We'll go ahead and get started. Welcome everyone. My name is Naza Makoum. I'm at the University of Washington. I'm happy to chair this exciting session this afternoon. My co-chair is actually in a different session and he's running over. He's Dr. Helmut Per Felner and he'll join us as soon as he's available. We're going to cover different assessments of atrial myopathy today and try to tie it with outcomes and results. And we have a great lineup of speakers and each will have about 12 minutes to tell us about their research and their findings. And then we'll have about 12 minutes at the end to ask questions. So when we get to the Q&A, please come up to the microphone and ask your questions or send them via the app and we'll be able to see them here and ask them for you. So with that, we'll go ahead and get started with our first speaker, Dr. Natasha Dekrut from Erasmus University Medical Center. And she's going to tell us about voltage mapping and atrial myopathy, lessons from unipolar mapping. So good afternoon, ladies and gentlemen. So the topic of this presentation is what can we learn from unipolar voltage mapping. We know that when atrial fibrillation changes from paroxysmal atrial fibrillation to the more persistent types of atrial fibrillation, it also changes from a trigger-driven to a more substrate-mediated arrhythmia. And the substrate can be influenced by the non-modifiable, modifiable parameters, but also by, for example, sex, age, and your genetics. And of course, the problem is the ablation of the patient with persistent atrial fibrillation. And over time, many approaches have been introduced, and voltage mapping is just one of them. And I think we all know this trial where patients were randomized to pulmonary vein isolation only or pulmonary vein isolation in addition to ablation of low-voltage areas. But I think there are still a lot of questions. What is exactly the definition of a low-voltage area? Because if you take a look at literature, many cut-off values have been proposed, and they include 0.5, 0.05, 0.2, and also 0.1. The second question is, what type of electrogram do we need to choose? And how does the electrogram type affect your voltage map? And most straightforward is, I think, the unipolar voltage. This is what we can understand. This is the morphology of a typical unipolar potential. You have a wave propagating towards the recording electrode, moving away from it after the initial deflection. So you have an R and an S wave, and the voltage is the peak-to-peak amplitude. For bipolar electrograms, it becomes a little bit more difficult. It's a subtraction of two unipolar electrograms. And we already know for many years that the peak-to-peak amplitude of a bipolar potential can never be larger than the two original unipolar electrograms. The types where we can nowadays choose from is the unipolar electrograms, and we see now that it's luckily increasingly being used at the cath lab. But we also have the bipolar electrograms, and lately, there is also a tendency towards using the multipolar electrograms, and particularly the omnipolar electrograms. But the peak-to-peak amplitudes of potentials are influenced by not only properties of the atrial tissue, as you see here summarized for the unipolar and the bipolar electrograms, but also by technical aspects, for example, the contact, the distance between the electrodes, the filtering, the sampling, but also, for example, the cycle length. So how can you measure the peak-to-peak amplitude of the unipolar potentials? And we know that if we do that at the cath lab, it depends on what you annotate as the local activation time. So for example, here, this is a clear local activation time here, so you measure from peak-to-peak. Here, it's also clear, and here also. But then you have potentials like this. This represents asynchronous activation of the tissue beneath the recording electrode, but it contains multiple deflections. And here, it might be difficult to understand exactly what the local activation time of the tissue is. And also, if you have synchronous activation of the tissue, you have a certain degree of your voltage, but does that actually correspond to the moment that the same tissue is activated asynchronously? Is it simply the case that if you add these peak-to-peak amplitudes of all the separate deflections, is that then the same as this amplitude here? And that is something we do not know. And when you use bipolar electrograms, and you want to calculate the potential voltages, it even becomes more complex, because this is a unipolar fractionated electrogram. This is a unipolar fractionated electrogram, and that results in these bipolar electrograms. So there, what's the local activation time, and what is then what you choose for your bipolar voltage? And that's not what you really realize when you make voltage map at the catalab. So what can you learn from apicardial mapping? So in our hospital, we routinely perform intraoperative mapping procedures. So every patient undergoes a apicardial mapping procedure. And because the surgeon actually put the electrode on the heart, and he actually can see it, you know for sure that there is good contract between your electrode and the atrial tissue. And this is how we do the mapping. So we have the mapping array. We shift it across the right atrium in an orderly fashion, the Pachman's bundle, the left atrium appendage, and finally, the left atrial posterior wall. And in this way, you end up with a scheme which looks like this. So this is a schematic presentation. And you have about 10 positions per patient, of course, depending on the size of the atria, which results in over 1,900 sites, recording sites, with an intra-electrode distance of only two millimeters. So by using this technology, my colleague, Dr. Foshee, investigated the unipolar voltage distribution in patients with mitral valve disease. And he had patients with no history of atrial fibrillation, 44, and he compared it with 23 patients who had paroxysmal AF episodes. And this is, again, the methodology. So we measured the peak-to-peak also, and we classified them for different potentials consisting of one, two, or three or more deflections. And what he found was quite remarkable, because he found that even in sinus rhythm, there's already a difference in the voltages between patients