In the Secondary Analysis of FIRE and ICE, Focus on the Results, Not the Discussion

John Mandrola, MD


June 12, 2016

In presenting secondary analyses from the FIRE and ICE trial, Prof Karl-Heinz Kuck unequivocally delivered the biggest news from the European Heart Rhythm Association (EHRA) EUROPACE-CARDIOSTIM 2016 meeting. The room was packed; his words forceful; the applause strong; the debate spirited.

The total time Kuck spent at the podium was 28 minutes—13 minutes to present the data and 15 minutes to promote his view of the data.

I rarely do this, but I will begin with quotes from Prof Kuck. As you read his words, which I recorded, remember three things:

  • The primary results of the FIRE and ICE trial showed no difference in efficacy or safety between cryoballoon and RF ablation, including predefined-subgroup analyses of the composite primary efficacy end point[1].

  • Medtronic sponsored the trial.

  • Millions of patients worldwide have AF.

Here are the quotes:

"These guys were born to do RF ablation—their character, their personality, everything. These are the best RF ablation guys in the world. If there was a bias in the trial, it was in favor of RF."

"In the setting of a multicenter randomized controlled trial [RCT], you can't do better than they did [in RF ablation]."

"I truly believe these data will change our strategy in Hamburg."

"Does time to first AF recurrence tell us the whole story for patients? It does not."

"Not only did the cryoballoon-ablation strategy lead to less AF burden, it has a shorter procedure time than RF. That means we can add another patient to the schedule that day." (Author's note: AF burden was measured indirectly by the secondary end points mentioned below.)

"I truly believe in 5 years' time there's no more point-to-point RF ablation to isolate the pulmonary veins."

"Everything will be done with balloon technology. I'm not saying everything will be done with a cryoballoon technology, but I truly believe everything will be done with balloon technology."

"I have no question about it. It [cryoballoon] is more practical, shorter, easier, and there is no reason to continue to do a procedure [RF ablation]."

"Even in the hands of experienced operators, the follow-up is significantly better, the clinical benefit better, the relative event reduction is 25% to 50%. There is a clear trend, and it's one direction, and it's highly significant in favor of the cryoballoon ."

In the news story on heartwire from Medscape, Kuck said:

"The PVI-balloon–based procedure is such a safe procedure that I do believe that a very experienced interventional guy who is well trained in an [electrophysiology] EP lab can do it."

The Results

Now that you have heard one eminent professor's beliefs, here are the actual results of a secondary analyses of an industry-sponsored noninferiority trial that showed equivalence of the tested strategies.

Outcomes of FIRE and ICE
Outcome Cryoballoon, N=374, n (%) RF, N=376, n (%)
Absolute difference (%) P Prespecified
Rehospitalization for CV cause 89 (23.8) 135 (35.9) 12.1 <0.01 Yes
Rehospitalization for any cause 122 (32.6) 156 (41.5) 8.9 <0.01 No
Cardioversion 12 (3.2) 24 (6.4) 3.2 0.04 No
Repeat ablation 44 (11.8) 70 (17.6) 5.8 0.03 Yes

In a subgroup analysis of the secondary end point cardiovascular rehospitalization, cryoballoon performed slightly better in some subgroups.

Mental and physical quality of life, assessed at baseline and every 6 months after ablation, improved equally in both groups at 6 months and was maintained throughout 30 months of follow-up.


FIRE and ICE is an exemplary trial. This is the way we should be resolving clinical questions. It's multicenter and randomized and included blinded-follow-up.

In the Q & A after his presentation, Kuck responded to a criticism from a physician who said RF ablation has improved since the initiation of FIRE and ICE. "At the time of an RCT you have to make a decision to use the technology available. This is a limitation of an RCT; you have to stick to a protocol; you have to accept limitations; if you don't accept limitations, you will never do an RCT. It's better to do a limited RCT than no RCT," Kuck said.

He is correct. And the AF-care community should pay close attention to this comment. We have ablated AF for more than 15 years and have precious little RCT evidence for our strategies. Consider that FIRE and ICE randomized fewer than 800 patients; and it's the largest RCT yet in AF ablation. How is that possible, given the hundreds of thousands of procedures done worldwide?

That said, I interpret the results much differently. My take aligns well with session cochair Dr Helmut Puererfellner (Elisabethinen Hospital, Linz, Austria), who said, "These data do not give a definite answer whether cryo is better because the primary result still remains the primary result; that it was noninferior."

It's not problematic to look at secondary end points and subgroups. The problem comes in how one interprets and translates them to practice. Remember, these are secondary end points and subgroups of a trial that showed equivalence of the two strategies in efficacy and safety. Two of the four main findings Kuck showed were not even prespecified secondary end points. One of his visually strongest slides was a subgroup analysis of a secondary end point. I'm no statistician, but that seems like a stretch.

One thing these data do suggest is the need for knowing one's results. I've written previously about a paper that found a center's experience affects the results of AF ablation done by either cryoballoon or RF[2]. Namely, centers that had extensive experience with RF performed similarly with either procedure, whereas centers with less experience in RF did better with the cryoballoon. Part of doing AF ablation in 2016 should be a mandate to know your data.

I also object to Kuck's idea that an experienced interventional cardiologist "well-trained in EP" could use cryoballoon ablation. I can't vouch for the European situation, but in the US, a shortage of people capable of doing AF ablation is hardly a public-health problem.

My strongest dissent to Kuck's comments is his implication that being able to add another AF ablation on the daily schedule is a good thing. We mustn't forget that these costly procedures—whether done with freezing or burning—fail in more than one in three patients. If you extend follow-up to 5 years or more, failure rates increase. And, crucially, we have zero evidence these procedures lower the rate of stroke or death.

Ablation may have a role in some patients with atrial fibrillation, but the notion that treating AF by adding scar to the atria belies the growing understanding that AF is less a disease than a manifestation of inflamed, stretched, and scarred atria. The key to making a difference for patients with AF is unlikely to be found in an EP lab.



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.
Post as: