From the Department of Medicine, Montreal Heart Institute, Montreal, Canada Atrial fibrillation is the most common sustained arrhythmia.1 In patients with heart failure, its prevalence increases with the severity of the disease, and it is responsible for significant morbidity and increased mortality.2,3 Recent experimental studies have shown electrical and structural atrial remodeling with increased fibrosis in animals with heart failure.4,5 This form of remodeling is, in part, prevented by angiotensin-converting enzyme (ACE) inhibitors.6,7 To verify the hypothesis that ACE inhibitors prevent the development of atrial fibrillation in patients with heart failure, we retrospectively analyzed the data of patients from the Montreal Heart Institute who were included in the Studies of Left Ventricular Dysfunction (SOLVD) trial.
Methods Our study population consisted of patients who were enrolled in the SOLVD trial at the Montreal Heart Institute. SOLVD was a multicenter, double-blind, randomized, placebo-controlled study examining the effect of the ACE inhibitor enalapril on survival in patients with left ventricular dysfunction (left ventricular ejection fraction [LVEF] ≤ 35%).8,9 For the purpose of this study, clinical charts were reviewed, and a single experienced cardiologist, blind to treatment, interpreted every available electrocardiogram (ECG). Atrial fibrillation was defined according to the American College of Cardiology, American Heart Association, and European Society of Cardiology guidelines.10 The episodes were described either as paroxysmal, in which the patient converted to sinus rhythm spontaneously with medical therapy or with a single cardioversion, or persistent, in which the patient remained in atrial fibrillation despite changes in medical therapy or cardioversion. Exclusion criteria included patients with significant supraventricular arrhythmia on the baseline ECG (atrial fibrillation or flutter). We used Kaplan-Meier curves to analyze the time to occurrence of atrial fibrillation and Cox regression analysis to assess enalapril as an independent predictor.
A total of 391 patients from the Montreal Heart Institute were enrolled in the SOLVD trial. Of these, 17 patients (4.3%) exhibited arrhythmia on baseline ECG and were excluded (16 atrial fibrillation and 1 flutter), leaving 374 patients for analysis. There were 251 patients in the prevention arm of the study (New York Heart Association [NYHA] functional class I or II not requiring diuretics) and 123 patients in the treatment arm (NYHA class II to IV). The baseline characteristics were similar between the enalapril (n = 186) and placebo (n = 188) groups, except for a higher prevalence of previous myocardial infarction (MI; P = .026) and a trend toward an increase in current smoking in the enalapril-treated group. A nonsignificant higher prevalence of supraventricular arrhythmia in the placebo group (7.5% versus 3.8%) was also observed. A majority of patients were men with NYHA class II symptoms and severely depressed left ventricular function (mean LVEF, 26%).
In the enalapril group, 798 ECGs were examined (4.3 ± 5.0 per patient), and 693 ECGs were examined in the placebo group (3.7 ± 4.1 per patient). A total of 43 Holter examinations were performed, 24 in the enalapril group and 19 in the placebo group. During the 2.9 ± 1.0 years of follow-up, 55 patients developed one or more episodes of atrial fibrillation, 10 patients (5.4%) in the enalapril group and 45 patients (24%) in the placebo group (P < .01), which was an absolute risk reduction of 18.6%.
The probability of remaining free from atrial fibrillation was significantly higher in the enalapril group than in the placebo group (hazard ratio [HR], 0.22; P < .01; figure 1). Most of the episodes were paroxysmal, requiring hospitalization for worsening heart failure, but electrical cardioversion was performed in a minority of patients. In a multivariate analysis using the Cox regression model, which included age, ischemic cause of left ventricular dysfunction, and history of supraventricular arrhythmia and diuretic use, treatment with enalapril was the most powerful predictor for reduction in the incidence of atrial fibrillation (HR, 0.22; 95% confidence interval [CI], 0.11—0.44; P < .01), followed by the presence of an ischemic cause of left ventricular dysfunction (HR, 4.9; 95% CI, 2.32–10.41;
P < .01).
