Earlier studies have suggested that immediate percutaneous coronary intervention (PCI) following thrombolytic therapy for acute myocardial infarction (AMI) is associated with an increase in adverse events and that routine PCI in this setting has offered no advantage over a conservative strategy. To reassess this issue in a more recent era, we evaluated 1,938 patients from the Thrombolysis in Myocardial Infarction (TIMI) 10B and 14 trials of AMI. Patients in TIMI 10B were randomized to receive tissue plasminogen activator or TNK tissue plasminogen activator, whereas patients in TIMI 14B trial were randomized to receive thrombolytic therapy with or without abciximab. All patients underwent angiography 90 minutes after receiving pharmacologic therapy. Patients who underwent PCI were classified as having undergone a rescue procedure (TIMI 0 or 1 flow at 90 minutes), an adjunctive procedure (TIMI 2 or 3 flow at 90 minutes), or a delayed procedure (performed >150 minutes after symptom onset, median of 2.75 days). Among patients with TIMI 0 or 1 flow, there was a trend for lower 30-day mortality among patients who underwent rescue PCI than among those who did not (6% vs 17%, p = 0.01, adjusted p = 0.28). Patients who underwent adjunctive PCI had similar 30-day mortality and/or reinfarction as those who underwent delayed PCI. In a multivariate model both had lower 30-day mortality and/or reinfarction than patients with "successful thrombolysis" (i.e., TIMI 3 flow at 90 minutes) who did not undergo revascularization (p = 0.02). Thus, early PCI following AMI is associated with excellent outcomes. Randomized trials of an early invasive strategy following thrombolysis are warranted.