Hi, and welcome to the second episode of The Lead, Heart Rhythm Society's new podcast, which is put together by the Digital Education Committee. This is meant to be a mini journal club review with experts in the field discussing the latest and greatest publications in cardiac electrophysiology. I'm your host for this episode, Jason Jacobson. I'm the Director of Complex Arrhythmia Ablation at Westchester Medical Center in Valhalla, New York. My co-host today is Dinesh Kella. He's the Senior Associate Consultant at Mayo Clinic Jacksonville. Today we'll be discussing a recent paper that was both presented here at Heart Rhythm 2023 and published simultaneously on Jack EP called Epicardial Roof-Dependent Macro-Reention Tachycardia After Ablation of Atrial Fibrillation, Prevalence, Electrophysiological Characteristics, and Ablation Strategy by Lai and colleagues from Beijing Angen Hospital, Beijing, China. We have the great pleasure today of being joined by Apoor Patel from Houston Methodist Hospital, who's the Director of Electrophysiology at Houston Methodist Sugar Land Hospital, and Pierre Jais, Professor of Cardiology, Bordeaux University Hospital, and CEO of Larrick Institute. Thank you all for joining us today. Dr. Patel, why don't you summarize the paper for us? Sure. Thank you, Jason and Dinesh, for the kind invitation to speak today. It's really a pleasure to be here. So this paper is a retrospective study, and their focus is looking at patients with roof-dependent epicardial atrial tachycardia. They looked at 44 patients with roof-dependent AT, and of these, about 15 or a third had a presumed epicardial circuit. Now when we think about how do they define what is an epicardial circuit, first they looked at entrainment. So they would entrain from the posterior wall, and if the entrainment was long or if they couldn't capture, they thought that this kind of excluded an endocardial circuit. In addition to that, you had to be able to entrain from other anterior and posterior wall sites, you know, indicative of roof-dependent reentry. They also defined that all these patients had a slow conduction or no conduction zone on the posterior wall, and again, if you pace here, you either won't capture, or if you do capture, you have a long PPI minus TCL, but if you pace at high output, then you might entrain in the circuit, you might have a short PPI minus TCL, because now you're capturing an epicardial component, and they show that nicely with electrograms. They also describe a pseudo-focal activation pattern from the posterior wall, so you make your activation map, and it looks like you have a focal AT from the posterior wall, but actually it's just because you're missing, you're just mapping the endocardial breakthrough of an epicardial circuit. And then they described that a lot of these circuits were missing more than 10% of the tachycardia cycle length when you made your activation map. So this is kind of how they defined, excluded an endocardial circuit and defined the inclusion of an epicardial atrial tachycardia. They then looked at the patient characteristics between patients with endocardial roof-dependent AT and epicardial roof-dependent AT, and essentially the patients with epicardial circuits had a longer duration of AFib, they had been in AFib longer, and not surprisingly, had a higher percentage of prior posterior wall ablation. The authors describe three activation patterns of these ATs. So they describe a counterclockwise view and a clockwise view. It's kind of hard to describe it, but imagine looking at the left atrium from a right lateral view. In a clockwise AT, you're going cranial to caudal down the anterior wall and caudal to cranial up the posterior wall. In a counterclockwise rotation, you're just going in the opposite direction. And then they describe a biatrial tachycardia, which involves both atria obviously, and both of those patients who had biatrial tachycardia had prior anterior lines. So in those cases, you're using Bachman's bundle to jump from the left to the right atrium, and you're using the septopulmonary or septoatrial bundle to jump from the anterior to the posterior wall. They then describe how these tachycardias terminate, and they show that you can sometimes ablate them at the site of focal endocardial breakthrough. So you make your map, and wherever you see the focal breakthrough, if you burn there, it may terminate. If that didn't work, they would try roof or floor ablation. And what they found was that one third of the patients could not be terminated with either roof or floor ablation, which was interesting. And that translates to six patients. Out of those six patients, three of them terminated with anterior wall ablation, and the other three had to be cardioverted, which obviously is frustrating for any electrophysiologist not to be able to terminate the tachycardia. So I think their main point is that anterior roof lines or roof lines are hard. It's hard to get a transmural lesion. You can try to reinforce the line, but that often doesn't work. And they speculate that this has to do with tissue thickness at the roof. And in addition, they reference Pierre's work and the Bordeaux group's kind of work from a few years ago that elegantly shows that you have adipose between the septopulmonary and septoatrial bundles, which is not conductive and makes it harder to get a transmural lesion. You can do a floor line, but that's complicated because then you have to deal with the esophagus. So those are kind of the main points of the paper. And I'll kind of turn it over to Jason to generate the discussion now. Thanks, Apar. I think that was a really nice summary of the paper and some of the issues. And Pierre, I'd like to get your thoughts on what these epicardial connections are. I know there have been those who believe that it's really not any separate distinction between the endo and epicardial layers and that it's really just our inability to create a transmural lesion. However, as Apar alluded to, you do have some very interesting data, both autopsy and imaging. And when you look at that, both autopsy and imaging, it is pretty clear that it's not just a thickness issue. Thickness matters a lot and RF typically can give you five millimeters depth. But beyond that, it is really more challenging. And we've seen that