You're listening to This Week in Cardiology from TheHeart.org, Medscape Cardiology. This podcast is intended for healthcare professionals only. Any views expressed are the presenter's own and do not necessarily reflect the views of WebMD or Medscape. Hi everyone, this is John Mandrola from TheHeart.org, Medscape Cardiology, and this is This Week in Cardiology for January 30th, 2026. This week, I'll talk about some listener feedback that was quite good, huge news in the rapidly expanding world of PFA AF ablation, obesity numbers, a beautiful trial studying an AI-enhanced diagnostic tool in the office. First is that I want to say thanks to the British Cardiovascular Intervention Society for inviting me to speak at the Advanced Cardiovascular Intervention Meeting here in London. I'm actually recording in my hotel room before I get on the plane to Chicago. My assignment at the meeting was to lecture on the three worst interventional trials that have affected practice nonetheless. I also had a talk about how wasteful AF ablation can be. These were easy talks for me to put together. I love this meeting, and I have to say that British meetings are a lot less stuffy than U.S. ones. It was great fun, and thanks to Dr. David Hildick-Smith and others for inviting me. It was wonderful. Okay, first some listener feedback. This is on the matter of TEE versus no TEE before ablation. Professor Joachim Ehrlich writes to me about ICE versus TEE coverage two weeks ago. His group in Wiesbaden, Germany, published a research letter in the Heart Rhythm Journal. I'll link to it. And they were chronicling the use of a risk score designed to avoid TEE before cryoballoon AF ablation. The gist of their series of 1,400 ablations was that both TEE and no TEE, if negative risk score, ended up with a 99-plus percent chance of having no post-procedural strokes. This is a nice report of real-world data. Now, the low stroke rates support our policy of requiring TEE only in super-high-risk patients, such as those with persistent AFib and unreliable DOAC use or a previous left atrial appendage traumas. Another note came from my friend and former partner in private practice here in Louisville, who's now out in San Diego, Dr. Anthony Pearson. He's also author of the Skeptical Cardiologist Substack, and he writes that the problem with TEE is that docs are not trained well enough in it. I've trained dozens and watched lots of incompetent clinicians doing TE. In the right hands, he writes, it is beautiful and elegant, and the images are way more magnificent than any other procedure. He says that the lack of training also results in many errors in interpretation, the most common being over-calling of left atrial thrombus and of valvular regurgitation. Another note from Dr. Stefan Asbach from southern Germany, near Lake Constance on the Swiss border. And I have to say some people choose geography wisely. He writes that another notable result from the ICE versus TEE trial was a very low rate of bleeding related to the transseptal puncture. I went back and actually looked. It was actually two of 906 patients in the ICE arm and 12 of 904 in the TEE arm. And Dr. Osbach writes that it is the first randomized trial proving that the use of ice increases safety by reducing the number of major bleeding events related to the transeptal. This would possibly avoid around 1,000 potentially life-threathing complications in Germany every year. Unfortunately, due to reimbursement issues, the use of ice in Germany is rare. Now, I'm not sure I would use the verb prove, but my point of view of ice has changed. Now that I've been medically legally coerced into using ice for the past five to eight years, I've come to like the reassurance it gives me during the transeptal. Now, I simply confirm that there's good tenting before puncturing. I mean, 95% of the time, the transeptal is easy without ice, and it wouldn't have been necessary, but it's still reassuring nonetheless. Now, whether it's worth $500 or $1,000 extra per procedure, that is a much harder question. Okay, another topic that I received feedback on was the study on procedural volume with TAVI and mitral transcatheter edge-to-edge procedures. Now, while it seems obvious that more procedures would lead to better outcomes, I was struck by how strong the association was. And it caused me to say that maybe it was better for patients to travel a little bit farther to go to a big center, rather than having it done closer to home at a lower-volume center for just convenience. Dr. Thomas Maybaum from Rostock, Germany disagreed. He rightly notes that some of these people are frail and elderly and being far from family is not so easy, and it's a good point, of course. And then finally, I received a comment on LGE in older athletes. Dr. Christopher Schennewice from Kolen, Germany, wrote regarding my coverage of the Ventu author's paper on 106 older endurance athletes who had loop recorders and CMR studies done. That paper found the strong correlation between