754-5 A Multi-Centre Double Blind Crossover Comparison of Disopyramide, Atenolol, and Placebo for Symptomatic Paroxysmal Atrial Fibrillation

• Published in: Journal of the American College of Cardiology 1995-02, Vol.25 (2), p.231A-231A
• Main Authors: Murgatroyd, Francis D., O’Farrell, Dympna M., Foran, John P., Kelly, Stephen J., Ward, David E., John Camm, A.
• Format: Article
• Language: English
• ISSN: 0735-1097; 1558-3597
• DOI: 10.1016/0735-1097(95)92418-5

Class la drugs and β-blockers are commonly used to treat symptomatic paroxysmal atrial fibrillation (PAF), and patients with “vagal” and “adrenergic” patterns of PAF onset are held to respond to anticholinergic and β-blocking agents, respectively. However, neither class has been subjected to controlled clinical trials in this condition. CRAFT-2 was a double blind three-way crossover comparison of atenolol (At) (initial dose 100mg daily). disopyramide (Dp) (initial dose 450mg daily), and matched placebo (PI). The primary endpoint was the time to the first and second attacks of PAF, documented by transtelephonic monitoring, with a maximum of two attacks or two months on each treatment. All patients completed a 10-item symptom questionnaire aimed at identifying vagal and adrenergic patterns of arrhythmia onset. 75 patients from 7 hospitals were screened for CRAFT-2, and 30 were randomized; their clinical characteristics closely approximated those of 334 patients with PAF in the CRAFT registry. The numbers of patients with 0 and<2 attacks on placebo were 0, and 1, respectively. On At, these were 6 and 7, and on Dp 5 and 6, respectively. The median times to the first and second attack were 4.2 and 7.5 days on PI, 4.4 (p=0.08) and 16.4 (p=0.031 days on Dp, and 12.4 (p=0.007) and 29.1 (p=0.008) days on At, respectively. The mean (SD) heart rates during attacks were 130 (21) beatslmin on PI, 135 (26) on Dp. (p=NS), and 115 (23) on At (p=0.02). Dose reductions due to side-effects were required in 0, 2, and 4 cases, and treatment withdrawal was required in 0, 1, and 1 cases during PI, Dp, and At treatment, respectively. There was no clear relationship between autonomic patterns of symptom onset and the response to treatment. We conclude that disopyramide and especially atenolol reduce the frequency of symptomatic episodes of PAF. The efficacy of atenolol may relate to symptom reduction because of a lower heart rate, but that of disopyra-mide is likely to be a true antiarrhythmic action. The pattern of symptom onset does not easily predict response to individual antiarrhythmic agents.