To identify predictors of postoperative congestive heart failure (CHF), a high-risk population, mainly hypertensive and diabetic patients undergoing elective general operations, was studied. Of the 254 patients, 6 per cent had postoperative CHF. Among patients with preoperative cardiac disease (that is, previous myocardial infarction, valvular disease or CHF), 17 per cent had postoperative CHF; in contrast with less than 1 per cent of those without cardiac disease (p less than 0.001). Patients with diabetes were also at high risk (12 versus 2 per cent, p less than 0.005), particularly those with cardiac disease. Patients who had equal to or greater than 40 millimeters of mercury increases or decreases intraoperatively in mean arterial pressure in relation to preoperative baseline had increased postoperative failure rates (p less than 0.02). Of note, postoperative failure rates were highest among patients with less than 500 milliliters per hour of net intake (input and output) (p less than 0.03). Risk for postoperative CHF was restricted to patients with preoperative symptomatic cardiac disease and was especially high if patients also had diabetes. Intraoperative fluctuations in mean arterial pressure increased the probability of postoperative failure, while the intraoperative administration of higher net volumes of fluid was associated with decreased risk.