having non-sustained VT and LGE on CMR. His point was that the LGE pattern in the study's athletes was non-ischemic, and it affected the inferolateral basal wall in 88%, which is suspected for prior myocarditis. And this would be in line with CMR studies on myocardial inflammation with persistent LGE and prognosis. But, but, he adds, it is important to distinguish the patterns of LGE because we know that some patterns in athletes seem to be benign, like, for instance, the ventricular hinge point fibrosis that seem, and others have a greater correlation to relevant ventricular arrhythmias. And I say thanks for the comments. Again, I don't know about you all, but these comments really help me because it induces me to go back and look at the coverage that I did, and then by looking at it again and looking at the paper again, it reinforces the findings of the paper into my memory. So thank you. Keep it up. All right. The big topic today is actually in my world of PFA, pulse field ablation, for AFib. The first bit of news is there was a big trial called BEAT-PEROX AFib Trial Published. And I say before you EP docs tune out, the news I'm about to tell you speaks not only to the world of AFib ablation, but it also transcends that to early adoption of any new technology, especially in the setting of intense marketing. This is a problem for all of modern cardiology actually The European Heart Journal has published results of the BEAT AF trial comparing PFA to RF ablation in patients with paroxysmal AF This was a multicenter European RCT of 292 patients with PAF. The first author was Professor Pierre Jace from Bordeaux. The PFA was with the pentaspline ferrapulse Boston Scientific System, and RF was with the modern contact force sensing system. The PFA system is important because I'm convinced that PFA systems are not all interchangeable. The primary endpoint was single procedure success at one year, and this was defined basically as the absence of any arrhythmia occurrence greater than 30 seconds. So really strict criteria. There had to be no cardioversions, no resumption of antiarrhythmic drugs, or no repeat ablation. So this is one of the situations where you want higher numbers in the primary endpoint. and indeed the primary endpoint occurred in 77.2 percent of the pfa group and 77.6 percent of the rf group so clearly no difference you don't need statistics there were also no notable subgroup interactions the mean total procedure time was shorter for pfa 59 minutes versus 98 minutes so that's substantial the mean left atrial dwell time was less 39 versus 77 minutes and there was slightly more minutes of fluoroscopy, 12 versus 6 for PFA versus RF. Procedure-related adverse events were 5 of 145 in the PFA group, that's 3.4%, versus 11 of 144, 7.6% in the RF arm. That lower rate in the PFA group did not reach statistical significance, and importantly, there were no death strokes or cases of persistent phrenic nerve palsy noted in either arm. in the rf group there were two cases of cardiac tamponade in one case of esophageal bleeding and this was felt to be related to the tee additionally they had two cases of severe pv stenosis were reported in the rf group all involving the left inferior pv only cases only transient cases of phrenic nerve palsy were observed in the trial two in the pfa group and one in the RF group. Okay, so my comments on this BEAT-PEROX-AF trial. From the names on the list of the authors, this trial clearly compared PFA to super highly experienced RF ablation in RF operators, which has probably been PFA's toughest comparison to thermal ablation to date. The trial did not use ILO monitoring, so success rates may be a little better than the last PFA versus thermal ablation trial, this single shot champion trial, which used ILRs. Of course, an ILR is going to pick up more AFib, but I don't see the lack of ILR use as a weakness because in real life, we don't use ILRs in everyone. It's normal to do intermittent monitoring and use symptoms, and trials that emulate this are easier to translate. So in sum, like so many other studies, this one shows that PFA is faster, but no better in terms of efficacy or really common safety measures. The lack of difference in adverse events, though, is somewhat deceptive, right? Because while we don't love tamponade, we really worry about death from atrial esophageal fysterea or permanent phrenic nerve. And that PFA likely reduces these terrible five sigma events is a huge positive. But stay tuned for the last report I have coming up. Now, this lack of difference in efficacy could be construed as disappointing for PFA, but I would submit that this trial represented probably the hardest test for PFA because the RF control arm was performed in some of the best labs in the world. Had PFA versus RF been done in the average American EP lab, it's possible PFA would be dominant because RF is much more dependent on expertise. Now finally, the ceiling of success continues to be the most consistent theme in EP evidence. In