My name is Olu Adujola from the UCLA Cardiac Arrhythmias Center. I'm Associate Professor of Medicine and I also serve as the Associate Director for the Arrhythmias Center. It's my pleasure to talk today about neuromodulation for ventricular arrhythmias theory and practice. I'd like to start by thanking Dr. John Miller and the rest of the session organizers for the kind invitation to be here today. Here are my disclosures and my funding sources. As well as an acknowledgement of my colleagues at UCLA who make it a terrific place to work and who have been really instrumental in helping develop some of the concepts that I will be talking about today. I'd like to start the talk today with a slide that is a clinical presentation that really underscores the importance of neuromodulation for ventricular tachycardia therapy. This is a 52-year-old male patient with non ischemic cardiomyopathy from prior drug use. He had a low ejection fraction at about 35% and he presented to his local emergency room with monomorphic ventricular tachycardia at 180 to 200 beats a minute. What's shown on this somewhat busy slide is the patient's hospital course over about the next 30 days or so. On the y-axis here is the severity of this VT. What you can see is that the patient, after being transferred to UCLA, had ventricular tachycardia ablation twice, more medications, more shocks, another ablation, treatment with enteriotic balloon pump, thoracic epidural anesthesia. It wasn't until we instituted left and right cardiac sympathetic denervation that the patient really had complete cessation of this arrhythmias, even without PVCs. We collected our initial experience in the first six patients in whom we applied this therapy showing really strong suppression of ventricular tachycardia. Now, moving on to my outline today, I've split it up into two parts. I'd like to talk first about the theory of neuromodulation for VT, briefly review neuroanatomy and how the nervous system controls the heart in health and disease. And then I'll move on to practice. We'll talk about some practical approaches to neuromodulation and some clinical considerations. To start with, I'd like to emphasize just how densely innervated the heart is. The heart is just chock full of nerves. What you're looking at in this video, which I'll replay again, is the heart of a small animal, but the same is true for larger mammals and humans. And what you can see are these red bundles here that are each indicating nerves that are in the heart. And you can see just how densely innervated the heart is. In addition, there are neurons that sit within epicardial fat pad that are on the ventral and dorsal surfaces of the heart. A magnified image of one of those ganglia is shown here, and you can see several neurons here with nerve fibers coming into the ganglion and also exiting the ganglion. And together, the nerves that you see and the neurons on the heart control the function of the heart, as shown here on the left, via a series of feedback loops, where information from the heart, shown here in blue, travels from the heart to various sites within the nervous system, even reaching the forebrain. That information is processed and then relayed back to the heart via two main limbs, sympathetic and parasympathetic. What this does is it ensures that the heart is working on a beat-to-beat basis without any perturbations, and if there are perturbations, it allows the heart, it allows the nervous system to respond to that. An example of what's shown here is in the bottom right here, where you can see individual neurons firing in a very specific timing relative to left ventricular pressure, shown here, and the ECG shown here. And you can see just how critically timed neural activity is to cardiac activity. These are recordings from the intrinsic cardiac nervous system, which are those ganglionated plexi that I showed you a second ago. Some brief terminology that will be useful for this talk. We will talk about the sympathetic nervous system, which is the flight or fight response. We will talk about parasympathetic nervous system, which is the rest and digest response. Then we'll talk about sensory afferent nerves, which is how the body's nervous system, both peripheral and central, sense what's going on in the organs, including the heart. And what they do, these sensory or afferent nerves, is they mediate neurotransmission from the body or organs, such as the heart, to the central nervous system. There are two routes for this. One are what we call the spinal afferents. These tend to be run with sympathetic fibers, or the vagal afferents, which run with parasympathetic fibers. So, that's sensory information from the body or organs, such as the heart, to the nervous system. Then the opposite also occurs, which is information from the nervous system to the heart. These are known as efferent nerves. They mediate neurotransmission from the nervous system to the body or to organs, such as the heart. And there are two limbs of this. These are, again, efferent sympathetic, which