Preventing atrial fibrillation recurrence

Cardiology Review® Online, April 2007, Volume 24, Issue 4

In patients with lone paroxysmal atrial fibrillation (AF) and normal cardiac function, in the absence of hypertension, myocardial infarction, and cardiac insufficiency, perindopril (Aceon) or losartan (Cozaar) and low-dose amiodarone (Cordarone) therapy are more effective in preventing recurrence of AF than low-dose amiodarone alone. Adding losartan or perindopril to amiodarone therapy can inhibit left atrial enlargement.

In patients with paroxysmal atrial fibrillation (AF), preventing the recurrence of this condition and maintaining sinus rhythm is a common problem. The onset, recurrence, and persistence of AF are mainly due to the electric and structural remodeling of atrium. Research indicates that the renin-angiotensin-aldosterone (RAA) system has an important effect on both electrical and structural remodeling of the atrium.1-3 It also has been shown that AF-induced atrial electrical and structural remodeling can be prevented or reversed by therapy with angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs).1-7 Two studies showed that persistent AF was reduced when patients received ACEI or ARB therapy.4,5 But no study has compared an ACEI with an ARB for the prevention of lone AF recurrence. Therefore, we designed a prospective, randomized trial to compare the long-term efficacy of low-dose amiodarone (Cordarone) with that of low-dose amiodarone combined with perindopril (Aceon) or losartan (Cozaar) to prevent recurrence of lone paroxysmal AF.

Subjects and methods

All subjects included in this study had symptomatic lone paroxysmal AF documented by Holter or 12-lead electrocardiogram (ECG). A total of 177 patients were randomly assigned to 1 of the 3 therapeutic groups, out of 280 patients screened for eligibility. The therapeutic groups were as follows: amiodarone only (group 1, n = 59), amiodarone plus losartan (group 2, n = 59), and amiodarone plus perindopril (group 3, n = 59).

All patients took amiodarone 600 mg/day for the first week. The dosage was reduced to 400 mg/day for the next week, and reduced to 200 mg/day thereafter. Patients receiving losartan took 50 mg/day for the first 2 weeks and 100 mg/day thereafter if hypotension did not occur. Patients receiving perindopril took 2 mg/day for the first 3 days and 4 mg/day thereafter if hypotension did not occur. In cases of hypotension, the dosage of losartan was reduced to 50 mg/day, and the dosage of perindopril was reduced to 2 mg/day. Echocardio­graphy and Holter monitoring were done at 6, 12, 18, and 24 months in all patients after randomization. All follow up was performed by telephone, visiting the clinic regularly for physical examination, 12-lead ECG, and Holter monitor.

The primary endpoint was the onset of AF (≥30 s in duration) documented by 12-lead ECG or Holter monitor. To evaluate the differences in continuous variables, we used analysis of variance with the post hoc Bonferroni’s test. The χ2 test was used to test categorical data. To estimate patient survival time without recurrences of AF, we used the Kaplan-Meier method. To test differences in patient survival times without recurrence of AF, we used the log-rank test. To examine the association between treatment groups and AF recurrence after adjusting for covariates, we used the Cox proportional hazards model. A 2-sided value of P ≤ .05 was considered significant.


As shown in the

, patients did not differ in their baseline clinical characteristics, except for a shorter history of AF in the patients treated with amiodarone plus perindopril. Follow-up was achieved in 100% of patients. At the end of the study, there was no difference in the change of systolic and diastolic blood pressure among the 3 groups. As shown in the





, the Kaplan-Meier survival analysis demonstrated a significant reduction in AF recurrence in group 2 (amiodarone plus losartan; = .006, log-rank test) and group 3 (amiodarone plus perindopril; = .04, log-rank test) when compared with group 1 (amiodarone only). The overall risk reduction was 54% and 41% in groups 2 and 3, respectively. After adjusting for age, sex, frequency of AF episodes, history of AF, body mass index, left ventricular ejection fraction, and left atrial diameter, a significant beneficial effect of losartan (relative risk [RR] = 0.36; 95% confidence interval [CI] = 0.17, 0.75; = .006) and perindopril (RR = 0.39; 95% CI = 0.20, 0.78; = .008) on preventing AF recurrence could be shown through the Cox proportional hazards model.


