Background: Previous studies of rescue percutaneous coronary intervention (PCI) for failed thrombolysis yielded conflicting results. In the current era of newer thrombolytic agents, coronary stents, glycoprotein IIb/IIIa inhibitors, and aggressive hemodynamic support, the outcome of this high‐risk patient group has not been characterized. Methods: From January 2000 to October 2004, 214 consecutive patients were transferred and underwent emergent coronary angiography following failed thrombolysis. One hundred and fifty five (72%) underwent immediate PCI, 23 (11%) underwent delayed PCI, and 36 (17%) received surgical revascularization or medical therapy. Medical records and angiograms for the entire PCI cohort (n= 178) were reviewed for in‐hospital events including bleeding complications, stroke, recurrent ischemia or myocardial infarction (MI), target vessel revascularization (TVR), and death. Results: Time from symptom onset to thrombolysis (mean ± standard deviation) was 5.6 ± 11.9 hr, and time from thrombolysis to angiography was 7.0 ± 5.5 hr. The study cohort was critically ill, with 9.6% experiencing cardiac arrest, 21% in cardiogenic shock, and 12% intubated prior to transfer. Coronary stents were placed in 88%, Rheolytic thrombectomy was used in 21%, an intraaortic balloon pump was placed in 17%, and a glycoprotein IIb/IIIa inhibitor was administered in 92%. Patients receiving delayed PCI had higher TIMI 3 flow grade at initial angiography than those receiving immediate PCI (83% vs. 34%, respectively, P < 0.0001). Angiographic success was 90% for the entire PCI cohort, 89% for the immediate PCI group, and 100% for the delayed PCI group. Clinical success (angiographic success and freedom from major adverse cardiac events) was 85% for the entire PCI cohort, 83% for the immediate PCI group, and 100% for the delayed PCI group. Severe and moderate bleeding complications occurred in 7.3%, stroke in 1.7%, recurrent ischemia or MI in 7.3%, and TVR in 3.4%. Overall, in‐hospital mortality for the entire PCI cohort was 3.4%. Conclusions: This observational, consecutive, real‐world study of contemporary rescue PCI for failed thrombolysis shows a high use of coronary stents, Rheolytic thrombectomy, glycoprotein IIb/IIIa inhibitors, and intraaortic balloon pump placement. Angiographic and clinical success was high with low bleeding complications and low in‐hospital mortality, suggesting that prospective, randomized trials using contemporary interventional therapy for rescue PCI be considered. © 2006 Wiley‐Liss, Inc.