Although emergency coronary artery bypass for complications of percutaneous transluminal coronary angioplasty (PTCA) has proved to be a relatively successful 'bail-out' procedure, little is known about the durability of revascularization under these potentially disastrous circumstances. The authors therefore retrospectively examined their results with this procedure. Emergency coronary artery bypass for complications of PTCA was performed in 112 patients between 1 January 1984 and 19 May 1992. Fifteen patients underwent PTCA for acute myocardial infarction. Eleven patients (9.8%) were stable, and underwent emergency coronary artery bypass after PTCA because of suboptimal angiographic results from percutaneous transluminal coronary angioplasty. None of these stable patients died. The remainder of the patients underwent emergency coronary artery bypass after PTCA because of ongoing documented ischemia, including cardiac arrest requiring cardiopulmonary resuscitation during transit to the operating room in 11 patients (9.8%) and preoperative intra-aortic counterpulsation in 24 (21.4%). The average number of coronary arteries bypassed at emergency coronary artery bypass was 2.2, and 19 patients (17%) received at least one mammary artery conduit. The perioperative incidence of myocardial infarction was 8.9% (10/112), and the operative mortality rate 8% (9/112). During follow-up, which averaged 55 months, the survival rate (including operative mortality) was 85% while 98% of patients experienced freedom from reoperative coronary bypass, 89% experienced freedom from myocardial infarction (including postoperative) and 90% experienced freedom from subsequent catheterization or PTCA. In conclusion, emergency coronary artery bypass for PTCA complications successfully avoids subsequent untoward cardiac events. When compared with published results of PTCA without emergency coronary artery bypass, emergency coronary bypass is more reliable for avoiding subsequent cardiac catheterization (with or without PTCA) than PTCA alone.