other words, we're the same efficacy in AF ablation as we were about 25 years ago, around 60 to 70 percent. And that, I propose to you, is due to our lack of knowledge regarding the pathophysiology of AF. It's shocking to think about it, so I, like you, avoid thinking about it. but we literally do the same procedure in everyone with AF. Unlike everything else in EP, we have almost no targeting mechanisms for AFib. And I think it's because our research has been much more interested in selling ablation devices than discovering causes of AF. All right, there's more news on PFA, and these are sobering reports. Circulation has published a paper from the Hamburg CCB Group. First author is Melanie Gonerwardine titled, Life-Threatening Delayed Myocardial Ischemia and Malignant Arrhythmias Occurring After PFA Ablation of Atrial Fibrillation. And boy, that's a scary title. This paper was easy to miss since it was published on Christmas Day, actually. This was a case series from seven big centers. It included patients who had delayed or prolonged ischemic or arrhythmic adverse events after PFA for AFib. The total number of consecutive ablations was large, 6,721, and life-threatening adverse events of either ST segment elevation, malignant ventricular arrhythmias, or sudden death occurred in 11 patients, or 0.16%. Nine of these 11 patients had had ablation, PFA ablation, outside just the PVI, so extra ablation, not just PVI. Five patients presented with delayed ST segment elevation at a median of 20 minutes after the last PF delivery. But although the median was 20 minutes, the range was 4 to 238 minutes. One patient had VF and died. Two patients presented with VF during the procedure both required CPR and shocks. Two patients had prolonged hemodynamically relevant vagal responses after PFA remote from the conduction system. And this next data point may be the most scary. Two patients had sudden cardiac death three in 22 days after PFA. Three in 22 days. Autopsy was available for two patients. In patient one, the ILR data showed AF followed by VF progressing to asystole on day 22. That autopsy revealed no evidence of acute MI. There was an intact LAD stent, only mild atherosclerosis. There was no stroke. Cardiac tamponade, pulmonary embosur bleeding. So the cause of death was determined to be spontaneous VF. Notably the heart showed mild left ventricular dilation and hypertrophy In patient two the autopsy revealed an acute inferior infarction in the setting of a dominant right coronary without evidence of plaque rupture, thrombosis, or coronary artery disease, fistula or embolism. A very delayed right coronary artery spasm was suspected by the treating physicians in the absence of any coronary disease. So then, in the research letter, the authors speculated on this novel and scary report where major arrhythmic events occurred in minutes or days after PFA. Now, I say this is novel because we have long known that PFA adjacent to coronary arteries can cause spasm immediately, but both the delayed effect and the fact that coronary events with ablation far from the coronary geography, this is notable. Another common theme in this paper is that 9 of 11 patients had PFA outside the normal PVI lesion set. But one fatal case, in fact, involved just PVI. PFA, PVI, ablation, and sudden death afterwards. That's remarkable. Ultimately, the true cause, if there is just one cause of this, is likely via many different possible mechanisms, things like muscle stimulation with ionic imbalance, particularly calcium. How about electroporation-induced release of vasoactive mediators? how about hemolysis related nitric oxide depletion how about autonomic imbalance how about underlying coronary abnormalities such as subclinical plaque microvascular dysfunction or prince metalangina all these things may contribute and i would say in other words it's really unknown what the cause of these are now perhaps you'd be reassured by the low incidence of 0.16 percent but I'm not reassured by that at all, and I'd say for two reasons. One is that this is a voluntary survey of only seven centers. Surely if there is a signal here, it is also probably in other places. Second is that death from an AE fistula with RF ablation or cryoablation is sort of in this category of rarity, but it still keeps us up at night because people in their 50s, 60s, or 70s are not supposed to die from AF ablation. Low incidence events can be quite scary when they have such terrible consequences. Now, the authors made no comments about the PFA systems, but inspecting the table of the 11 patients, I saw that 9 of 11 of these events occurred with the pentaspine PFA system. Now, this is nowhere near proof, but I wonder about whether similar injury would occur at the same frequency with other systems. I'm asking the question. I think it may, but I don't know. The authors write that rare, life-threatening, ischemic, and arrhythmic adverse events highlight the need for better recognition, reporting, and understanding of underlying mechanisms to improve PFA safety, and I could not agree more. Yet there was one more piece of sobering PFA news. There was also a well-circulated MAUD adverse event of a death from AE fistula following AF ablation from a PFA system, the Aferra Medtronic system. And so on the MAUD report, it was reported that 20 days after cardiac ablation, the patient was diagnosed with an atrial softgial fistula and hospitalized. The ablation procedure was performed on the anterior aspect of the pulmonary veins using RF energy, and the procedure was completed successfully. There were no issues reported during the procedure, and no further patient complications have been reported as a result of this event. Now, this event involved the Medtronic Sphere 9 dual lattice tip catheter, where you can deliver PFA or RF. It's the most common PFA catheter that we use at our place, and it's a pretty amazing catheter. And you can switch back and forth from PFA to RF with a simple switch, but this requires the operator telling the mapper to switch the type of delivery. Now, a number of docs speculated on social media about the mechanisms of this. how could you have AE fistula if you're using RF on the anterior ridge of the pulmonary veins? That's unusual. I mean, so perhaps the operators were more posterior than they thought. Perhaps they accidentally delivered RF in the posterior wall instead of PFA because this is a simple switch. I strongly hope that the team with this case writes up the details because we need to know. It's remarkable. Overall, though, in this report today, we learn that PFA did not outperform RF, and two reports of serious safety events warn us to stay vigilant. While tens of thousands of PFA cases have been done worldwide, signals of this magnitude will take a lot of vigilance to sort through. I guess the bottom line is that when you're destroying large areas of the myocardium, the saying that there is no free lunch holds true. I think we need to stay tuned for more on this matter. And gosh, while PFA is likely here to stay, and it is surely great for patients, the fact that it has done nothing to tell us the causes of AFib or help direct individualized ablation still remains true. As I've said many times, a real breakthrough in AFib care will most likely come from a basic science lab, not an EP lab. Okay, a quick report on obesity trends. This is scary. JAMA has published an update on U.S. state-level prevalence of adult obesity by race and ethnicity, and this comes from the Institute for Health Metrics and Evaluation at the University of Washington. It'll be a short topic, but I want you to pause on the numbers. The data source involves 11 million U.S. participants, and it comes from three different surveys of behavioral risk factor surveillance system, the Gallup Daily Survey, and NHANES. The main results are that the obesity crisis continues. Here are some numbers. In 1990, 19% of U.S. adults had obesity. That was 34 million people. In 2022, 42% had obesity. That's 107 million people. And in 2035, the authors forecasted that 47% will have obesity. That's 126 million people. And they also note very large racial ethnic disparities, especially among women. For instance, non-Hispanic black females have the highest rates of obesity. 57% in 2022 projected to reach 59.5% by 2035. Hispanic females are close behind at 49% in 2022 projected to go over 53% by 2035 and males across all groups have more similar rates 40 to 43% in 2022 Not surprising is that the highest rates of obesity were in the Midwest and southern parts of the U.S. but there were also alarming age trends Peak obesity occurs in middle age 45 to 64 years The biggest increases since 1990 are among women under age 35 which of course indicates an earlier onset of obesity. Now, I don't know what to say about this, and you already know it's a problem, but the fact that half, half of U.S. adults are expected to be obese in the coming years is shocking. I'm not a public health doctor. We treat patients individually. but this sure seems like a public health catastrophe, one that may take many generations to reverse, and I hope it's reversible. The GLP drugs are going to help, but it's going to take a lot more of a societal change to really reverse these numbers. All right, here's a story about a new tool in the office for diagnostic purposes. Let's just say a shiny, happy industry rep brings in lunch to your office so that he or she can show you this super duper AI enhanced stethoscope that can record an ECG over 10 to 15 seconds and then do heart sounds. And then it can output a diagnosis of say heart failure, valvular heart disease, or AFib. And this would sound fantastic. And I actually see tons of fancy stethoscopes out there on the ward. So this would be the fanciest of all stethoscopes. But what would you really want to know before you spent your money on it? Obviously, you would want to know if it actually works. Does it enhance diagnosis of conditions like heart failure or AFib? If