Patients with paroxysmal atrial fibrillation (PAF) are prone to recurrence. We compared the efficacy and tolerability of flecainide acetate in the control of PAF by means of conventional treatment with quinidine in 19 patients without structural heart disease. A randomized, placebo-controlled, crossover protocol consisting of 8 weeks of treatment with either agent was used. The recurrence of PAF was documented in a symptom diary and confirmed by event ECG recording. Complete control of symptoms was achieved in 4 of 19 and 2 of 11 of patients with flecainide and quinidine, respectively. Both agents prolonged the time to the first recurrence and significantly reduced the total duration of PAF recurrence by 40% and 47%, respectively (p less than 0.05 compared with placebo). Compared with quinidine, flecainide significantly reduced the frequency of recurrence and the rate of PAF during a recurrent episode. However, treatment with flecainide was associated with a higher incidence of symptomatic sinus pauses and visual disturbances compared with a higher incidence of gastrointestinal side effects with quinidine. During a follow-up period of 32 months, satisfactory control was achieved in 74% of patients with the use of these two antiarrhythmic agents.