Reducing the incidence of atrial fibrillation by inhibiting angiotensin-converting enzyme

Hospital admissions for atrial fibrillation, which is the most significant cardiac arrhythmia, have increased markedly over the past few years.1 Because of the aging population and because the prevalence of atrial fibrillation is age-related,2 the incidence of atrial fibrillation is expected to increase further. Atrial fibrillation is also associated with significant morbidity (mainly stroke-related) and mortality (the rate doubles compared with patients in sinus rhythm).3 In addition to causing stroke, atrial fibrillation may decrease functional status and quality of life because of symptoms of heart failure due to increased heart rate, atrioventricular asynchrony, and left atrial or left ventricular dysfunction and because some patients experience significant discomfort as a result of irregular rhythm.

Standard therapies include chemical and electrical cardioversion and antiarrhythmic drugs for maintenance of sinus rhythm. However, the recurrence rate of atrial fibrillation after successful cardioversion is as high as 50% at 1 year for patients who are prescribed long-term antiarrhythmic drugs, which are associated with significant side effects. An interventional approach to atrial fibrillation is showing promising preliminary results in selected patients using radiofrequency catheter ablation around the junction of the pulmonary veins in the left atrium. Several catheter ablations are often needed because of recurrences, however, and rare but clinically significant complications do occur with this strategy.

New insights into the pathophysiology of atrial fibrillation have emerged from basic science and animal models. These data indicate that the renin-angiotensin system may play a role in the pathogenesis of atrial fibrillation. Specifically, angiotensin II (AT-II) appears to be implicated in structural and electrophysiological remodeling of the atria, therefore contributing to atrial fibrillation. AT-II shortens the effective refractory period and lengthens intra-atrial conduction time favoring fibrillation, possibly through mechanical stretch of the atria.4,5 In addition,

AT-II is a potent promoter of fibrosis through reduced collagenase activity, cardiac myoblast proliferation, and activation of extracellular signal-regulated and mitogen-activated protein kinases.6,7 Atrial fibrosis is frequently found

in patients with atrial fibrillation,4

providing a possible explanation for

the characteristic conduction disturbances and persistent susceptibility. Consequently, the angiotensin-converting enzyme (ACE) inhibitor enalapril (Vasotec) and the A-II receptor blocker (ARB) candesartan (Atacand) were studied in dog models with favorable results for the prevention of atrial structural and electrical remodeling, with a subsequent decrease in the occurrence of atrial fibrillation.8-10 Early clinical trials evaluating the preventive effect of ACE inhibitors in patients with reduced left ventricular function after myocardial infarction11 and ARBs in patients with atrial fibrillation who underwent successful cardioversion have shown significant reductions in the recurrence of atrial arrhythmias.12 More recently, our group found a beneficial effect with enalapril for the

prevention of atrial fibrillation in pa-

tients with chronic left ventricular

dysfunction.13

Hypertension is known as the most prevalent and potentially modifiable risk factor for atrial fibrillation. It affects approximately 20% of the adult population worldwide. Cardiac adaptive responses to hypertension, such as left ventricular hypertrophy, left atrial enlargement, modifications in mechanical function of the atria, and increased filling pressures, have been found to independently predict development of atrial fibrillation.14 Because experimental and early clinical studies have suggested a role for ACE inhibitors in the prevention of atrial fibrillation and because they are among the most frequently prescribed antihypertensive medications, we assessed the effect of this class of drugs on the occurrence of atrial fibrillation in a large cohort of hypertensive patients. We used long-acting calcium channel blocking agents as the blood pressure—lowering comparator because they represent the most frequently prescribed antihypertensive drug class after ACE inhibitors and because it has been suggested that they have antiarrhythmic properties.

