Background Time-to-treatment is important for survival in patients with acute myocardial infarction (AMI) treated with fibrinolytic therapy, but the importance of time-to-treatment with primary percutaneous coronary intervention (PCI) is controversial. Previous studies evaluating the importance of time-to-treatment with primary PCI have not analyzed patients with cardiogenic shock separately. Methods Consecutive patients with AMI (n = 1843) treated with primary PCI were prospectively enrolled in the LeBauer Cardiovascular Research Foundation Registry. Late clinical follow-up was obtained in 98% of patients, at a mean time of 6.1 years. Results Reperfusion times were longer in women and patients with diabetes mellitus and shorter in patients with prior myocardial infarction. In patients with shock (n = 138), the inhospital mortality rate increased progressively with increasing time-to-reperfusion (<3 hours, 31%; 3-<6 hours, 50%; ≥6 hours, 62%; P = .01), whereas in patients without shock (n = 1705), inhospital and late mortality rates were similar across 3 categories of time to reperfusion (<3 hours, 5.8%; 3-<6 hours, 4.6%; ≥6 hours, 4.8%; P = .46). After adjusting for differences in baseline variables, reperfusion time was a significant independent predictor of inhospital mortality in patients with shock, but not in patients without shock. Conclusions Reperfusion time with primary PCI is important for survival in patients with shock, but appears to be less important in patients without shock. These data emphasize the importance of achieving early reperfusion in patients with shock and have implications on the triage of patients without shock for mechanical reperfusion and the mechanism of benefit of reperfusion therapy with primary PCI. (Am Heart J 2003;145:708-15.)