Safety and efficacy of propafenone and disopyramide for long-term maintenance of sinus rhythm after electrical cardioversion was studied in 56 patients with chronic atrial fibrillation (median arrhythmia duration, 5 months). After cardioversion, patients were randomly assigned to receive double-blind propafenone 300 mg tid (25 patients) or disopyramide 250 mg tid (31 patients). Downward dose adjustment was allowed in case of intolerable side effects. The endpoints were arrhythmia recurrence; and side effects not amenable to dose reduction. For patients randomized to propafenone (mean dose, 878 +/- 65 mg/day), 66% [95% confidence interval [(CI) 46-85%] and 55% (95% CI, 34-76%) remained in sinus rhythm at 3 and 6 months, respectively (Kaplan-Meier method). Similar figures were found with disopyramide (mean dose, 704 +/- 81 mg/day): 71% (95% CI, 54-87%) and 67% (95% CI, 50-84%) at 3 and 6 months, respectively (p = NS). In the patients with a relapse of atrial fibrillation, the ventricular rate while still using the prophylactic agents did not increase significantly compared with precardioversion. However, one patient on disopyramide had an excessively high relapse heart rate (170 vs. 100 beats/min). Side effects were more frequent on disopyramide. Side effects necessitating drug discontinuation occurred in 12 patients: 4 patients (16%) on propafenone and 8 (26%) on disopyramide. Severe adverse effects occurred in two patients, who developed heart failure while on disopyramide. There were no proarrhythmic events or deaths. Thus, propafenone and disopyramide are equally effective for maintaining sinus rhythm after cardioversion of atrial fibrillation. Propafenone is, however, better tolerated than disopyramide, which may cause heart failure.