When evaluating the effect of enalapril on the incidence of atrial fi-brillation in the prevention and treatment arms of the SOLVD trial, the beneficial effect of enalapril was more marked in the less symptomatic patients. In the prevention arm of the study, four patients (3.2%) had atrial fibrillation in the enalapril group versus 31 patients (24.6%) in the placebo group (P < .001). In the treatment arm, six patients (9.8%) had atrial fibrillation in the enalapril group versus 14 patients (22.6%) in the placebo group. Figures 2 and 3 show the Kaplan-Meier curves for time to occurrence of atrial fibrillation for the two arms of the study.
The results of this study show that long-term ACE inhibitor therapy with enalapril significantly decreased the risk of atrial fibrillation development in patients with left ventricular dysfunction. To our knowledge, this is the first study to show such a beneficial effect in a chronic heart failure population.
Similar results were shown by Pedersen and colleagues in patients with reduced left ventricular function (mean LVEF, 33%) 3 to 7 days after MI, using trandolapril compared with placebo.11 The clinical situation in their study was different, however, because treatment was started at the time when myocardial changes were occurring and thus may not have reflected the established congestive heart failure (CHF) condition in which increased atrial pressure and left ventricular remodeling have been present for a long time. Furthermore, after 2 and 4 years of follow-up in the TRAndolapril Cardiac Evaluation (TRACE) study, there was only a small absolute risk reduction (2.5%) in the incidence of atrial fibrillation. Increasing evidence suggests that, in the failing heart, left ventricular and atrial remodeling, increased left atrial pressure and fibrosis, and neurohormonal activation interact and may precipitate atrial fibrillation, which can be prevented in part by ACE inhibitors.6
The manner in which ACE inhibitors exert their beneficial effects against the development of atrial fibrillation in heart failure is not thoroughly understood. The inhibition of the neurohormonal activation that occurs in heart failure, principally the renin-angiotensin-aldosterone system, may be one possible explanation. Cardiac fibroblast proliferation and reduced collagenase activity occur as a result of fibrosis, which is promoted by angiotensin II.12 Increased atrial expression of ACE and extracellular signal-regulated kinase (ERK) has been shown in heart failure and in atrial interstitial cells of patients with atrial fibrillation.13 The amount of activated ERK2 was decreased after treatment with ACE inhibitors.
Experimental CHF, induced by rapid ventricular pacing, produced abnormal local atrial conduction properties associated with atrial structural changes, which increase the ability of atria to sustain atrial fibrillation.4 In a rapid ventricular pacing model, Shi and colleagues showed that the atrial remodeling changes were attenuated when enalapril was given at the onset of pacing to animals that were followed up for 5 weeks, leading to decreased vulnerability to atrial fibrillation.7
Atrial stretch secondary to increased atrial pressures, partially due to increased levels of angiotensin II in heart failure, may also be involved in the initiation and pathogenesis of atrial fibrillation through shortening of the refractory period and lengthening of intra-atrial conduction time.14 ACE inhibitors cause a reduction in left atrial and left ventricular end-diastolic pressures.6 As a result of these effects and attenuation of left atrial enlargement, these agents may reduce the susceptibility to atrial fibrillation.
Recent studies in animal models and humans have shown that atrial fibrillation induced by rapid atrial pacing itself produces shortening
of the atrial effective refractory period and reverses the normal physiological rate adaptation of refractoriness. This phenomenon of electrical remodeling increases the induci-
bility and stability of atrial fibrillation15 and was prevented by treatment with candesartan or enalapril.16 Moreover, a beneficial effect of irbesartan on atrial fibrillation recurrence was shown clinically when the drug was administered 3 weeks before cardioversion and used in combination with amiodarone.17
These experimental and clinical data suggest that treatment inter-
fering with the renin-angiotensin-aldosterone system, either with ACE inhibitors or angiotensin II recep-
tor blockers, has a protective role against development of atrial fibrillation in patients with heart failure, improves their short- and long-term prognosis, and permits avoidance of the potential risks of anticoagulant and antiarrhythmic therapy.
Our study shows that ACE inhibition with enalapril significantly decreases the risk of development of atrial fibrillation in patients with left ventricular systolic dysfunction.