is that flight or fight, or efferent parasympathetic, which is the rest or digest. Again, the feedback loops are shown here on the right. Now, in the setting of disease, this system that I've just described undergoes incredibly intense remodeling and changes to its structure and function. In this nice review from Circulation Research, there are a few… The slide is busy, but there are some very specific alterations that happen within the nervous system. These include increased afferent neurotransmission. Again, that's information from the heart to the nervous system in green here. There's increased efferent sympathetic signaling. So, the sympathetic drive to the heart is enhanced, and this is what we try to antagonize with beta blockers and calcium channel blockers and other forms of neurohormonal modulation. There's decreased efferent parasympathetic signaling. So, the normally protective response that you get from the parasympathetic system gets decreased in the setting of chronic cardiac injury. There are other alterations, such as reset baroreflex, which is the target of baroreflex activation therapy. There's altered chemosensation, and there are even intra-abdominal compartment shifts, which have been shown in recent studies to be a potent cause of volume overload and symptoms in heart failure. Here is a working model that allows, I think, an easy conceptualization of what happens. Taking the example of myocardial infarction, following chronic infarcts, there are alterations that are mediated by a variety of factors, two of which are shown here, that cause remodeling within the ganglia that control… that house neurons that control the heart. There are also alterations that happen at the level of the injured heart itself, where, in the process of recovery from the infarct, there are heterogeneous nerve sprouts at the level of the heart, in the scar border zone primarily. And together, alterations within neural elements and alteration of neural elements within the heart, together drive sympathetic activation that we know promote ventricular arrhythmias. For example, when we have looked at stellar ganglia from humans with chronic cardiomyopathy and arrhythmias, we find a variety of pathologic remodeling. These include increased oxidative stress, which suggests that the neurons are under increased activity. We see activation of support cells of these neurons called glial cells. Their data have been reported regarding switching of neurotransmitters, and there's also inflammation that's seen within these ganglia, all because of chronic cardiac injury and the enhanced neural remodeling that's occurred. In fact, the neural remodeling not only occurs within the peripheral nervous system, it actually goes all the way to the brain. In this study here that's highlighted in this review in Journal of Physiology in 2016, you can see that there are alterations in the insula, insula cortex, which is a major center for receiving information from the body in the setting of chronic heart failure. So, that remodeling extends all the way to the brain. Now, if you asked whether such remodeling is important for clinical outcomes, the answer is yes. I'd like to highlight neuropeptide Y, which is a neurotransmitter that's released from sympathetic nerve endings in the heart. And what we show is that in the setting in normal controls, there are very low levels of neuropeptide Y release. But as you go from acute coronary syndrome, stable angina, to ST elevation MI, to chronic heart failure, you can see just how much greater the release of neuropeptide Y. Again, this is another neurotransmitter that we do not antagonize with our current therapies in the setting of chronic cardiac injury. You can see just how increased the levels of neuropeptide Y are. And in fact, when you look at the relationship between the levels of neuropeptide Y and outcomes in patients with stable heart failure, so these were sampled from the coronary sinus, and if you look at survival in these stable patients undergoing CRT, you see that those with a lower neuropeptide Y level, so less than 130 picograms per ml, had a somewhat slower pace of mortality compared to those with much higher neuropeptide Y levels. You can see that virtually almost all those patients were deceased by 38 months following this sampling of neuropeptide Y from the coronary sinus. So, the remodeling that happens in the setting of injury is very important for clinical outcomes. So, now coming back, so I showed you changes that happen in the stelic ganglion and how the factors that are released by the stelic ganglion into the heart are altered and can impact mortality. I'd also like to show you at the level of the heart some of the effects of the remodeling that happens in terms of the nerve endings in the heart. So, this preparation here is from a large animal, but we can record from the beating heart. You can see here we have stretchable, flexible arrays that can record from the epicardium in the scar border zone, and you can see here this is a map of action potential duration known as the activation recovery interval. You can see what this looks like at baseline, and you can see that it's heterogeneous in this area here because of the injury and the fact that this is a scar border zone. But doing sympathorexcitation, you can see just how dramatically altered the action potential duration is in the same region, this exact same region that we're looking at here. And importantly, you can find that there are now increased and very steep repolarization gradients, which we know to be a marker for arrhythmogenic tissue. In addition, the same thing occurs during activation. You can see just how the same region in the scar border zone that gets activated late becomes activated quite early and even creates potential substrates for monomorphic VT here, where you see two regions of delayed activation with an isthmus-type region of earlier activation. Again, we know that these electrical characteristics are harbingers of ventricular arrhythmias. So, the working concept, now that I've shown you the reason why neuromodulation is important, the working concept is that we would like to delink the heart from the higher centers on the sympathetic side. So, that means that that sympathetic drive that's coming from the brain down to the heart, we would like to reduce that. Or we would also like to enhance the protective signaling that comes from the parasympathetic nervous system down to the heart. And this is, in general, the concept for neuromodulation, is to attenuate increased information going from the heart to the nervous system and reduce efferent sympathetic drive to the heart. While enhancing parasympathetic drive to the heart. This is the concept of neuromodulation for VT. This slide is a nice summary slide, I think, that really highlights most of the clinically applicable therapies today for neuromodulation for VT. They include thoracic epidural anesthesia, stellar ganglion block or sympathectomy, renal artery denervation, tragus stimulation or vagus nerve stimulation, spinal cord stimulation. And again, and in fact, at the level of the heart, not for VT, but ganglionated plexi have been targeted for AFib as well. And so just to really illustrate the concept of delinking the heart from the higher centers or mitigating sympathetic drive to the heart. This study here shows that it's a preparation where you can measure norepinephrine release in the heart, while occluding one of the coronary arteries. And what you can see is that in animals that are resistant to ventricular fibrillation, shown here in the lower panel in blue, you can see that the amount of norepinephrine that they release is much lower than those animals that are susceptible to ventricular fibrillation, where the amount of norepinephrine they release is much higher. The reason this is important is that when you start to now intervene in terms of neuromodulation, what we find is that we can reduce the amount of norepinephrine that's released and that directly relates to the susceptibility to VF. So as shown here, for example, if you delink the heart at this initial level, as shown here in the upper scissors, you can reduce the amount of norepinephrine released from a 152% increase in the setting of coronary occlusion to 114%. And then when you completely denervate the heart, which mimics cardiac transplantation, you can see that that's reduced to only a 16% increase in the setting of coronary occlusion. And in the animals where we see 152% increase, there's a 30% risk of VF, but once you now intervene, either by decentralizing or mimicking heart transplantation, what you see is that the risk of ventricular fibrillation is now dramatically reduced. And again, this is a form of neuromodulation really emphasizing the role of delinking the heart from the central nervous system in reducing VF. So now I will talk about the practice of neuromodulation, which is now really focused on clinical application of neuromodulation for your patients with VT storm, incessant VT, or patients for whom you've tried other forms of therapies for whom you're not able to get good control of the VT. There are several neuromodulatory approaches, as I've shown, that can be employed. So let me review some of the evidence for some of these. These approaches include pharmacologic approaches. We, again, should not forget that beta blockers and ACE inhibitors are a form of neuromodulation. There are toxins that can be used. These are now getting to clinical practice. There are bioelectronic approaches, such as vagal nerve stimulation or tragus stimulation. There's use of light therapy, which is not clinically applicable yet. Some folks are using ultrasound therapy to ablate different regions of the nervous system. This is getting to clinical practice. There's certainly radiofrequency ablation of various nodes within the nervous system and the surgical denervation, for example, sympathectomy. Again, this is the slide I showed earlier. I've now called out the different parts of the nervous system in terms of neuromodulation that I will