The Validity Of The Global Registry Of Acute Coronary Events (GRACE) Acute Coronary Syndrome Prediction Model For In Hospital Mortality In A Sub Population Of Chongqing

Objective: To determine the validity and applicability of the GRACE prediction model for in-hospital mortality in all forms of acute coronary syndrome (ACS) in a sub population of Chongqing.Methods: Data of 667 ACS patients were collected retrospectively from Jan 2005 to Apr 2008 and were recorded on a standardized case report form. For each patient the GRACE risk score was calculated (using the GRACE calculator available from the grace website) using specific variables collected at admission. Patients with missing data and those transferred from other hospitals were excluded. Receiver operating characteristic (ROC) curves was plotted for the GRACE risk score.Results: Among 576 ACS patients, 98 (17.01%), 36 (6.25%), and 442 (76.74%) presented with ST-elevation myocardial infarction (MI), non-ST elevation MI and unstable angina, respectively. The GRACE risk score could not be determined in 91 (9.3%) patients due to missing data or for patients who were transferred from other hospitals and were excluded from the analysis. The median GRACE risk score was 133 (interquartile range: 92-174) and, the in-hospital rates of death and death/(re-)MI were 6.1% and 7.6 %, respectively. The GRACE risk score demonstrated excellent discrimination (c-statistic=0.86, 95% CI 0.79-0.91, P<0.001) for in-hospital death/ (re)-MI.Conclusion: The GRACE RS studied had a good predictive accuracy for death or MI across the wide range of ACS in this population. It may be a useful risk stratification tool that helps identify high risk patients who will benefit most from myocardial revascularization and low risk patients who may be spared from undergoing more aggressive interventional treatment.