very well in ventricles. So this could be an explanation, but I think that the presence of fat and the interposition of fat is certainly even more powerful than just the thickness. This is why the mitralismus line is also sometimes very difficult because it's bridged epicardially by the martial vein that is surrounded by fat. And every time we have this anatomical configuration, we struggle. And you know, the other point I wanted to discuss, and I don't know if you want to do that now, but the thing that I was most surprised in their paper is the fact that those patients in whom they had epicardial component of the circuit were also those with the longest exposition to atrial fibrillation. I found this puzzling because the way the atrium remodels with atrial fibrillation is that it dilates, but not necessarily that the thickness increase. It should in fact be the opposite when you look at specimens and imaging. And so maybe the fat component is in fact increasing as time goes by because fat is really part of the pathophysiological mechanism with atrial fibrillation. I mean, there is an infiltrate that comes from the epicardial layers and that progressively invades the endocardial or mid-wall layers. So that's just an hypothesis I wanted to run, but I've been surprised. I see no reason why this would be associated, and it was pretty significant. That's an interesting point. There's another question I've had, and I haven't, again, this is I think more hypothetical, I haven't seen or I'm not aware of any data on this. We've talked about the heat sink of the coronary sinus, but what about the pulmonary artery as it goes over the roof of the left atrium? And the aorta sometimes is in direct apposition to the back wall of the left atrium. Do you think these could have any effect on our ability to create these lesions elsewhere? Honestly, I don't know. I tend to think that the proximity is not important enough. They are typically not close enough to be the explanation, but their patience in womb, it is really close, so that may very well be an explanation as well. The thing I tend to believe more when it comes to the aorta is the compression onto your aspect of the left atrium where you see sometimes low voltage and reentries that are localized really in this area, and you can tell that for sure from imaging. It's really matching well. And I think sometimes we can see the atrial signal coming across the right pulmonary artery as it goes across the left atrial roof. Maybe this is a way to potentially see what's going on in the roof in some of these flutters to try to map better, and I doubt we can really capture from there, but at least we can try. That's a good point. We're certainly not looking at that systematically. I remember doing it in perhaps two patients in my entire career and didn't make a change, didn't ablate from there or whatever. Apoor, I'd like to get your thoughts on approaching this without radiofrequency. I know your group has really brought forth the alcohol ablation approach to these patients, and sometimes there's veins that drain this area. What do you think about that as an option? Yeah, I think it's an interesting hypothesis. When you think about the veins that we see, in addition to the vein of Marshall, there's septal veins, inferior veins, appendage veins, anterior veins. All of those veins come directly off the coronary sinus. Then there are posterior wall veins and roof veins, which don't come directly off the CS. You see them when you inject these veins, basically. I think I was talking to my colleague, Miguel Valderrabano, about have we ever had a case where we've given alcohol and gotten roof block, and we haven't seen that, but I think you can see some low voltage extending up to the roof, because the vein of Marshall will often collateralize with the roof vein, where these other veins will. I think more importantly, you can see extension on the posterior wall, and it can help you get posterior wall isolation and transmurality on an inferior line. I think there is benefit to alcohol in these cases, but it's not necessarily going to be pure roof block, but it helps in other ways, and helps extend that scar where you need to be on the posterior wall. Thanks. This is very interesting. I'd like to make two points here. The first one is that if we're talking about non-RF, I think PF is a fantastic solution. I know you don't have yet access in the US, but it will come soon, fortunately. The benefit of using it is that you're not going to injure the esophagus. You can have your roof line really low, where the septal pulmonary bundle is no longer on the way. I did, I don't know, probably close to a hundred of these cases now. It's really solid and durable posterior wall isolation. It has been reported by Vivec also, there has been some systematic reassessments at three months. It's really solid lesions, because the wall thickness is not really important, low in the posterior wall, and you don't have the fat issue or the septal pulmonary bundle issue. I have one case, we've published that recently, where using electroporation, we failed to block the posterior wall. Then after that, we did the martial vein alkalization and it was blocked. Sometimes you have some branches that are going posterior out of the martial that are bridging and driving you nuts. When you guys do the Bordeaux approach or the martial plan to ablation, with PFA, you're kind of saying you're doing a low roof line, but you could just also do an inferior line at that point, right? Because the safety factor is there with electroporation in terms of not having to worry about the esophagus. The authors in this paper, their solution was, in a few of the cases, they deviated the esophagus to alleviate those concerns, but with electroporation, it seems like it doesn't really matter where that line is, right? Absolutely right. The thing, I guess, and I did not really test that hypothesis, plus it's difficult because the flower diameter is really big. When you look at that on the posterior wall, it's pretty much the entire posterior wall. But you still can have a rather superior or inferior position. We didn't really test what the success would be and the durability would be with a superior position because from the beginning, we knew that the esophagus was safe, so we started