Should We Refer Symptomatic HF Patients for AF Ablation?

September 7, 2017

That depends, as suggested by results of the CASTLE-AF study, presented at ESC Congress 2017, in Barcelona, Spain.

Barcelona, Spain saw the convergence of world leaders and thinkers in cardiology to discuss the latest topics in contemporary cardiovascular care. There were many studies addressing atrial fibrillation (AF) and some potential “game-changers” made their debut as well.

Among them was CASTLE-AF, presented at the hotline sessions. The study garnered some attention as the only randomized controlled trial of catheter ablation vs pharmacologic therapy in patients with heart failure (HF) and AF that actually found a benefit in terms of “hard” outcomes (eg, death, heart failure hospitalization).

CASTLE-AF is the only RCT of catheter ablation vs drug therapy in patients with AF and HF to find benefit as measured by hard endpoints.

AF and HF are common co-morbid conditions and exacerbation of one can often worsen the other. Physicians struggle with how to best reduce the burden of AF – with either catheter ablation or pharmacologic therapy. This randomized trial, although small (N=397), had important implications and took over 9 years from enrollment to completion and screened ~3000 patients. The study enrolled patients with symptomatic paroxysmal or persistent AF, LVEF ≤35%, ICD with home monitoring capacity, NYHA Class 2 or worse, and failure of or intolerance to/reluctance to take anti-arrhythmic medication. Subjects were randomized to either radiofrequency catheter ablation (RF-ablation) with pulmonary vein isolation or conventional AF medical therapy after a run-in period.

The trial was positive based on both its primary (composite of all-cause mortality and heart failure hospitalization) and secondary (individual endpoints of all cause mortality, heart failure hospitalization, cardiovascular death, cardiovascular hospitalization) outcomes, as displayed in the Table.  In addition to these hard outcomes, there was also a significant reduction in AF burden and an improvement in LVEF of 8% in the ablation arm.

Table.

Ablation Group

Anti-arrhythmic (control) Group n=184

Primary endpoint (composite of all-cause mortality, HF hospitalization)28.5%44.6%0.62 (0.43-0.87), p=0.007

All-cause mortality

(K-M curves started separating at 3 yrs)13.4%25%0.53 (0.32-0.86), p=0.011

HF hospitalization

(K-M curves started separating at 6 mo)20.7%35.9%0.56 (0.37-0.83), p=0.0004


Adverse outcomes were mostly procedure-related and mostly confined to the ablation arm and included pericardial effusion (n=3), severe acute bleeding (n=3), pulmonary vein stenosis (n=1), groin infection (n=1), worsening heart failure (n=1), stroke/TIA (n=7 in ablation arm vs 12 in the conventional arm)

Next: Results similar to study of ablation vs amiodarone

Results similar to AATAC trial

Results are somewhat similar to the previously published AATAC trial (ablation vs amiodarone), which had a similar population (although LVEF criteria was <40%) and 203 patients. The study is one of the many to date that is providing a growing body of evidence that AF ablation may be preferable to anti-arrhythmic medications for HF patients with symptoms from their AF.

Prior to this one, the majority of AF ablation studies have focused on AF burden and not as much on hard outcomes. Although CASTLE-AF has important implications and has been well-received by the cardiology community at large, some express caution. In an online review of the study, clinical electrophysiologist John Mandrola, MD, of Baptist Medical Associates in Louisville, Kentucky, points out that these results may not be applicable clinically for a variety of reasons:

  ► Young study population (mean age 64 years)

  ► Highly selective group (~3000 patients were screened to get enough for the study and only LVEF ≤35%, mostly NYHA Class II)

  ► RRR is much larger than expected with ablation, suggesting that the difference in event rates may have been spuriously inflated due to worse outcomes in conventional therapy arm rather than better performance of the ablation arm.

  ► Procedural outcomes, although small in number, are also a concern

There are outstanding questions to be answered as well (pending larger studies) such as whether there was any difference in ischemic vs non-ischemic HF.

So, perhaps ablation is not quite ready for prime-time for all types of symptomatic HF patients (ie, HFPEF, mildly reduced LVEF, elderly, those without ICD/CRT, NYHA Class I, etc). But, one thing is for sure… we are slowly pushing the envelope to make ablation an important therapy with potential to improve outcomes by decreasing AF burden and avoiding the toxicity of anti-arrhythmic medications.

References: