|Home | About | Journals | Submit | Contact Us | Français|
Our poor understanding of the mechanisms of atrial fibrillation (AF) is humbling. Although valuable insights have been gained after decades of active research, the simple question of whether it is caused by reentry or focal activity remains unanswered, even if we tried to do so just in relative terms. More than a century ago, Winterberg suggested that AF was the consequence of multiple ectopic foci.1 Decades later, the idea that a single rapid-firing site could lead to AF was proven by Scherf2 and Prinzmetal,3 who showed that focal injection of aconitine (a sodium channel opener) produced rapid regular focal activations at the injection site, but global atrial fibrillatory activation patterns. However, when Moe and Abildskov4, 5 proposed the multiple wavelet hypothesis, it became accepted that AF was caused by self-perpetuating activation wavelets that propagated on heterogeneous atrial tissue. Indeed, mapping studies were able to demonstrate such multiple wavelets,6 and it became accepted that a minimum of 4–6 wavelets were required to sustain AF.7
However, Allessie et al had also previously demonstrated functional reentry in the atrium in the absence of an anatomical substrate,8 which supported the idea that reentry –by definition a self-sustainable process- could underlie the mechanisms of perpetuation of AF. How could a periodic phenomenon such as reentry underlie a chaotic and aperiodic one such as AF? Rapidly activating rotors may lead to global fibrillatory activation patterns if drifting,9 or if activations stemming from the rotor fail to conduct 1:1 to neighboring tissues. This so-called fibrillatory conduction has been shown to be caused by wavebreak in anatomically-determined locations, such as pectinate muscles.10 Support for relatively stable rotors as the engine of AF came from a series of works from Jalife’s laboratory, showing fast, local periodicity11 and reentry in the left atrium12 that led to stable left-to-right frequency gradients13 during sustained AF. A predilection for left atrial reentry to occur in the neighborhood of the pulmonary veins was also evident: Arora et al14 showed detailed optical mapping of such local reentry, and Chen’s group emphasized the complex underlying histological patterns in this region (including the ligament of Marshall), that could lead to reentry,15 a concept that had been predicted by Spach decades earlier16 and that seems to be relevant in ventricular fibrillation as well.17 In the clinical realm, the concept of focal discharges as a cause of AF got enormous support with Haissaguerre et al’s seminal finding that ectopic beats from the pulmonary veins initiated AF.18 Mechanistically, focal beats triggering AF were shown to arise from electrogenic sodium/calcium exchange in situations of calcium overload.19, 20 During ongoing AF, however, focal activations have been harder to prove. Indeed, Atienza et al21 postulated that frequency acceleration by adenosine administration during AF supported reentry as the primary mechanism of AF maintenance.
In this issue of Heart Rhythm, Yamazaki and colleagues22 report on the effects of several pharmacological interventions in the activation patterns of AF induced by mechanical stretch (stretch–related AF, SRAF). This is a well-established model, previously described by the authors, that leads to sustained AF for hours. The central purpose in this work is to assess the relative relevance of focal activations vs. reentry in the maintenance of AF. Using this model of sustained AF, the authors then attempted to suppress intracellular calcium-derived afterdepolarizations with caffeine or ryanodine and showed that either of these two drugs succeeded at terminating AF (10/13 cases). When on top of the physiological alterations created by stretch, the authors added the combined administration of acetylcholine and isoproterenol, then caffeine or ryanodine failed to terminate AF (1/11 cases). The authors correlate these differential effects with the specific activation patterns of AF, suggesting that without acetylcholine/isoproterenol, AF is maintained by focal activations (which would be suppressed by caffeine or ryanodine), whereas with acetylcholine/isoproterenol, reentry plays an increasingly predominant role, which becomes seemingly exclusive when caffeine or ryanodine are added and focal activations are suppressed. Although the overall interpretation seems rather simple and it is not without important caveats, this is a commendable, rigorous effort to attempt to correlate mapped activation patterns with the underlying physiological conditions leading to AF. Furthermore, the authors complement their results with computer simulations to show how focal discharges interact with rotor core meandering dynamics. While answering some important questions, the paper raises more additional questions that remain unanswered.
Important caveats relate to the methods of suppression of intracellular calcium-derived afterdepolarizations. The authors chose ryanodine or caffeine. A better drug regime would have included thapsigargin along with ryanodine to completely disable the calcium cycling, which has been shown to suppress focal discharges in AF models.23 The calcium chelator BAPTA would have also been an uncontroversial eliminator of calcium-induced arrhythmogenesis.24 Additionally, it is unclear whether conclusions drawn from a very specific form of AF (stretch-related) apply to others.
The overall picture remains complex. One conclusion that can be safely drawn from this paper is that AF is an ever-changing phenomenon whose mechanisms vary depending on the underlying physiological conditions. Thus, rotors and focal activations do not seem to have a fixed hierarchy. AF may be sustained by a rotor from which daughter wavelets emanate but propagate in an irregular pattern due to fibrillatory conduction. However, the rotor’s stability is subject to continuous threats by focal activitations that may arise at any given time by a triggered activity mechanism.19, 20 Arguably, the irregularity of fibrillatory conduction would enhance the generation of triggered activity by promoting relative pauses, and focal activations would still owe their existence to their triggering beats and indirectly to the mother rotor. Thus, the mother rotor seems to be a promiscuous source of offspring wavelets, both by direct emanation and by indirect triggering. What this paper illustrates is that AF seems to be a continuous struggle between two related phenomena: 1) the tendency of wavelets to self-organize as rotors, 2) the triggered beats that can invade the rotor and extinguish it, or form a new rotor. Each can lead to one another, and either can predominate depending on the underlying conditions.
Additional questions remain regarding the mechanistic, clinical, prognostic and therapeutic relevance of focal- vs. rotor-driven AF: Are activation patterns merely a reflection of the underlying atrial physiology -as in this paper- applicable to clinical scenarios where AF develops, for example hypertensive AF vs lone AF? Clinically, do the activation patterns impact the long-term stability of AF? Do they impact the clinical course of AF (paroxysmal or persistent)? Or its thrombogenic potential? Or the susceptibility to different therapies, ablative, antiarrhythmic or substrate-based (i.e. ACE inhibitors)? All electrophysiologists ablating AF have experienced the enormous variability of activation patterns in different patients, as reflected by intracardiac recordings. The current paper illustrates mechanistic variations of AF caused by different underlying physiological conditions, and perhaps suggests that our focus should be not so much on AF itself, but the conditions that generate it.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.