For decades, conventional wisdom suggested that sinus rhythm was preferred over rate control in the treatment of patients with atrial fibrillation (AF). Recent randomised trials have been unable to substantiate this viewpoint, and the occurrence of serious outcomes for patients has been similar for both treatment strategies.1,2
More specifically, several large, randomised, prospective clinical trials have compared morbidity and mortality outcomes of rhythm control versus rate control, only to show no distinct advantage for one strategy over the other in terms of clinical outcomes. The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM), Rate Conrol versus Electrical Cardioversion for Persistent Atrial Fibrillation (RACE), Pharmacological Intervention in Atrial Fibrillation (PIAF), Strategies of Treatment of Atrial Fibrillation (STAF) and How to Treat Chronic Atrial Fibrillation (HOT-CAFÉ) studies have been unable to show conclusively the superiority of one strategy over the other.2-6 The AFFIRM and RACE studies underscored the importance of continued anticoagulation as a major therapeutic intervention for reducing the risk of stroke, which is achieved independently of whether a rate or rhythm control strategy is chosen.2,6 However, all of these aforementioned studies were largely conducted on patients without heart failure (HF) and included patients with paroxysmal AF. Notably, only 23% of patients enrolled in the largest rate versus rhythm AF trial to date, AFFIRM, had HF. However, there appear to be specific advantages of sinus rhythm in patients who have AF and HF. These might include a salutary effect of sinus rhythm due to at least three factors: regularisation of the rhythm, physiological rate control and restoration of atrial contribution to cardiac output. In support of this concept, Ahmed et al. evaluated outcomes in patients with systolic HF with and without AF.7 They concluded that a history of AF was associated with HF hospitalisations but did not have an intrinsic association with mortality. Furthermore, a meta-analysis of the prognostic significance of AF in patients with chronic HF concluded that there was an increase in mortality in the presence of AF in patients with systolic dysfunction and in those with preserved left ventricular function.8 Bycontrast, the recent Atrial Fibrillation and Congestive Heart Failure (AF-CHF) trial, which compared rhythm versus rate control strategies in patients with systolic HF, showed no advantage of rhythm control over rate control in terms of the primary end-point, death due to cardiovascular causes, or the secondary end-points of total survival, occurrence of stroke and exacerbation of HF (see Table 1).9
To gain a better understanding of the apparent conflicts of the value of sinus rhythm in HF, it is important to analyse in more detail a variety of studies that have evaluated clinical outcomes in patients with AF and HF. This article further reflects the debates on the treatment of AF in the setting of HF at the 2008 CREATE AnnualAdvisory Meeting.
Heart Failure and Atrial Fibrillation – Maintain Sinus Rhythm or Rate Control?
The treatment of patients with both AF and HF remains difficult. AF may be a major factor or a simple marker of the risk of HF. While anticoagulation treatment is necessary in all cases, the definition of the optimal therapy prescribed to control arrhythmia itself is much more difficult. Choosing between AF strategies, whether in the presence of HF or not, can be summarised as a dilemma between rhythm and rate control. The problem remains that it is not known whether it is better to maintain sinus rhythm, which involves pharmacological or non-pharmacological therapies that are not devoid of adverse effects, or to choose simple rate control to avoid the deleterious effects associated with excessively high heart rates.
It is well known that persistence of AF can cause electrophysiological and anatomical changes in the atria.1 The presence of a persistent rapid ventricular response in humans can lead to a tachycardiamediated cardiomyopathy. A canine model of rapid pacing-induced HF has shown reduced left ventricular fractional shortening, reduced left atrial contractility, increased atrial fibrosis and alterations in ion channel expression during persistent AF.10 Thus, it is clear that persistent AF can have a deleterious effect on both atrial and ventricular function, depending on the parameters evaluated.
Hsu and colleagues prospectively evaluated the outcomes of catheter ablation of AF on left ventricular function in patients who had congestive HF with systolic dysfunction.11 The left ventricular ejection fraction (LVEF) was less than 45% in the 58 patients studied. After a mean follow-up of 12 months, 78% of patients with congestive HF remained in sinus rhythm. Importantly, LVEF increased by 21±13% and LV dimensions were significantly reduced. Equally important, there was a significant improvement in exercise capacity, symptoms, and quality of life. The authors stated that patients undergoing ablation had adequate ventricular rate control prior to ablation, suggesting that reduced LVEF was not tachycardia-mediated. The improvement in LVEF is therefore more likely to be related to the restoration of atrial contribution to cardiac output as well as the regularisation of cardiac rhythm.
At least two other studies support the observations of Hsu et al.12,13 Gentlesk and colleagues determined the effect of catheter ablation in patients with AF on left ventricular function in 67 individuals who had a mean LVEF of 42%.12 The LVEF increased to 56% during followup. These authors stress that a reversible cardiomyopathy in patients with AF may be present even when ventricular rate control is adequate. The effects of surgical AF ablation in patients with HF were evaluated by Grubitzsch et al.13 At a mean follow-up of 13 months, 66% of patients with HF were in sinus rhythm. Among patients who had LVEF Ôëñ30%, sinus rhythm conversion was found to significantly improve LVEF.
The aforementioned data from non-pharmacological approaches to restoring and maintaining sinus rhythm in patients with AF and HF strongly suggest a benefit with sinus rhythm control towards this patient population. However, the large randomised prospective AF-CHF study evaluating rate versus rhythm control in patients with AF and HF failed to substantiate this conclusion.9 Of 1,376 patients enrolled, 682 were assigned to the rhythm-control group and 694 to the rate-control group for a mean follow-up of 37months. Entry criteria included an LVEF of Ôëñ35%. No difference was found between treatment strategies in terms of cardiovascular mortality, all-cause mortality, stroke or worsening HF. While this study claimed to compare strategies of rate versus rhythm, it is noteworthy that during follow-up visits, sinus rhythm was oftenelectrocardiographically observed among patients assigned to the rate-control strategy. This is therefore clearly not a 'strict' rate versus rhythm control evaluation. Furthermore, a higher proportion in the rhythm-control group required electrical cardioversion compared with the rate-control group (59 versus 9%).
The issue of rate control versus rhythm control in patients with HF was also evaluated by Shelton and colleagues.14 This randomised controlled study enrolled only 61 patients with HF and persistent AF. Amiodarone was used to maintain sinus rhythm, which was achieved in 66% of patients assigned to the rhythm-control group at one year. The rate-control group received digoxin and beta blockers; adequate rate control was achieved in 90% of these patients. At one-year follow-up, there was no difference between the treatment strategies in New York Heart Association (NYHA) class or six-minute walk test (6MWT). However, patients with rhythm control exhibited significant improvements in LV function and quality of life indicators compared with those who were assigned to the rate-control group.
The data from these studies exploring rate versus rhythm control strongly suggest that clinicians should keep an open mind on whether sinus rhythm is important in patients with HF. While sinus rhythm may not always be needed, it is clear that many patients with HF may benefit from restoration and maintenance of sinus rhythm. There are now several studies that have confirmed improvements in LV systolic function among patients with AF in whom sinus rhythm has been restored. This may be more difficult to accomplish with antiarrhythmic drugs (AADs) such that nonpharmacological approaches may be preferred in some of these individuals. The cause of improvement in left ventricular function is likely multifactorial, and not simply due to reversal of a tachycardiamediated cardiomyopathy. Regularisation of cardiac rhythm has been shown to benefit cardiac output, just as restoration of sinus rhythm has led to the recovery of atrial contractility. Thus, it would be prudent to assess carefully whether patients presenting with both AF and HF will benefit from improved outcomes following restoration and maintenance of sinus rhythm, whether achieved by drug therapy or catheter ablation.
Modalities of Rhythm Control
The latest American College of Cardiology (ACC)/American Heart Association (AHA)/European Society of Cardiology (ESC) recommendations propose the use of amiodarone or dofetilide for the maintenance of sinus rhythm in the setting of HF.1 Ablation has been proposed as a possible alternative in cases where maintenance of sinus rhythm by these drugs fails. Studies that had surrogate end-points such as ejection fraction or diastolic ventricular diameter have shown this technique to be effective in HF patients.11 The comparison between pulmonary vein isolation and atrioventricular (AV) nodal ablation with biventricular pacemaker implantation in the recent Pulmonary Vein Antrum Isolation versus AV Node Ablation with Biventricular Pacing for Treatment of Atrial Fibrillation in Patients with Congestive Heart Failure (PABA-CHF) trial has shown benefits in favour of pulmonary vein isolation, in terms of ejection fraction and 6MWT.15 However, limitations of this study include its small size (~40 patients per group), the need for a second ablation among many patients, the increased risk of complications, in particular the occurrence of pulmonary vein stenosis, very short follow-up (less than six months) and, importantly, the fact that the clinical end-points assessed in the study provided no information on morbidity or mortality.
New therapeutic approaches are now available for treating AF since the availability of dronedarone. The first trials of dronedarone (i.e. the American-Australian Trial with Dronedarone in Atrial Fibrillation or Flutter Patients for the Maintenance of Sinus Rhythm [ADONIS] and the European Trial In Atrial Fibrillation or Flutter Patients Receiving Dronedarone for the Maintenance of Sinus Rhythm [EURIDIS]) assessed the efficacy of the drug versus placebo in patients without recently decompensated HF.16 The ANtiarrhythmic trial with DROnedarone in Moderate to severe Congestive Heart Failure Evaluating morbidity DecreAse (ANDROMEDA) is a more specific trial involving HF patients, where dronedarone was shown to have serious adverse effects in patients with severe HF.17 It is therefore not possible to prescribe dronedarone for this population of patients. Nevertheless, this drug can be used in patients with AF and underlying heart disease such as hypertensive cardiomyopathy or coronary artery disease, as supported by results from A Trial With Dronedarone to Prevent Hospitalization or Death in Patients With Atrial Fibrillation (ATHENA).18 According to data from ATHENA, it is also possible to prescribe dronedarone in patients with mild HF (NYHA functional classes I and II). Caution is recommended in prescribing this drug for patients with NYHA functional class III HF.