be speaking more fully about. These include thoracic epidural anesthesia, stelic ganglion block, cardiac sympathetic denervation, renal denervation, transcranial magnetic stimulation, tragus stimulation, and vagus nerve stimulation. Again, there are many other forms of neuromodulation, but I've highlighted these seven approaches here as they are much more clinically applicable or very close to the clinic. So, starting with thoracic epidural anesthesia, the way this is done is much in a way that epidurals are provided for women in labor. This takes a similar approach where we target the first thoracic to second thoracic or second to third thoracic interspace. Using a standard loss of resistance approach, we can then place a catheter here at this level, start with a bolus injection of bupivacaine or rapivacaine, and then followed by an infusion, which would then titrate to arrhythmic response or adverse effects. And you can see in this study here led by Dr. Duc Do, one of our colleagues at UCLA, they took 11 patients, four of which had VT storm, the remaining seven had incessant VT. And we tried to, we either shocked the patient or reprogrammed the ICD, applied antiarrhythmic drugs or sedated the patient. And then following that in those patients in whom we did not get good control, we went on to thoracic epidural anesthesia. And you can see that as a temporizing measure, it allowed us to get patients to catheter ablation, to sympathectomy. In some patients, there was actually resolution of VT that's shown here in this block here. And in some patients, there was no resolution. And then even some patients were temporized to a heart transplant. So in the patients in whom we did not have resolution of VT, out of the three, two of them passed away. Of these two, one was actually temporized by bilateral sympathectomy. The other unfortunately passed away. And in this group here, where we went on to catheter ablation, we were able to successfully control arrhythmias in one patient, but not in the other. And of this very difficult to control patient population, you can see that 64% of the patients we applied this therapy to survived to discharge, where we were able to utilize thoracic epidural anesthesia to control their VT. And you can see here, even independent of the outcome of the patient, you can see here that when you compare before in terms of VT episodes to after, you can see a strong suppression of ventricular tachycardia and SIM for ICD shocks instituted by TEA. So this is a potent way to really acutely suppress ventricular arrhythmias and temporize patients to other forms of therapy. What about stelaganglion blockade? Very similar concept. The idea here is that we instill an anesthetic next to the stelaganglion, which I showed earlier. This is a very nice study from the group at Duke University led by Murad Fadim. And what they did is to introduce a neurocardiology service where they took patients who were admitted for VT storm, try to manage them with antiarrhythmics, including sedation. And in those who had continued VT ablation, they were evaluated for catheter ablation and mechanical support. And in parallel, they were evaluated for stelaganglion block. And you can see that they then took some patients for stelaganglion block and looked at whether they had neuroreduction in VT, where they then went back to this evaluation here. And in those in whom there was reduction for VTVF, they either did repeat stelaganglion block or surgical sympathectomy. And let me show you their outcomes here. You can see that in the first 24 hours, there is a strong reduction in defibrillation events, as well as sustained VT episodes. And this lasted out even to 48 hours. So the effect of stelaganglion block was really quite potent in helping suppress the acute arrhythmias that you see in your patients with VT storm. And it didn't matter whether the arrhythmias episodes were monomorphic or polymorphic in nature, or ischemic or non-ischemic. There were really significant effects that were seen for stelaganglion blockade. Now, that study had about 20 patients. We did a systematic review and meta-analysis where we collated many studies that actually did not include the study I just showed. And we got up to nearly 40 patients or so. And again, across all these studies, this was led by Ling Meng, who is a resident here at UCLA. Across the patients in that study, you can see here a strong reduction in arrhythmia episodes per day following stelaganglion block. And you can see here the number of shocks was really just substantially reduced. And again, the LVEF, the ejection fraction, did not influence the relative reduction in arrhythmia episodes. And it didn't matter whether the patients had no cardiomyopathy, whether they had ischemic cardiomyopathy, or whether they had non-ischemic cardiomyopathy. We saw fairly potent effects across the board. And you can see here the type of anesthetics that were used. Lung and short actin, bupivacain, gropivacain, lidocaine, and bupivacain were used. And some patients had bolus