We compared the efficacy of flecainide versus quinidine in preventing paroxysms of atrial fibrillation in a randomized open crossover study. Twenty-six patients with weekly attacks of atrial fibrillation during the last 3 months, objectified by 24-h holter monitoring or 12-lead electrocardiogram (ECG) were treated for a period of 3 months with flecainide 100 mg b.i.d. or quinidine 500 mg b.i.d. Efficacy was assessed by 24-h holter monitoring and a questionnaire at the end of each month. Dosage was adjusted to flecainide 100 mg t.i.d. or quinidine 500 mg t.i.d. if patients still had symptomatic paroxysms of atrial fibrillation according to a questionnaire or on holter monitoring. In 46% of the patients, flecainide 100 mg b.i.d. caused total abolition of supraventricular tachycardia; after dose adjustment it caused 50% total abolition. For quinidine, the figures are 16% (p less than 0.05) and 32% (NS), respectively. Side effects occurred with flecainide only after dose adjustment (23%), but on quinidine they occurred before (8%) and after dose adjustment (20%). We conclude that flecainide suppresses paroxysms of atrial fibrillation significantly more often as compared with quinidine in the lower dosage regimen. Optimal treatment dosage of flecainide is 100 mg b.i.d. After quinidine dose adjustment, the difference in efficacy is no longer significant. However, side effects necessitating discontinuation of quinidine developed in 20% of the patients as compared to none in patients treated with flecainide 100 mg b.i.d.