The knowledge base of AF ablation gets a jolt from the DECAAF trial

Dr John Mandrola


September 01, 2013

Things are happening in the world of atrial fibrillation ablation. Thank goodness. We've been stuck doing the same thing (PV isolation), settling for the same mediocre results, and not having answers to the same basic questions for too long. Perhaps there is an association here?

An analysis of the Delayed Enhancement-MRI Determinant of Successful Catheter Ablation of Atrial Fibrillation (DECAAF)trial, released today as a hot-line late breaking trial at the European Society of Cardiology (ESC) 2013 Congress , adds to the disruption that began with FIRM ablation .

First, the basic questions

When patients consider AF ablation, the first question they ask is fundamental: What is the success rate? Here we can quote only general statistics. Clearly, though, some patients stand better chances than others. One would think after a decade of ablating this disease, better predictors would exist. They don't.

Second, AF patients often ask me how do I know where to ablate. My answer is dated: "We electrically isolate the pulmonary veins." And then the response: "How do you know this area is the problem?" Alas, we don't know that either.

These questions are why I am drawn to research that teaches us more about the mechanisms of AF ablation. If we knew mechanisms, AF ablation would be more like SVT ablation. It would be easy!

Today, in Amsterdam, we learned a lot about AF and AF ablation.

Dr Nassir Marrouche (University of Utah) presented results from the DECAAF trial, a multicenter, observational, one-year prospective cohort study designed to assess the relationship between atrial fibrosis and AF ablation success. Today's findings add to what he presented earlier this year at the Heart Rhythm Society sessions.

This analysis included 329 patients recruited from 15 experienced AF ablation centers worldwide. Patients had delayed-enhanced MRI scans performed 30 days before the index procedure, and 177 patients had follow-up scans at 90 days postablation. Patients were typical of a general AF ablation cohort—mean age 59, and 64% had paroxysmal AF. (See this heartwire coverage for the details.)

Atrial fibrosis was graded using the four stages of the Utah classification system. Follow-up was one year. There were no requirements on ablation technique. Operators were blinded to the findings of the MRI scan.


  • Hypertension was the only predictor of atrial fibrosis. Typical risk factors (age, gender, AF type, smoking, CHADS, CHF, etc) did not predict fibrosis.

  • In multivariate analysis, the degree of atrial fibrosis was the only predictive variable.

  • There was a graded relationship between fibrosis and success: patients whose preablation fibrosis was stage 1 (defined as <10% damaged atrial tissue) had an 85.8% success rate, those with stage 2 (10%–20% damage) had a 63.3% success rate, those with stage 3 (20%–30% damage) had a 55% success rate and those with stage 4 (>30% damage) had a 31% success rate.

  • Atrial fibrosis predicted postablation symptoms: For every increased percentage of fibrosis before ablation, there was 6.3% increased risk of recurrent symptoms after ablation (hazard ratio [HR] 1.063).

Analysis of postablation MRI images is where DECAAF gets its jolt

First, the number of PVs completely encircled was sobering. (Recall these are the most respected AF centers on the planet.) Only 6.7% of patients had all four veins completely encircled, while 39% of patients had zero PVs completely encircled. (Wow.)

Second, when PV encirclement was used as a predictor of postablation recurrence, there was no relationship. This statement deserves a pause: In DECAAF, PV isolation did not matter.

What did matter was the extent of residual fibrosis left over after ablation. Residual fibrosis was quantified from the baseline MRI by subtracting ablated scar area from preablation scar. Fibrosis can be easily distinguished from ablative scar on the MRI images. The percent of residual fibrosis was split into four quartiles, and Kaplan Meier plots showed well-demarcated graded risks of recurrence.


These provocative findings allowed the researchers to conclude: 1) atrial fibrosis is a strong and independent predictor of procedural outcome, and 2) lower degrees of residual fibrosis after ablation were associated with improved outcomes.

Lead investigator Dr. Marrouche sums it up well in a quote taken from the press release:

"We do ablation around the pulmonary veins because we have assumed for years that the trigger for AF comes from the vein—that's the standard of care," he said. "But what we found in DECAAF is that ablation of the veins did not predict outcome. In fact, the most important predictor of outcome, along with stage of atrial fibrosis, was the degree of ablation of the fibrotic tissue. Rather than targeting the pulmonary veins, procedures that ablated fibrotic tissue produced better outcomes—the more that was targeted, the better the outcome."

DECAAF tempts us to individualize ablation. It moves us away from simple anatomic procedures that shoot at easy targets, like PVs.

My take home

This, my friends, is progress. DECAAF gives us an incredibly good individual predictor of success—atrial fibrosis can be easily visualized on painless scans without radiation exposure. Expense will be a barrier in the real world, but let's not spoil the party.

Most important, DECAAF sheds light on AF mechanisms.

When I spoke with Dr Sanjiv Narayan at EuroPace 2013, he encouraged me to think of ablating AF similarly to paroxysmal supraventricular tachycardia (PSVT) ablation. When we target atrioventricular node reentry tachycardia (AVNRT), for example, we ablate the slow pathway, not the premature atrial complexes (PACs). Granted, if you could eliminate every PAC, it wouldn't matter if there were two pathways. But it is far easier to ablate the slow pathway than every PAC.

It's the same thinking for AF: it is better to target the critical areas of the atria rather than PACs. DECAAF and FIRM teach us similar lessons: Get to the source, not the initiators. In FIRM, it's rotors. In DECAAF, it's scar. For sure, it's not just about PV isolation.

Yes, we need to know more. But we are getting warmer. Finally.

The day when AF ablation becomes more like SVT ablation is coming.

Is that a flickering light at the end of the tunnel?



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: