Subscribe
Share
Share
Embed
Monthly clinical deep analysis in cardiac electrophysiology: AF, VT, SVT, ablation, devices, antiarrhythmic drugs, and high-impact trials. The Signal: physician-level analysis that identifies what matters in EP and translates evidence into clinical practice.
EP Edge™: The Signal is the flagship monthly podcast from EP Edge, delivering structured, expert-level interpretation for electrophysiologists, cardiologists, fellows, and clinically engaged practitioners. Each episode goes beyond summaries to integrate evidence across trials, guidelines, mechanisms, and real-world practice.
Episodes cover the full spectrum of electrophysiology, including atrial and ventricular arrhythmias, supraventricular tachycardias, antiarrhythmic pharmacology, pacing and defibrillator strategies, mapping and ablation technologies (including pulsed field ablation), and emerging data shaping clinical decision-making. This is not a news recap—it is a curated synthesis focused on what truly changes practice.
Content combines mechanistic insight, cross-trial evidence review, critical appraisal of methodology and outcomes, and practical application in the EP lab and clinic.
EP Edge™: The Signal complements the weekly EP Edge™: Journal Watch by providing deeper analysis and clinical synthesis. For patient-focused education, explore EP Edge™: Heart Talk.
Available as both podcast and newsletter via EP Edge on LinkedIn and Substack: https://epedge.substack.com/
Welcome back to the EP Edge podcast. I am Doctor. Sharma and this is part three of our Pulse Field Ablation series. If you're just joining, here's the context. In part one, we went under the hood, engineering and biophysics.
Niraj Sharma:What electroporation really means, why waveform matters, why catheter geometry matters, and how you can have tissue selectivity without heat being the primary weapon. Then in part two we talked pivotal trials and the clinical narrative that grew out of them. Safety looked strong, especially for the classic thermal complications, and procedural workflows got faster. But efficacy in a lot of datasets looked more like comparable rather than clearly superior. So what's left for part three?
Niraj Sharma:Here's the honest answer: once a technology scales, the real safety profile stops being defined by what trials were designed to capture and starts being defined by what the real world is willing to report, investigate, and publish. Today, we're going to focus on a set of non randomized but very important studies: case series, mechanistic work, and technology comparisons that point to something I'll call a PFA complication phenotype. Let me frame the thesis in plain language. The early safety story for pulse field ablation was built on trials and registries that, very reasonably, prioritized the complications we feared most from thermal energy: esophageal fistula, pulmonary vein stenosis, phrenic paralysis, stroke. But the newer literature suggests something subtly different.
Niraj Sharma:With pulse field energy, you can introduce complications that are: one) waveform and catheter dependent two) often mediated by blood electric field interactions stimulation three. Sometimes delayed. And that last part, delayed, is the one that should make all of us pause. Because the case ended well and the patient went home well are not always the same thing as the physiology is done reacting. Think of it like this: with thermal injury you're often watching for structural collateral damage: burns, swelling, perforation.
Niraj Sharma:With PFA, some of what we're seeing looks more like physiologic destabilization: vascular tone changes, autonomic swings, and blood interaction effects that can show up after you've already moved on. The uncomfortable implication is this: procedural success and the absence of classic thermal injury do not automatically guarantee safety, especially with higher lesion burden, posterior wall, CTI, extra lines, and certain catheter geometries. I am going to walk through the reported complication categories in a practical order: life threatening ischemia and malignant arrhythmias, coronary spasm, hemolysis and kidney injury, silent cerebral lesions, then phrenic issues, esophageal temperature signals, airway complications, and finally troponin and atrial stunning. Let's start with the highest acuity dataset in this packet, an international multicenter case series describing life threatening delayed myocardial ischemia and malignant arrhythmias after PFA. The denominator matters across 6,721 consecutive PFA procedures, life threatening adverse events occurred in zero point one six percent, so yes, rare, but the numerator matters too because the events are severe.
Niraj Sharma:There were eleven patients, mean age (mid-60s), and a heavy representation of persistent AF and coronary disease. The authors categorized events into three buckets: delayed ST segment elevation, malignant arrhythmias like ventricular fibrillation or profound bradyarrhythmias and sudden cardiac death. Two sudden death cases in that report are worth describing. This is about patent recognition. In one case, a sixty one year old man, paroxysmal AF, prior LAD stent, He died twenty two days after the last pulse field delivery.
