|Home | About | Journals | Submit | Contact Us | Français|
Non pharmacological approaches for the management of atrial fibrillation are rapidly emerging as the mainstay for the definitive management of this arrhythmia. Over the past several years, numerous studies have appeared in the literature that have highlighted various aspects of the pathophysiological mechanisms underlying this arrhythmia. The purpose of this brief review is to place the current apporaches being ulitized for arrhythmia management in context of what is known about arrhythmia mechanisms.
Atrial fibrillation is an important cause of morbidity and in recent years has received increasing attention in the fields of interventional electrophysiology and cardiac surgery. This is predominantly due to the success of ablative techniques in managing this arrhythmia, and the disappointing results of pharmacologic therapy. This has prompted an expanded use of these techniques in patients with paroxysmal forms of the disease especially in younger patients and those without any structural heart disease1. From a procedural standpoint, the favorable long-term results of the surgical maze procedure2 as well as the recognition that pulmonary veins are critical to the initiation of paroxysmal atrial fibrillation3, 4 have also prompted renewed interest in the management of this arrhythmia.
Currently, diverse techniques are being used for the non-pharmacological management of AF. These include: (i) surgical maze and its variants 2 (ii) catheter-based pulmonary vein (PV) isolation (antral or ostial)4, 5 (iii) left atrial ablation6 (iv) ablation of complex left atrial (LA) electrograms7 and (v) ablation of fat pads (epicardial autonomic ganglia). The purpose of this manuscript is to review the relevant aspects of genesis of arrhythmia especially from the pulmonary veins and attempt to place the current procedures for the management of this arrhythmia in perspective.
At the outset, it would be useful to place the left atrium in its developmental context and review the anatomy of the atria in relation to various structures in its vicinity that have been implicated in the genesis of atrial fibrillation.
The anatomy and electrophysiology of the PVs is complex and still incompletely understood. During human embryologic development, the myocardium around the common PV shows expression of an antigen (HNK-1) is associated with the developing conduction system8. This raises the possibility that these sites may harbor cells with pacemaker characteristics; indeed, PV myocytes seem to show ion channel expression similar to that of sinoatrial nodal cells, with significant If current and lack of Ik1 current9. A recent study has shown that venous tissue is incorporated into the smooth walled body of the LA10. This is in contrast to the sinus venarum portion of the right atrium and the remaining parts of the left and right atria (Figure 1). The pulmonary vein left atrial junction definition can be difficult to define and the ‘antral regions’ of the pulmonary veins could encompass vast portions of the posterior left atrial wall5. Further mapping studies of atrial fibrillation have shown complex patterns of activation in these regions of the atrium (Figure 2). Most catheter based ablative procedures that are being utilized these days tend to target myocardium in this region of the left atrium11. Such lesion sets could affect multiple mechanisms potentially driving this arrhythmia (Figure 3).
Besides having intrinsic pacemaker activity, PV muscle sleeves differ from LA tissue in other important ways. Even in normal controls, effective refractory period (ERP) of PV myocytes is shorter than that of neighboring LA myocytes, favoring reentry or propagation of rapid triggered activity12. However, in patients with AF, PV ERPs are even shorter than in normal controls 13, with significant heterogeneity in conduction properties and refractoriness14. Autonomic stimulation, especially parasympathetic3 but also sympathetic15, can further shorten PV ERPs. These changes are likely to be arrhythmogenic in patients prone to AF, and may be one reason why vagal denervation seems to improve outcomes in surgical and catheter-based ablation procedures.
Thoracic veins, especially the pulmonary veins, are electrically active and have been the subject of several studies3, 16–19. Several mechanisms have been proposed for pulmonary vein arrhythmogencity. These structures can generate abnormal impulses, and the complex micro-architecture of muscle bundles in this region lead to abnormalities in impulse propagation. These factors set the stage for arrhythmogenesis20. Focal (triggered) activity and reentry have been implicated in mapping studies of PVs18, 19. A detailed review of these mechanisms is beyond the scope of this review. Table 1 summarizes some of the key studies in this field and Figure 2 shows a schematic interpretation the published data21.
Classically, triggered activity is divided into early after depolarizations (EADs), occurring before the end of the action potential, and delayed after depolarizations (DADs), occurring after the end of the action potential. These have been defined based on single cell recordings and response of cells to stimulation protocols and pharmacological perturbations19. In the case of PV arrhythmias, triggered activity has been implicated along with reentry19, 22. The data for triggered activity were predominantly generated using pharmacological methods, such as sympathomimetics to initiate PV ectopy19, 23. Electrophysiological characteristics of PVs have also shown to be different from those in the atria. Beta-adrenergic receptor blockers, calcium channel blockers, and sodium channel blockers have been shown to suppress PV ectopy23. Traditionally, EADs are seen in the setting of a prolonged action potential duration that favors reactivation of inward calcium or sodium currents during the plateau phase of the action potential. However, recent work has provided a new paradigm for EADs in the atria, resulting from a complex interplay of the autonomic nervous system’s effects on action potential duration and calcium current in PV myocytes. In this setting, APD prolongation is not required for EADs. Instead, the PV muscle sleeve is associated with a peak calcium transient that occurs much later in the action potential, a phenomenon augmented by parasympathetic stimulation induced shortening of the APD. The result is increased intracellular calcium at a time that the membrane potential is nearly repolarized, providing the necessary conditions for inward current via the sodium calcium exchange, which can trigger arrhythmias24, 25.
