We sought to evaluate the underlying coronary pathology of fatal postoperative myocardial infarction (MI). It has been hypothesized that most MIs following noncardiac surgery occur in the setting of increased oxygen demand that exceeds coronary blood supply. However, most MIs not associated with surgery are caused by plaque rupture and intracoronary thrombosis. In a retrospective cohort study, we reviewed 1841 consecutive autopsy records from 1981 to 1995 at two institutions and identified 26 cases of postoperative MI with coronary arteries available. Plaque rupture was present in 12 cases (46%, 95% confidence interval [CI] 27%-67%). Of the 9 (35%) patients with intracoronary thrombus, 5 (56%; 19% of entire group) had total occlusion. Thrombus occurred on a >50% stenosis (by cross-sectional area) in a total of 33% (95% CI 16%-55%) of patients. The only statistically significant difference in clinical variables between patients with and without plaque rupture was longer interval from surgery to death in patients with plaque rupture (7.8+/-4.4 days versus 4.4+/-4.8 days; p = 0.047). In this autopsy series, coronary plaque rupture was associated with almost half of fatal postoperative MI cases. Strategies aimed at reducing triggers of plaque rupture with coronary occlusion might reduce postoperative MI fatality.