Fifty six adult patients were randomised to treatment with flecainide (group 1, n = 29) or a combination of digoxin and disopyramide (group 2, n = 27) for acute atrial fibrillation and flutter after cardiac surgery. Intravenous flecainide was given as a 2 mg/kg bolus over 20 minutes followed by an infusion (0.2 mg/kg per hour) for 12 hours. Group 2 were given digoxin (0.75 mg) intravenously followed two hours later by an intravenous bolus of disopyramide (2 mg/kg) and an infusion (0.4 mg/kg per hour) for 10 hours. Within 12 hours sinus rhythm was restored in 86% of the group 1 (25 patients) and 89% of the group 2 (24 patients). The median time to reversion was significantly shorter in group 1 (80 minutes, range 30-180 minutes) than group 2 (220 minutes, range 138-523 minutes). None of the patients in group 1 and four of the patients in group 2 had transient relapses into atrial fibrillation during the 12 hours of intravenous treatment. There were five late relapses in group 1 and seven in group 2 during subsequent oral treatment. Two group 1 patients and two group 2 patients showed adverse drug effects. Intractable ventricular arrhythmias occurred after five days of oral treatment in one patient (group 1) who had poor left ventricular function, hepatic impairment, and toxic concentrations of drugs at the time of death. Flecainide was as effective as the combination of digoxin and disopyramide and it acted significantly faster and was associated with fewer relapses. Monitoring of blood concentrations of flecainide is essential in patients with poor left ventricular function and hepatic impairment.