There was no difference in severe adverse effects among the groups: they occurred in 5 patients (8.5%) in group 1, 5 patients (8.5%) in group 2, and 8 patients (14%) in group 3. These effects included intolerable, unproductive cough, severe sinus bradycardia, hypotension, significant QT prolongation, hyperthyroidism, and hypothyroidism. No torsade de pointes occurr­ed in the study.


The left atrial dimension of patients treated with amiodarone plus losartan (group 2) and with amiodarone plus perindopril (group 3) was significantly smaller than that of patients treated with amiodarone (group 1) at the end of follow-up (36 ± 2.3 mm and 35 ± 2.4 mm vs 38 ± 2.4 mm; < .001 for both comparisons).


Many animal studies show that ACEIs and ARBs can inhibit the RAA system, preventing AF by attenuating changes in cardiac structure and function such as atrial dilation, atrial fibrosis, and conduction velocity slowing.8-11 Another animal study that treated to the same hemodynamic targets with hydralazine and isosorbide mononitrate did not get these benefits, suggesting that the effect on preventing AF is specifically related to inhibition of the RAA system.10 The TRACE trial,5 which followed patients for 4 years, demonstrated that trandolapril reduced the incidence of atrial fibrillation by 55% versus placebo in patients with left ventricular dysfunction after acute myocardial infarction (MI). Madrid et al4 and Ueng et al12 reported that treatment with irbesartan (Avapro) or enalapril (Vasotec) resulted in a significant reduction in AF recurrence in patients with long-standing persistent AF after external direct-current cardioversion. The apparent effect appeared within weeks of the cardioversion.

Most previous studies focused on AF in patients with structural heart diseases (eg, heart failure, MI, hypertension). Hansson et al13,14 and Wachtell et al15 demonstrated that ACEIs or ARBs compared with other agents could not significantly reduce AF in patients with hypertension. Therefore, it may be asked whether the reduction in AF recurrence really is completely attributable to the ACEI or ARB itself. Besides the effect of antiarrhythmia drugs, the reduced AF recurrence may be partly or even completely the result of improved left ventricular function and cardiac remolding of primary heart diseases, as well as lowering of blood pressure. The mechanism was hard to distinguish. One randomized trial performed by Madrid et al4 showed that AF recurrence in patients with lone persistent AF after cardioversion was reduced by combination therapy with irbesartan and amiodarone. So far, no study has examined the effect of ACEIs or ARBs on prevention of lone paroxysmal AF recurrence. Our study is the first one to use an ACEI or an ARB (both combined with amiodarone) compared with amiodarone alone for the prevention of AF recurrence in patients with lone paroxysmal AF.

The results of our study showed the beneficial effect of adding ACEIs/ARBs to amiodarone to prevent AF recurrence in patients with lone paroxysmal AF. Our study also showed that amiodarone plus losartan and amiodarone plus perindopril were similarly effective in preventing AF recurrence in these patients. Because only patients with lone AF were included, the study showed that ACEIs and ARBs have direct antiarrhythmia effects (presumably they reverse AF-induced atrial modeling and have sympatholytic activity). Finally, for the first time it was shown that patients with lone paroxysmal AF receiving ACEIs/ARBs had inhibition of left atrial enlargement.

The beneficial effects of losartan and perindopril may have several causes. By shortening refractoriness due to a high atrial rate, these agents may reverse AF-caused electrical remodeling. (Atrial electrical remodeling, which worsens gradually with time, perpetuates AF.) Also, AF-induced structural atrial remodeling has been observed in patients with lone AF, and losartan and perindopril may inhibit this remodeling. The inhibition of left atrial enlargement observed in this study may at least partly be explained by the ability of losartan and perindopril to reverse AF-induced structural remodeling. Finally, the risk of AF recurrence may be decreased because these agents reduce the plasma norepinephrine concentration, resulting in sympatholytic effects.


In patients with lone paroxysmal AF and normal cardiac function (ie, without hypertension, MI, or cardiac insufficiency), this study showed that perindopril plus low-dose amiodarone or losartan plus low-dose amiodarone prevents AF recurrence more effectively than low-dose amiodarone alone. Left atrial enlargement in this group of patients can be inhibited by addition of losartan or perindopril to amiodarone therapy.