" DJ Lakkireddy: Hi, I'm DJ Lakkireddy, Executive Medical Director from the Kansas City Heart Rhythm Institute in Kansas City, Kansas. We are here at the AFib Symposium in Boston.
Pasquale Santangeli: And I'm Pasquale Santangeli, VT Director of the Cleveland Clinic. Pleased to be here.
DJ Lakkireddy: Pasquale, the world of atrial fibrillation ablation has been exploding and we have a number of tools at our disposal. We've really overcome a lot of the past complications, but now we have a new set to contend with. Can you walk us through the existing data on some of the classic complications we used to worry about with radiofrequency ablation — did we really overcome those? And what are some of the new things on the horizon that concern you and where we need more data?
Pasquale Santangeli: Yes — we learned over the years that the major issue with radiofrequency or thermal ablation is bystander injury to non-cardiac tissue. The phrenic nerve is one of them, pulmonary vein stenosis another, and of course esophageal damage, which was rare but fatal in up to 50–60% of cases. So we really feared those. The hope was that by moving to non-thermal energy — like pulsed field ablation — we would overcome all of this. But as your studies like NEMESIS-PFA showed us, new things have come up. The question is to what extent these new complications offset the old ones — is it really safer? We're seeing hemolysis, potential for phrenic nerve stunning — self-limited, but real — and this new fear around vasospasm that can be both acute and late. So we know the old devil well. How about the new devil? What do you think?
DJ Lakkireddy: I think it's actually very interesting that we traded the old devil for a new one — which I think is relatively small in scale. But in this day and age, with all the scientific advancements we have, any complication is intolerable. And with the rapid cycling of technologies and form factors — a new catheter on the horizon every single quarter — we really need to keep our antennas up and study this rigorously.
On hemolysis specifically: I think it's largely a function of catheter design and the electric field we create. Earlier generations of PFA catheters had open architecture and larger electrical fields, so the hemolysis risk was significantly higher. With the current generation of catheters — better form factors, occlusive technologies — that risk is lower. That said, any time there is a change in form factor, with around 20 variables going into the mix, we need to be very stringent about studying these things before putting them into humans and before granting approval. That's where the guardrails come in.
I worry about haemoglobinuria and pigment-related cast nephropathy, and also about long-term pigment-related interstitial nephritis. Glomerulonephritis is a problem in vulnerable populations — something we should pay close attention to and study long term.
When it comes to myocardial injury, I don't think we really understand why we see such significant troponin leaks. It's hard for me to believe that atrial injury alone accounts for so much troponin elevation. Are we impacting the adjacent ventricular myocardium? How much of this is reversible electroporation that resolves completely? What are the impacts on long-term atrial function? Our acute studies show it definitely affects acute function, but the long-term picture remains uncertain.
Then there's the elephant in the room — coronary spasm. We understand the pathophysiology of acute coronary spasm reasonably well. But the cases of sudden cardiac death — are they emerging because we're paying more attention and they're surfacing more often? Or is there a delayed impact on the coronary vasculature? Is it altered calcium handling? The way these vessels repair themselves through the reversible component of electroporation? Autonomic nervous system innervation? Or broken-down haem-related vasospastic effects from nitric oxide depletion in the vasa vasorum? There are many unknowns here, and we really require more rigorous bench work and long-term clinical follow-up. That's precisely why I think we should all participate in the Heart Rhythm Society's PFA registry platform — bringing all of these platforms onto one singular monitoring unit so we can make better sense of it all.
Pasquale Santangeli: That's a key point. Some of this will only become evident through large registries, because the event rates are so low that single-centre or even randomized studies won't have the power to detect them.
A few comments about PFA for VT specifically, which I've been using for roughly the past year. We've learned that the current waveform needs to be optimized — in other words, it's not as deep as animal studies suggested. We've also seen that when PFA is delivered close to implantable rhythm devices — pacemakers or defibrillators with exposed conductors — there have been cases of defibrillator failure, including failure to generate a pacing stimulus. The question of whether we need to replace the generator on the table is a real one. These are entirely new complications that we simply haven't encountered with thermal energy.
DJ Lakkireddy: On that point — how close does one need to be to a device to risk generator failure? Even when you're across the septum in the left ventricle with a defibrillator coil — essentially an open circuit — is there a much higher probability of that strong electrical voltage being sent back into the can and destroying it?
Pasquale Santangeli: The closer you are, the worse it is in terms of energy being transmitted back to the defibrillator. None of the device cans have been engineered to withstand this type of voltage — the only surge protection mechanism is designed for cardioversion, which delivers much lower energy. There's also a question of whether there's a meaningful difference between bipolar PFA versus monopolar delivery to a ground electrode, with the latter potentially more prone to this problem. As for how close is too close — we're still figuring that out. In our series, which we hope to present at HRS, the distance was 7.5 millimetres — not trivial, but more than the standard 5 millimetres we typically consider safe.
DJ Lakkireddy: You've done a lot of pioneering work in the PFA-VT space. There's a lot of excitement about using PFA in the ventricular myocardium, and there's also growing conversation about using a combination of RF and PFA sequencing. Can you tell us more about how that works and how it relates to minimizing complications?
Pasquale Santangeli: A couple of issues with PFA in this context. The effect on electrograms or arrhythmia termination can be very non-specific — you may be right on top of the VT circuit or adjacent to it and still see termination. Radiofrequency has the benefit of giving you a more definitive answer as to whether you're truly at the critical site. That's point one.
Point two: in some areas, because of the lateral spread of the field, the post-ablation effect is less controllable. When ablating the border zone in particular, you may create myocardial stunning — which in the ventricle means reduction in ejection fraction. And when you're close to the conduction system, you risk inadvertently damaging it. In patients without devices or with low EF, taking out the left bundle causes desynchrony, which is hemodynamically harmful. So there's a lot to learn. I think dual energy remains a good approach for these cases until we better understand how to titrate PFA in specific areas.
DJ Lakkireddy: Knowing what we know, and given the race towards bigger, deeper, higher electrical delivery to the target tissue — do we really need that much power?
Pasquale Santangeli: That's a great question. There's a push from both industry and many physicians to go transmural. But the reality is that for many years we were nowhere near transmural and still achieved very good outcomes in most patients. The early surgical sub-endocardial resection series only removed 2 or 3 millimetres of sub-endocardium and got the job done. Being deeper matters for a subset of patients — but not for all of them. For the large majority, having something that achieves similar results to radiofrequency but much faster, with a quicker and larger footprint — cutting procedure time by perhaps 50% — would be genuinely beneficial. Transmural lesions should be reserved for specific cases, not applied universally.
DJ Lakkireddy: Excellent. Thank you so much.
Pasquale Santangeli: Thank you. Thanks for having me.