To assess the cardiac and extracardiac safety and efficacy of flecainide versus propafenone in patients suffering from episodes of paroxysmal atrial fibrillation (AF) or atrial flutter, 97 patients were enrolled in a randomized, open-label, long-term, parallel, comparative multicenter study. The diagnosis of paroxysmal AF or atrial flutter had to be fully documented prior to inclusion in the study. Of the 97 patients enrolled in the study, 48 patients (25 men, 23 women, mean age 62.4 +/- 12.3 years) received flecainide; 49 patients (26 men, 23 women, mean age 63.6 +/- 12.2 years) received propafenone. The initial dose of flecainide was 50 mg twice daily, and this could be increased in steps of 50 mg twice daily every 4 days to a maximum of 300 mg/day. The initial dose of propafenone was 300 mg twice daily and this could be increased in steps of 300 mg every 4 days to a maximum of 1200 mg/day. At each visit, medical events, vital sign measurements (blood pressure, pulse rate), concomitant medications, adverse experiences, and study drug dosage changes were evaluated. Routine clinical laboratory tests were evaluated at the month 6 visit, and a 24-hr Holter recording was obtained at the month 1 visit. Almost half (45) of the patients were discontinued from the study before completing 1 year of therapy. The probability of successful treatment versus time--i.e., the proportion of patients who remained on therapy over the course of 1 year therapy--was 0.619 for the flecainide group and 0.469 for the propafenone group (p = 0.079; difference not significant). The difference is largely attributed to the higher proportion of patients in the propafenone group (9) than in the flecainide group (2) who experienced side effects important enough to stop the treatment. (The incidence of side effects was not statistically different between treatment groups, although it was higher in the propafenone group.) The proportion of patients who discontinued treatment due to inadequate response was similar in the 2 groups: 11 patients (22.9%) in the flecainide group and 12 patients (24.4%) in the propafenone group withdrew from the study, primarily because of an inadequate response, i.e., they experienced an increase in duration, frequency, and severity of attacks of AF or atrial flutter. Neurologic signs, central and peripheral, were mostly encountered in the flecainide group (8.5%), and, gastrointestinal effects were more often reported in the propafenone group (16.7%). In paroxysmal AF and paroxysmal atrial flutter, flecainide and propafenone are equally effective. However, in this study the probability of a patient's staying on flecainide after 1 year had a tendency to be higher than the probability of staying on propafenone, due to a greater proportion of secondary effects with propafenone.