Preoperative left ventricular systolic function is an important predictor of postoperative prognosis in patients with aortic regurgitation. Although left ventricular dysfunction is reversible after aortic valve replacement to a greater extent in patients with good preoperative exercise capacity compared with patients with impaired exercise capacity, not all patients with preserved exercise capacity demonstrate improved left ventricular function after aortic valve replacement. To determine the influence of duration of preoperative left ventricular dysfunction on postoperative reversal of left ventricular dysfunction, we studied 37 patients with aortic regurgitation who preoperatively had left ventricular dysfunction, defined as subnormal echocardiographic fractional shortening (less than 29%), and good preoperative exercise capacity, defined as completion of stage I of the NIH treadmill protocol without limiting symptoms. Eight patients were asymptomatic. In 11 patients left ventricular dysfunction was documented 18 to 57 months preoperatively (prolonged); in 10 patients left ventricular dysfunction developed in an interval of 14 months or less preoperatively (brief); in 16 patients duration of left ventricular dysfunction was unknown. Patients with brief vs those with prolonged left ventricular dysfunction did not differ with respect to severity of preoperative symptoms or exercise tolerance, echocardiographically determined left ventricular dimensions or fractional shortening (25 +/- 3% [SD] vs 25 +/- 3%), or radionuclide angiographic ejection fraction (42 +/- 5% vs 42 +/- 5%).(ABSTRACT TRUNCATED AT 250 WORDS)