you were especially clever, you would also want to know if it worked during everyday regular practice. Namely, the clinicians liked it, it didn't slow people down, and they accepted it. Well, a large team of researchers from the NHS actually studied these questions in an elegant cluster randomized trial of 200 plus practices, primary practices in the UK. They called their trial the Tricorder trial and Lancet published a paper. It's really nice and worth reading. The trial randomized 205 primary care practices covering 1.5 billion patients and you either use the AI stethoscope into practice or you just do regular practice. Again, patients were not randomized. Practices were. The primary endpoint of new heart failure diagnosis and secondary endpoints of AFib and valvular heart disease detection sadly showed no significant difference between the groups. Heart failure incidence rates was 0.94, not near statistical significance. There was no shift toward community-based diagnosis versus hospital-based diagnosis, and they also found similar results for AF diagnosis and valvular heart disease. Now, the researchers, I think, learned a lot more than just the null primary results. Other things they learned, that new stuff brings implementation challenges. For instance, only 49% of examinations were labeled with patient identifiers for data linkage. By 12 months, 40% of practices had stopped using the device entirely. Clinicians cited workflow integration barriers as the primary obstacle. Now, the authors then did a per-protocol analysis. In other words, in patients this device was used on, how did it actually work? But here they had a problem because in most trials, a per- protocol analysis compares patients who stayed with the protocol versus those that stayed in the control group, there is still, in this setting, a least partially randomized comparison. But here, the practices were randomized, not the patients. So the authors had to do a very strong adjusting with propensity scoring system to match patients who had the AI stethoscope used on them versus similar patients that did not have it used on them. Now, it was obvious that clinicians chose to use the AI stethoscope in sicker patients as the mean age was older, there was more hypertension and more diabetes in those who had the stethoscope used on them. And so in this adjusted analysis, the stethoscope crushed it with 2.2 times higher rates of heart failure diagnosis, 3.4 times higher rate of AF diagnosis, and 1.9 higher rate of valve or heart disease diagnosis. But again, this was an observational non-random analysis, and even with propensity matching, there was surely selection bias because clinicians chose to use it versus not choosing to use it. So my overall comments is, I highlight this trial because it is so well done and so honestly reported. It's refreshing. These new tools are nice, and reps can make great PowerPoint presentations and videos, but the true test is not whether they can work in an idealized patient, but whether they work in a real-world setting when doctors are busy and they require something as seamless as it is nifty. Now, I would be happy if all or most new tools had to pass muster in this such a tough test. These authors show that it can be done. For instance, when handheld AI-enhanced echo come out and are pocket-sized, there should also be a similar study just like this. Now, another thing this paper underscores is the difficulty in screening for conditions that have population incidences that are about 1%. Again, basic common sense is that when a condition is this low in prevalence, even a super great test, 95 plus percent specificity is going to deliver mostly false positive results just based on a low disease prevalence. And perhaps the broader lesson here is that new tools and drugs and devices have to pass two bars. One bar is that they have to work in idealized conditions, but the other is that they have to work in the setting of the real world, in all comers. And this is why, of course, the intention-to-treat analyses are so important. A per-protocol or as-treated analysis does not preserve randomization, as those who get off-protocol are often different than those who stay on. So I say congratulations to the authors. It's a really nice piece of science and refreshing. And so that's it for this week in cardiology. As always, I'm grateful that you listened, and I say thank you to all the people here in London who came up and said hello. It really helps me to know that people listen. Thank you. And remember, if you like the podcast, please take the time, give us a rating on whatever app you use. These sorts of things help others find us. Until next week, this is John Mandrola from the heart.org Medscape Cardiology. You're listening to This Week in Cardiology from the heart.org Medscape Cardiology. This podcast is intended for healthcare professionals only. Any views expressed are the presenter's own and do not necessarily reflect the views of WebMD or Medscape.
