Finish What You Started?

Mounting evidence suggests that atrial Ž brillation (AF) in man is principally a “left atrial disease.”1 AF triggering, disproportionately associated with myocardium in and about the pulmonary veins (PVs), has held center stage of late given the catheter ablation modalities currently in vogue. However, the evidence pointing to the left atrium (LA) as the key to AF sustenance is no less convincing. Apart from the fact that they appear to occur in contiguous space, comprehension of the interface between AF triggering and AF sustenance remains poor. In this issue of the Journal, O’Donnell et al.2 describe the phenomenon of “paroxysmal short cycle length recordings,” periodic episodes during ongoing AF in which local endocardial atrial interelectrogram intervals diverge (shorten) markedly from the dominant interval. They localize this phenomenon to myocardium investing the PV, although (as the authors point out) a paucity of extravenous sites were evaluated. Within PV, the paroxysmal short cycle length recording phenomenon was not ubiquitous, and its presence was correlated with sites that subsequently were demonstrated to be associated with AF triggering. In some of these patients, the periodicity of paroxysmal short cycle length recording matched that of the atrial ectopic activity occurring at the same site during sinus rhythm. Based on order of activation of two contiguous bipolar recordings from a single catheter, the authors suggest that paroxysmal short cycle length recording is due to activity originating within the PV from which it was recorded. Similar electrographic phenomena have also been reported previously.3 ,4 Is it possible that electrical activity originating within the PV serves a dual role as both AF initiator and perpetuator? This is clearly one possible interpretation of the data reported by O’Donnell et al., whereby activity originating within one or more PVs serves as a “wavelet generator,” with the atrial body following passively. Alternatively, or additively, the activity could serve as a conditioning stimulus, promoting remodeling of atrial body myocardium that results in intrinsic reentry (multiwavelet, rotor, etc.) similar to recurrent atrial tachypacing in the goat model of Wijffels et al.5 An alternative interpretation of these data portrays the PV myocardium in a more passive, yet central, light. Based on anatomic/histologic and/or cellular/ionic properties (possibly aided by extramyocardial in uences, e.g., neuronal), it is conceivable that this tissue/region is particularly hospitable to reentry, thus perpetuating AF by serving as a “base of operations,” permanent or transient. In this regard, it is interesting that at each site where it was observed, the atrial interelectrogram interval during paroxysmal short cycle length recording was consistent, approximately half the dominant interval. This may have been indicative of dissociation between the fascicles of PV myocardium, resulting in reentry with a particularly rapid cycle length that may have engendered its own stability by resisting invasion of activity from slower activating entities (e.g., LA body). Unlike the Ž rst interpretation, in this one PV myocardium has divergent roles in AF initiation and perpetuation. Like any quality “phenomenologic” report, the present study asks more questions than it answers. To begin with, it is a small study utilizing a group of young men (primarily) with preserved cardiac structure/function and “recent” persistent AF. As such, this group is not representative of the majority of patients with persistent AF. It is possible that the mechanism underlying paroxysmal short cycle length recording may vary between and within patients; the latter may occur in an evolutionary fashion. Second, the duration of recordings at each site was highly limited, and the methodologydid not permit insight into whether multiple sites could exhibit the paroxysmal short cycle length recording phenomenon simultaneously and/or whether there was temporal/spatial sequencing among sites. Third, with the exception of coronary sinus recording, there was no information on the electrographic properties of the remainder of the atrial body. It is possible that paroxysmal short cycle length recording is not peculiar to (nor dependent upon) myocardium in and about the PV. The lack of a curative ablation outcome observed in the majority of patients even after isolation from the atrial body of PV myocardium encompassing paroxysmal short cycle length recordings is of interest in this regard. It is worth remembering that it was triggering and not sustenance that caused these investigators to sample PV myocardium in the Ž rst place. In addition, the PVs have a practical attraction because they are one of the few areas of the LA where standard mapping catheters can be placed with the expectation of physical stability. Finally, there was no effort to manipulate the atrial mechanical or electrical substrate (e.g., pharmacologically) so as to better assess the link between PV paroxysmal short cycle length recording and AF sustenance. In summary, although I Ž nd the phenomena observed by O’Donnell et al. very interesting, I remain unclear as to their mechanism and unconvinced about their relevance.