low anyway. Low means that we have the shaft, so the middle of the flower, that is at the level of the line between the two inferior veins. That's pretty low. Okay. It's pretty low. One thing I'd like to point out is it seems these are purely iatrogenic flutters that we're creating in our attempts to carve up the left atrium. How do we prevent this without the availability of PFA right now? We talk about best ablation approaches. Is it the fast and furious, high power, short duration, or is it the low and slow so we can get a more transmural lesion, obviously taking into account the esophageal damage that can occur? When it comes to epicardial targets, you're better off going with long duration and lower power to prevent from popping, for example, but then for sure, the esophagus is in the way and that's also ... You have this goal of having a deeper lesion and then you put the esophagus at higher risk, of course. Very clearly, the pulse field ablation is a very good way to approach that. This roof line gave us hard time for 20 years. It's really changing our life. It's pretty much the same change as what we had with the martial vein alkalization. It's completely changing. When we use martial vein alkalization, the amount of RF that is needed to block the mitral isthmus is five minutes, as compared to something like 25 minutes in the early reports. The percentage of block was much less at that time. I agree. I think for us, in the US where we don't have electroporation, I think it has to be some type of esophageal protection strategy if we're going to be more aggressive on the poster wall, whether it's deviation, cooling, collapsing the esophagus, whatever it is, or just temperature monitoring and then waiting for the temperature to settle down before you burn some more. There has to be some strategy in place to do it safely. Do you think imaging has any role, pre-ablation imaging, to help choose your lines wisely? I'm not sure at the present stage because we're limited. MRI has a spatial resolution issue. The poster wall thickness may be as thin as two millimeters. MRI pixels are at 1.5, so we're not there. Common CT is better, but then it is hard to identify where the wall stops. I hope that photon counting CT scanner, which is having a much better spatial resolution that is down to 200 or 100 microns plus some tissue characterization, because it looks at the spectral power of each photon individually. You may have a chance to do a much better job here. Maybe one comment about that paper before I forget. The one thing that has not been, I think, mentioned, unless I missed it, is that typically when you map on the posterior wall those patients with a picardial component of the circuit, you have a far-field component that is early and that is delayed on the superior part of the posterior wall. As you go lower, they get closer, and finally, the two components are fusing where it connects. That's also something people can look at that's useful, and annotation systems are not good at that. They would go for the SHARP and LVDT. It's something that can be recognized by someone looking carefully at the signal. One question I had for both of you, do you think epicardial mapping, maybe ablation, would it start moving from VT to these difficult flutters, or do you think, yes, it may be for some cases, still something might be mid-myocardial over the adipose tissue and all those things. We may not get anything out of epicardial mapping, especially these patients being on anticoagulation. It's always a challenge to do it before planning. I think in the vast majority of cases, it's probably not necessary. I think in very refractory cases, certainly, Rod Tong has published a good article about that, and I think there's a role for it, but I think overall, and especially with electroporation, if we can continue to get good transmural lesions, hopefully, we'll just be able to avoid that altogether, especially, I think, for a roof-dependent tachycardia, maybe for other regions like Bachman's or the mitral isthmus, although, again, with mitral isthmus, with alcohol, it's become so much more readily available to get mitral block. I completely agree with that. One more point, maybe, when we are facing those difficult cases, we have to find a solution better than cardioversion, as you said, which is very frustrating, obviously, and expose the patient to redo at least recurrences, and so this is where it starts to be a little bit of a problem, in my opinion. What happened is that we fail superiorly, then we go lower, and we may fail as well, and then we go anterior, and at the end of the day, the amount of destruction starts to be really significant, plus we have created, probably, further obstacles in heterogeneity for other tachycardia to develop, because, as you said, most of these are man-made, in a way. I suspect not all of them. I mean, there are some signals that you may also have rotations around the veins or around the mitral during AF, and you don't see it, but it might be there for one rotation or two, and it may participate. I'm pretty convinced about that, but anyway, it's true that most are created by incomplete RF lesions, and at the end of the day, the impact on the atrial function is significant, and particularly, I don't like anterior lesion much, unless there is a significant amount of spontaneous destruction there, because you're going to, you know, put everything that is lateral, and particularly the appendage, out of the game, and no longer useful from a hemodynamic standpoint. I think we also tend to see biatrial flutters when we do these anterior lines, especially de novo. Just one other point, I think, before we should probably wrap up, is that the authors do mention high output pacing, and whenever I do end up doing posterior wall isolation, I try to pace as high as I can, and if I capture something, I assume it's epicardial, and try to hit that harder if the esophagus allows. Absolutely, makes sense. All right, well, I'd like to thank Dinesh, Pierre, Apoor, thank you all for joining us today for this second episode of The Lead. Thank you for listening and watching.

epicardial roof-dependent macro-reentrant tachycardia

ablation of atrial fibrillation

entrainment

mapping activation patterns

prior posterior wall ablation

roof region ablation