OBJECTIVE To identify clinical criteria predictive of underlying coronary artery disease in patients with cocaine-associated myocardial infarction. METHODS Using a retrospective cross-sectional study design at 29 acute care hospitals, we identified 70 patients with cocaine-associated myocardial infarction who had a determination of the presence or absence of coronary artery disease. Clinical characteristics of patients with coronary artery disease (> 50% stenosis on cardiac catheterization or reversible ischemia on stress test) were compared with patients without coronary artery disease (< 50% stenosis on cardiac catheterization). RESULTS Compared with patients without coronary artery disease (n = 21), patients with coronary artery disease (n = 49) were older (42 versus 31 years; P < 0.001), had more traditional cardiac risk factors (2.3 versus 1.5; P < 0.001), more frequent history of hypertension (odds ratio [OR], 5.3; 95% confidence interval [CI], 1.4 to 20.4); more frequent family history of myocardial infarction (OR, 4.4; 95% CI, 1.3 to 15.1), more bradydysrhythmias (OR, 8.0; 95% CI, 1.0 to 65.5), and more likely to have an inferior infarct location (P = 0.04). CONCLUSIONS Age, number of cardiac risk factors, location of myocardial infarction, and bradydysrhythmias predict underlying coronary artery disease in patients with cocaine-associated myocardial infarction. If validated, this knowledge may be used to develop a medically appropriate, cost-effective evaluation strategy for patients following cocaine-associated myocardial infarction.