Experimental and clinical studies have suggested that late opening of an infarct-related artery (IRA) after myocardial infarction (MI) could improve clinical outcome. However, the suggestive observational data are limited by selection biases. Indeed, most small randomized studies have not demonstrated benefit. Thus, there is no recommendation for routine late opening of the IRA in current national guidelines for management of stable post-MI patients. The OAT is designed to test the hypothesis that opening a totally occluded IRA 3 to 28 days after MI in high-risk asymptomatic patients will improve clinical outcome and be cost-effective. The primary end point is the first occurrence of recurrent MI, hospitalization/treatment of New York Heart Association class IV congestive heart failure, or death. Trial background, design, and preliminary baseline characteristics of 2027 randomized patients are presented. Eligible patients are randomly assigned in equal proportions to optimal evidence-based medical care or optimal care plus late opening of the IRA using percutaneous coronary intervention of the occluded IRA. Treatment groups will be compared using intent-to-treat analysis. The results of OAT should have broad clinical impact by defining an evidence-based approach to the asymptomatic, high-risk, post-MI patient with an occluded IRA. If the efficacy and cost-effectiveness of percutaneous coronary intervention are established, then a policy of routinely seeking and opening persistently occluded IRAs could be advocated. If not, this strategy should be avoided in this large subgroup of post-MI patients.