This was a double-blind, placebo-controlled, crossover study designed to determine the influence of digitalis treatment on left ventricular (LV) response to physical exercise in patients with congestive heart failure (CHF). In 10 patients with CHF (ejection fraction 29 +/- 2%), LV function was assessed during upright bicycle exercise using an ambulatory radionuclide detector for continuous noninvasive monitoring of cardiac function. Exercise was performed during control conditions and after a 3-week treatment with digoxin (0.25 mg/day orally) or placebo. Ten normal volunteers matched for sex and age constituted the control group. In normals, exercise ejection fraction and end-diastolic volume increased (both p <0.001), while end-systolic volume decreased progressively (p <0.001). In control conditions, patients with CHF had a sharp increase in heart rate during exercise, while ejection fraction did not change; both end-diastolic and end-systolic volumes increased significantly (both p <0.001) during exercise. During digoxin treatment, heart rate response to exercise recorded in patients with CHF was comparable to that recorded in normal subjects. In addition, a significant increase in ejection fraction during exercise was detected (P <0.001), and the increase in end-systolic volume was significantly smaller than that observed in control conditions (p <0.05). When patients received placebo, the responses of LV function to exercise were comparable to those observed in control conditions. These findings demonstrate that digitalis has a favorable influence on LV functional adaptation to exercise in CHF.