From 1981 to 1984, 224 patients undergoing isolated coronary artery bypass grafting were analyzed with preoperative and early postoperative ejection fractions. Changes in ventricular function were assessed utilizing technetium-99m, pertechnetate radionuclide ventriculography. 106 patients (47%) showed an improvement (≥5%) in postoperative resting left ventricular function (p = 0.03). Those patients with resting preoperative ejection fractions less than 50% were those found to have an improvement in resting postoperative performance (p = 0.05). Patients with a resting preoperative ejection fraction greater than 50% were found to have a decrease in postoperative resting ejection fraction (p = 0.008). The 30 patients ( 13 %) of the cohort having both resting and exercise ejection fractions determined were evaluated independently. Covariables were univariately examined to determine factors related to exercise-induced deterioration or improvement in postoperative left ventricular performance. Longer aortic cross-clamp time (p = 0.09), greater total cardiopulmonary bypass time (p = 0.003), preoperative Q waves or STT wave changes (p = 0.07), and failure to revascularize importantly diseased (> 50% stenosis) right (p = 0.03) or circumflex marginal artery systems (p = 0.05), were found to be associated with an important decrease (> 5%) in postoperative exercise left ventricular ejection fraction. Only less severe angina preoperatively (NYHA class 0–2) was found to be associated with an improvement (> 5%) in postoperative exercise ventricular function (p = 0.08). We conclude that coronary artery bypass grafting results in important changes in ventricular performance. Risk factors for deterioration in ventricular function are identified and methods for their neutralization outlined. Radionuclide ventriculography is a good method for assessment of patients in the perioperative period.