Thirty-three patients who had aortic valve replacement (AVR) for isolated aortic stenosis (AS) were retrospectively evaluated to determine the influence of preoperative ventricular function on the results of operation. Patients were arbitrarily divided by ejection fraction (EF) into Group I having EF 0.46 or above and Group II having EF 0.45 or below. Group II had significantly lower mean EF and cardiac index and higher left ventricular end-diastolic volume index and end-diastolic pressure. There were two early and no late deaths in 21 Group I patients. There were no early and five late deaths in 12 Group II patients. All Group I survivors and six of 12 Group II patients were New York Heart Association (NYHA) functional Class I or II at follow-up evaluation. No preoperative hemodynamic index was predictive of which Group II patients would do well. Mortality was substantially higher in patients with severe preoperative disability. Early or late death occurred in five of 12 patients who were NYHA functional Class IV before operation but in only two of 21 patients who were functional Class II or III preoperatively. We conclude that patients with isolated AS and poor left ventricular function have a low hospital mortality but an increased risk of late death or poor functional result following AVR. Excellent functional results can be obtained after AVR in most patients with AS and good left ventricular function and approximately half of the patients with AS and poor left ventricular function.