Validation of the GRACE score for prognosis in Indian patients with acute coronary syndromes

Aim: To validate the global registry of acute coronary events (GRACE) score in acute coronary syndromes (ACS) patients and study its angiographic correlation. Methods and results: Two-hundred and thirty-five ACS patients were studied for the combined endpoint of all-cause in-hospital mortality and non-fatal infarction/reinfarction. We tested the predictive accuracy of the composite GRACE score using the receiver operating characteristics (ROC) curve. Lower systolic blood pressure (SBP) (odds ratio [OR] 7.93, P=0.005), ST-segment deviation (OR 7.79, P=0.02) and cardiac biomarker positivity (OR > 6.52, P=0.01) were significantly associated with events. Serum creatinine > 1.4 mg/dL showed a trend towards statistical significance (OR 4.14, P=0.05), whereas age > 50 years (OR 3.62, P=not significant [NS]) and Killips class 4 (OR 2.71, P=NS) showed good association. The best value for predicting events was a GRACE score of > 217 and these patients were more likely to have double/triple vessel disease (P = 0.0009). The C statistic for the GRACE score was 0.75. Conclusion: Higher GRACE score predicts in-hospital events and more severe angiographic coronary artery disease (CAD).