Niraj Sharma:Autopsy did not show acute myocardial infarction, pericardial effusion, pulmonary embolism, or bleeding. An implanted monitor captured a rhythm sequence of AF transitioning to VF, that's a chilling sequence because it does not require a visible structural complication to end badly. Another case, a 68 year old man with persistent AF. Three days after PFA, he had out of hospital arrest with pulseless electrical activity. He required CPR and ECMO.
Niraj Sharma:Angiography reportedly did not show obstructive coronary disease. He died. Now, separate from sudden death, the delayed ST elevation phenotype is also striking. Five patients had delayed ST elevation at a median of about twenty minutes after the last PFA delivery, with a wide range minutes to hours. The table includes scenarios like inferior ST elevation on telemetry, complete heart block, cardiogenic shock, and coronary spasm responsive to nitrates, And importantly, at least one delayed ST elevation case progressed to ventricular fibrillation and death.
Niraj Sharma:Management in these cases was all hands on deck, EP, and critical care medicine. Intracoronary nitrates, temporary pacing, CPR, and defibrillation emergent angiography. The mechanistic discussion in this series is careful. They're not claiming causality for every late event. But they do propose a multi mechanism frame: coronary vasomotor effects, spasm electroporation induced ionic imbalance, especially calcium handling, predisposing to ventricular arrhythmias, and autonomic perturbation, vagal extremes, and bradyarrhythmias.
Niraj Sharma:Here is the practical point. This paper is essentially warning us that ventricular fibrillation may occur with or without ST elevation and that some catastrophic events can present minutes to weeks after pulse field delivery. That has implications for same day discharge pathways, because the observation window we choose may not map perfectly to the biology. Okay, let's zoom in on coronary spasm spasm because it shows up as a recurring theme across datasets, and because it's one of those complications that can look like a classic STEMI until you realize it isn't. First case: a 53 year old man with persistent AF underwent PFA using a pentaspline system.
Niraj Sharma:The lesion set was extensive, dozens of pulmonary vein applications plus posterior wall applications. Forty five minutes after the procedure, he deteriorated chest pain, hypoxia, hypotension, bradycardia. The ECG showed inferior ST elevation and complete heart block. He progressed to pulseless electrical activity arrest and mechanical CPR was started. In the cath lab, angiography initially looked like complete RCA occlusion, so the team faced the most stressful kind of uncertainty.
Niraj Sharma:Is this thrombosis? Is this spasm? Is it dissection? What is happening? They gave intracoronary nitrate and manipulated wires and balloons under CPR, and then the vessel reappeared as a diffusely narrowed RCA, (about two millimeters in caliber) with restored flow and then re expanded to roughly four millimeters.
Niraj Sharma:He went to the ICU, underwent targeted temperature management, and ultimately did well, discharged home. The case also noted lab signals consistent with some intravascular hemolysis, hemoglobin drop, and bilirubin rise. Tying this spasm narrative to the blood interaction narrative we'll cover in the next segment. Second case, a 77 year old woman, persistent AF, treated with a circular array PFA system. Posterior wall PFA was performed and CT IRF was also performed.
Niraj Sharma:About seventy eight minutes after PFA, she arrested. She achieved ROSC with CPR and epinephrine. ECG showed inferior ST elevation. Angiography revealed diffuse severe spasm in the RCA and diffuse moderate spasm in the left coronary system, both improving after intracoronary nitrates. What's interesting is that the authors also did a CT distance analysis showing large separation between pulmonary vein ostia and the coronaries, arguing against a simple proximity only model.
Niraj Sharma:They emphasized autonomic dysfunction as a plausible mechanism, particularly with the posterior wall serving as autonomic substrate. Now here's where it gets even more practical: a prospective single center technology comparison during CTI ablation. Fifty one patients underwent CTI pulse field ablation using either a pentaspline catheter or a circular array catheter. The incidence of RCA spasm was dramatically different, about ten percent with the circular array versus fifty percent with the pentasplain catheter. The authors propose a mechanistic explanation that sounds very Part one: Higher voltage, roughly 2,000 volts versus 1,500 volts, and a broader electric field, plus electrode configuration that may place more field exposure adjacent to the RCA depending on how the catheter sits.
Niraj Sharma:They also describe nitrate strategies, none, IV, or intracoronary, and how that intersects with spasm severity. So if you take the catheter comparison data and then you read the multi center life threatening case series, it starts to look like the catastrophic tail might be an extreme expression of the same physiology: delayed ST elevations, diffuse spasm, ventricular fibrillation shock, CPR, sometimes ECMO and sometimes death. Next, hemolysis. This is one of those topics that sounds labby until you realize it can translate into real clinical management questions. So let's slow down and define it.
