Do amiodarone and dronedarone prevent thrombo-embolic stroke by treating the atrial myopathy of patients with atrial fibrillation? A provocative hypothesis

Patients with atrial fibrillation (AF) are at markedly increased risk of thrombo-embolic events, and this risk can be reduced with the use of oral anticoagulants. However, many patients with AF with a heightened risk of stroke are not prescribed these drugs. Furthermore, if warfarin is used for anticoagulation, many patients do not receive doses that provide international normalized ratios that are adequate to reduce the risk of systemic thrombo-embolic events.1 Importantly, even when they are prescribed oral anticoagulants appropriately, patients with AF are likely to suffer from small repetitive cardiogenic thromboemboli (clinically overt or inapparent) that can cumulatively lead to a vascular dementia. Is the enhanced risk of stroke in patients with AF directly related to the arrhythmia or is it caused by the underlying left atrial disease? Traditionally, physicians have believed that the disorganized electrical activation of the atria seen in AF drives thrombus formation. However, it is the decreased flow velocity in the atrial chambers due to intrinsic structural disease that predisposes to thromboembolization; mitral regurgitation protects against left atrial stasis and stroke even though it promotes chamber dilatation and increases the likelihood of AF. The fibrotic remodelling process in the left atrium that underlies the genesis of AF is the primary driver of the impairment of the chamber’s reservoir and conduit functions, whether or not AF is evident. Furthermore, inflammation and fibrosis of the left atrial wall may directly promote the thrombogenic potential of the atrial endocardium. Accordingly, atrial fibrotic remodelling is independently associated with left atrial thrombus formation, and the extensive atrial fibrosis seen in patients with long-standing AF predisposes to stroke, independently of left atrial chamber size or the dyssynchrony of contraction (Figure 1).2 Contrasting effects of Class III antiarrhythmic drugs and catheter ablation on atrial myopathy and the risk of thrombo-embolic stroke. In many patients with chronic atrial fibrillation, an underlying inflammatory and fibrotic atrial myopathy predisposes to atrial thrombus formation and to thrombo-embolic stroke. Catheter ablation can abolish the atrial arrhythmia, but it may exacerbate the atrial myopathy, and thus, it appears to have little effect on stroke. In contrast, in randomized controlled clinical trials, Class III antiarrhythmic drugs (amiodarone and dronedarone) prevent stroke independently of their effect on atrial fibrillation and even in patients receiving oral anticoagulants, and this benefit may be related to a direct effect of these drugs to ameliorate the atrial myopathic process. Additional doubts about the primacy of AF in causing stroke is derived from longitudinal studies that have utilized continuous electrocardiographic monitoring devices to detect AF in patients at risk for or with a history of stroke. Taken collectively, these studies do not support a close relationship between episodes of AF and the occurrence of stroke.3 Most patients with a stroke do not show evidence of AF in the weeks preceding the event, and if AF is detected, the episodes are extremely infrequent and transient. Importantly, most patients who suffer a stroke have no evidence of AF at all or exhibit AF for the first time only after the stroke has occurred. These observations lend further support to the premise that the severity of left atrial disease (as reflected by the degree of atrial enlargement or fibrosis) may be the determinant factor for stroke and for vascular brain injury, whether patients have or do not have AF.2 If the presence of an atrial myopathy is the principal source of the cardiac thrombi that cause embolic strokes, the well-established epidemiological association between AF and stroke may be related in large part to the fact that patients with AF typically have more advanced left atrial disease. How can the atrial myopathy that drives the risk for systemic thrombo-embolic events be treated? In certain patients with AF, the atrial myopathy may result from the expansion and inflammation of epicardial adipose tissue that adjoins the left atrium. The transmission of the inflammatory process to the underlying myocardium causes electroanatomical remodelling. Drugs that exert anti-inflammatory effects on this fat depot (e.g. statins) seem to ameliorate the severity of the left atrial myopathy. Such an action may explain why statins have reduced the incidence and burden of AF in randomized controlled clinical trials, and their use has been accompanied a decrease in systemic thrombo-embolic events in observational studies.4 However, most physicians do not regard statins as agents to be prescribed for the management of AF. In contrast, amiodarone and dronedarone are chemically similar Class III antiarrhythmic drugs that are widely prescribed for the treatment of AF, either to control the ventricular response or to suppress the arrhythmia entirely. In the large-scale randomized placebo-controlled ATHENA trial, patients who were randomized to dronedarone experienced a 34% reduction in stroke (P = 0.03); importantly, this benefit was incremental to that produced by oral anticoagulants or antiplatelet drugs, which were prescribed in 60% of patients.1 Interestingly, this benefit could not be ascribed to the effects of the drug on AF, since