with and without atrial fibrillation, but it was particularly located at Pachman's bundle. And that's, of course, a structure we cannot reach from the endocardium. And you see that in the patient without fibrillation, nearly five millivolts, and here in the patient with atrial fibrillation, it goes from nearly five to nearly three. And also the amount of low-voltage areas increased considerably from two to almost 12 percent. Here you see six typical examples of these voltage maps. So every 1,200 electrograms is represented by a color code in voltage. And here you can see it for the patient, typical examples for patients without atrial fibrillation, and here for patients with atrial fibrillation. And these maps clearly show that there's quite considerable variation in potential voltages within a patient, but also between the patients. And every patient has low-voltage areas, and they're distributed throughout the atrium. And most importantly, in these patients, we did not find any predilection sites, also not in the patients who had paroxysmal AF episodes. Then there is an increasing tendency to use omnipolar electrograms, and how does that relate to the unipolar voltages? And Dr. Van Schie also evaluated the omnipolar, unipolar, and the bipolar voltages using the clique methodology applied to our electrode array, our high-density electrode array. And he found that the unipolar voltages are the largest. The omnipolar are somewhat between the unipolar and the bipolar. And here you see it also illustrated, so these are the voltage map, and each electrode is color-coded to the voltage. And you can see that if you have the bipolar calculated in the X direction and in the Y direction, there's a clear difference. That's also because there's a direction dependency in the bipolar electrograms. But if you compare the unipolar of the, sorry, the omnipolar with the maximum bipolar, they are more or less comparable. So there is a difference in this methodology. And here there's another interesting finding. So here he took all the electrograms for the bipolar, omnipolar, and the unipolar. He took the fifth percentile, which corresponded to a bipolar voltage of 0.55, which corresponds nicely to the clinical studies, which most often use 0.5 millivolts as a threshold. And he found no relation between potential voltages and conduction velocity. So conduction velocity could either be high in low-voltage area, but it would also be low. Then the next question is, if you measure on the endo and the apicardium, is that then the same? What are actually the difference in opposite potentials on the endo and apicardial side? So we made an incision in the right A2 appendage. We used a clamp-like electrode, and we clamped the tissue between the endo and apicardial electrode array. There were 128 electrodes on both sides, again, with an inter-electro distance of only 2 millimeter. And here you see the opposite activation map, but this is what you expect. If you have sinus rhythm and the wave propagates nicely, there is no difference in between these maps. And if you calculate the voltages, so if you plot all the apicardial voltages and the endocardial voltages, we did it for 26 patients, which resulted in over 100,000 potentials. We found a positive relation between the opposite endo and apicardial voltages. But if we compare them separately from the endo and the apicardium, we see that the apicardial voltages were higher than the endocardial voltages, and that was the case for the single potentials and also for double potentials or fractionated potentials. And then the different type of electrogram of the recording modes were also compared for on the endo and apicardium. And there we found that 75% of the endocardial low voltage areas have normal apicardial voltages. So what does it mean? If we target on the endocardial side, the low voltage area, you're also ablating healthy apicardial tissue. And this is nicely shown in these maps where the white crosses indicate areas of electrical asynchrony between the endo and the apicardium, and the white lines indicate the areas of low voltages. And here you can see it when you do omnipolar voltage mapping. In this example, you see it on the endocardium and not on the apicardium. And in this bipolar voltage map, you again see it on the endocardium but not on the apicardium. So this is important to remember. The next step would be that some of these studies do not only use voltage mapping during sinus rhythm, but there are also studies who reported a voltage mapping during atrial fibrillation. And then the discrepancy between the endo and apicardium voltage becomes more important, and this is what you see here. So if you just compare the colors, you can see that there's quite a difference between the endo and apicardial voltages. Also there, we do not know what it exactly means and why we should target either the endo or apicardium low voltage area. So how should you measure voltage mapping during F on its own? I think that is already not known. There is a spatial temporal variation in the voltages, and there is no consensus how we should measure this. And then as a last step, I would like to remind you that voltage cannot be easily translated to just scar tissue. In the study where we did endo-apicardial mapping of the left atrial appendages, we clamped the tissue between the electrodes. We then cut off the electrical appendage. We marked where the electrode was positioned on the appendage, and we then performed a histological analysis. And surprisingly, we found no correlation between areas of fibotic tissue and low voltage areas. So what can we learn from unipolar mapping? So there were no differences between endo and apicardium voltages during sinus rhythm. That is in general, but the localized, if you really look into the localized areas, you see that there is a difference. Patients with paroxysmal atrial fibrillation have low voltage areas, and particularly located at Bachmann's bundle, which is quite an inconvenience because that's a structure which can only be accessed from the apicardium. There is no simple relationship between unipolar, bipolar, and omnipolar voltages and conduction velocity, which makes the role of what exactly the substrate of resistant atrial fibrillation is quite questionable. But we do see differences even during sinus rhythm. And also very important, low voltage cannot just be translated to fibrotic tissue. So I think there's still a lot of things to learn. But I think the most important thing is that we still don't understand what exactly the mechanistic relationship of low voltage areas is. Because we can compare what the difference are between patient with and without atrial fibrillation. We can target it during an ablation procedure and see whether AF terminates. But I think the mechanistic relation why actually low voltage areas are responsible for perpetuation of atrial fibrillation is not yet understood. So if you're more interested in this topic, I can recommend you this era position paper where we go in detail into all the aspects of voltage mapping. So I'd like to thank you for your attention. Thanks a lot. Sorry for being a bit late. I'm Helmut Purafil and I was already introduced. I was a bit late because of another session. But nonetheless, it's a pleasure for me to call the next speaker for his presentation. And it's Dr. Yamaguchi. Thank you for your kind introduction. I'm Dr. Yamaguchi from Saga University, Japan. I'd like to talk about histology of the atrial biopsy samples and its relationship to voltage. So far, we performed 1,000 cases atrial biopsy in 1,000 cases without any complications. So we got a lot of new findings in terms of histology. So I'd like to share my knowledge today about atrial biopsy histology. So atrial structural remodeling progresses along with atrial cardiomyopathy. So you know LA size increases, LA strain decreases. And we electrophysiologists know left atrial voltage decreases or low voltage area frequently appears along with structural remodeling. And the mortality rate, heart failure rate, stroke rate, recurrence rate increases. And histological evaluation in autopsy cases shows a diffused fibrotic process. Even though we always find the low voltage area locally, it looks like a local phenomenon. But histological evaluation shows a diffused fibrotic process in autopsy cases. To examine the discrepancy between histology and the voltage mapping, we performed high density voltage mapping study by H.D. Greed. And we analyzed left atrial mean voltage, global LA voltage in 140 AF patients and classified into quartiles, Q1 lowest and Q4 highest. And we also analyzed control group without atrial fibrillation. And we also analyzed regional LA voltage including anterior wall, roof, posterior wall, atrial wall, and inferior wall. And regional LA voltage uniformly decreased in all regions as global LA voltage decreased, suggesting voltage reduction is a diffused process. So the regional voltage at the anterior wall linearly correlated not only with global LA voltage, but also regional voltage at septum, roof, inferior wall, posterior wall, and lateral wall. And why does a low voltage area appear locally, like anterior wall or septum? So look at the patient with low voltage area less than 0.5 millivolt. Patient with low voltage area was only classified in, identified in Q1 quartile. Q1 quartile is our lowest LA voltage group. So the presence of low voltage area less than 0.5 millivolts suggest the patient had a diffused low, diffusely low voltage. So why the low voltage area frequently appears, locally appears, anterior wall and septum? Look at the Q4 group and the control group. In patients with Q4 group and the control group, so all patients had a higher voltage, but the regional voltage at the anterior wall, septum, red-yellow, significantly lower than the regional voltage at the posterior wall, lateral wall. So suggesting the regional voltage at the anterior wall and septum is innately lower than other parts. That's why a low voltage area locally appears, predominantly appears at the anterior wall and septum, along with global voltage reduction. So voltage reduction is a diffused process. So we challenged atrial biopsy from the limbus of fossil ovaries, the right atrial septum here under ICE guidance. You can see limbus of fossil ovaries here, and you can see biotome here, and we got atrial samples like this. So our first and second biopsy showed only white, grossly white samples. That means only endocardium because the endocardium atrium is thick, very thick, thicker than we expect, about, in some cases, one millimeter. So we needed to remove the endocardium first, and then we can dig a hole and reach to the myocardial layer here, and you can get a myocardial layer sample, like red samples. So the histology shows there are histological factors associated with voltage reduction is not only fibrosis, but also increased intercellular space, myofibrillar loss, and a reduced nuclear density. All these factors are significantly associated with voltage reduction, bipolar voltage reduction. Look at these examples. This patient had a very normal histology, and he had high voltage in left atrium and also biopsy site. This patient had a severe fibrosis, severe intercellular spacing, myofibrillar loss, and severely decreased nuclear density and amyloid deposition. All these patients had a low voltage in the left atrium and also at the biopsy site in the right atrium. So let's see this case. This patient had 60 years old and with proximal a-fib, and he had a low voltage area at the anterior wall here, and less than 0.5 millivolt. And after PBI, he had a micro-reentrant atrial tachycardia pedometrophilata, and I ablated the low voltage area and terminated this atrial tachycardia. In the past, electrophysiologists probably considered the histological background of voltage reduction in these patients is fibrosis. But actually, his histology showed very minimal fibrosis, but he showed very severely decreased nuclear density, like very small numbers of myo-nuclei here. This is compared to normal patients. The same scale. This histology is from 35 years old patient without a-fib. So you can recognize severely decreased nuclear density in this patient. This is a primary histological change in this patient, this patient with low voltage area. So look at examples of cases with recurrence after ablation, recurrence as persistent a-fib. Even after ablation, this patient had a recurrence, persistent a-fib recurrence. The histology showed very severe fibrosis interstitial changes, fibrosis, fibrosis, fibrosis, compared to control patients. So suggesting that recurrence is not our fault. This is primarily due to histological change of these patients. Even when we use PFA or very nice, sophisticated devices, I think the recurrence is due to patient's histological background. So we cannot cure histology. So this slide showed a summary of histological change and structural remodeling and interstitial change. Cellular space, fibrosis, myofibular loss, decreased nuclear density, and amyloid deposition are all associated with voltage reduction. And along with voltage reduction, that means interstitial remodeling, low voltage areas appears, frequently appears, fractionated electrograms appears increased, slow conduction zones increase, macrolentrant at the tachycardia induced, and the atrial tachyarrhythmia recurrence increases. So amyloid deposition is another important factor. This is an ATTR case. And this patient had a very low voltage and atrial tachycardia in the second session, the atrial tachycardia around lateral reach. And the histology showed amyloid deposition, you can see here, and amyloid typing identified AL amyloidosis in this patient. This patient also had amyloidosis. Here you can see the red deposit here in Congo red staining, and the histology amyloid typing identified ANP amyloidosis. It is so-called isolated atrial amyloidosis. We performed 578 atrial biopsy in a few ablation cases. And 40 patients among 578 cases had amyloid deposition, about 7%, 7%. And their type is mostly ATTL and AL, 15%, and ANP, 5%. And half of the patients with atrial amyloidosis had ventricular amyloidosis, and mostly 85% ATTR. And a patient with low voltage area, less than 0.5 millivolt, and over 60 years old, 21% of those patients had amyloid deposition in atrial samples. And if patient is more than 80 years old, and a low voltage area, less than 0.5 millivolt, 31% of those patients had amyloid deposition in atrial samples. And also, if patient had LV posterior wall thickness more than 12 millimeters, and those patients, half of them had amyloid deposition in atrial samples. So amyloid deposition are higher, prevalence of atrial amyloidosis is higher than we expected. And a patient with amyloid deposition, atrial amyloidosis, had poor outcomes in terms of cardiovascular events, including heart failure, hospitalization, death, and stroke, than patient without atrial deposition, amyloid deposition. The histology of the atrial samples are summarized in this review article. I hope this review article will be helpful for you. Thank you for your attention. Thank you very much, Dr. Yamaguchi. Our next speaker is Dr. Kassar from the University of Washington. He's going to talk about epicardial adiposity and left atrial functional and electrophysiological remodeling. So, good afternoon, okay. Good afternoon, everyone. Thank you for the kind introduction. So, my name is Ahmed Qassar. I'm from the University of Washington Medical Center. And today, we're going to have a small discussion about epicardial adiposity and left atrial functional and electrophysiologic remodeling. So, I do not have any pertinent disclosures. So, as a brief introduction to our talk today, we know that epicardial adiposity is usually located between the visceral pericardium and the atrial myocardium. So, during early life, we know that this epicardial adiposity has cardioprotective and thermogenic functions. However, as we age, there is a loss in this thermogenesis and more secretion of pro-fibrotic and pro-apoptotic factors, which ultimately lead to various pathological conditions, which are not restricted to atrial fibrillation, but also include coronary artery disease, diabetes mellitus, and heart failure. So, how does epicardial adiposity do so? It's usually through an increased inflammation, as well as deposition of fibrosis and decrease of cardiomyocyte continuity, as well as the presence of ganglionated plexi, which increase autonomic output, ultimately leading to both electrophysiological, structural, and functional remodeling, which is depicted by the presence of re-entrant circuits and prolonged action potentials, which form this niche of atrial fibrillation substrate. Now, in terms of assessment, how could we assess epicardial adipose tissue? And there are multiple ways and multimodality imaging that we could resort to in order to assess epicardial adipose tissue. Now, usually at the University of Washington, what we resort to is cardiac magnetic resonance imaging, and specifically, Dixon sequences, which allow us to highlight epicardial adipose fat, as you can see here, in white against a background of black, which depicts blood, the blood pool, as well as the atrial myocardium. So, in these slices, you could notice that we usually segment and contour in a semi-automated manner epicardial adipose tissue in blue, which allows us to have a volume of epicardial adipose tissue. Now, moving on to atrial function. So, very similar to how we assess epicardial adipose fat, we resort to multimodality imaging. So, we could use echocardiography, CAT scans, and cardiac magnetic resonance images, which have been previously validated across multiple cohorts. And this is some of our work at the University of Washington, where I show on the left side of the panel four-chamber and two-chamber cinematic view images of cardiac magnetic resonance scans, where we contour the endocardium and epicardium across the entire cycle in order to be able to generate strain curves as shown in the middle panel. And this allows us to plot strain graphs where we could extract reservoir, conduit, and booster pump strain functions of the left atrium. Now, this is some of our work at the University of Washington, where we wanted to look at the association between function of the left atrium as depicted by atrial strain and the presence of epicardial adipose tissue. So, we resorted to cardiac magnetic scans, where we recruited 101 patients undergoing AFib ablation at the University of Washington. And in a very similar manner, what we did is that we contoured the left atrial endocardium and epicardium in both two-chamber and four-chamber cinematic views in order to be able to generate strain curves. And what we noticed is that there we had a negative correlation between global longitudinal conduit strain and left atrial epicardial adipose tissue volume. Now, switching gears to the other discussion that I would like to talk about, and this is the relationship between epicardial adipose tissue and electrophysiologic remodeling. So, very similar to the function that I spoke about in terms of the pro-apoptotic factors that are secreted by epicardial adipose tissue, this not only leads to structural and functional remodeling, but also leads to electrophysiologic remodeling through ion channel and gap junction modulation as well as with the deposition of fibrosis. This creates a pro-arrhythmic myocardial remodeling. Now, a very interesting manuscript that was published by Nalia et al in JAG in the year of 2020 looked specifically at the secretome of the epicardial adipose tissue, and as well as the structural remodeling and functional remodeling, which I touched base on, they noticed electrophysiologic remodeling depicted by prolongation of action potential, conduction velocity slowing, as well as connexin 40 lateralization and intramyocyte disruption, which led to aberrant excitability and conduction heterogeneity. Now, this is one of the initial work that we did at the University of Washington related to epicardial adipose tissue and structural remodeling as depicted by atrial volume and fibrotic remodeling. We utilized LGE MRI scans in order to highlight fibrotic areas in green, as you can see here, and healthy atrial tissue in blue, and what we did on next is that we segmented left atrial epicardial adipose tissue using cardiac magnetic resonance scans, specifically the Dixon sequences, and superimposed the epicardial adipose tissue on top of the left atrial cells. Now in terms of the findings, so what we noticed is that there was a strong correlation between structural remodeling depicted by atrial fibrosis and left atrial dilation, as well as left atrial epicardial adipose tissue index, even after stratifying the patient for body mass index. Now to move on further a bit, we also wanted to look at the correlation between voltage abnormalities, and specifically unipolar and bipolar voltage, and this is why we wanted to look at endocardial mapping where we contoured low voltage zones, so we looked at bipolar voltage and chose a threshold of 0.5 millivolts, and unipolar voltage and chose a specific threshold of 1 millivolt, and what we noticed is that there was a significant correlation between both low voltage zone areas, unipolar and bipolar voltage, as well as left atrial epicardial adipose tissue volume index. Now in a subsequent analysis, we wanted to see how does this epicardial adipose tissue or does this epicardial adipose tissue predict AF recurrence following ablation. We ran a sensitivity analysis and noticed that patients who had increased epicardial adipose tissue had more recurrence, faster recurrence within the first year post-ablation therapy. In a subsequent analysis that we published in Circulation, Arrhythmia, and Electrophysiology, we wanted to see how does this epicardial adipose tissue lead to AF recurrence. So our initial hypothesis was that this recurrence is usually due to post-ablation fibrosis, and our secondary hypothesis was that we wanted to see if thermal energy affects epicardial adipose tissue. Now there was no clear association between epicardial adipose tissue and post-ablation scar formation. However, what we noticed is that due to possible thermal ablation, we noticed a significant reduction in epicardial adipose tissue volume from the pre-ablation to the post-ablation group. However, we can notice that the BMI was almost constant in the patients, and specifically the form through which we noticed epicardial adipose tissue volume decrease was in the form of disappearance in some areas as we can see here in panel C, as well as a decrease in thickness as we noted in our spatial analysis. Now in conclusion, what are we trying to deliver here in regards to epicardial adipose tissue? It's clearly that it's not a static substrate of atrial fibrillation. It's a dynamic, metabolically active component of atrial fibrillation. We have shown previously that an increase in left atrial epicardial adipose tissue volume leads to more impairment of atrial function as shown in the paper that we published with a reduced strain. Previous literature showed a decrease in conduction velocity with the increase in left atrial epicardial adipose tissue, and we were able to show that we had a decrease in voltage and an increase in low voltage zones whenever we had more epicardial adipose tissue in patients with atrial fibrillation. So where do we want to take all of this work on epicardial adipose tissue and atrial fibrillation? So we talked about global epicardial adipose tissue, but we want to resort to multimodality imaging in order to investigate regional distribution of epicardial fat and how does this affect the mechanistic function of the left atrium in certain areas and not just the global function of the left atrium as depicted by reservoir, conduit, and booster pump, but also look at specific areas such as the appendage. We would also like to tie that to outcomes of atrial fibrillation, which include stroke, for example. How does epicardial adipose tissue disrupt the mechanistic properties of the left atrium and how does that feed into thrombus formation and stroke? And ultimately, Dr. DeGroote touched base on the multiple thresholds of bipolar and unipolar voltages. So across the literature, there has been a lot of discussion in regards to bipolar voltage of using a 0.5 millivolts, but is 0.6 healthy atrial tissue? Is a 0.7 millivolts healthy atrial tissue? And what about unipolar voltage? I'd like to thank you all for your time. I'd be more than happy to answer any of your questions. Thank you. And so this is now the last presentation for this session, endocardial voltage mapping and outcomes from Dr. Masuda. Okay, thank you for inviting me for such a special session. It's my honor to present here. My topic is endocardial voltage mapping and outcomes. Today I would like to talk about two topics. First, let me start with low voltage areas and clinical outcomes, results from a retrospective cohort study. Atrial myopathy is known as underlying pathophysiology in the majority of AF patients and to be possibly related to development of complications of AF such as heart failure and stroke. Voltage mapping during the AF aberration procedure depicts the atrial myocardial degeneration