injections, some had continuous. And the approach was taking in some patients landmark only, so this is very applicable to the bedside. Or with ultrasound, similarly. Or with fluoroscopy in patients that are brought down to the EP lab or cath lab. Now, one really interesting point I'd like to make regarding stellar ganglion block relates to the study published by Dr. Babak Nazer at Oregon. His group took an approach where they studied patients who had single injection, so that's SI here, or continuous infusion. And you can see here that they took patients with VT storm, they try to temporize them with antiarrhythmic drugs. Those patients went into stellar ganglion block. In an initial experience, those patients received single injections. In a follow-up experience, some patients received continuous infusion. And you can see that compared to continuous infusion, where they saw a 94% reduction in ventricular arrhythmia episode without an increase in complications, only those, only 54% reduction was seen in those patients with single injection, and in fact, 44% of them required repeat injection. And so I think that for those who now be utilizing stellar ganglion block as part of their treatment algorithm for patients with VT storm, I think this study is a worthy and very strong consideration because again, not only are the effects really quite potent, they are more sustained and reduce the risk of repeat injection. You can imagine many of these patients are on anticoagulants and are quite ill, so the need for repeat procedures in such patients can come at a really high cost. So the idea of using a continuous infusion as suggested by Dr. Nazer and their group, I think is a really, really important concept to consider in using stellar ganglion block. You can see here again that in the patients that got single injection, four of them required repeat stellar ganglion block, while in those who received continuous infusion, while two of them had some additional arrhythmia, certainly not to the degree of what they had before the block, none of them required repeat injection or additional intervention from stellar ganglion perspective. Now, moving on to bilateral cardiac sympathetic denervation or cardiac sympathectomy, I should say. The first thing to point out is that the therapy is actually being utilized for angina and ventricular arrhythmias for over a hundred years now. This is the first report by Dr. Tomas Ionesco, who was the first to attempt this therapy. And he did this actually in 1918 and published this work in 1921. And this has been followed up on by multiple groups, including James White from the Massachusetts General Hospital in 1952, and then somewhat more recently by Doug Zipes and their colleague at Duke, as well as Gaetano De Ferrari and our group and others in more recent work. So this form of therapy has a very long history and rationale for its use in a variety of arrhythmia, ventricular arrhythmia subtypes and etiologies. One important point I really want to make regarding cardiac sympathetic denervation for VT is that the etiology of the VT and the type of patient that you're treating is very important to consider the type of sympathetomy that you apply. So in patients with channelopathy, such as CPVT and long QT, a left cardiac sympathetic denervation alone is probably appropriate for those patients. There's a wealth of data from Art Moss and Peter Schwartz that really established this as a therapy for such patients going back decades. And only in the setting of failure or recurrence do those patients then go on to get right cardiac sympathetic denervation to complete bilateral cardiac sympathetic denervation. This is in contradiction, distinction from patients with structural heart disease, such as ischemic or non-ischemic cardiomyopathies, where, as I showed you, there is remodeling of the neural elements bilaterally. So in patients with structural heart disease, if you apply just the left cardiac sympathetic denervation alone, that would be incomplete therapy because the right stellate ganglion has also undergone substantial remodeling and can continue to drive VT. So this is the preferred approach for patients with structural heart disease, again, non-ischemic cardiomyopathy, ARVC sarcoidosis, and other examples. Now, how is this performed? A surgeon, typically using video-assisted thoracoscopic surgery, identifies the sympathetic chain. Here's the first rib here. So the stellate ganglion, which is a fusion of the eighth cervical and the first thoracic ganglion of the sympathetic chain, lies just right beneath the first rib. And what is done is a resection of the lower half of the stellate ganglion, or lower third, to the fourth thoracic body. Importantly, again, I want to emphasize importantly, one must verify that there are neurons that have been removed from once the sympathetic chain is cut out. And the reason for this is because at times the anatomy can be quite tricky, and unless you verify that you have ganglia neurons have been removed, not just nerves, that would