Catheter ablation for atrial fibrillation in congestive heart failure: consider the upside, consider the downside Free

This editorial refers to ‘Catheter ablation of atrial fibrillation in patients with heart failure: impact of maintaining sinus rhythm on heart failure status and long-term rates of stroke and death’ by W. Ullah et al., 2016;18(5):679–686.

When patients present with deteriorating heart failure (HF) symptoms associated with atrial fibrillation (AF) development, there is understandably considerable clinical momentum to deal definitively with the arrhythmia. The adverse impact of AF on HF symptoms and outcomes including mortality has been extensively documented. However, whether this trend can be successfully reversed by rhythm management has not yet been clearly established. Definitive evidence that return of sinus rhythm results in improved outcomes including stroke risk and mortality has been lacking. Indeed, a randomized study of pharmacologic rhythm control vs. rate control demonstrated no benefit, possibly in part due to relative drug inefficacy and in part to pro-arrhythmia.¹ A number of small randomized studies with 6- to 12-month follow-up have shown that compared with a medical rate control strategy, catheter ablation of AF in HF patients results in improved NYHA class² and increase in ejection fraction.³ However, the latter finding has not been universal,⁴ and the studies were too small to consider stroke or mortality outcomes.⁵ Further complicating this picture is the recent STAR AF (Substrate and Trigger Ablation for Reduction of Atrial Fibrillation) 2 study⁶ finding that there is no proven ablative approach to substrate management in persistent AF. Yet pulmonary vein antral isolation alone for persistent AF (particularly when present >1 year) has quite modest success rates. In a HF population, particularly those with persistent AF, atrial substrate is highly likely to play an important mechanistic role.⁷

The study by Ullah et al.⁸ in this issue of EP-Europace addresses many of these important issues. It is a retrospective international registry of consecutive patients undergoing AF ablation. One thousand two hundred and seventy-three patients were included: 171 with HF and 1102 without, and as such it is one of the largest studies to date on the subject. The median follow-up was also considerably longer than in previous reports at over 3 years. The study made a number of important findings as follows.

Firstly, the final procedure success rate was no different for patients with paroxysmal AF whether they had HF (79%) or not (86%; P = 0.186). This implies that even in the context of HF, when AF is paroxysmal the predominant mechanism still relates to pulmonary vein foci.

Secondly (and in contrast to the paroxysmal AF group), patients with persistent AF had a significantly lower success rate when HF was present (57%) than when it was not (76%; P < 0.001). This too is an important result in a very large series of patients. This significantly lower success rate should be viewed in the context of evidence that the HF atrium demonstrates advanced remodelling.⁷ Basic studies have demonstrated that the hallmark of this remodelling is atrial fibrosis.⁹ Although this may be present in patients with persistent AF without HF, it is more advanced when significant left ventricular dysfunction is present.

Thirdly, HF was an independent marker of adverse procedural outcome indicating that this result was not dependent on secondary phenomena such as atrial enlargement or associated conditions but rather primarily related to congestive HF. Numerous factors may be at play in HF promotion of remodelling beyond just atrial enlargement. For example, activation of the renin angiotensin system may lead to development of atrial fibrosis.¹⁰

NYHA class decreased from 2.3 at baseline to 1.5 at follow-up (P < 0.001) and left ventricular injection fraction increased from 34 to 46% (P < 0.001). Prior small studies of AF ablation impact on ejection fraction have yielded disparate results with the majority showing improvement³ but others failing to demonstrate a significant change.⁴ The current large but observational study has confirmed dramatic symptomatic benefits and ejection fraction improvement from restoration of sinus rhythm in this refractory population. Is this the best long-term strategy? A small study with short-term follow-up demonstrated both symptomatic and ejection fraction benefit when AF ablation was compared with an ‘ablate and pace’ strategy.¹¹ However, a recent systematic review did report an overall mortality benefit of the ‘ablate and pace strategy’ in patients with AF and HF.^12,13 It is probable that both strategies have a role and patient selection for AF ablation remains dependent on factors predicting long-term success.

One of the key issues in ablation of persistent AF is indeed the long-term result. In patients with persistent AF even without HF, the long-term results have been disappointing. This is particularly important as systematic long-term monitoring was not performed in this retrospective study, and it might be argued that the recurrence rate was therefore under-estimated. Numerous studies have demonstrated that interval visits and symptom-guided follow-up will significantly underestimate the true recurrence rate.¹⁴ For how long can sinus rhythm be maintained in a population who are likely to have progression of substrate in relation to persistence of impaired left ventricular function. Numerous studies have informed our understanding of long-term results in AF ablation. That is, if underlying causative factors (e.g. obesity, sleep apnoea, hypertension) are not dealt with the late recurrence risk is significantly higher.^15,16 Whether or not more aggressive HF therapy (perhaps coupled with modern approaches to detect fluid overload) will have an impact on progression of remodelling remains unclear.

Aetiology also seems to be important. Two-thirds of this population were idiopathic dilated cardiomyopathy, so the number of patients with ischaemic cardiomyopathy was relatively small. This group had a smaller improvement in the ejection fraction following ablation indicating the significant irreversible component of left ventricular dysfunction in this aetiology. An aggressive ablation strategy may yield the best results in the non-ischaemic, idiopathic population.

In this study, recurrent AF was strongly predictive of stroke or death in HF patients, an observation predominantly driven by increased cardiovascular mortality.⁸ This is an important result but of course begs the question: does recurrence of AF after an AF ablation in a HF population simply indicate an intrinsically higher-risk population or should increasingly aggressive attempts be made to restore sinus rhythm? These propositions may not be mutually exclusive. In this study, the mean number of procedures in the HF population was over 2 to achieve a relatively modest success rate of just over 50%. Furthermore, the major complication rate was high at 5% consisting mostly of tamponade at 3.5% and stroke/TIA at 1%, although the incidence was similarly high in the non-HF population. Significantly, these data are from the era prior to contact force catheters and uninterrupted anticoagulation both of which may have considerably reduced the complication rate. Nevertheless, the complication rate, required procedure number, and long-term efficacy will all be critical issues when physicians and healthcare providers consider the place of AF ablation in the HF population.

The authors appropriately conclude that this multicentre registry highlights the need for a randomized outcome study of ablation in the HF population. Their data along with prior studies strongly support an approach of ablation when AF is paroxysmal. However, this study demonstrates that further evidence is needed before we can routinely recommend ablation for AF in HF patients when the arrhythmia has been persistent for more than 1 year. Such studies are ongoing. Until they are available, we will necessarily continue to make individualized decisions that consider the possible positive vs. negative outcomes.

References