Atrial Fibrillation Ablation: Two Concerning Trends

John M. Mandrola, MD


October 01, 2019

Two recent trends in atrial fibrillation (AF) ablation worry me.

The Business Side

I have a Google alert for "AF ablation." The number of alerts has risen sharply in recent years. But it's not from advances in science. Rather, the alerts point to stories mostly from business analysts who tell of an expanding market for AF ablation technology.

My practice bucks this trend—I have not increased the number of AF ablations I do each year. My conservative approach stems from three lines of reasoning.

In 2015, I argued for less AF ablation after Prashanthan Sanders, MBBS, and his team in Adelaide, Australia, published a randomized controlled trial (RCT) showing that risk factor modification reduced AF episodes.[1] They have confirmed these findings in numerous observational studies.[2,3,4,5] Why destroy atrial tissue when weight loss, reduction of alcohol intake, sleep apnea therapy, and exercise could improve atrial (and overall) health?

The second reason for a conservative approach to AF ablation stems from decidedly modest data on efficacy. The pivotal CABANA study, comparing ablation vs antiarrhythmic drugs in more than 2000 patients, found no difference in hard outcomes.[6] Yes, ablation reduced AF episodes (a surrogate marker) and improved quality of life,[7] but the trial had no placebo control arm. Most sobering was that approximately half the patients in the ablation arm still had AF after 4 years.

The third reason for caution is the potential for serious harm. Most patients with AF who are referred for ablation are not dying of their arrhythmia. AF ablation is elective. Proponents of AF ablation call the procedure "safe" because of low complication rates in trials of AF ablation. They are not entirely wrong; the procedural adverse events are low—on average.

But, as mathematician Nassim Taleb teaches, one must consider not only the average risk of harm but the consequences. Despite advances in ablation technology, the risk for thermal damage to the esophagus, which can lead to the fatal condition atrial-esophageal fistula, remains.[8] Many expert ablation doctors tell stories—mostly off the record—of fatalities. Atrial-esophageal fistula occurs weeks after the procedure; thus, death from AF ablation is surely underreported.

Of course, complications are part of doing procedures. But it's one thing to have a fatal complication when doing a necessary procedure (such as inserting a pacemaker for heart block); it's yet another during an elective procedure. Consider also the notion of causing a death from a procedure that has yet to be tested in a proper sham-controlled trial.[9]

More Aggressive AF Ablation

The other worrisome trend in AF ablation is the creep of more aggressive techniques, such as the strategies of electrically isolating the posterior wall of the left atrium and/or the left atrial appendage.

Some background: The foundation of AF ablation involves electrical isolation of the pulmonary veins (PVI), accomplished by radiofrequency or cryoballoon ablation. The seminal STAR-AF II study found that even in patients with advanced AF, additional lesions beyond simple PVI did not improve outcomes.[10] Thus, if you are going to ablate AF, the best evidence supports doing PVI alone.

Yet two recent studies and comments online might tempt some to accept a more (ablation)-is-better approach.

Left Atrial Posterior Wall Isolation

Professor Andrea Natale, MD, from the Texas Cardiac Arrhythmia Institute, teaches that the pulmonary vein musculature extends posteriorly on the left atrial wall and that "empirical isolation of the left atrial posterior wall should be performed in all patients undergoing AF ablation."[11] Many centers agree and routinely add this posterior wall isolation step. My two concerns with this strategy are that the esophagus lies immediately behind—often millimeters from—the posterior wall, and no trial-level evidence shows that it improves efficacy.

Enter a recently published systematic review and meta-analysis that studied the feasibility, safety, and efficacy of posterior wall isolation (PWI) during AF ablation.[12] The lead sentence of the conclusion leaned positive: "PWI as an end point of AF ablation can be achieved in a large proportion of cases with good rates of 12-month freedom from atrial arrhythmia."

The senior author of the analysis, the same Professor Sanders who inspired me to do less ablation, wrote in a tweet that he was "increasingly convinced this [posterior wall isolation] is the way to go." Luigi Di Biase, MD, Albert Einstein College of Medicine, and others strongly concurred with this add-on strategy.

Yet the trial-level data in that systematic review do not support posterior wall ablation. Here I cite the authors' own words: "RCTs comparing posterior wall isolation to PVI were limited by small sample sizes and heterogeneous methodology and were unable to confirm any incremental benefit to PWI."

Left Atrial Appendage Isolation

The second addition to ablation stems from the thinking by some that the left atrial appendage is a frequent source of nonpulmonary vein triggers and that electrical isolation of the appendage improves ablation efficacy.[11] Findings from the BELIEF trial support this claim.[13] When it was presented in 2015, I offered five reasons to be skeptical. Meta-analyses have also been cited in support of appendage isolation, but these include observational studies; BELIEF stands as the only RCT.[14,15]

While I have heard electrophysiologists question the frequency of the left atrial appendage as a trigger of AF, all agree that isolating the appendage increases the risk for embolic stroke.[16,17] This is because the isolated appendage receives no signal to contract or eject pooled blood and the pooled blood is a serious nidus for stroke.

How best to mitigate this increased risk was the focus of a recent observational case series of more than 1800 patients who had electrical isolation of the appendage.[18]

All patients in this series had transesophageal echocardiography at 6 months to determine contractile function of the left atrial appendage. In the 336 patients who had recovered appendage function (translation: appendage isolation was not durable), oral anticoagulation (OAC) was stopped and no strokes occurred. In the 1516 patients who had impaired appendage function, the rate of stroke in those who discontinued OAC (n = 432) was 16.7% vs 1.7% in those who continued OAC (n = 1086).

Of the 90 patients in this series who had a stroke or transient ischemic attack, 84 subsequently had appendage exclusion with a Lariat device (n = 12) or occlusion with a Watchman device (n =72). Most eventually stopped OAC. After a year of follow-up, only one patient had an embolic event.

These findings led the authors to conclude that electrical isolation of the appendage is associated with a significant risk for stroke that can be effectively reduced by optimal uninterrupted OAC or left atrial appendage occlusion (LAAO) devices.

This observational nonrandomized study, which was published in an influential journal, will likely be cited as evidence for both electrical isolation of the appendage and then appendage occlusion. For the record, that's not one extra procedure but two—beyond standard PVI.

Isolating the left atrial appendage concerns me. First, as Drs Boris Schmidt and Julian Chun wrote in an accompanying editorial, isolating the left atrial appendage is a human-made risk factor for stroke.[19] Doing that routinely requires a high bar of evidence. One RCT and observational data are not enough.

Second, I would not conclude that LAAO can effectively reduce stroke risk after electrical isolation from a tiny subset (84 patients; 5%) of an observational cohort. LAAO might mitigate stroke risk after appendage isolation, but we need more data.


The evidence to support the expanding business of AF ablation is not convincing. Given the role of lifestyle in AF, the lack of hard outcome evidence in patients without heart failure, and the potential for severe harm, I see no reason to veer from a conservative approach to AF ablation.

In addition, when the procedure is performed, the evidence for destroying extra atrial tissue beyond that required to accomplish PVI is also lacking. Given the downside risk of additional atrial ablation, the burden of proof for net benefit lies with those who promote these aggressive strategies. Clearing this high bar requires more than one trial, observational data, and eminent voices.

My mind is open to change, but it needs more persuasive evidence. Until then, everything about AF ablation suggests less is definitely more.


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.