Modalities of Rate Control
If the maintenance of sinus rhythm is not feasible, whether as first-line therapy or in the case of treatment failure, rate control becomes necessary. The therapeutic possibilities in this regard, however, are relatively limited. Digitalis can obtain a good control of heart rate at rest but its efficacy at exercise is insufficient. Beta blockers can also be used. It is generally better to prescribe drugs that have demonstrated efficacy in HF, such as carvedilol, bisoprolol, metoprolol or nebivolol. Calcium antagonists are theoretically contraindicated in this situation owing to their negative inotropic effect. Lastly, amiodarone may be used; although able to exert an effect on AV node conduction, this drug is usually not recommended because of the risks associated with its long-term use, particularly extra-cardiac side effects. There is therefore an authentic need for the development of new drugs to be used in this indication and increased research in this field. Dronedarone, as demonstrated in the Efficacy and safety of dronedarone for the control of ventricular rate during atrial fibrillation (ERATO) trial, is able to slow AV conduction and to have a role in the rate-control strategy.19 Heart rate can also be controlled via the nonpharmacological ablative techniques. Although this technique can elicit true improvement in some patients with AF and HF, this method requires the definitive creation of an AV block, rendering a patient independent on a pacemaker.
Non-antiarrhythmic Upstream Therapies
The limitations of AF therapy in patients with concomitant HF imply a need to explore alternatives to AAD therapies in order to improve the status of these patients. The utilisation of renin-angiotensin system blockers is an important element in this regard. A metaanalysis published by Healey et al. suggests that these renin-angiotensin-system blockers may be efficient therapies in AF patients with concomitant HF,20 but direct evidence from randomised prospective trials is still lacking. The Atrial Fibrillation Clopidogrel Trial with Irbesartan for Prevention of Vascular Events (ACTIVE I) shows that the only benefit conferred by irbesartan towards AF patients with at least one risk factor for stroke is a decrease in the number of HF-related hospitalisations. Moreover, the meta-analysis by Nasr et al. states that, among all the therapies with potential preventative effects on AF in HF patients, beta blockers were associated with lower occurrences of AF in this patient population.21 This meta-analysis took into consideration the results of the Cardiac Insufficiency Bisoprolol Study I (CIBIS I), Metoprolol CR/XL Randomized Intervention Trial in congestive Heart Failure (MERIT-HF), Beta-Blocker Evaluation of Survival Trial (BEST), Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS), Study of Effects of Nebivolol Intervention on Outcomes and Rehospitalisation in Seniors with heart failure (SENIORS) and Carvedilol Post-Infarct Survival Control in Left Ventricular Dysfunction (CAPRICORN); these studies have enrolled 13,000 patients with ischaemic or non-ischaemic cardiomyopathy and an ejection fraction ~25%.
The ideal preventative treatment of AF and, by association, any progression towards the development of HF is likely to involve the optimal control of arterial pressure. Emerging pathophysiological and epidemiological data point towards a cascade of pathological and deleterious events evolving from hypertension to HF, with the occurrence of diastolic dysfunction and AF.
Expert Opinions
During the 2008 CREATE Annual Advisory Meeting, experts in the field deliberated over four questions concerning the treatment of patients with AF and HF. The first question asked, 'Is it necessary to maintan sinus rhythm in patients with HF or is rate control preferred?' Despite the results of the AF-CHF trial, a majority of cardiologists consider maintenance of sinus rhythm to be preferable, with 62.5% of experts voting for maintenance of sinus rhythm and 37.5% voting for rate control.
The second question asked, 'Is amiodarone the only safe AAD that can be used in HF to suppress atrial fibrillation?' The majority that took part in the debate believed so, with 60% voting yes and 40% voting no. It is clear from this result that some cardiologists clearly consider dofetilide to have potential as a therapy for patients in HF, drawing on data from the Danish Investigations of Arrhythmia and Mortality ON Dofetilide (DIAMOND) CHF study.22
The third question asked, 'Are drugs inferior to radiofrequency ablation to maintain sinus rhythm in HF?' Interestingly, despite many publications and studies on ablation in AF, many cardiologists believe AADs are still at the forefront in treating AF, with 67.5% backing pharmacological therapy against the 32.5% of experts supporting catheter ablation.
Finally, the fourth question asked, 'Are there differences in the need for sinus rhythm in systolic versus diastolic HF?' The experts were clear in emphasising the problem underlying diastolic HF, with 87.5% voting yes. Discussion concerning this specific topic raised the possibility of dronedarone being effective in this patient population.
Conclusion
The treatment of patients having both AF and HF remains extremely complex. The major problem is the lack of AADs, with the exception of amiodarone, that can be used in these patients, particularly if the HF presented is severe. The choice of strategy, whether rhythm or rate control, has to be made for each individual case, as neither strategy has shown any clear superiority over the other. The strategy of rhythm control is preferred in highly symptomatic patients. Nonpharmacological techniques also have their place, notably when pharmacological treatment has failed.