including fibrosis as low voltage areas, LVS. However, few studies have examined the long-term prognostic impact of atrial myopathy in AF patients. The purpose of this study was to delineate differences in long-term prognosis among patients stratified by LVS size. This was a single center retrospective observational study including consecutive patients who underwent initial AF aberration. The study included 1,488 patients who underwent initial aberration for AF. The patients were divided into three groups according to LVS size, 1,136 patients for no LVS and 250 for small LVS that was 20 square centimeter, less than 20 square centimeter of LVS size, and 102 for extensive LVS that was 20 square centimeter or more. Clinical outcomes including heart failure, stroke, and death were followed up for five years. Let's move on to patient characteristics. Patients with more extensive LVS were older, more likely to be female, had lower body mass index, more likely to have persistent AF, and larger left atrium. And patients with more extensive LVS underwent more frequently non-pivotal aberration, left atrial roof and bottom lines, CTI aberration, and LVS guided aberration. So let's look at the risk outcomes. As we all know, patients with more extensive LVS had poor risk outcomes, higher FAT recurrence. And this is a main result of this study, composite endpoint of heart failure, stroke, and death. Patients with more extensive LVS had poorer composite endpoint outcomes. As you can see, the risk of composite endpoint increases as LVS size becomes more extensive. So this slide shows instance of each endpoint. Annual incidence rate of heart failure and stroke increases as LVS became more extensive. And this slide shows prediction of composite endpoint using multivariate analysis. Clinical factors independently associated with composite endpoint were diabetes, heart failure, low EGFR and AFAT recurrence, and LVS presence. So looking at the pathophysiological association between atrial myopathy and heart failure development, there would be three possible mechanisms why atrial myopathy was associated with heart failure development. First, atrial myopathy acted as a marker of ventricular dysfunction because advanced atrial remodeling reflects ventricular dysfunction through long-term elevated atrial pressure. In addition, several upstream factors causing atrial myopathy impair ventricular function as well. And second, atrial myopathy acted as a predictor of AFAT recurrence, which can lead to heart failure development. And finally, atrial myopathy might act as a cause of heart failure development, possibly due to increased left atrial stiffness, mitral regurgitation, and ANP deficiency. Next, the association between atrial myopathy and stroke, which can be explained by virtual triad. Patients with extensive LVAs would have states of blood flow and endothelial injury in left atrium, resulting in increased thrombogenesis in left atrium and increased risk of stroke. So in conclusion so far, LVA extension was associated with increased composite endpoint of deaths, heart failure, and stroke. So let's move on to the next topic, efficacy of low-voltage area ablation results from randomized controlled suppressed F-trial. This was an investigator-initiated multi-center open-label randomized controlled trial. The purpose of this study was to evaluate the efficacy of LVA ablation in patients with persistent AF and left atrial LVAs. The study included persistent AF patients undergoing initial AF ablation, and the patient had LVAs five square centimeters or more. The patients were randomly assigned to PBI plus LV ablation group or PBI alone in a one-to-one fashion. The primary endpoint was one-year AF80 recurrence free rate after initial ablation. And the flow is shown here. 1,347 patients underwent voltage mapping after PBI. Among them, 342 had left atrial LVA. These patients were randomly assigned to PBI plus LVA ablation group in a one-to-one fashion, and final number of patients assigned to each group was 170 for PBI plus LVA ablation group and 171 for PBI alone group. During one-year follow-up, 10 and 7 patients dropped out, resulting in protocol completion rate of 94.7 percent. This is the baseline characteristics. First of all, this was a randomized controlled trial, and there was no difference between groups. Well balanced. Mean age was 74 years old. Nearly half of patients were female. 20 percent of patients had long-standing persistent AF, and the left atrial diameter was 43 to 44 millimeter. Total procedure time was 30 minutes longer in PBI plus LVA ablation group, and LVA size was 13 to 14 square centimeters. Complete LVA elimination was achieved in 78 percent patients assigned to PBI plus LVA ablation group. Ablation of regular AT was more frequently performed in PBI plus LVA ablation group, 70 percent, which was 7 percent in PBI alone group. So now I'd like to present the result of primary endpoint, AFAT recurrence. As you can see all on left-hand side panel, Kaplan-Meier curve appear to show superiority of PBI plus LVA ablation group compared to PBI alone group. We observed 11 percent difference. However, this was not statistically significant. On the other hand, if we look to AF recurrence on the right-hand side, the superiority of PBI plus LVA ablation group become a little bit more obvious. However, the difference was not statistically significant. So the pie chart shows AT recurrence as a proportion of total recurrent arrhythmias. As you can see, AT recurrence rate in PBI plus LVA ablation group was 36 percent, which was twice as high as 18 percent in PBI alone group. So if we look to subgroup analysis, we found several subgroups demonstrating the efficacy of LVA ablation. That were patients aged 75 years old or more, left AT diameter 45 millimeter or more, LVA size 20 square centimeter or more. These results suggest patients with advanced atrial remodeling would benefit more from LVA ablation. So in conclusion, LVA ablation in addition to PBI did not reduce one-year AF recurrence in patients with persistent AF and left AT LVS. Numerical superiority was observed in patients with adding LVA ablation to PBI, suggesting the worthwhile benefit of LVA ablation for patients with persistent AF. This study was conducted in eight centers in Osaka, located in Osaka, and I'd like to thank all