Niraj Sharma:Hemolysis means red blood cells are being disrupted in the bloodstream. When that happens you see lab signatures, haptoglobin drops, LDH rises, bilirubin can rise, and free hemoglobin can stress the kidneys. This is why urine color, hydration status, and post procedure creatinine checks can matter in certain patients. The largest real world technology comparison in this section included five fifty two PFA procedures with systematic hemolysis labs from baseline to post op day one. The headline finding is almost provocative.
Niraj Sharma:Using their definition, haptoglobin drop greater than 10 mgdL, hemolysis occurred in ninety five percent overall. But the more important message is that the degree of hemolysis was catheter dependent. By catheter, the hemolysis signal was: about 88 with a focal lattice tip system, ninety seven percent with one pentasplin system, and one hundred percent with a circular array system, again using their lab definition. They also stratified significant and severe hemolysis and that clustering by system was even more striking. For example, one pentasplene platform had roughly forty six percent significant and twenty one percent severe hemolysis, while the focal lattice tip system was far lower, single digits significant, and essentially zero severe in that dataset.
Niraj Sharma:And here is the Part I linkage. They observed a clear dose response. The more applications you deliver, the more haptoglobin drops, linearly. But the slope of that relationship differs a lot across systems. So this is not PFA causes hemolysis in a generic way.
Niraj Sharma:It's a given waveform, plus a given electrodegeometry, plus a given number of applications produces a predictable blood interaction footprint. Now kidney injury. In that same cohort, after exclusions, about four point four percent developed acute kidney injury within twenty four hours by KDIGO criteria. No one required dialysis but some needed prolonged hospitalization for volume optimization and monitoring. The analysis noted AKI risk concentrated with more intense hemolysis phenotypes.
Niraj Sharma:The teaching point is simple but important. When creatinine bumps after PFA, the hypothesis is pigment related injury and vasomotor effects after intravascular hemolysis. There's also a prospective registry comparison looking at focal versus circular systems. In that data set, haptoglobin decreased more with circular systems than with the focal lattice tip approach and per application decline differed strongly, again reinforcing that geometry and waveform matter. Notably, that smaller cohort did not report hemolysis related clinical complications, which is a reminder that lab signals and clinical outcomes are related but not identical.
Niraj Sharma:Finally, a mechanistic review in this section tries to connect the dots using the concept of microbubbles. The idea is that high voltage pulse fields can generate microbubble formation through electrolysis or cavitation like effects depending on waveform and electrodegeometry. Micro bubbles can amplify hemolysis by increasing shear stress, disrupting red cell membranes, and creating local hotspots in the electric field. If you want one practical sentence, the more energy you deliver into the blood pool and the more your catheter design encourages blood pool exposure, the more hemolysis you should expect and that may have downstream implications for renal monitoring in selected patients. Now let's talk about silent cerebral lesions detected on diffusion MRI.
Niraj Sharma:This is always a tricky topic because it sits at the intersection of imaging sensitivity, anticoagulation strategy, procedural technique, and what we choose to call clinically meaningful. The key dataset here is a catheter shaped comparison with diffusion MRI assessment. There were 16 patients and diffusion MRI detected silent cerebral lesions in half (eight out of sixteen) and when lesions were present they were multiple. The median was about 7.5 lesions with a wide range. No clinical neurologic symptoms were reported.
Niraj Sharma:Here is the geometry signal. Silent lesion incidence was much higher with a variable loop circular catheter (six of seven patients) compared with a circular multi electrode array catheter (two of nine). They also note differing ACT targets between groups, which is a confounder worth remembering. The study makes a larger conceptual point. There may be a gap between microembolic signals, lesion detection on MRI, and what translates into symptoms, depending on technique, anticoagulation, and monitoring.
Niraj Sharma:So what do we do with this? Don't interpret no clinical stroke as no cerebral footprint, and don't interpret PFA as one monolithic thing. Catheter geometry may matter. Three quicker but important topics to bundle here: phrenic nerve injury, esophageal temperature rises, and an airway event. First, phrenic nerve injury.
Niraj Sharma:A prospective study used sequential compound motor action potential monitoring during PFA in sixty four consecutive AF patients. They reported phrenic nerve injury in forty point six percent. Now that number is attention grabbing, but here's the nuance: the phenotype was often incomplete. Not classic paralysis, you can't miss. At the end of procedure, about nineteen percent had incomplete recovery.
Niraj Sharma:In a later subset with discharge fluoroscopy, incomplete dysfunction persisted in about a quarter at discharge, and by three months most recovered, but one patient still had dysfunction. The translation is not automatically phrenic proof. If you aren't monitoring objectively, you may undercount partial injuries. Second, the esophagus. One prospective evaluation with luminal esophageal temperature monitoring found temperature elevations to or higher in 58%.
Niraj Sharma:The mean maximal temperature was around 40 degrees, with a maximum around 42.5. The key spatial finding was that rises occurred when energy delivery was within a few millimeters of the temperature probe. In the small subset that crossed 42 degrees and underwent endoscopy within forty eight hours, they did not see ulcerations or transmural lesions, and there were no fistulas during short follow-up. Now I want to pause on that because this is exactly where PFA can fool us if we are not careful. Nonthermal doesn't mean the esophagus is irrelevant, it means the mechanism is different.
Niraj Sharma:So you can still see luminal esophageal temperature signals, especially when your ablation is essentially right on top of the probe. In that prospective luminal esophageal temperature experience, temperature rises showed up when ablation was within just a few millimeters of the probe. And that's a practical lesson, not a theoretical one. If you're close enough, you can heat whether the primary lesion mechanism is electroporation or not. A second study compared luminal temperature kinetics across multiple PFA platforms and RF strategies.
Niraj Sharma:And the takeaway is not that PFA burns the esophagus the way RF can. The takeaway is more subtle. Non thermal does not mean no heat. And platform waveform appears to influence temperature kinetics. Think of it like this: we're not just asking, did the temperature rise?
Niraj Sharma:We're asking, how fast did it rise, how high did it peak, and how long did it hang around. That's what kinetics means and it's one reason why different systems may not be interchangeable from a collateral risk standpoint. Okay, let's move from the esophagus to something that surprised a lot of people when it was first reported: an airway complication. Here's the rationale in plain language: PFA uses high voltage, high current, ultra short electrical pulses, and even though the therapeutic goal is irreversible, electroporation in myocardium, those electric fields can also induce or concentrate energy in places we don't want it, like device leads and the generator. Now an important context point: most pivotal PFA trials excluded patients with implantable devices, so the evidence base here is more real world registries, observational cohorts, and case reports.
Niraj Sharma:That's not a weakness, it's just the reality of how new technologies mature. Mechanism number one is field induced coupling, basically electromagnetic induction. The PFA field can induce voltages and currents with transvenous leads, especially when the ablation catheter is within a few centimeters of lead electrodes or defibrillation coils. And what does that look like clinically? Oversensing, where the device interprets PFA pulses as intrinsic cardiac activity.
Niraj Sharma:Pacing inhibition, where the device decides it doesn't need to pace, when in fact it does. Or inappropriate detection of ventricular arrhythmias. Practically, this has been seen when delivering PFA near structures that tend to sit close to right sided leads, right pulmonary veins, posterior wall, CT isthmus, and SVC. The theme is proximity. Mechanism number two is more consequential: conducted energy via direct contact or near contact interactions.
Niraj Sharma:This can happen when the PFA catheter contacts or comes very close to pacing electrodes or ICD shock coils. There is also the vector issue. If the ablation vector aligns with a low impedance pathway, for example an SVC shock coil to the generator can, you may essentially create a preferential conduit for energy into the system, and that is the mechanism implicated in irreversible ICD generator damage failure of protective components in the high voltage protection circuitry, which can lead to device reset, battery depletion, or complete generator failure. Okay, so that's mechanism. Next, let's talk about which devices are most vulnerable and what does risk mitigation look like in a real lab workflow.
Niraj Sharma:Device specific vulnerability. Start with pacemakers and CRTP systems. Observational cohorts of left atrial PFA generally show preserved function. No sustained changes in sensing, impedance, or capture thresholds, but generally preserved doesn't mean nothing happens. Transient pacing inhibition, on the order of a few seconds, has been documented during PFA delivery near right sided structures, and the clinical relevance changes completely if your patient is pacemaker dependent.
Niraj Sharma:A two-four second pause can be hemodynamically meaningful in the wrong patient. Now ICDs and CRTD systems, these are uniquely susceptible because they include long conductive shock coils, especially SVC coils, low impedance, high voltage conductors, and very sensitive protection circuitry. Multiple case reports confirm permanent ICD generator damage associated with PFA reported during SVC isolation, left sided septal ventricular tachycardia ablation, and CTI ablation when you are close to RV or SVC leads. And here is the unsettling part: irreversible damage has occurred even without direct catheter lead contact, which reinforces the role of extreme proximity and field concentration. There's also emerging evidence that leadless or intracardiac defibrillation systems, including temporary EP lab connected defibrillation systems, can be vulnerable when bipolar multi electrode PFA is delivered within millimeters, with reports of irreversible internal circuit damage affecting recording and pacing capability.
