CNA Conference - Dr. Narayanguru Kowlgi - Experimental CNA at Mayo Clinic

Uh, next, I'd like to welcome, uh, Doctor, uh, Guru Kagi. He's attending clinical electrophysiologist and assistant professor of medicine at the Mayo Clinic in Rochester. Uh, he's actively involved in various ablation experiments and techniques, techniques, uh, AI and electrophysiology, and obviously neurocardiogenic syncope. Uh, his presentation will be telling us more about experimental cardio neuroablation at the Mayo Clinic. Everyone, my name is Guru Kogi. I am an assistant professor of medicine and a senior associate consultant in cardiac electrophysiology here at Mayo Clinic in Rochester, Minnesota, and it is my distinct pleasure to be invited to present the experimental cardioural ablation at Mayo Clinic, and I want to thank Dr. Peron and Dr. Clark for the kind invitation, and I'm very happy to be here. I have no disclosures. So a little bit about my background. I did my medical schooling in New Delhi in India, and then I went to the University of Connecticut where I did internal medicine and chief residency. From there I went south to Virginia Commonwealth University, did my cardiology fellowship and chief fellowship there. And then ended up at Mayo Clinic for my EP fellowship and stayed back on staff after that. I also completed a Master of Science in artificial intelligence in healthcare and I was extremely fortunate to spend some time with Dr. Pashon and his team in Sao Paulo last summer. where I learned a lot about the intricacies of cardio neural ablation and learned from some of the best people in the field. So I like to use that experience to sort of delve into what we've done in the translational lab at Mayo Clinic and hopefully we all can learn something from each other in that process. So people are often ask me this question, why is translational work needed? So we are at a point where we've been doing clinical work for a while. So why do we need to do work in animals and understand this better? So essentially we do understand the benefit of cardio neural ablation and various applications, but there are many variations, as we've heard today, of how these are done. So variations in how we localize the ganglia, what are the ideal ablation energy settings? Do we ablate endocardial versus is there a benefit in doing epicardial ablation? What kind of energy source is ideal for this, for increasing efficacy and improving safety outcomes? And how do we check our endpoints to know that we've played it adequately or not? Beyond that, you know, is there any role for sympathetic modulation and cardiac arrhythmia treatment? And what about ventricular arrhythmias? Can we treat those with cardio neural ablation? So really we have the capability of pushing the boundaries of what we know already to learn new principles about CNA. So I feel there is always a space for translational work. For today, because time is limited, I'm going to focus on energy source because that is a hot topic indeed and we've heard a lot about PFA and AFib. So I'd like to show you some of the findings that we've seen in our acute for experiments. So speaking of electroporation, so it is the application of pulsed electric fields or DC current for short periods of time to create membrane pores, and that's how it creates a cell death. So there are reversible and irreversible forms depending on how much energy you're applying and if the pores are completely generated leading to nuclei destruction or not. So the advantages are supposed to be high tissue specificity. So with regard to AFib, for example, we can ablate the pulmonary vein antrum, the myocardium specifically, by sparing some of the other structures to reduce collateral damage. And because there are irreversible and reversible forms, we could potentially apply less energy in areas of concern, such as when we are close to the conduction system, and then see if we can get damage while sparing some collateral structures. So the work for this has been going on for a while at Mayo Clinic. I've been fortunate to partner with some colleagues who've been working on this for many years, Dr. Asharwattem, Dr. De Simone, Dr. Padma Nahand, who authored this paper. So this particular one in 2019, they came up with a novel catheter design to map in the epicardium and then a blade both in the transverse sinus as well as the oblique sinus, and they had different catheter designs, as you can see. So the one in the transverse sinus was linear, and they had this two-pronged catheter to go into the oblique sinus posteriorly, and then they used some of these fractionated signals to identify areas of ganglia in addition to what we know anatomically. Applied unipolar pulsed field ablation here, 1000 volt, 100 microsecond pulse width, and just 10 pulses, which is, you know, quite less compared to what we're doing with PBS these days. And then they looked at pathology to see if they achieved cellular destruction or not, a ganglia destruction or not. And then the areas were kind of the usual suspects, so in the oblique sinus and the inferior left inferior right in the venal martial region, superior left area, and then in the transverse sinus you could access the aortic cable ganglia as well. So you can see on the left here that this is a control in panel A where you can see the ganglia, I'm going to zoom in over here. There's intact ganglia in this slide, and then moving to the right we can start to see some damage to the ganglia over here. Tyrosinous staining, and on the right we can see some further destruction and a different magnification. And with H&E staining here you can see there's damage to the ganglia with nuclear nuclei pattern within the ganglia. So there's the nuclei sort of merging into each other. And then to the right here we can see that there's some minimal myocardial damage and the ganglia completely destroyed. So this is to show that it is possible to be selective, selectively destroy the ganglia while sparing the myocardium, which is our goal. And then uh these are uh uh my and traum straining here for the esophagus showing intact esophagus when abating in the oblique sinus and the posterior wall. And intact myocardium in other areas as well, with some fibrosis seen near the ganglion. These were canine experiments and they were survived for 4 months. So, so both acute and chronic changes seen with the PFA in the ganglion. So in our studies we hope to replicate some of this, but then try to do the PFA endocardially because because again, doing an epicardial ablation for young patients with vasovagal syncope or those with Afib seems a bit aggressive if we can achieve the same effect endocardially as we, as what we've seen with radiofrequency. So a couple of points of differences in pigs versus humans is that I noticed that the site of stimulation to achieve a reproducible vagal response was more cordial compared to what we do in humans. So this is where we started out and went higher up even and did not see a good response. But then about 2 to 3 vertebral spaces lower is where I've seen consistent response in most of the pigs. Because of variations in anatomy. And we also did, I'm not going into details of this, but then we had, we performed those thresholds with increasing voltages and changing frequencies, and then what Dr. Pershon has recommended, about 1 volt per kilogram and 50 Hz, kind of worked for most of the pigs. And lately I've been going higher on the voltage, so even if the pigs are about. 30 kg. I'll go to 50 or 60 volts for voltage and I've seen a more reproducible response there. We compared these to direct nerve stimulation as well and noticed that there's less of a dose curve when it comes to endovascular stimulation because we tend to go higher on voltage to get a more reproducible response, whereas with direct stimulation we could see a dose response curve with more stimulation with higher outputs and then it plateaus after about 25 to 30 volts with direct stimulation. But we use RF ablation as control, as we know that these work in humans as well. And then similarly in pigs we saw when we updated in the standard areas we saw changes in cycle length were abolished after RF ablation. The decrease in heart rate stimulation was less obvious with post RF ablation, and then the drop rate at 20% baseline of vagal stimulation also went away after ablation. So this is just to show that RF fibrillation works similarly in pigs as compared to humans. But then what was interesting was when we tried to do PFA endocardially and then we used a much higher voltage than what was used epicardially, assuming we had to, you know, traverse more thickness of the myocardium, so we need a higher voltage, and we were OK. These were acute experiments, so we were OK causing some myocardial damage, but then we did want to see ganglia damage and we started lower and went up to 1500 volts, and we did some studies at 1750 volts as well. Uh, pulse rate of 20 microseconds, much lower frequency than what you know we've seen with pulmonary vein isolation experiments, but just to see if this was enough to creating GP ablation. And we did see that the vagal response did not go away. In fact, in this, in the first set of animals, we actually saw that there was more vagal response after PFA. Not clear why that happened, but then I think at that point we were still sort of fine tuning our initial stimulation, which may not have been the best. But either way we did not see the vagal response go away in this initial set. And In these, in some of these animals, we went back and did RF ablation in the same animals and then just in the right atrium, so aortic cable ganglia and then the ganglia opposite our GP-2 site in the anterior right, so higher in the septum, posterior septum, and then inferior right septum as well. And with the ablation there we did see the vagal response go away. So RF was still working, but PFA at those settings was not effective. And there was some degree of denervation because our AV nodal ERP numbers did go down with with PF ablation, as was seen with PF plus RF brillation, but interestingly our atrial ERP did not increase, you know, so we've heard that with vagal denervation, your atrial