Patients and methods

We performed a retrospective, longitudinal cohort study to assess the effects of ACE inhibitors on atrial fibrillation in patients with hypertension compared with long-acting calcium channel blocking agents. We used an integrated medical and pharmacy claims database of more than 8 million people in the United States. To be eligible, patients had to be diagnosed with hypertension from one or more medical visits during the period 6 months before and after the index prescription of ACE inhibitors or calcium channel blocking agents. Any additional antihypertensive medication prescribed during the follow-up period was permitted to be taken, except calcium channel blocking agents for the patients taking ACE inhibitors and ACE inhibitors for the patients taking calcium channel blocking agents. The index prescriptions for ACE inhibitors or calcium channel blocking agents were filled between January 1, 1995, and June 30, 1999, and the final cohorts were evaluated until June 30, 2002, with an average length of follow-up of 4.6 years for the ACE inhibitor group versus 4.2 years for the calcium channel blocker group. Propensity scoring and logistic regression were used to match cohorts and minimize potential baseline differences that could bias final outcomes.

Results

Of 10,926 patients, 5,463 eligible patients were matched in each group. The major baseline characteristics

showed a mean age of 65 years in both groups, a 3.7% versus 3.6% prevalence of congestive heart failure, a 9.9% versus 10.1% prevalence of diabetes, a 6.4% versus 6.7% prevalence of stroke, and a previous history of atrial fibrillation in 2.3% versus 2.4% of patients in the ACE inhibitor

and calcium channel blocking agent groups, respectively. About 48% of patients in both groups received at least one other antihypertensive medication, with a higher rate of thiazides being taken in the ACE inhibitor group and a higher rate of beta blocking agents being taken in the calcium channel blocking agent group (17.6% versus 12.4% for thiazides and 15.5% versus 19.6% for beta blocking agents in the ACE inhibitor versus calcium channel blocking agent groups, respectively).

The incidence rate of new-onset atrial fibrillation was significantly lower in the group treated with ACE inhibitors than in the group treated with calcium channel blocking agents (hazard ratio, 0.85; 95% confidence interval [CI], 0.74—0.97). The average number of months to the first onset

of atrial fibrillation was 29.5 versus 26.1 (difference, 3.4 months; 95%

CI, 0.72—6.04) in the ACE inhibitor versus calcium channel blocking agent groups, respectively. Time to occurrence using the Kaplan-Meier method showed a significant decrease in new-onset atrial fibrillation with the ACE inhibitor over the whole observation period (P = .018; Figure 1).

The potential benefit of treatment with ACE inhibitors over calcium channel blocking agents in patients with hypertension was also assessed by analyzing the rate of atrial fibrillation— related hospitalization. The rate of hospitalization related to atrial fibrillation was 8.5 per 1,000 patient-years for patients taking ACE inhibitors compared with 11.9 per 1,000 patient-years for patients taking calcium channel blocking agents, with an incidence ratio for patients treated with an ACE inhibitor of 0.74 (95% CI, 0.62–0.89; Figure 2). Further analysis showed an incremental benefit of ACE inhibitors over calcium channel blocking agents in patients with a previous history of atrial fibrillation compared with patients without a previous episode of atrial fibrillation. The incidence ratio for patients with a previous history of atrial fibrillation treated with ACE inhibitors after the entire follow-up period was 0.55 (95% CI, 0.38–0.78) compared with an incidence ratio of 0.82 for patients without a history of atrial fibrillation (95% CI, 0.67–1.01). Therefore, ACE inhibition was associated with a reduced incidence of atrial fibrillation and related hospitalizations for hypertensive patients in a managed care setting.

Discussion

ACE inhibitors are extensively prescribed to lower blood pressure in hypertensive patients and, for patients with impaired left ventricular function, to control neurohormonal activation and remodeling and to decrease mor-

tality. Recent developments in our understanding of the pathophysiology of atrial fibrillation have allowed their evaluation as antiarrhythmic drugs. This study, as well as others, strongly suggests that ACE inhibitors have significant antiarrhythmic properties, especially with regard to atrial fibrilla-

tion, that distinguish them from other antihypertensive drugs. They are now emerging as a potential adjunct to the current treatment options for atrial fibrillation. The socioeconomic effect of these results warrants a dedicated prospective randomized trial. Because of the high prevalence of hypertension and atrial fibrillation, any class of drugs that can normalize blood pressure and prevent atrial fibrillation is likely to have a significant impact on health care.

Conclusion

Our study showed that hypertensive patients taking ACE inhibitors had a lower incidence of atrial fibrillation than those who were taking calcium channel blocking agents. These findings indicate that ACE inhibitors have considerable antiarrhythmic properties, especially with regard to atrial fibrillation.