.2% (p<0.001), respectively. The GRACE score had similar accuracy in predicting mortality in the COVID-19 (AUC 0.7) and control (AUC 0.71) groups. In multivariate analysis, the odd ratio of COVID-19 on mortality was 3.62 (CI: 2.65-4.94, p<0.01). Conclusion: The Grace score had similar accuracy in predicting mortality in NSTE-ACS patients with and without COVID-19 despite a higher mortality among the COVID-19 population. It remains a viable tool for risk stratifying COVID-19 patients with NSTE-ACS.