Is atrial fibrillation so common after supraventricular arrhythmia ablation as to require prophylactic treatment?

This editorial refers to ‘Incidence and predictors of atrial fibrillation and its impact onlong-term survival in patients with supraventricular arrhythmias’ by C. Ozcan et al., on page 16(10): 1508–1514. The relationship between atrial fibrillation (AF) and cavotricuspid isthmus (CTI)-dependent atrial flutter (AFL) has long been appreciated. The presence of common triggers and substrate, particularly in the context of structural heart disease with concomitant structural and electrical remodelling, helps explain why both arrhythmias are often seen together and considered two sides of the same coin. In fact, functional conduction block, especially along the crista terminalis, and slowed conduction allow the transition from multiple unstable wavefronts in AF to a single organized circuit in typical AFL.1,2 Likewise, the same triggers that initiate AF, often arising from the pulmonary veins, may also be the triggers for AFL, via transitional AF. The roles of AF and AFL in arrhythmia initiation and maintenance are not straightforward: AF can initiate AFL or organize to AFL, especially in patients taking antiarrhythmic drugs, but electrical remodelling caused by AFL may also serve as the substrate for subsequent AF occurrence.3 The relationship between AF and other supraventricular arrhythmias (SVA) has not been addressed as thoroughly. However, electrical remodelling is not exclusive to AFL. The phrase ‘AF begets AF’ has been used to describe the pro-arrhythmic effect of prolonged rapid atrial rates. It is an established electrophysiological principle that atrial myocytes exhibit shortening of the action potential duration and effective refractory period with rapid atrial electrical activation. Atrial tachycardia (AT) induced electrical remodelling occurs both in the intact whole heart in vivo and in cultured atrial myocytes exposed to rapid stimulation.4 Thus, one can hypothesize that electrical remodelling caused by incessant or recurrent AT, atrioventricular nodal re-entrant tachycardia (AVNRT) or atrioventricular re-entrant tachycardia (AVRT) may also serve as the substrate for AF development. On the other hand, after SVA substrate elimination, or ablation of the more stable and dominant arrhythmia such as AFL or AT, bursts of atrial ectopic activity may now manifest as AF (unmasking phenomenon). This notion is supported by the observation that pulmonary vein isolation (PVI) alone in patients with AF and AFL can control both arrhythmias.5 In this issue of EP-Europace, Ozcan et al. evaluated the prevalence and predictors of AF and the impact of AF on long term survival in patients with SVA who underwent catheter ablation for the treatment of symptomatic AFL, AT, AVRT or AVNRT.6 The authors concluded that the presence of any form of SVA was associated with an increased risk for the development of AF, especially in specific subgroups such as those with AFL, older individuals, those with low left ventricular ejection fraction (LVEF), prolonged PR interval and dilated left atrium.6 Importantly, long-term survival among patients with SVA and no comorbidities was no worse than that of age- and sex-matched members of the general US population. In patients with SVA and comorbidities, such as older age, coronary artery disease, valvular heart disease, diabetes, abnormal creatinine and low LVEF, survival was lower and AF more frequent.6 Atrial fibrillation was not an independent predictor of mortality in the multivariate analysis in their study, in contrast to what has been reported in larger epidemiological studies.7 In the absence of comorbidities, the long-term survival among younger patients with SVA who have undergone catheter ablation was similar to the expected survival in the general population.6 Interestingly, the prevalence of clinically apparent AF was higher before compared with after catheter ablation, raising the possibility that treatment of SVA with catheter ablation may reduce susceptibility to AF in some patients.6 This study corroborates the results of previous series and meta-analysis that established a relationship between typical CTI-dependent AFL and AF8,9 and further supports the concept of electrical remodelling. Supraventricular arrhythmias predispose to AF and treatment of SVA such as AT, AVNRT, and AVRT through radiofrequency ablation effectively decreases subsequent risk of AF. In particular, Ozcan et al.6 report a much higher prevalence of AF before vs. after AT ablation (24 and 4.5%, respectively). However, successful AFL-substrate ablation does not seem to significantly modify the natural history of AF. In fact, the incidence of AF after CTI-dependent AFL ablation was 15.2 per 100 