Thank you, Rod. Thank you, David. Thank you, Heart Rhythm Society, for the invitation. And thank you all for coming, despite all the trials and tribulations you had to face while coming here. So appreciate it. My task is, in 15 minutes, to talk about something that probably all of you know more about than I do. But I'm going to try to build it ground up again, because it's a challenging topic, and there is a varied level of understanding of it. But in the next 15 to 20 minutes, I'll talk about why we need to understand the pathophysiology of heart failure, because there are different kinds of heart failure. That's one. Second is, recognize the importance of conduction defects, and how they can actually determine outcomes in patients with heart failure. Thirdly, what are the different pacing strategies that are important that can actually prevent progression of heart failure and reduce sudden cardiac death? And then, obviously, how do we individualize our electrical therapies in 2020 and 2025 for treating heart failure? So I'm going to throw out some ideas out there, and then we'll talk about them over the course of the discussion. It goes without saying that LVADs and transplant are not the answer. And it also goes without saying that the escalating costs for treating heart failure is increasing exponentially. Now, having said that, I think I'm going to take a step back and just very quickly revisit what happens in heart failure. As the heart fails, as it dilates, as it remodels, there is myocyte hypertrophy. There is interstitial fibrosis. There is myocyte necrosis and apoptosis. But what's really important to understand is that there is a gradual change in the cardiac skeleton of the heart. And when we look at the cardiac skeleton, we have the sinus node, the AV node, the bundle branches, the Purkinje network. All of that actually undergoes a remodeling process. So somebody who didn't have a conduction defect may actually land up having a conduction defect because of this remodeling process. But on the flip side, there are patients who start off with an electrical problem and then develop mechanical dyssynchrony and develop heart failure. And I'll show you some pictures related to that, too. So it's important to understand that. But it's also important to understand that there is an electromechanical interface. And we don't really understand this whole concept of electromechanical uncoupling because you can provide electricity, but a good myocyte may not contract. And there is something that is happening at that interface that, again, we don't understand that very well. So very quickly, HEF-PEF is heart failure with preserved ejection fraction. That is very different from HEF-REF, which is heart failure with reduced ejection fraction. Now, if I were to look cross-sectionally today, on one day at Mass General Hospital in a 1,000-bedded hospital, there are 110 patients admitted with heart failure. 110 patients carry the diagnosis of heart failure. Now, if I were to substratify those patients as to how many of them have HEF-PEF and how many have HEF-REF, 60% of them have HEF-PEF, preserved ejection fraction. That has not got enough attention today, which is the next era that I think a lot of us need to be paying attention to. And then, obviously, 40% of patients have HEF-REF. Why is HEF-PEF different from HEF-REF in the pathophysiology side of things? Completely different. HEF-REF occurs at the myocyte level. There are more changes at the myocyte level. There's apoptosis. There's fibrosis. There's collagen deposition. HEF-PEF is different. It is associated with hypertension, diabetes, obesity, aging. There are endothelial changes out there. There are inflammatory processes that actually occur out there. There's oxidative stress. All these things then subsequently lead to renal injury and myocyte dysfunction and also result in a stiffening of the heart. So the ejection fraction is preserved, but the heart is stiff. Patients go into heart failure. So we have two different entities. And if you look at the pathological changes out there, you can see the yellow is the endothelium. There's really not a whole lot of endothelial changes in the HEF-REF population unless they have accompanying comorbidities that we just talked about. But the HEF-PEF population has largely changes occurring at the endothelium, at the vascular level. And then you have the myocyte changes as a consequence of that. So two very different entities. Treatment strategies are widely different. Now, let's throw in a little bit of AFib into this. Changes the whole picture out there, right? Another electrical issue. 40% to 50% of patients with heart failure have atrial fibrillation, right? It's a chicken and the egg situation. People ask you, did the AFib lead to heart failure? Did the heart failure cause the AFib? And they're tied together. We know that when you have AFib, you have a fast ventricular rate. You have RR variability. The heart doesn't really feel that well. There's AV dissociation. That can result into a myopathic process. That myopathic process then itself can cause all the interstitial fibrosis, can cause atrial remodeling. That itself can lead to AFib. So it is a vicious cycle that can occur in patients both with HEF-PEF and HEF-REF. And oftentimes, patients with HEF-PEF, with preserved ejection fraction, when they have atrial fibrillation for a long period of time, they develop reduced ejection fraction because their myocytes kind of give way. They start developing a myopathic process. And that's when patients have a kind of an intermingling of the two processes. And it's really important. That's the pathophysiology. And I'm going to come back to it. But it's really important to recognize that patients who have HEF-REF have a very high incidence of sudden cardiac death. So if you look at the prevalence of sudden cardiac death, the prevalence of sudden cardiac death is highest in the general population. But the incidence is really small. But if you look at the incidence in patients who have an ejection fraction less than 35%, it's pretty large. But the prevalence overall is very less because our risk stratification strategies in the large population are not that good. But we know that if patients have a low ejection fraction, and we know if they've had an MI in the past, they are more than 10 to 12-fold at a higher risk of sudden cardiac death compared to somebody who does not. So it's important to recognize that there's a difference between prevalence and incidence. But it's important to recognize that we also need to be able to risk stratify these patients a little better. Now, looking at this very simplistically as an electrophysiologist, you kind of say, hey, which patients of mine are more predisposed for arrhythmic deaths? Which patients of mine are more predisposed for pump failure? The simplest way of