patients in the trial experienced a decrease in the risk of stroke with dronedarone even if they remained in AF. Similarly, in the SCD-HeFT trial (the largest randomized trial that compared amiodarone with placebo in patients at risk of stroke who did not have AF), patients randomized to amiodarone had a 56% reduced risk of systemic thromboembolism (P = 0.01), which could not be accounted for by an effect of the drug to prevent new-onset AF.5 In light of these observations, it is noteworthy that both amiodarone and dronedarone have been shown to ameliorate the development of adverse myocardial remodelling, and thus, prevent the development of experimental myopathy in response to various stresses (Figure 1). Amiodarone retards the development of atrial myopathy produced by rapid atrial pacing, even if tachypacing is maintained.6,7 Similarly, dronedarone reduces the degree of ventricular remodelling produced by experimental pressure overload.8 If these drugs are capable of directly ameliorating the evolution of atrial myopathy, such an effect can lead to a reduced risk of systemic thromboembolism independently of any benefit that these drugs might exert on the accompanying atrial arrhythmia. It is therefore noteworthy that the benefit of dronedarone seen in the ATHENA trial on the risk of stroke was particularly notable in patients with a meaningfully increased CHA2DS2-VASc score, which identifies individuals who are likely to have a meaningful underlying atrial myopathy. Catheter ablation can effectively abolish AF for meaningful periods of time in a large proportion of patients. However, the effectiveness of ablation on AF largely depends on the absence of a meaningful atrial myopathy. Those who have evidence of an atrial myopathy prior to the procedure (as reflected by increased left atrial dimensions or pressures or atrial fibrosis) are at increased risk of AF recurrence, and they often require repeat ablation procedures. Furthermore, if a patient with AF has an underlying atrial myopathy, ablation may add to the pre-existing fibrotic burden of the left atrium, further compromising chamber capacitance and its transport function (Figure 1).2 The atrial injury produced by ablation has an important thrombogenic potential, explaining why the procedure itself carries a high immediate risk of systemic thrombo-embolic events.9 The recently completed CABANA trial highlights the challenges of reducing the risk of stroke in patients with AF. The trial randomly assigned patients to catheter ablation or to treatment with antiarrhythmic drugs, which commonly included amiodarone, but also allowed other antiarrhythmic agents. Most patients were treated with oral anticoagulants, but it was not clear whether they were maintained at doses that could reliably reduce the risk of stroke. It is therefore noteworthy that there were 19 hospitalizations for stroke or a transient ischaemic attack in the catheter ablation group and 16 such events in the drug treatment group, even though AF was controlled far more effectively in the patients who had ablation.10 The investigators have not yet publicly reported the stroke rates in patients who were treated only with amiodarone or dronedarone. Yet, the similar risks of stroke in the two treatment arms may have been related to (i) the absence of a direct causal relation between AF and the risk of stroke; and/or (ii) an additional effect of amiodarone or dronedarone to ameliorate the severity of the underlying myopathy and thus prevent stroke in a manner that is independent of the actions of these drugs to suppress AF. Aside from the relief of arrhythmia-related symptoms, the major reason to treat AF is to prevent the occurrence of systemic thromboembolism, since there remains uncertainty as to the role of AF in the progression of cardiomyopathy or heart failure. There is a growing interest in the use of catheter ablation as a first-line treatment for AF, since it is generally more effective than drug therapy in abolishing the arrhythmia, and it often spares patients from receiving potentially toxic antiarrhythmic agents. However, any preference for catheter ablation should be tempered by the possibility that certain Class III antiarrhythmic drugs may have an effect to reduce the risk of stroke because of a direct benefit on the underlying atrial myopathy. Oral anticoagulation is the most important treatment for the prevention of stroke, but many patients with AF are not prescribed these drugs or remain at risk of stroke despite receiving long-term anticoagulation. Additionally, because it exacerbates the underlying atrial myopathy, AF ablation without concomitant Class III antiarrhythmic drugs may not provide optimal protection against stroke, even in those who are prescribed oral anticoagulants. Therefore, in patients with AF who are at meaningful risk of stroke, physicians may well be advised to assess the presence and severity of an underlying atrial myopathy as a means of guiding their treatment decisions. The opinions expressed in this article are not necessarily those of the Editors of Europace or of the European Society of Cardiology. There is a single author who takes full responsibility for the entire content of the manuscript. Conflict of interest: Dr M.P. has recently consulted for Abbvie, Actavis, Akcea, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Cardiorentis, Daiichi Sankyo, Johnson & Johnson, NovoNordisk, Pfizer, Sanofi, Synthetic Biologics, and Theravance. None of these relationships are relevant to the topic of this manuscript.