investigators for the completion of the study. And the results will be published in Nature Medicine next week, April 30. Thank you for your attention. Fantastic. Thank you for all the speakers for actually keeping us on time. And now it's time for questions and discussion. Please come up to the microphone if you have a question or send it over via the app and we can read it out. Go ahead, Dr. Kalman. Hi, thank you. Great session. Just a question for Natasha. Wonderful data. And, you know, interesting that you find no relationship between voltage and fibrosis, which is, I think, is really important. Is there, I have two questions, is there, you know, have we just got the voltage cut point wrong? Obviously, it's with a variable wall thickness, it's very unlikely to be a single cut point. Did you find a relationship between conduction and complex signals and fibrosis? Is that a better marker? Should we be looking at something else rather than just using these very simple voltage cut points? Yeah, we also analyzed the relation between the fibrotic tissue and fractionation and conduction abnormalities, and also there I did not find a relation. So what we measure, what we call electropathology, it might be something different, and the person I collaborate with is Professor Brandel from Amsterdam. She looked into, in successive papers, what actually is changing in the age of structure of these patients, and she found mainly DNA damage and proteostasis as a role in the perpetuation of AF. So it might be that we think too much, it's fibrosis, but it could be something like DNA damage. Interesting, thank you. Yeah, it's like a very great session, a question practically for all speakers, and related as a follow-up of John Kalman's question, how exactly, when we're analyzing electrodes from AP or from endocardium, how exactly it's matching on the layer? Because all representation, it's not represent clear AP, middle, endo layer of fibrotic tissue, as well as adipocyte infiltration. Is it, Natasha, is it affecting adipocyte infiltration, also affecting conduction abnormality? Because when you map from apicardium, you have definitely more also fat infiltration from apicamin. Yeah, I could imagine that is the case from a theoretical point of view. But we found lower voltages on the endocardial side and not on the apicardial side. So there's a discomfiture. I could not explain it in this way. Yeah, and how regional difference seen, because, again, it's regional in terms of right atrial appendage, which is commonly used initially than left atria, posterior wall versus appendage. There are whole regional differences exist in the human atria and especially in remodeling. It's patient-specific. Yeah, exactly. But I think there the thickness of the myocardial bundles also play an important role. And that is sometimes, if you can, you can have a low voltage, which doesn't mean that it's thyroidic tissue, but it's just a small bundle which is activated. And particularly on the right atrial free wall, you have the multiple bundles which goes in multiple directions. I can imagine that will affect the potential voltage tremendously. And perhaps that you measure on the apicardium, which is a smooth layer, that you measure the summation of everything which is transmurally. But yeah, to know that for sure, then we'd love to measure it transmurally. But then, yeah, you cannot measure voltage anymore. Yeah, fiber inpatient, maybe, and the difference in my fiber. I think it's one of the major contributor. And I guess both fibrosis and fat infiltration affecting on this twist. Yeah, and I can imagine if the fiber direction changes transmurally, that it might also cancel out. And that, therefore, you could also measure smaller voltage. Thank you. And for, I may not pronounce correctly your name. Yamaguchi. Yamaguchi, Yamaguchi, Yamaguchi. So you show and publish multiple cool stuff. When you're taking biopsy, you're taking always from some specific region. So how, again, the question, if you have any option, compare your biopsy regional, because indefinitely your voltage coming from entire atria, but the tiny snapshot which you're taking are very regional. And it may not represent the entire remodeling in this human patient. Yeah, it's a very important problem. But based on voltage electrophysiological study, we proved that voltage reduction is similar. It's a uniform process in left atrium and also right atrial septum. So that suggests that histological change is also uniform. But we have not confirmed it in human. But only in autopsy cases, we confirmed the histology of the right atrial septum is similar to other part of left atrium. So, but not confirmed in human. And as far as I understand, it's more subendocardial biopsy. Again? Subendocardial. Yeah, subendocardial, it's more subendocardial. We naturally have more fibrotic tissue in human atria. Thank you so much. Hi, from Metro Health in Cleveland. Question for the last speaker. So the data, I think, suggests that the main issue was ATAC after, if you take away the ATAC, it looks like you guys have a positive side, which probably there is, it looks the curves are split. So just slightly underpowered for the effect. But I'm wondering about your, so ATAC suggests that may have the right area, just not fully burned or whatever you did to it. And so the question is, how did you deal with block? So especially, you know, across the roof, across the anterior wall, a lot of this low voltage is in the anterior wall. So how did you guys deal with that? Okay. Thank you. Important question. So first of all, if the sample size was double, the difference, the difference became statistically significant. We confirmed that. And your question, the weak point of this study was the protocol did not specify what to do if we found slow conduction isms after low voltage aberration between metal isms. So that's why ATAC carrier recurrence was relatively more. So if we eliminate this heterogenic AT, the benefit of low voltage aberration become higher. So we have a similar study, actually, that we're doing. You guys beat us, but hopefully it will be finished soon. But we, we, we specified that you have to demonstrate block across the anterior wall, whatever you, you, and if you can't leave an isthmus, and we're hoping that we'll see less ATs, but that ATs is kind of the