Hypertension begets hypertrophy begets atrial fibrillation? Insights from yet another sheep model

The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology. The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology. Atrial fibrillation (AF) affects approximately two million patients in the USA and an equal number in Europe and increases morbidity and mortality in affected patients and population worldwide. Although acute conversion of AF to sinus rhythm can be achieved in almost all patients, therapeutic options to maintain sinus rhythm (repeated cardioversions, anti-arrhythmic drugs, catheter ablation, or surgical procedures) are often ineffective. Understanding the different mechanisms that contribute to AF may guide us towards a more effective ‘rhythm control’ therapy. Arterial hypertension, found in 65–70% of AF patients1,2 but only in 25–50% of the population,3 is the most common co-morbidity found in AF registries in Germany and Europe. Although this suggests a causal link between hypertension and AF, the mechanisms by which hypertension predisposes to AF are not well understood. Electrical, structural, and ultrastructural changes usually concur in the atria before AF develops. Some of the electrical changes that precipitate AF, shortening of the atrial action potential and refractory period and/or local conduction disturbances, have been delineated in detail during the past decade. The underlying pathophysiological concept of electrical remodelling (AF-induced shortening of atrial refractoriness) stems from studies in a goat model with pacing-induced AF (‘AF begets AF’).4 In this issue of the European Heart Journal, Kistler et al.5 report observations in a sheep model that provide insights into potential mechanisms by which chronic ‘isolated’ arterial hypertension can beget AF. Kistler et al. used maternal treatment with corticosteroids to induce post-natal, life-long arterial hypertension in sheep. Although the cause of hypertension in men might differ from the mechanisms in this animal model, the study delineates mechanisms that promote the initiation and perpetuation of AF in the hypertensive heart. First of all, the study provides experimental evidence that hypertension creates a substrate for AF. Prenatal steroid treatment not only caused life-long hypertension, but also markedly prolonged AF episodes induced by aggressive pacing (mean AF duration 84 vs. 0.5 s) in old sheep (4.5 years or age). There was no evidence for a shortening of atrial refractoriness, but the atrial structure was altered, with cellular hypertrophy, patchy fibrosis, increased collagen content, and myolysis. Altered atrial structure and atrial fibrosis can form a ‘substrate’ for AF and are the most likely explanation for conduction slowing, conduction barriers, and increased inducibility of AF in this model, comparable with the changes found in failing dog hearts.6 Although it has long been known that hypertension provokes a hypertrophic response in the ventricular myocardium, the paper by Kistler et al. extends this finding to the atria and suggests that atrial hypertrophy, provoked in response to chronically elevated blood pressure, may be the pathophysiological link that connects arterial hypertension and AF. The effect of elevated blood pressure on atrial conduction described by Kistler et al. was even more pronounced than conduction disturbances demonstrated by others in structurally remodelled atria. In models of atrial dilatation or heart failure,6,7 conduction was slowed exclusively at high rates or during premature atrial stimulation. In hypertensive sheep, conduction is already slow at low rates, more resembling electrophysiological remodelling in dilated atria of patients with heart failure.8 This difference might be related to different mechanisms inducing structural remodelling or to the long time during which the substrate could develop (4.5 years). The authors suggest that activation of the renin–angiotensin system, one of the most powerful stimuli for cardiac hypertrophy, could provoke the atrial hypertrophic response in their model. The morphological and electrophysiological changes in the atria of failing dog hearts subjected to ventricular tachypacing6 are of striking resemblance to the changes induced by long-standing arterial hypertension in the present model: atrial hypertrophy, increased atrial fibrosis, conduction slowing, and prolonged duration of induced AF. Blockade of the renin–angiotensin system can prevent structural remodelling associated with AF in dogs with pacing-induced heart failure.6 In the LIFE study, regression of cardiac hypertrophy—assessed as left ventricular hypertrophy—was associated with prevention of new-onset AF, irrespective of classical risk factors for AF or treatment allocation.9 The sequence of events suggested by the data of Kistler et al. and by these studies—hypertension activates the renin–angiotensin system, induces (atrial) hypertrophy, and thereby causes structural remodelling of the atria—could explain why inhibition of the renin–angiotensin system is so effective in the prevention of AF.9,10 Further studies are warranted to assess the activity of the renin–angiotensin system and the effects of its pharmacological blockade in this model. In addition, the effect of lowering blood pressure without interfering with the renin–angiotensin system on the structural remodelling process and the substrate for AF might be of interest. As in all models, there are limitations. The authors did not search for spontaneous AF. The role of atrial stretch as a trigger for atrial or pulmonary vein automaticity has not been explored. Furthermore, although pre-natal steroid treatment is accepted as a model for hypertension, increased blood pressure is a consequence of a foetal intervention, causing altered steroid-regulated gene expression. Cardiac hypertrophy could be a consequence of this prenatal intervention rather than a response to hypertension. Last but not least, other signalling cascades than the renin–angiotensin system can provoke atrial hypertrophy, fibrosis, and cell death; these should be studied. The paper by Kistler et al.5 establishes a link between hypertension and AF and thereby introduces atrial hypertrophy as a potential new target for rhythm control treatment in patients with AF. This study was supported by the German Ministry of Education and Research BMBF (AFNET, Gi020407). Conflict of interest: The authors have received research grants and honoraria from pharmaceutical companies for studies related to AF.