There is evidence that coronary collaterals improve the prognosis in patients with acute myocardial infarction (MI). However, there is limited clinical information on the protective role of collaterals in patients with stable coronary artery disease. This information may help risk stratification and the development of novel therapies, such as arteriogenesis and angiogenesis. The relation between collaterals and cardiac death or MI at 1 year after coronary revascularization was studied in 561 patients who were enrolled in a randomized study that compared stent implantation with bypass grafting. Collaterals were assessed on an angiogram using Rentrop's classification and considered present with a Rentrop grade >1. Unadjusted and adjusted odds ratios for cardiac death or MI at 1 year were calculated using univariate and multivariate regression analyses. In addition, determinants of collaterals were assessed using univariate and multivariate analyses. Collaterals were present in 176 patients (31%). The adjusted odds ratio of cardiac death or infarction was 0.18 (95% confidence interval 0.04 to 0.78) in the presence of collaterals. Independent determinants of collaterals were age (odds ratio 0.97, 95% confidence interval 0.95 to 0.99), multivessel disease (odds ratio 1.60, 95% confidence interval 1.02 to 2.51), impaired ventricular function (odds ratio 1.85, 95% confidence interval 1.04 to 3.29), type C lesion (odds ratio 3.72, 95% confidence interval 2.33 to 5.95), and stenosis severity >90% (odds ratio 9.08, 95% confidence interval 4.65 to 17.73). In conclusion, in patients with a low risk profile, the presence of collaterals protects against cardiac death and MI at 1 year after coronary revascularization. Variables that reflect the duration and severity of the atherosclerotic and ischemic burden determine their presence.