lead to a more efficacious procedure. And so a wealth of data, I showed you our initial experience here, a wealth of studies have now shown this, and importantly, if you look in this multicenter retrospective study that included UCLA, Hopkins, and several other groups, you can see that the number of ICD shocks was substantially reduced following cardiac sympathetic denervation, and that's bilateral. This is in patients with structural heart disease, with modest adverse effects that we found to be temporary. And importantly, when you compare, again, left cardiac sympathetic denervation shown here in blue to bilateral sympathetic denervation shown here in red, you can see that the freedom from shock was, or I should say shock-free survival was substantially improved in patients who had bilateral sympathectomy for structural heart disease compared to patients who had left cardiac sympathetic denervation alone for structural heart disease. So an important concept to keep in mind. Now, one sort of point that comes up, not infrequently, was investigated by the Hopkins group, Hari Tandri and his colleagues. What can typically be seen is that in the initial period following sympathectomy, patients may have brief episodes of ventricular arrhythmias or some ICD shocks, and that often has been seen as a, or at least viewed as a failure of the procedure. But this study really educated us that perhaps we shouldn't think of it that way. What Dr. Tandri and his colleagues did was that they took 22 patients who underwent bilateral cardiac sympathetic denervation and had long-term follow-ups. So this is now greater than 18 months. The study I showed just before this had really intermediate outcomes. These are now long-term outcomes following sympathectomy. Two patients were excluded because they had less than 18 months of follow-up, but 20 patients were studied. And what they did was they looked at the outcomes of patients who had early VT or ICD shock recurrence within 12 weeks, so three months following the procedure, and then compared them to patients who had recurrence greater than three months out. And the important finding here is that 60% of those patients, again, this is a small subset of patients, but I think this is really quite applicable to the larger patient pool because we have the same experience, 60% of these patients had no late recurrence. And it makes sense that in the early period following sympathectomy, when the nerves in the heart are being resorbed and are going away, it tends to be a potentially irritable period in the heart such that once that period has been elapsed, then the true effect of sympathectomy is seen, and those patients have really a low risk, or at least those patients have no late VT recurrence, again, 60% of such patients. And when you compare early to late recurrence shown here on the right, it was no statistically significant difference. Again, small pool of patients, but the clinical experience that we have really has borne this out as well. And again, what they found is that compared to following cardiac sympathetic denervation, there was a reduction in the use of class one and class three antiarrhythmic drugs. Another important study that I'd like to highlight is from the Vanderbilt Group led by Travis Richardson and Arvind Kanagasandram. They looked at patients where they had specifically tried catheter ablation and those patients were refractory. And the sites that they studied include difficult to access sites, such as the LV summit and the cardiac crux, as well as intramyocardial circuits that are notoriously difficult to ablate, such as within a very thick septum or within the papillary muscles, which can also be difficult to approach. And they applied sympathectomy as shown in this table here. So these are seven patients. You can see the variety of substrates from hypertrophic cardiomyopathy to normal heart to coronary artery disease to ARVC. And the burden of the patients is shown here. The mechanism of the VT, so increased automaticity, short coupled PVCs, re-entry, triggered activity, and the drugs that the patients were on. In fact, you can see that some of these patients had undergone two cardiac catheter ablation procedures. And those patients, most of them had bilateral sympathectomy because they typically had structural heart disease. And then in one patient that was left only. And you can see here that they had fairly significant success following a sympathectomy in these patients. Some of them had up to 49 months free of sustained ventricular arrhythmias. One patient went to orthotopic heart transplantation for a separate reason. But overall, what this study really highlights is that sympathectomy can be applied to a variety of substrates and VT mechanisms, particularly in patients in whom catheter ablation has failed. Now, moving on to tracheostimulation, which is now more on the parasympathetic side, the idea here is to stimulate the concha of the ear using low-level tracheostimulation, so sub-threshold stimulation that's below the pain