In an attempt to assess cardiac risk in non-cardiac surgery, 1001 patients over 40 years of age who underwent major operative procedures were examined preoperatively, observed through surgery, studied with at least one postoperative electrocardiogram, and followed until hospital discharge or death. Documented postoperative myocardial infarction occurred in only 18 patients; though most of these patients had some pre-existing heart disease, there were few preoperative factors which were statistically correlated with postoperative infarction. Postoperative pulmonary edema was strongly correlated with preoperative heart failure, but 21 of the 36 patients who developed pulmonary edema did not have any prior history of heart failure. Nearly all of these 21 patients were elderly, had abnormal preoperative electrocardiograms, and had intraabdominal or intrathoracic surgery. In the absence of an acute infarction, bifascicular conduction defects, with or without PR interval prolongation, never progressed to complete heart block. Spinal anesthesia protected against postoperative heart failure but not against other cardiac complication. By multivariate regression analysis, postoperative cardiac death was significantly correlated with (a) myocardial infarction in the previous 6 months; (b) third heart sound or jugular venous distention immediately preoperatively; (c) more than five premature ventricular contractions per minute documented at any time preoperatively; (d) rhythm other than sinus, or premature atrial contractions on preoperative electrocardiogram; (e) age over 70 years; (f) significant valvular aortic stenosis; (g) emergency operation; (h) a 33% or greater fall in systolic blood pressure for more than 10 minutes intraoperatively. Notably unimportant factors included smoking, glucose intolerance, hyperlipidemia, hypertension, peripheral atherosclerotic vascular disease, angina, and distant myocardial infarction.