Randomized crossover study of the long-term effects of pilsicainide and cibenzoline in preventing recurrence of symptomatic paroxysmal atrial fibrillation: influence of the duration of arrhythmia before therapy.

BACKGROUND

There is little information on the selection of antiarrhythmic agents for long-term prevention of paroxysmal atrial fibrillation (PAF). In the present study the preventive effects of pilsicainide (Pil) and cibenzoline (Cib) were compared in patients with PAF that was defibrillated at <48 h or >or=48 h after onset.

METHODS AND RESULTS

A total of 60 patients (45 men, 15 women, mean age 66+/-10 years) were divided into 2 groups: Group I consisted of 22 patients in whom atrial fibrillation (AF) lasted for <48 h before cardioversion and Group II consisted of 38 patients in whom AF lasted for >or=48 h. A randomized, crossover protocol of treatment with Pil (150 mg/day) and Cib (300 mg/day) was used. Mean follow-up was 35+/-18 months. In Group I, the mean duration of maintenance of sinus rhythm was 12.3+/-2.9 months in patients treated with Pil, compared with 12.9+/-2.5 months in those givem Cib (p=NS between 2 groups). Actuarial event-free rates at 1, 3, 6, 12 months were 82%, 68%, 59% and 41%, respectively, in patients treated with Pil, and 91%, 77%, 68% and 50%, respectively, in those givenh Cib (p=NS between 2 groups). In Group II, the mean duration of maintenance of sinus rhythm was 1.6+/-0.5 months in patients treated with Pil, compared with 5.9+/-1.7 months in those given Cib (p<0.01). Actuarial event-free rates at 1, 3, 6, 12 months were 45%, 18%, 8% and 3%, respectively, in patients treated with Pil, and 63%, 45%, 29% and 16%, respectively, in those given Cib (p<0.05, at 12 months).

CONCLUSIONS

Prolonged tachycardia (>or=48 h) in patients with PAF seems to cause electrical remodeling. Cib, a multichannel blocker, is considered to be more effective in preventing the recurrence of PAF in the electrically remodeled atria than Pil, a pure sodium-channel blocker.