threshold. And the idea with this is that it enhances vagal output to the heart. And while there are no studies that have looked at tracheostimulation for VT, I wanted to highlight a couple of points. Tracheostimulation has been shown in a randomized control study for atrial fibrillation. The results I summarize here by Stavros Stavrakis and his colleagues, where they show that in sham patients, you can see the increase in AF burden in the patients, but in those in whom tracheostimulation was employed, you see here that there's no increase. So that's here in blue, no increase in the burden of AF. We sampled blood from these patients and showed that consistent with the clinical finding that in patients who got sham treatment, there was an increase in neuropeptide Y. This is the same neurotransmitter that I showed you earlier that was associated in the setting of chronic heart failure with adverse outcomes. Similarly, in these patients, we see that there's an increase in neuropeptide Y levels in those who are sham treated, whereas those who are active treated with tracheostimulation, they did not have an increase in neuropeptide Y. Additionally, Stavros and Antonis Armandas from Mass General have gone on to study other electrophysiologic properties that we know are associated with ventricular arrhythmias, such as microvoltaic T-wave alternates. And they showed very nicely that compared to sham, active tracheostimulation reduces T-wave alternates both in sinus rhythm and with atrial pacing. So the hope, and there are ongoing studies at this time for tracheostimulation for ventricular arrhythmias, the expectation is that this will also be borne out in terms of ventricular arrhythmia episodes and ICD shocks once those studies are completed. Now, one of the last methods of neuromodulation that I want to focus on here is vagus nerve stimulation. And before that, I wanted to address some of the negative studies that are out there in terms of vagal nerve stimulation. And I wanted to really emphasize one important point that I think we as electrophysiologists need to understand when it comes to neuromodulation. And that is that in patients where we put in pacemakers, we put in a pacemaker lead, implant the device, we set our pacing parameters, and for the next 10 years or 15 years or however long, as long as there aren't any major issues that occur with the lead or with the heart, those patients can be paced at those same parameters for as long as is needed. However, neuromodulation does not work that way. And I think that the two of the negative vagal nerve stimulation for heart failure studies, which include NECTAR and the Innovate CardioFit studies were negative because the parameters that were selected for vagal nerve stimulation did not really engage the efferent parasympathetic limb as much as is needed to. And so in this really three-dimensional plot here, what you're looking at is the heart rate response to vagal nerve stimulation here on the Y-axis and on the X-axis, you can see the vagal nerve intensity and the frequency here on the Z-axis. And what you can see is that when you stimulate at high frequencies and high intensities, you get a strong heart rate reduction. But if you stimulate at high frequency and low intensity, you actually have the opposite effect, which is an increase in heart rate. And the same with low frequency and high intensity, again, a paradoxical response. The Anthem Heart Failure and Encore studies stimulated at parameters that actually enabled reduction in the heart rate as shown here. And that's why this is the only positive study we have for vagal nerve stimulation in heart failure. While this hasn't been applied to VT yet, there are several animal studies that very, I think, nicely indicate that the, or nicely illustrate the strong suppressive effect of vagal nerve stimulation on VT and on the structural abnormalities that lead to VT say following infarct. The same has been shown for heart failure and such studies are coming for VT. So in the initial Anthem study published now in 2014, 78 patients were randomized to left or right vagal nerve stimulation. And you can see that there's a strong improvement in heart rate and a overall reduction in LV encystolic volume and encystolic dimension. So this is another form of neuromodulation that is coming to the fray for VT already established in heart failure. And we're looking forward to being able to offer this to patients. I will very briefly talk about renal denervation, which has been really shown in a randomized prospective clinical trial for atrial fibrillation. We have some small data sets for VT. This was published by my colleague, Jason Bradfield here at UCLA, where in patients who had undergone cardiac sympathetic denervation and who had a refractory VT, the renal neural signaling was targeted by renal denervation. And you can see here that we showed a strong suppression in terms of ICD therapies and ICD shocks. If you look at shocks specifically, you can see a strong reduction here across all 10 patients that were