Atrial Fibrillation (AF) and Heart Failure (HF) are major, and often coexisting, public health burdens, which are increasing in prevalence and incidence [1-3] and affect equally patients with impaired and preserved Left Ventricular (LV) function [4,5].

Atrial fibrillation, which affects 1% - 2% of the population of the western world [6], is associated with increased risk of stroke, hospitalization and overall mortality and has a significant impact on quality of life [7-12]. The prevalence of AF increases in patients with more advanced HF symptoms [13]. Heart failure was present in 34% of AF patients, and atrial fibrillation in 42% of HF patients according to The European Heart Surveys [14,15]. There is a direct relationship between New York Heart Association (NYHA) functional class and the presence of AF. Atrial fibrillation prevalence increases from 4 to 40 percent as the NYHA functional class increases from I to IV [16-23]. This association of AF and HF exists also in those with isolated diastolic dysfunction with preserved LV function [24].

Different mechanisms could explain the co-existence of atrial fibrillation and heart failure including: 1) rapid ventricular response leading to rate-related cardiomyopathy [24]; 2) reduced cardiac output from loss of atrial transport [25,26]; 3) heart failure causing increased atrial filling pressure and atrial dilatation promoting atrial fibrillation by ionic remodeling and atrial fibrosis [27]; and 4) atrial stretch-related effects on atrial electrophysiologic properties such as refractoriness and triggered activity [28]. In fact, treatment to relieve chronic stretch has been demonstrated to reverse this remodeling process [29,30].

Treatment to eliminate AF in patients with impaired LV function improves LV function [31,32]. Several studies addressed the prognostic influence of the presence of AF in HF. Mamas et al., in a systematic review of 53,969 patients [33], showed that AF is associated with worse prognosis in HF patients whether (OR 1.49) or not (OR 2.0) LV function was impaired.

Pharmacologic therapy of AF in heart failure is challenging in symptomatic patients despite conventional HF therapy. Drug-based AF rhythm control strategy in HF patients has failed to show any survival benefit compared with a rate control strategy in randomized studies such as RACE [34] AF-CHF [35] and ANDROMEDA [36]. This may be because of potential pro-arrhythmia with all antiarrhythmic drugs, including amiodarone [37-40]. In addition, in these rate vs. rhythm control studies, which are intention to treat, there is a significant number of rate control patients who achieve sinus rhythm and conversely a significant number of rhythm control patients in whom AF persists. The somewhat similar rhythm outcomes in the two treatment arms, though valid within the intention to treat, undoubtedly affects the veracity of the results if survival in sinus rhythm vs AF is the desired end point.

Data from AFFIRM [41,42], DIAMOND [43] and other studies [44] suggests that patients maintaining sinus rhythm may have better survival, but the benefit may be offset by the adverse effects of anti-arrhythmic drugs. CAFÉ II study supports this further by showing improved LV function and NT-pro BNP in patients assigned to rhythm control [45].

Since the initial reports of catheter ablation for AF rhythm control in 1998 [46], there has been substantial progress in the understanding, technique and technology of catheter ablation such that the procedure is becoming mainstream treatment in patients with drug refractory symptomatic AF. Catheter ablation for AF rhythm control has been found to be a relatively safe, effective and costeffective therapy in both clinical trials and in a worldwide survey [47,48] and has superior efficacy compared to anti-arrhythmic drugs [49,50].

The improvement of LV function, following catheter ablation for AF, may be due to restoration of atrial transport, better ventricular filling and enhanced contractility through Starling mechanism. Sinus rhythm also usually results in better rate control.

Extending this therapy to patients with LV systolic dysfunction is appealing, because of the multiple potential effects of AF on the multiple pathophysiologic mechanisms of HF [51]. However, because of differences in atrial substrate and co-morbidities, ablation in these important patients may be associated with higher risk for recurrent AF or complications.

Despite encouraging reported results for catheter ablation in patients with HF, ablation is not widely used in this substantial patient population. Even in experienced high-volume centers, HF is present in only a small proportion of patients undergoing ablation [52-54]. In this review we summarize all the data available from recent AF ablation publication enrolling patients with HF

Cha et al. [64] assessed catheter ablation for anti-arrhythmic drug-refractory AF in 368 patients with HF. They compared 3 groups prospectively: 111 HF patients with systolic dysfunction, defined as LVEF ≤ 40%; 157 HF patients with isolated diastolic dysfunction but preserved LVEF ≥ 50%; and 100 control patients without HF and with normal LV function. The primary end point was freedom from AF and anti-arrhythmic drugs at one year post ablation. This end point was achieved in 62% of patients with systolic dysfunction, 75% of those with diastolic dysfunction, and 84% of controls (P = 0.007). AF control on or off drug was achieved in 76% of patients with systolic dysfunction, 85% of those with diastolic dysfunction, and 89% of controls (P = 0.08). In the systolic dysfunction group, 49% experienced an increase in LVEF by ≥5% after ablation. Of these, 64% achieved normalization of LVEF. In the diastolic dysfunction group, 30% of patients demonstrated at least one grade improvement in diastolic dysfunction. They concluded that an ablation approach for AF has the potential for substantial quality-of-life improvement and LV functional benefit.

