The benefit of thrombolytic therapy given late after the onset of acute myocardial infarction (AMI) has been controversial because of low reperfusion rates and limited myocardial salvage. Persistent chest pain has been used as a criteria for late intervention, but there is little documentation to validate this practice. Clinical outcomes and myocardial salvage were evaluated in 74 patients with AMI and persistent chest pain who underwent late reperfusion (> 6 hours) with direct coronary angioplasty, and these were compared with outcomes in 460 patients with early reperfusion (< or = 6 hours). Patients with late reperfusion had a high infarct artery patency rate (96%), a low hospital mortality rate (5.4%), and a low incidence of reinfarction (1.4%) and recurrent ischemia that were similar to patients with early reperfusion. Patients with late reperfusion had surprisingly good recovery of left ventricular function with improvement in ejection fraction from 50% to 60% at follow-up angiography. Patients with late reperfusion had a greater incidence of collateral flow (45% vs 22%, p < 0.001) and a lower value of peak creatine kinase (1,357 vs 2,057 U/liter, p < 0.001) than patients with early reperfusion. This study emphasizes the importance of persistent chest pain as a marker of continued myocardial viability in patients who present late after AMI. These data suggest that the probable mechanism of continued viability is preserved flow to the infarct zone. Patients with AMI and persistent chest pain may benefit from reperfusion therapy beyond 6 to 12 hours.