affective refractory period should go longer, sort of the rationale for their use in afib, but then we did not see this in most of the pigs we did the denervation for. So in gross pathology with RF lesion, there's clear tissue destruction, uh, and clear myocardial damage as well. Um, and, um, with PF lesions, these were the ones that were not effective, uh, not nearly enough uh damage as what we see with the close pathology. And as a microscopy slides to follow, but as we were fine tuning this, we thought we were getting some arching at higher voltage, so we came down to 1250, changed the frequency, kept the pulse rate the same, increased the number of pulses. And with this we started seeing some response with PFA. So we also made our vagal stimulation quite robust, so we got better at doing that. And then we were seeing that post PFA and in a sequential manner as we see with RF, we were losing our vagal response with extra cardiac vagal stimulation. And now we sort of seeing more tissue damage. On the left. You see ineffective PFA lesion. You can make out somewhat make out the outline, but on the right you can see clear lesions. In fact, here we are actually seeing some degree of hemorrhage as well. So this is epicardial hemorrhage seen in H&E stain here. There's epicardial inflammation and there's some collagen lysis as well. And then when you look at a higher magnification of the ganglia, the ganglia are not completely destroyed. There's some destruction, but some of the nuclei are still preserved. So again, we are getting better, but we're not at a point where we are not damaging the myocardium and getting pure ganglia destruction with the power settings we have. So we need to understand further how to titrate pulse field ablation, and there are different ways of doing this. All vendors use some kind of simulation software like this one here. So like I visited in Brazil, Dr. Pon's lab, I did get the opportunity to visit some of our colleagues in Madrid and Barcelona who work with simulation for pulse field ablation. And then it can be quite complex, and these are engineers who do this for a living, but you know it is something we can get better at, or at least we can incorporate into our imaging systems and mapping systems where depending on the tissue thickness for a specific individual, we can then. Decide what kind of PFA settings we need to penetrate the myocardium and to achieve an epicardial lesion, especially at, uh, for, for neural structures, because we can select the resistivity, the permeability, uh, and conductivity for each of these tissue elements. And then we can see what the tissue temperature is and then we can sort of keep it to a minimum to reduce tissue damage, but get the electric field we need for the threshold of the ganglion. So Limitations. So we know that endocardial PFA for GP is feasible, and we know that it can be as effective as radiofrequency ablation. And we know that PFA can be tissue selective, but as we've seen in PBI, you know, it's not something that we can just ramp up the power and then hope not to see any damage. It can be thermal. There can be thermal damage at higher power settings, and so we really need to tune, fine tune this and understand this better. So finding the optimal thresholds is more challenging for GP as we understand compared to pulmonary vein isolation. We do need more experiments and validation in other centers. Simulation software, in my opinion, should be integrated with imaging for easier workflow and also because every patient is different, you know, it's for PVI we have like a low setting, high setting. I don't think ablating GP would be that simple. We need to have much more flexibility in changing the. Settings as an EP community, we must demand that industry disclose PFA parameters. I know this is easier said than done, but then we need to know what works if someone figures out. So then we can do further study and try to optimize this better like we've done in the past for radio frequency and other energy sources. So I'll stop with that. Thank you so much for your attention. Again, it's a pleasure to be part of this conference, and I look forward to the other talks. Thank you so much. Welcome back. Uh, we've kind of been communicating with Doctor Kogi. Uh, he just got on an airplane, so we've been, uh, texting back and forth, uh a few questions, um. You know, this, we feel is, uh, there's still a lot to learn on PFA, uh, you know, what voltages, what frequencies to use, uh, and, and how they can affect the GPS. Um, we, we are really excited to see what additional research he can provide us on that. So, yeah, I agree with you, uh, Brian. I think, you know, he, he didn't have time to talk about this, but he had mentioned the use of computer software and, and simulation in this sort of, you know, in this endeavor. And I think we can even push that further. Maybe at the next Congress, we can talk about, you know, artificial intelligence and machine learning and deep learning and how it's gonna affect cardio neuroablation as something that's that's really on the, on the forefront.