Brain Safety After Pulsed Field Ablation for Atrial Fibrillation

The Study of the Week reports on a study of brain safety during pulsed field ablation

The richer a society gets the higher the rates of atrial fibrillation. Obesity, lack of exercise, alcohol use and advancing age are the main drivers of AF.

The AF-treatment market size in the US is measured in the billions of dollars. It’s expected to grow at high rates, powered largely by expansion of catheter ablation.

The new technology in ablation of AF is something called pulsed field ablation or PFA. PFA destroys myocardium with electricity (shocks). Every “application” is actually a shock.

It creates pores in heart cells—and they die. Before PFA, most centers used thermal ablation (radiofrequency (heat) or cryoballoon (cold) to destroy atrial myocardial cells.

PFA has become quite popular. There are purported advantages. The main one being its cardioselectivity.

PFA is supposed to work only on voltage-gated myocardial cells. That means the dreaded complication of thermal damage to the esophagus can be avoided. This is not a small thing. While atria-esophageal fistula is super rare (≈1 in 10,000), it is often fatal. I ablate AF and potential thermal damage to the esophagus keeps me up at night. Fortunately, we now know about the complication, and have active avoidance measures.

Another potential advantage of PFA is that it can be faster than standard thermal ablation. Europeans do PFA ablation in well under an hour. Americans can’t seem to replicate that. Nonetheless, most of my local colleagues and I do thermal ablation in about 60-80 minutes. It’s hard to be much faster than that.

The Neutral Martian Take of PFA Evidence

Neutral martians are not influenced by marketing. Neutral martians don’t gain status by being early adopters. Neutral martians look only at data.

FDA required a comparison trial between PFA and thermal ablation. The ADVENT trial randomized ≈600 patients to either of the two strategies.

Here were the efficacy results for elimination of AF.

You can see there are no differences. Many observational studies have shown similar results. Everyone agrees that PFA offers no efficacy advantage. One caveat is that proponents have some hope that the PFA’s ability to more safely destroy myocardium will lead to better results in more advanced cases of AF. Emphasis here on hope.

ADVENT also recorded safety events, though trials are not often the best way to measure safety—because of the low number of randomized patients.

6 of 305 patients in the PFA arm vs 4 of 302 patients in the thermal arm had serious safety events. Two patients in the PFA arm had cardiac tamponade (due to perforation of the heart) vs 0 in the thermal arm. One of these patients died. (Death is very uncommon in AF ablation. The authors report that it was not specifically due to the PFA.)

One surprising finding was that 4 patients had transient phrenic nerve injury in the PFA arm. It’s surprising because PFA is not supposed to effect non-cardiac cells.

We really can’t say much about safety when only 600 patients are randomized.

Dr Vivek Reddy presented the results of a 17,000-patient registry study last year. The MANIFEST-17k registry revealed quite standard AF-ablation complications. Stroke occurred in 7 patients (0.12%). They observed no esophageal or phrenic nerve events.

MANIFEST did show a new and unexpected complication: 5 patients had hemolysis and kidney failure. Hemolysis is busting up of red blood cells. Subsequent studies have confirmed that the number of PFA lesions correlate with the risk of hemolysis.

The Newest Study on Brain Safety

The journal Heart Rhythm published the results of 77 of 607 patients in the ADVENT trial who had brain MRIs after the procedures.

Quick background: It’s sobering but the fact remains that any time a doctor moves catheters in the left side (arterial) of the heart for ablation or stents or valve procedures, debris can travel north to the brain and cause white spots on MRI. Thus far, we have not noted long-term clinical impact in things like cognition.

PFA is especially concerning because the shocks in the left atrium create intense microbubbles. Proponents wrote back in 2018 that longer-term safety studies are needed that specifically assess…”cerebral microemboli assessed by magnetic resonance imaging.”

Yet only 13% of the regulatory trial (34 PFA and 37 thermal) had MRI assessment. No patient in the thermal group had an abnormal MRI. Three or 10% in the PFA group had lesions on the MRI.

The paper is open access so I took two screenshots. It’s a bit sobering.

Note the last phrase in patient 3: “newly detected 12 mm subclinical lesion in the left temporal lobe.” Here is the MRI.

Neurologic assessments of these patients revealed normal findings. We would call these silent brain lesions.

Since the overall findings were only 3 vs 0, statistical assessment was not possible. Notable also that in the main paper, there was one TIA in the PFA arm and one stroke in the thermal arm but neither of these were in the MRI subgroup of the study.

Comments

There isn’t enough signal to declare definitive knowledge.

There is enough, though, to worry. The authors cite a few other studies of MRI after ablation. One single-center study included only 30 patients, one of which had a silent abnormality. Another study of 121 patients who had PFA, only 18 had MRI scans after the procedure, two of which had lesions.

I reiterate that any left-sided cardiac procedures can cause these sorts of lesions. But. The bubbles seen on ultrasound imaging during PFA was enough for the proponents to write in 2018 that brain safety was an issue to be studied.

As of 2024, I am not sure this signal has been adequately studied.

The reason I led with the size of the AF ablation market is that millions of patients will have AF ablation. Even a small signal of brain damage after the procedure could be important. Many of the patients getting AF ablation are middle-aged.

What’s more, the lack of esophageal damage has encouraged PFA operators to deliver more (not less) ablation. That may increase the risk of brain lesions.

If I were a regulator, I would want more brain safety data. It’s likely ok, but we don’t know. And a lot is at stake.