Duration of heart failure and the risk of atrial fibrillation: different mechanisms at different times?

Chronic heart failure increases the risk of atrial fibrillation (AF), with the prevalence of AF paralleling the severity of heart failure. Factors that underlie this increased susceptibility to AF may include electrical, structural, and neurohumoral changes. In AF, it is recognized that atrial electrophysiological remodelling occurs and contributes to the perpetuation of the arrhythmia, most notably the decrease of effective refractory period (ERP) which predisposes to re-entry by shortening the wavelength. Does heart failure cause similar changes in atrial electrophysiology that predispose to the arrhythmia? An extensively studied dog model, in which heart failure was induced by rapid ventricular pacing, has suggested not: atrial electrophysiological changes were either absent, or were in the opposite direction, with prolongation of the action potential duration (APD). Susceptibility to induced AF was increased, but this was more related to structural changes, in particular increased fibrosis, than electrophysiological properties. In contrast, our recent study of human right atrial isolated myocytes found that left ventricular (LV) systolic dysfunction was associated with electrophysiological changes—decreased APD and shortened refractoriness—which would predispose to AF, contrary to the findings in the dog model. Sridhar et al. shed some light on this discrepancy, indicating that the key factor may be the duration of heart failure. The main methodological difference in the dog model used in this study was the duration of ventricular tachypacing—at least 4 months, compared with a maximum of 6 weeks in previous studies. In contrast to the studies with short-term heart failure, this study of chronic heart failure found that the LV dysfunction was not fully reversible, the induced AF was sustained rather than self-terminating, and there was atrial electrophysiological remodelling with shortening of left atrial myocyte APD (both at 50 and 90% repolarization) and of right atrial ERP in vivo. They also found increased fibrosis similar to that described with shorter duration heart failure, suggesting that the progressive changes were electrophysiological rather than structural. This study provides a new insight into the association between chronic heart failure and AF, but also raises a number of questions. What are the ionic mechanisms underlying the shortening of APD and ERP with chronic heart failure? An increase in Ito was found, consistent with the observed shortening of early repolarization, although it might be expected to have less influence on late repolarization or ERP. However, a mathematical model of a canine atrial cell showed that such an increase in Ito would shorten both early and late repolarization. 8 Alternatively, the reported decreases in potassium currents (IKur, IKs, IK1) would tend to oppose this effect. There was no change in ICaL, although the shortened early APD due to the Ito increase might decrease the duration of ICaL, and thus shorten the late repolarization. However, this would also affect other currents, such as the delayed rectifiers, which might be activated less strongly, opposing late APD shortening. Why did Ito increase, and not decrease as found in numerous models of atrial pathology, including short-term heart failure in dogs, and chronic AF or LV systolic dysfunction in humans? There was no change in Kv4.3 protein, the a-subunit which carries Ito, but the Kv channel interacting protein 2 (KChIP2), which affects Kv4 surface expression and functional properties, was increased. This would be consistent with the observed Ito increase, as well as the IKs decrease, since Ito was decreased and IKs increased in mice in which the gene encoding KChIP2 was deleted. The role of increased oxidative stress was investigated, and an anti-oxidant prevented the Ito increase—but did not reverse the shortening of terminal repolarisation in chronic heart failure. What happens to atrial intracellular calcium handling with chronic heart failure? In the short-term heart failure model, after 2 weeks of ventricular tachypacing, there was APD prolongation and cellular calcium overload with resultant triggered or spontaneous arrhythmic activity. This is in contrast to the chronic heart failure model where the shortened action potential resulted in reduced ICaL and thus less calcium entry. However, the changes in calcium handling and related regulatory proteins observed with short-term heart failure have yet to be studied in chronic heart failure. The opinions expressed in this article are not necessarily those of the Editors of Cardiovascular Research or of the European Society of Cardiology * Corresponding author. Tel: þ44 141 211 4833; fax: þ44 141 552 4683; E-mail address: [email protected]