SUMMARY There are few data on the long-term effects of new Q waves on survival and morbidity after coronary bypass graft surgery (CABG). We followed 1340 patients who underwent CABG in 1978 at 10 hospitals participating in the Coronary Artery Surgery Study (CASS). The incidence of perioperative Qwave infarction was 4.6% (range 0.0-10.3% by hospital). The rate of infarction was higher in patients who had an increased left ventricular end-diastolic pressure or cardiomegaly on the preoperative chest radiograph. Patients who received more grafts or who had longer cardiopulmonary bypass time were also at higher risk of infarction. In a stepwise discriminant analysis of 44 clinical, angiographic and surgical variables, cardiopulmonary bypass time, topical cardiac hypothermia and cardiomegaly entered the stepwise selection of variables.Long-term survival was adversely affected by the appearance of new postoperative Q waves. The hospital mortality was 9.7% in the 62 patients who had new postoperative Q waves and 1.0% in the 1278 patients who did not (p < 0.001); the 3-year cumulative survival rates were 85% and 95%, respectively (p < 0.001).In patients who survived to hospital discharge, the presence of new postoperative Q waves did not adversely affect 3-year survival (94% and 96%, respectively). The survival rates were worse in patients who had a history of infarction or who had impaired left ventricular function preoperatively. The number of readmissions to hospital after CABG among the patients who had a transmural perioperative infarction was similar to that among patients who did not.We conclude that the appearance of new Q waves after CABG adversely affects survival. The major impact on mortality occurs before hospital discharge. Patients who are destined to have a perioperative infarct cannot be predicted from commonly measured preoperative and angiographic variables.LEFT VENTRICULAR function is a major determinant of long-term survival in patients who undergo coronary artery bypass graft surgery (CABG).1-' Numerous studies have documented new wall motion abnormalities after CABG in patients who have sustained a perioperative infarction. Although worsened left ventricular performance theoretically should affect survival adversely, few studies have compared cumulative survival of patients who had a perioperative infarction with that of patients who have not.'0 '5 Perioperative myocardial infarction can be difficult to define. The postoperative development of new Q waves on the ECG, elevation of serum cardiac enzymes, or the finding of myocardial uptake of technetium-99 pyrophosphate have been proposed as diagnostic criteria for perioperative infarction.9' 15-18 However, the clinical significance of a single abnormal test result is still debated, particularly when the results of several tests are discordant.The reported ECG incidence of perioperative infarction ranges from 5.0% to 23.8%.'9-39 Several studies