Objectives: Practice guidelines still recommend antiarrhythmic drugs (AADs) as first line treatment for symptomatic atrial fibrillation (AF). However, the association between the efficacy of AADs and LA size remains unclear, whereas most of AADs have been well known to be ineffective to prevent ventricular arrhythmias in patients with reduced LV function. We assessed the hypothesis that the efficacy of AADs is dependent on LA size.
Methods: Among 1301 patients in AF registry of Severance hospital, 690 patients with symptomatic paroxysmal or persistent AF (503 male, 61 ± 12 years) who have taken AADs, were finally selected for this study. Inclusion criteria were 1) preserved LV systolic function (EF ≥45 %), 2) no structural heart disease, 3) no history of hyperthyroidism. Patients were treated with flecainide (n=620), amiodarone (n=252), sotalol (n=144) and propafenone (n=77). AADs responsiveness was defined as no AF recurrence documented by 24 hr Holter monitoring or ECG every 3 to 4 months during follow-up period of 4 years.
Results: 307 of 620 (49 %) responded to flecainide, and 88 of 252 (35 %) responded to amiodarone. 26 % of sotalol and 18 % of propafenone were responded. Fecainide response group had more paroxysmal AF (90 % vs 64 %, p<0.01), more non-dihydropyridine(DHP) calcium channel blocker (CCB) (43% vs 26%, p<0.1), higher LVEF (65% vs 64%, p=0.01), smaller LA size (40 ± 6 vs 42 ± 6, p<0.01) and less LA AP diameter (LAd) ≥40 mm (41 % vs 55 %, p<0.01). There was no statistical difference in age, sex, diabetes, hypertension, stroke, LV chamber size and E/E’ (all p values >0.05). Amiodarone response group had less male (65% vs 77%), more paroxysmal AF (49 % vs 29 %, p<0.01) but was not related to LA size (p>0.05). Sotalol response group had more paroxysmal AF (79% vs 53%, p=0.01) and smaller LA size (42 ± 7 vs 40 ± 5, p=0.02). Propafenone response group had only more paroxysmal AF (92% vs 64%, p=0.04). By multivariate analysis, paroxysmal AF (OR 0.2, 95% CI: 0.1~0.3, p<0.01), non-DHP CCB (OR 0.4, 95% CI: 0.2~0.6, p<0.01) and LA AP diameter >40mm (OR 1.8, 95% CI: 1.1~2.7, p=0.011) were independent predictors of flecainide responsiveness, whereas only persistent AF was independent predictor of amiodarone responsiveness. In subanalysis, the patients taken with flecainide were divided into 4 groups (group 1: paroxysmal AF with LAd ≤40mm, group 2: paroxysmal AF with LAd >40mm, group 3: persistent AF with LAd ≤40mm and persistent AF with LAd >40mm). Flecainide failure rates were 37%, 49%, 76% and 79%, respectively.
Conclusions: The efficacy of flecainide is dependent on LA size and type of AF synergistically, whereas the efficacy of amiodarone is dependent only on type of AF.
|