studied. Finally, this is a recent study, randomized clinical trial from the UPenn group where they applied transcutaneous magnetic stimulation. You might recall that I mentioned earlier that the remodeling that occurs in the setting of chronic injury really reaches the brain. And so in this study, they applied transcranial magnetic stimulation, which is an outpatient procedure as shown here, they did this in patients who presented with VT storm. Their study looked at 30 patients, 26 of whom consented, and they randomized these 26 patients to 14 to get the transcranial magnetic stimulation and 12 patients to 60 minutes of sham stimulation. Those patients were studied for 72 hours, and they looked at an initial 24 hour efficacy and a 72 hour efficacy. And the data shown here in the first 24 hours, you can see here that more patients in the transcranial magnetic stimulation group have no VT compared to the sham group, while overall the data did not quite meet statistical significance in the first 24 hours. When you now look further out at now 36 to 48 to 72 hours, there was the use of transcranial magnetic stimulation was associated with a freedom from arrhythmias compared to sham, and that's shown very nicely in this curve here. So I've just taken you on a whirlwind tour of neuromodulation approaches clinically, and I just wanted to put this slide back up again to sort of summarize what we're trying to do. These therapies here on the left side, TEA, Steli-Ganglion block, cardiac sympathetic denervation and renal denervation have a goal of reducing sympathetic drive to the heart. While Tragus stimulation and vagus nerve stimulation have a goal of increasing parasympathetic tone, the mechanisms by which transcranial magnetic stimulation work are not well understood. But I thought, again, this figure is a nice way to really summarize the current approaches that are available for mitigating VT via neuromodulation. Importantly, I just wanted to point out that in the 2017 guidelines for the management of ventricular arrhythmias and prevention of sudden death, autonomic modulation is now a class 2B indication. You can, of course, follow the algorithm here. And one important concept to consider is that we may be waiting too long to apply these neuromodulatory strategies, and hopefully studies such as PreventVT as shown here will actually help make the point that perhaps autonomic neuromodulation should be offered earlier to patients with VT and VT storm. One point to make is just as our colleagues at Duke have done we similarly have a neurocardiology program, interventional neurocardiology program that's a multidisciplinary program where we have not only cardiac electrophysiologists but thoracic surgeons, cardiothoracic surgeons. It includes our cardiomyopathy colleagues, nursing, anesthesia and pain medicine, intensive care, psychiatry and interventional radiology where we really look to address a variety of arrhythmias and cardiac diseases via multiple neuromodulatory approaches. And such a program is very helpful for being able to offer patients a variety of the neuromodulatory approaches that I described to you. Here are some key takeaways. In patients with chronic structural heart disease there is pathologic remodeling of sensory, sympathetic and parasympathetic control of the heart. The goal of neuromodulation is to reduce cardiac efferent and efferent sympathetic neurotransmission as well as to enhance efferent parasympathetic neurotransmission. Thoracic epidural anesthesia and Steli-Gangli block can be used as temporizing measures for patients although continuous infusion may be superior to single injection. Application of cardiac sympathetic denervation in patients with channelopathy where we would start with a left cardiac sympathetic denervation versus patients with structural heart disease where the data shows that bilateral cardiac sympathetic denervation is superior to left cardiac sympathetic denervation. So the substrate and the etiology of arrhythmias matters. Cardiac bilateral cardiac sympathetic denervation is effective in a wide range of ventricular arrhythmias. So not only post-infarct non-ischemic cardiomyopathy but also idiopathic and inherited structural arrhythmias such as ARVC. Importantly, early recurrence following cardiac sympathetic denervation does not necessarily impact long-term success. So please do not view patients with an initial early recurrence as a complete failure. Following evaluation of such patients in the long-term there appears to be no major impact on long-term success and establishing a neurocardiology service enhances the range of neuromodulatory therapies that can be offered to patients with VT. And with that, I'll stop and thank you for your attention. And again, thank Dr. Miller and the session organizers for the kind invitation to share our work with you. Thank you.

neuromodulation

ventricular arrhythmias

theory and practice

neuromodulation therapy

case study

neural remodeling

neuromodulation techniques

comprehensive neurocardiology programs