Other studies in humans and animal models have suggested reentry as a mechanism of AF arising from the region of the PVs and the PV-LA junction 18, 26. The ultra structure of this region is well suited for mico-reentry27. Further, studies analyzing modulation of the action potential during AF and localizing sites of highest dominant frequency also suggest that paroxysmal AF is maintained by reentrant sources that may be located at the PV-LA junction28.
These mechanisms undoubtedly interact with other factors, such as atrial stretch, that profoundly affect the electrical behavior of atrial myocardium. In an animal model (sheep hearts), the rate and organization of fibrillatory waves originating from the superior pulmonary veins increased with elevation of intra-atrial pressures29.
Atrial myocardial sleeves extend a variable distance onto the epicardial surface of the pulmonary veins, often covering less than the full circumference of the vein30 (Figure 4). These outer circular or spirally arranged myocardial sleeves are formed from either a cardiac or an extra cardiac source. However, there is some level of agreement in the literature that the myocardium in the region of the primitive PV has different characteristics compared to the working myocardium of the LA appendage.
Several studies have shown that the atrial myocardial sleeves in human PVs have a highly variable architecture27, 31, 32. Further, the anatomical distinction between PVs and the LA can be difficult to discern either anatomically or by electrical recordings5. It can be argued that the entire posterior LA could constitute the ‘antrum’ of the pulmonary veins, as it embryologically contains pulmonary venous endothelium. Mapping studies have also shown that the actual conduction of impulses within the PVs can follow a complex course.20
Catheter ablation techniques for paroxysmal AF aim to prevent PV triggers from conducting to the LA by energy delivery at the LA-PV junction or within the ‘antral’ areas of the LA. Electrical measures using loss/reduction of PV electrogram amplitudes and electrical isolation of the PV’s (determined by pacing maneuvers) have been used as end points. In the early years of AF ablation, actual triggers within the PVs were targeted. However, due to reports of PV stenosis, the approach was modified to one that targets PV antra to achieve conduction block between the PVs and the LA33 It is intriguing that patients benefit from LA ablation even when PV isolation is not achieved, especially in persistent AF34. This highlights the importance of the LA in the genesis of this arrhythmia, and suggests that in chronic forms of the disease, the PVs might be somewhat less important in the pathogenesis.
Complex fiber geometry can provide factors such as structural heterogeneity, anisotropy, and slow conduction that favor initiation and maintenance of localized reentry within and around the PV’s. Micro-reentry is made possible by myocardial fibrosis that electrically isolates adjacent muscle bundles, forcing ‘zigzag’ conduction of impulses27. Further several studies have shown that atrial fibrosis can enhance the ability of the atrial myocardium to sustain arrhythmias in animal models35 and humans36. Human studies have also shown that the atrial myocardium may show areas with electrogram morphologies that identify sites from which autonomic responses can be elicited37; these may also serve as catheter ablation targets38.
In addition to the above findings, there are extensive experimental and clinical data showing dynamic electrophysiological changes that occur as soon as atrial fibrillation begins. These changes, collectively termed “atrial remodeling”, increase the likelihood of continued atrial fibrillation39. The mechanisms of these changes are beyond the scope of this review. However, it is important to place these pathophysiological events in perspective in order to develop a framework to understand different ablation procedures.
There has been a great resurgence of interest in the role of the autonomic nervous system in the genesis of atrial fibrillation40. Dedicated articles in this issue of Heart Rhythm will deal with the role of the autonomic ganglia, the fat pads and the role of structures such as the ligament of Marshall (LOM). With relevance to the discussion in this article the pulmonary veins have cells with specialized conduction properties: P cells, transitional cells and Purkinje cells have been demonstrated in the pulmonary veins41. Further epicardial nerves penetrate the PV walls transmurally and form a neural network beneath the endothelium of the PVs42. These structures, due to their anatomic location, may be modified during catheter ablation techniques that target the LA in general (Figure 3).
Finally, it should be pointed out that non-pulmonary vein foci can also drive AF43 and their mechanisms are incompletely understood.
Although several studies appearing in the literature have advanced our understanding of the mechanisms of atrial fibrillation, many questions still remain. Why PVs are arrhythmogenic in some people but benign in those without arrhythmias is an enigma. The mere fact that ectopic beats arising within the PVs can conduct to the atria and drive an atrial fibrillation is itself intriguing, considering the mismatch between a small source (PVs) and a large sink (LA) (Figure 5). Impulses from such sources would be expected to be blocked in tightly coupled tissues44. It is intriguing to hypothesize that a certain degree of uncoupling of conduction due to disease may be the responsible for PV to LA conduction. Differences in connexin (Cx40) expression between muscle sleeves and atrial myocardium have been reported in canine studies45. Recently, differences in the restitution characteristics and conduction properties of the PV-LV junction have been demonstrated in humans with paroxysmal AF compared to persistent AF46. Single PACs were shown to result in extreme APD oscillations and initiation of AF. Future studies will enable us to better understand how tissue excitability and conduction characteristics contribute to the genesis and persistence of AF. It is likely that there are important issues in cell-cell communication/coupling in this area that will shed light on the mechanisms of such electrical activity. Conceptual breakthroughs in understanding this arrhythmia along these lines are likely to have a major impact on our therapeutic strategies.
Supported by the NHLBI (R01HL084261 and RO1HL067647) to KS
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.