CASTLE-AF: Ablate Patients With HF Before It's too Late?

An Interview With Nassir Marrouche, MD

John M. Mandrola, MD; Nassir F. Marrouche, MD


September 25, 2017

John M. Mandrola, MD: Hi, everyone. This is John Mandrola from and Medscape Cardiology. I'm here in Barcelona, Spain, at the European Society of Cardiology 2017 Congress, and I'm pleased to be with my friend, Nassir Marrouche, primary investigator of the CASTLE-AF trial, which was just announced at the late-breaking clinical trials sessions. Welcome, Nassir.

Nassir F. Marrouche, MD: Thank you for hosting me, John.

Dr Mandrola: Tell us about the background of CASTLE-AF.[1]

Heart Failure and Atrial Fibrillation: Chicken and Egg

Dr Marrouche: CASTLE-AF started about 9 years ago. As you know, heart failure (HF) and atrial fibrillation (AF) are very well intertwined. We deal with these patients every day in our practice, and 30% or more of HF patients suffer from AF.

In 2005 and 2006, we came up with an idea, and we put a protocol together to study the effectiveness of catheter ablation on primary heart endpoints. We ablate AF almost every day, but we still don't, until today, know whether AF affects endpoints.

We put this study together to answer the question: Does AF ablation in a specific population—patients with HF, as we studied in CASTLE-AF—affect common hard endpoints of mortality and worsening HF admissions? We compared our data with a conventional guidelines-based approach for treating these patients, and today we presented the data.

Dr Mandrola: Mortality and HF hospitalizations were outcomes, not AF?

Dr Marrouche: Yes, those were being studied for the first time. Aren't you sick of looking at AF and atrial flutter as endpoints?

Dr Mandrola: Tell us about the patient population. These were patients with defibrillators?

Dr Marrouche: These were patients with defibrillators. We took a Sudden Cardiac Death in Heart Failure Trial (SCD HeFT)-like population who had an implanted implantable cardioverter defibrillator (ICD) or a biventricular ICD for primary or secondary prevention. Patients had paroxysmal persistent AF, a longstanding persistent AF, particularly any kind of AF. An SCD-HeFT population means that ejection fraction (EF) is ≤35%, and New York Heart Association (NYHA) class is ≥II. Patients were randomly assigned to conventional treatment versus ablation.

Dr Mandrola: And the ablation technique?

Dr Marrouche: As you know, we are still debating what the best ablation techniques are. But fortunately for CASTLE, we got buy-in by everybody in the study. One of the requirements to be in the study—there were over 31 sites—was that you had to have already done 50 ablations yourself over a period of a year.

We mandated pulmonary vein isolation. Anything else was left at the discretion of the investigators. Repeat ablation was allowed as well after a blind period of 3 months. In fact, we had 1.6 ablations per patient. Actually, 38% of the patients had a second ablation.

Study Results

Dr Mandrola: AF ablation in HF patients versus conventional therapy: What were the results? What happened?

Catheter ablation helps us keep patients with HF out of the hospital.

Dr Marrouche: We showed, for the first time, an improvement in the primary composite endpoint of all-cause mortality plus worsening HF admissions. Catheter ablation improved the primary composite endpoint (38% relative risk reduction), all-cause mortality (47% relative risk reduction), and worsening HF admissions (44% relative risk reduction). All of them were highly significant.

Dr Mandrola: If the composite was mortality and HF hospitalizations, was it driven by both?

Dr Marrouche: Exactly. That is what we saw. We thought we would find it driven by only one of them, but both of them are significant.

Dr Mandrola: What happened with the AF burden? These devices have AF monitors, correct?

Dr Marrouche: [In the past,] we never got to know what happens to our patients. We do Holter monitoring once every 3 months, and we get excited about seeing an episode in 7 days. Now we have continuous data over 7 years in some patients. It is amazing. We are so excited about that. But for this study, obviously we started looking at the data as we went. For the primary endpoint, AF ablation did cut AF burden by almost half at the close of the study.

At the end of the study, 63% of catheter ablation patients were in sinus rhythm. The device assured us that the burden was cut by almost half, so AF burden was around 25% to 30% at a follow-up of about 5 years.

Dr Mandrola: Decreased AF burden, improved mortality, improved HF hospitalizations: What did you conclude from this?

Dr Marrouche: Catheter ablation helps us keep patients with HF out of hospital. Of note, for the first time, catheter ablation was shown to reduce mortality and hard primary endpoints.

Clinical Implications of CASTLE-AF

Dr Mandrola: It sounds pretty big and new. What do you think the clinical implications of this are?

It seems that intervention sooner is better and before HF is well-progressed.

Dr Marrouche: We should ablate everybody; do you like that [joking]? First of all, this is a selected patient population. I showed the forest plot today, and almost every patient in this subgroup profits from ablation except patients with advanced disease, meaning left ventricular dysfunction (EF <20%-25%) and NYHA class III or IV. I cannot make a conclusion on class III, but the number is low as well for class IV. It seems that intervention sooner is better and before HF is well-progressed.

Dr Mandrola: That is an important point. You found that if the patient gets too sick with advanced HF class or depressed EF, ablation is probably not a benefit.

Dr Marrouche: That is what the data are showing so far from subanalyses, but as you know, a subanalysis is not a conclusion. After the data we showed today, I would recommend that you should try to take the patient to the lab as early as possible. And what's interesting as well, and confirms what other studies showed, is that if you look at the subanalyses, patients in the ablation arm—looking at an interactive analysis between the subgroups who did not take amiodarone—did better than patients who took amiodarone. This has been a dilemma for years; it never showed superiority. This was confirmed as well.

People who do well—I do not want to make that a firm conclusion—are people who are ablated early and not taking antiarrhythmics. Ablate the area during earlier disease stages of HF.

Dr Mandrola: Congratulations on this very important trial. We look forward to the publication, and thanks for being with us.

Dr Marrouche: Thank you for inviting me, John.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.