Amrinone, 100 mg orally every 8 hours, was administered to 13 patients with moderate-to-severe congestive heart failure (CHF) for 1 month on an outpatient basis to determine the beneficial and undesirable effects of this new cardioactive agent in this clinical setting. These subjects received conventional CHF medications during the course of study. Ten patients who received conventional CHF medications alone served as a control group. Changes in functional classification were not significantly different between the 2 treatment groups. Amrinone augmented exercise capacity 37% above baseline compared with a 12% improvement for the control group. Noninvasive indexes of resting left ventricular function (echocardiography and systolic time intervals) did not change significantly for either group, nor was there a significant change in the exercise ejection fraction. All patients treated with amrinone had greater than or equal to 1 symptom-related or laboratory-detected adverse effect. An increase in the frequency of ventricular ectopic beats was noted at rest in 4 and with exercise in 6 patients (salvos of nonsustained ventricular tachycardia in 2). Six subjects treated with amrinone had gastrointestinal symptoms and 8 developed a viral-like illness. Other adverse effects noted in the amrinone-treated group included near-syncope, headaches, marked anxiety, chest pain, palpitations, maculopapular rash, hypokalemia, and elevation of serum transaminase levels. The control patients had significantly fewer adverse effects. Although individual patients with CHF may benefit from long-term amrinone therapy, the low benefit-to-risk-adverse effect ratio does not warrant widespread application of this drug in the outpatient management of CHF and requires caution when prescribing.