person-years in a previous study, 25 times higher than the incidence of AF in the general population of the same age.8 This incidence is clearly time-dependent, with reported rates of 17–22% at 6 months, increasing to 50% at 2 years and 63% at 4 years.10,11 Although the authors also observed a fall in the overall AF incidence post-ablation (24.7%) compared with pre-ablation (45.3%),6 these percentages are likely to be significantly underestimated due to relatively short follow-up duration when compared with other studies and the lack of rigorous prospective surveillance for the detection of AF following ablation. In fact, their diagnosis of AF was based on documentation by electrocardiogram or Holter monitor recording during clinical follow-up, but it is well known that the incidence of AF after AFL ablation is proportional to the intensity of monitoring. Subcutaneous implantable cardiac monitors with dedicated AF detection algorithms have proved to be highly sensitive in detecting AF and measuring AF burden. In a small study of patients who had loop recorders implanted following AFL ablation, 55% developed new AF within the first year of follow-up.12 Thus, the incidence of AF post-AFL ablation is sufficiently high to warrant additional preventive or treatment strategies. Should isolated AFL ablation still be performed? What is the appropriate post-ablation anticoagulation strategy? Can reverse-remodelling be achieved through ablation and/or antiarrhythmic drugs? Whereas substrate and/or trigger ablation seems warranted for most types of SVA, especially AVNRT, AVRT and focal AT, to enhance the quality of life and prevent further electrical remodelling, ablation of typical CTI-dependent AFL may be somewhat more complex given the stronger relationship with AF. On the one hand, CTI ablation appears to have significant clinical benefit despite the relatively high rate of post-procedural AF. A decrease in antiarrhythmic drug utilization and cardioversion requirements and an improvement in quality of life over a long-term period can be expected after successful AFL ablation even in patients who develop AF. Given that the medical management of AFL is often more difficult than that of AF and the ablation of CTI-dependent AFL carries a low risk of complications, a CTI line is useful in most cases. However, in patients with both AF and AFL, PVI alone has been shown to be vastly superior to CTI ablation alone13 and not inferior to PVI plus linear CTI ablation.5 In the recent PReVENT AF Study I, a randomized clinical trial of patients in whom only typical AFL had been clinically observed, the addition of PVI to CTI ablation resulted in marked reduction of new-onset AF during clinical follow-up (12 vs. 52% in those with CTI ablation alone), as detected by an implantable loop recorder (ILR).14 This 1-year 40% reduction in the risk of ILR-detected AF may support the notion that patients with lone AFL but at high risk of AF benefit from prophylactic PVI, especially considering that 32% of patients in the isthmus ablation-only group subsequently underwent another ablation for AF, while those undergoing CTI ablation plus PVI did not need any further procedures. However, it is also noteworthy that 60% of patients were submitted to an unnecessary procedure carrying a small, yet non-negligible, incremental risk of complications. The added cost of PVI and the fact that ILR-detected AF is not necessarily an indication for PVI are also important considerations. It seems clear that this approach may be eventually beneficial in selected patients at very high risk of subsequent AF: according to Ozcan et al., perhaps those at an advanced age, with lower LV ejection fraction, prolonged PR interval and dilated left atrium.6 One should also remember that antiarrhythmic drugs taken before AFL ablation may mask the occurrence of AF until such medication is withdrawn following successful CTI ablation. The strong relationship between SVA, in particular AFL, and AF also highlights the need for a cautious benefit-to-risk ratio assessment regarding anticoagulation discontinuation. The relatively high incidence of AF post-ablation of SVA, especially typical AFL, and the fact that more than 75% of patients who experience AF after CTI ablation have a CHA2DS2-VASc score ≥115 should make us reconsider the currently accepted strategy of anticoagulation discontinuation after successful AFL ablation in patients without previous history of AF. In conclusion, the relationship between most SVA and AF should encourage clinicians to carefully evaluate longer-term anticoagulation in those undergoing an AFL ablation who are at higher risk of subsequent AF and to promote a more rigorous surveillance for the detection of AF post-ablation. Conflicts of interest: none declared.