To date, published, randomized trials involve only small numbers of patients but suggest benefit of catheter ablation for rhythm control for atrial fibrillation patients with heart failure whether or not LV function is preserved. Larger scale, randomized studies assessing a number of endpoints, including hard endpoints, are needed and indeed there are multiple ongoing studies looking into outcomes, predictors of success, and safety in this high risk population.

Pulmonary Vein Isolation for Rhythm Control in Patients with Atrial Fibrillation and Left Ventricular Dysfunction (Catheter Ablation for AF & HF) [69] is an observational study enrolling patients with NYHA IIII CHF on optimal medical therapy with planned catheter ablation for paroxysmal or persistent AF. Primary endpoint is improvement in LV end systolic volume by 15% or more frombaseline at 6 months. Secondary endpoints include remodeling as demonstrated by atrial and ventricular dimensions on echocardiography, NYHA class, quality of life and six minute hall walk test. Composites of hospitalization for heart failure, thrombo-embolism, major bleeding or all cause mortality, as well as absence of AF will also be assessed. This study was withdrawn prior to enrollment.

Catheter Ablation Versus Medical Treatment of AF in Heart Failure (CAMTAF) [70] is an open label, parallel assignment study randomizing patients with HF and persistent AF to medical treatment or catheter ablation to restore sinus rhythm. The primary end point is the difference in LVEF between groups. Secondary outcomes include differences in peak VO₂, NYHA class, BNP, quality of life, reduction in end systolic volume, and difference in HF symptoms. The study is complete but not yet published.

Catheter Ablation Versus Medical Rate Control for Atrial Fibrillation in Patients with Heart Failure (ARC-HF) [71] is a prospective study enrolling HF patients on optimal therapy, with documented persistent AF, and comparing the strategies of catheter-ablation and medical rate control in a 1:1 randomized fashion. The primary end point is difference in peak oxygen consumption at cardiopulmonary exercise test. Secondary endpoints includes changes in LVEF, quality of life score, 6 minutes walk test, BNP levels, and freedom from AF. The study is complete but not yet published AMICA Atrial Fibrillation Management in Congestive Heart Failure with Ablation [72] is a randomized, open label study in symptomatic AF patients comparing the effect of AF ablation plus ICD/CRT implant to best medical treatment plus ICD/CRT implant in patients with reduced EF of less than 35% who are felt to require ICD or CRT-D implantation. Primary endpoint is LVEF improvement as determined by echocardiography. This study is estimated to complete in 2013.

CASTLE-AF Catheter Ablation Versus Standard Conventional Treatment in Patients with Left Ventricular Dysfunction and Atrial Fibrillation [73] is a prospective, open label, randomized, multicenter study comparing the effect of catheter ablation to conventional treatment in HF patients with AF. Eligible patients are those with LV dysfunction (LVEF ≤ 35%) and NYHA class ≥ II who have already been implanted with a dual chamber ICD with Home Monitoring^® capabilities. The primary endpoint is the composite of all-cause mortality or worsening of heart failure requiring unplanned hospitalization using a time to first event analysis. Secondary endpoints are all-cause mortality, cardiovascular mortality, cerebrovascular accidents, worsening of HF requiring unplanned hospitalization, unplanned hospitalization due to cardiovascular reason, all-cause hospitalization, quality of life, number of therapies (shock and anti-tachycardia pacing) delivered by the ICD, time to first ICD therapy, number of device-detected ventricular tachycardia and ventricular fibrillation episodes, AF burden, AFfree interval, LVEF, exercise tolerance, and percentage of right ventricular pacing. This study should be completed in 2015.

Rate Versus Catheter Ablation Rhythm Control in Patients With Heart Failure and High Burden Atrial Fibrillation (RAFT-AF) [74] is a large multicentre, randomized, open label study comparing catheter ablation for AF rhythm control to rate control in patients with HF (NYHA II-III, on optimal medical therapy, and elevated NT-pro BNP) and high burden AF. Primary end point will be cardiovascular mortality. Cost-effectiveness and quality of life will be secondary end points. The study is currently recruiting, and anticipated completion is 2016.

AF and HF with impaired or preserved LV function coexist with increasing prevalence. AF in such patients is associated with increased mortality as well as deterioration of LV function, NYHA class, exercise tolerance, and quality of life. Pharmacological therapy of AF in this population is limited due somewhat to suboptimal efficacy and increased risk of pro-arrhythmia.

Observational data suggests AF ablation is a superior modality for rhythm control with significant improvement in LVEF, symptomatic state, and quality of life. This is further supported by data from small size randomized studies showing feasibility and safety of AF ablation with improved outcome in patients with HF.

Ongoing studies will hopefully shed light on a multitude of endpoints, particularly hard endpoints such as mortality and hospitalizations and further characterize the optimal heart failure patient subgroups.