Niraj Sharma:Catheter and energy delivery matter too. Monopolar systems may carry higher risk because of larger return pathways. Geometry matters. Large focal or lattice tip designs may concentrate energy. And lesion set matters, SVC, CTI and septal ventricular lesions appear higher risk than isolated left atrial PVI.
Niraj Sharma:So what can go wrong clinically? Pacing inhibition, over sensing with inappropriate detection, R on T, pacing leading to ventricular fibrillation, factory reset or software corruption, and in the worst case permanent generator damage requiring urgent replacement. Some of these present immediately, others can show up hours two days later. That brings us to practical protocol: how you run the case when a device is in the room. I'll frame it in three phases: pre, intra and post.
Niraj Sharma:Pre procedure: baseline interrogation thresholds, sensing, impedance identify pacemaker dependence deactivate ICD tachytherapies prior to ablation and in truly dependent patients, consider asynchronous pacing modes like VOO or DOO intra procedure. Maximize distance between the PFA catheter and leadscoils, especially in SVC and RV coil territory. Avoid catheter contact with ICD shock coils and in higher risk anatomies, I'd argue, for continuous surface ECG plus invasive arterial monitoring because if you get an abrupt pause or an ischemic phenotype, you want instant hemodynamic visibility. If real time device telemetry is feasible, use it. Oversensing and inhibition are much easier to interpret when you can correlate energy delivery with device behavior.
Niraj Sharma:Post procedure, immediate device interrogation is not optional and in higher risk anatomies if anything odd happened, repeat interrogation in twenty four to seventy two hours. Keep a low threshold for manufacturer consultation if there is reset behavior, abnormal impedances, unexpected battery changes, or sensing instability. Big picture future As PFA expands beyond left atrial PVI into right atrial work, SVC and ventricular ablation, these interactions are going to become more common. Risk appears anatomy, catheter and proximity dependent, not simply PFA is safe or PFA is dangerous. Okay, back to the final wrap up topics, troponin elevation and left atrial stunning using the nemesis framing.
Niraj Sharma:Troponin rises after PFA can be substantial, often exceeding what we are used to seeing with thermal ablation. Nemesis trial emphasizes three points: the troponin rise is universal, peak values can be striking, on the order of roughly 100 fold compared to RF in that framing, and there was no correlation with clinical myocardial infarction or coronary occlusion in uncomplicated cases. Mechanistically, that makes sense. PFA creates nanopores in cardiomyocyte membranes, you get calcium influx, loss of membrane integrity, and then programmed cell death or necrosis. Biomarkers leak out without requiring an ischemic event.
Niraj Sharma:So troponin is not a routine safety marker after PFA. It becomes clinically meaningful only when it pairs with ischemic physiology ST elevation, hemodynamic collapse, or ventricular arrhythmias, which by the way loops us right back to the coronary spasm and malignant arrhythmia phenotypes we discussed earlier in this series. On atrial stunning, this is transient left atrial mechanical dysfunction after ablation, particularly with extensive lesion sets and acute inflammation or edema. It is not unique to PFA, but because PFA can create large contiguous lesions rapidly, it may accentuate the phenomenon. The practical implication is conservative and familiar: continued anticoagulation early after ablation and humility about early rhythm success equaling immediate mechanical recovery.
Niraj Sharma:Absence of AF does not guarantee immediate atrial mechanical recovery. Let me close with a simple EP edge take. Three bullets. One. PFA.
Niraj Sharma:Safety is real but it's not uniform. Waveform and catheter geometry matter. Two. Watch for delayed physiology, ischemia phenotypes, autonomic swings, blood interaction effects, and now device interactions in the right patient. Three.
Niraj Sharma:The next phase of PFA maturity is risk discrimination and technique refinement, not just faster PVI. Thanks for listening. If you found this helpful, share it with your team because these complications when they occur are managed by the whole lab. If you are looking for references, graphs, more detail, they can be found on the LinkedIn newsletter as well as on Substack, epedge.substack.com. And as always if you have any questions or suggestions I can be reached via email epedgecastgmail.com
Niraj Sharma:Till next time I am Doctor. Sharma, bye for now.