Abstract

Primary outcomes were 30-day ACS and myocardial infarction. Results: There were 315 patients with 30-day ACS and 1833 patients without ACS. Likelihood ratio analysis revealed significant discrepancies in weight of the 5 individual elements shared by the HEART and HEARTS3 score. The HEARTS3 score outperformed the HEART score as determined by comparison of areas under the receiver operating characteristic curve for myocardial infarction (0.958 vs 0.825; 95% confidence interval difference in areas, 0.105-0.161) and for 30-day ACS (0.901 vs 0.813; 95% confidence interval difference in areas, 0.064-0.110). Conclusion: The HEARTS3 score reliably risk stratifies patients with chest pain for 30-day ACS. Prospective studies need to be performed to determine if implementation of this score as a decision support tool can guide treatment and disposition decisions in the management of patients with chest pain. © 2012 Published by Elsevier Inc. 1.