Achilles heel of this. Yes. Thank you. Thanks. I have a question to Nastassia, you know, looking at what you what you find and the answers you give to Jonathan's question means. We can't say anything about low voltage with regard to fibrosis, with regard to conduction velocity. So that's what I take from your answer. However, what we have clinically is for many years now that if we have low voltage, so-called low voltage areas, the prognosis in these patients for recurrent atrial fibrillation after ablation, for example, is worse. That's clear, completely clear. The other thing what we have is if we ablate, if we look for tachycardias, when they have tachycardias, regular tachycardias, then they have tachycardias between these scars and an anatomical boundary. So there must be something in. It can't be just nothing. So how do you interpret these clinical findings that we have with your results? Yeah, I get your question. And indeed, what I think it's not only low voltage, because like I say, of all the approaches we have so far for persistent AF ablation, there's always something. But as soon as we're going to apply a methodology, whether it's drivers, whether it's fractionation or voltage, if you're going to apply it to a huge amount of patients, it fails. So, and particularly because we also found like conduction velocity, high conduction velocity in low voltage area, perhaps it's something like the CHED-VAS score. If you take from all the electrical parameters and low voltage and slow conduction and conduction block or fractionation you added to it, then you might find that it's more specific. And on the other hand, I do believe that voltage is influenced by so many variables. It needs to be improved in order to get better results or better outcomes. And then again, I think we still haven't answered the question. I agree with you, Ed. So that low voltage, we also see it very clearly in patients with congenital heart disease, for example. There we know for sure if you have low voltage areas, you have interatrial tachycardias. If your atria dilates, you also have less tissue beneath your recording electrode. So isn't that a measure of atrial dilatation that we are measuring? But it doesn't mean that your conduction is affected. And still, then my question is to you, how does a low voltage area, how is it responsible for perturbation of atrial fibrillation? Is it just the core around which a re-entry circuit goes? Now, I don't think so because we rarely measure a real re-entry during AF. So what is it? And I think that's the question we have to answer. And combining it with different parameters, I think that's the way to go. I have a question. You've been so silent the whole time. On the histology that you showed us, it seemed like you mentioned there was no fibrosis, but there was bigger extracellular space. And there's also an increase in the myocyte size. First question is, is that empty space? What fills that space? And is the low nuclear density and bigger size something that causes a low current density? And is that why we're having low voltage or fractionated signals? Yes, thank you. Of the increased cellular space, there is a plasma component. It means edema. And increase in intercellular space precedes fibrosis collagen deposition in our observational study. So edema is the main cause of intercellular spacing. And I think that our early intervention may stop fibrosis to fill our intercellular space. And nuclear density is a very important histological factor, especially for elderly patients. And a decrease in nuclear density is significantly associated with voltage reduction, but not for fractionated potential or local slow conduction zones. So I think voltage is representative of residual myocardial mass. So fibrosis, intercellular space, myofibular loss, nuclear density, amyloid deposition, all associated with reduction in myocardial mass. So reduction in voltage represents all of the histological change and the residual myocardial mass. So you're saying that there are less nuclei and there's less overall myocardial mass? Yes. Despite a larger individual myocyte size? Yes. Atrial size, myocardial size is closely, very closely associated with nuclear density, decrease in nuclear density. So if the nuclear density decreases, myocyte loss increases. Residual myocyte, each myocyte increase, hypertrophic. So hypertrophic change of myocardial myocyte is not positive remodeling, but it's a compensation for loss of myocytes in our study. Do you have a question? No, sorry, just a comment. I was going to say that anyone who's ablating low voltage will tell you that there's not, it's not, low voltage is probably the wrong term because it's low voltage plus fractionation. That we always go after and see. So it's probably a misnomer. And not all low voltage is the same. So I think the, throw that in there. Sorry. If I may reply to that one. If I may reply to your remark, we recently published a study in Jack where we also demonstrated that if you take a look at the fractionated electrograms, that it's not per definition, low voltage fractionated electrograms, that you actually have, even around areas of conduction block, you have very high voltages. And still you have fractionation because there's block. And that's been described as discontinuous conduction by Rohr in somewhere in the 1980s. So that's also a different phenomenon. But also I would add that all of the randomized trials to date, including what we've heard today, did not take fractionation into consideration. They're just based on a voltage cut point. And I agree with you that I think that we need to target, we need to have a better way to understand this. But the randomized trials are based on a single voltage. I agree. But I guess the special sauce is the fact that as an operator, when you're looking at it, you're discerning. But I think it needs to be defined. Well, this sounds like a finishing phrase. My feeling is we are confused, but on a high level. And that's my finishing word. Thank you for this presentation. It was really, really very good. And thanks for coming and have a nice conference still. Bye bye.

atrial myopathy

voltage mapping

atrial fibrillation

low-voltage areas

atrial remodeling

epicardial adiposity

amyloid deposition

pulmonary vein isolation

SUPPRESS-AF trial

cardiovascular outcomes