There is a developing appreciation for the complex reciprocal pathophysiologic relationship between the development of AF and HF [5]. Both of these comorbidities share a common set of well-defined risk factors such as age, hypertension, diabetes, valvular disease and coronary disease, which partially explains these tandem epidemics. However, the interplay between these conditions likely involves a positive feedback loop whereby HF leads to a volume and pressure overload state promoting interstitial fibrosis leading to altered atrial refractory properties and heterogeneity of repolarization. This substrate eventually gives way to AF, which then causes a loss of atrial-ventricular synchrony, rapid ventricular response and irregularity of heart rhythm which further promotes atrial and ventricular remodeling thus continuing the cycle of AF and HF.

The onset of AF is known to impair cardiac performance particularly in patients with underlying HF [6]. A rapid ventricular rate can lead to inadequate stroke volume via suboptimal diastolic filling as well as the development of rate-related cardiomyopathy over time. Bradycardia may lead to syncope or other manifestations of inadequate cardiac output in the setting of low stroke volume. The irregular rhythm inherent to AF impairs both diastolic filling of the ventricles and leads to reduced stroke volume. Cardiac output may be further impaired by the loss of atrial systole and for some patients the transition from sinus rhythm to AF and sudden loss of atrial “kick” can exacerbate HF leading to acute symptoms prompting hospitalization. Atrial fibrillation activates the neurohormonal axis leading to further maladaptive atrial and ventricular remodeling as well as a pro-inflammatory state as evidenced by increased circulating levels of cytokines IL-6 and TNF-alpha [7]. Patients with HF who have persistent AF have demonstrated diminished exercise capacity (as measured by peak oxygen consumption during cardiopulmonary exercise testing) as well as decreased stroke volume index (as measured by oxygen pulse) when compared to patients with HF who are in sinus rhythm [8]. Nevertheless, one of the challenges of determining the hemodynamic and functional impact of AF on HF is distinguishing between true cause and effect versus the identification of AF as a surrogate marker of HF disease severity.

The role of AF as an independent contributor toward mortality in patients with HF is not well established. However, observational data from the Framingham heart study suggest that patients with AF and HF are 1.5–3 times more likely to die than patients in sinus rhythm [9]. On the other hand, subsequent cohort studies involving optimally medically managed HF patients found no independent association between AF and mortality after controlling for NYHA class, coronary artery disease, diuretic use, hemoglobin level and serum creatinine [10]. Clinical trials involving subset analyses of patients with AF and symptomatic HF have also presented conflicting data. Both SOLVD [11] (Studies Of Left Ventricular Dysfunction prevention and treatment) and CHARM [12] (Candesartan in Heart failure-Assessment of Reduction in Mortality and morbidity program) involved large samples of symptomatic HF patients (6,517 and 7,599, respectively) and demonstrated an independent association with all-cause mortality in patients with AF. However, these findings conflicted with the much smaller Veterans Affairs Vasodilator-Heart Failure Trials (V-HeFT-I, II) that evaluated a total of 1,427 patients with mild to moderate symptoms of HF in which 14 % had AF, yet there was no independent association with increased mortality or hospitalization [13].

In the setting of observational data as well as trial data suggesting the possible contribution of AF to increased mortality in patients with underlying HF, many have hypothesized that restoration of sinus rhythm may be a superior approach to rate control in patients with HF. To date, no large clinical trial has been able to demonstrate clear survival advantage of antiarrhythmic drug (AAD) therapy as a rhythm control strategy. However, the evidence-based medicine paradigm for following the intention-to-treat principle in clinical trials can make the interpretation of each trial more complicated than the summary headline; careful review of each relevant trial as well as a recent meta-analysis tells a more complete story of the net benefits of restoration of sinus rhythm in heart failure (Table 13.1) [14].

A clear observational trend has been the association of successful restoration of sinus rhythm with improved survival and reduced secondary endpoints such as hospitalization. What remains unclear is whether achieving sinus restoration resembles a marker of improved prognosis independent of AADs or if therapeutic intervention with AADs confers a survival benefit that is counter-balanced by known drug toxicity and proarrhythmia. Regardless, approved AADs (particularly amiodarone and dofetilide) used for the restoration and maintenance of sinus rhythm appear to be safe in patients with HF when administered appropriately in accordance with published guidelines. Dronedarone should clearly be avoided in patients with recently decompensated or advanced HF (NYHA class III–VI). The PALLAS trial suggests that it should be avoided altogether in patients with any heart failure [15]. No other AADs have been sufficiently studied to comment on safety or efficacy in HF. The following section summarizes select relevant trials of AADs involving patients with HF and/or reduced left ventricular systolic function. A general review of rate versus rhythm control trials involving patients without HF is described elsewhere [16].