Background: Clinical benefit of PCI is controversial in stable patients with ST-elevation myocardial infarction (STEMI) presenting between 12 and 72 hours after the symptom onset. We evaluated the efficacy and optimal timing of percutaneous coronary intervention (PCI) in early latecomers with acute STEMI. Methods: Employing data from the Korea Acute Myocardial Infarction Registry, we analyzed 2,640 stable STEMI patients with symptom-to-door time between 12 and 72 hours. Patients with cardiac arrest, ventricular arrhythmia, cardiogenic shock, hear failure (Killip class III or IV), fibrinolysis, or urgent PCI during initial conservative treatment were excluded. Patients were divided into PCI group (n=2,185) and medical treatment group (n=455). PCI group was further divided into two subgroups: early PCI (n=1,063; median door-to-balloon time=5.6 hours) and delayed PCI (n=1,122; median door-to-balloon time=42.6 hours). Twelve-month clinical outcome was compared, with composite of death and MI as the primary endpoint. Results: Patients in the PCI group were more often male and smokers; and more likely to have anterior MI with higher cardiac troponins, and history of MI, PCI, heart failure, stroke, and peripheral vascular disease. Patients receiving medical treatment were older and more often had renal dysfunction. After adjustment for confounders using propensity score methods, PCI group had lower mortality (3.2% vs. 10.1%; odds ratio [OR], 0.48; 95% confidence interval [CI], 0.32 to 0.73; P=0.001) and lower incidence of composite death/MI (3.9 vs. 11.2%; OR, 0.51; 95% CI, 0.35 to 0.74; P<0.001) at 12 months. Subgroup analysis between early and delayed PCI revealed that patients receiving early PCI more often had chest pain; shorter symptom-to-door time; anterior MI with higher cardiac troponins; and history of PCI, dyslipidemia, stroke, and chronic kidney disease. However, adjusted analysis using propensity score for early PCI found no statistical differences between early and delayed PCI in the rates of death (2.4% vs. 2.1%), death/MI (2.4% vs. 2.3%), and death/MI/repeat revascularization (3.5% vs. 3.3%) during 12-month follow-up. Conclusions: In stable patients with STEMI presenting 12 to 72 hours after symptom-onset, PCI was associated with significant improvement in 12-month clinical outcome. The optimal timing of PCI remains to be determined.