substratifying that is actually saying, OK, are they NYHA class 2, 3, or 4? Again, that's a very subjective classification, but it helps because NYHA class 2, where patients actually do reasonably well, have a higher propensity for actually having arrhythmic death, whereas NYHA class 3 can have both arrhythmic and pump failure, but there's more pump failure because their hearts have already given way a little bit. And then you have NYHA class 4, where they largely have pump failure. Now, what is the incidence of sudden death in HFREF? I'm showing you this slide largely to show you that over time, there has been a reduction in the incidence of sudden death. And we're going to come back to this in the discussion because all the clinical trials we've done have been done in an era where the incidence of sudden death was actually on the higher side. And with all the new medications that have come in, and if you look at paradigm HF, the incidence of sudden death with these newer evolving pharmaceuticals is actually better. Now, I don't know why the corona trial out there still had an incidence of sudden death that was a little higher. I guess it has to do something with what we're experiencing. Actually, the corona trial was a trial looking at statins in systolic heart failure. And therefore, it really didn't impact sudden death to that extent. Now, having said that, when patients with heart failure have a hospitalization for heart failure, this is the MATED-2 study. This was the study that actually resulted in the MATED-CRT study, where we found that if patients with heart failure actually got hospitalized, they actually had subsequently a much higher risk for sudden cardiac death and ICD therapy for VT or VF. And that created this reason for why MATED-CRT study was done. But before we get to that, I think it's really important that when we look at patients and treatment of heart failure, the goal of medical therapy is to prevent that trajectory. Every time a patient recovers from heart failure, they actually get a little worse the next time around, and they never come back to baseline. So what can we do from our electrical therapy armamentarium, from our sensor strategy armamentarium, from our pharmaceutical therapy armamentarium to prevent that drop down, prevent that trajectory, and make sure that we can actually prevent the first heart failure hospitalization so they don't have to eventually have that downward trajectory? Now, when you look at the course of heart failure, it's really important to understand the path of physiology out here. That patients actually, initially, may be hemodynamically stable, but they already start showing signs of developing pulmonary congestion. So their PA pressures go up, their LA pressures go up. They don't manifest with overt failure, but their filling pressures actually go up. And I'll talk about why that is important, especially when we look at this whole heart failure patient management over time. And then they're still not symptomatic, but their pressures are a little high. So the heart then starts adapting to it. It has autonomic changes. There's a change in the RR variability. There's a change in the sympathovagal balance. There are changes in the intrathoracic impedance. I think that all of us recognize that as fluid accumulates, the transthoracic impedance changes. And then patients start developing fluid accumulation, increase in weight, and developing heart failure. So the important thing out here to recognize is that there are stages before they actually manifest with overt heart failure. So again, as a part of our understanding of the pathophysiology of heart failure and for electrical therapies, how do we prevent that from actually occurring? So this is a model that is, when I first started thinking about sensors in 2009 or 2010, it said, for a patient with heart failure to actually develop heart failure hospitalization, it's important to recognize that they have a baseline neurohormonal status. And we found that patients who had an abnormal neurohormonal status with a slightly elevated or slightly depressed heart rate variability and abnormal sympathovagal balance were at more risk for developing heart failure hospitalizations. But also, they have an inherent myocardial risk depending on their substrate. So this is the baseline chronic heart failure patient. You throw in some comorbidities. Diabetes, hypertension, COPD, AFib. Now, all these conditions actually have sensors. You can measure blood glucose with one of those freestyle libros. You can measure atrial fibrillation. You can measure hypertension. You can measure O2 stats, all from peripheral sensors, to actually see which patients actually may have comorbid conditions that could influence their heart failure from actually progressing. And then you have triggers. You have triggers, whether it's ischemia, electrolytes, endothelial functions, arrhythmias. All of these triggers, again, have sensors that can be measured. And then you have sophisticated sensors that already exist in the devices that can actually pick up left atrial pressure. PA diastolic pressure has also been measured in the past. Impedance-based strategies, and there are newer sensors that are looking at biosensors out there. The reason for explaining this is because these are all interconnected in the pathophysiology of a patient going from a chronic heart failure state to an acute decompensation, where they actually then manifest with heart failure hospitalization. And then you have device-based measures that we all know that actually go out of whack when patients are admitted with heart failure. I'm going to shift gears out here to talk about mechanical dyssynchrony, because I think we all recognize that you need an abnormal activation pattern to have an abnormal contractile pattern of the heart. And that abnormal contractile pattern results in mechanoenergetic inefficiency, and that mechanoenergetic inefficiency translates into poor pump function. So that's mechanical dyssynchrony, as you can see out here. The heart, the lateral wall, and the septal wall are kind of wobbling. That's an activation pattern. And this out here is the levels of electrical dyssynchrony. So when we are talking about electrical dyssynchrony, we just say there's mechanical dyssynchrony between the lateral and the septal wall. But actually, when the heart is failing and there's electrical remodeling occurring out there, there is interatrial dyssynchrony, there's atroventricular dyssynchrony, there's intraventricular dyssynchrony, there's intramural dyssynchrony, and there's interventricular dyssynchrony. So when we talk about resynchronization therapy and we say, OK, we can put in three leads and we're going to resynchronize all of this, it is a simplistic strategy for a complex electrical problem. And you can only imagine what the number of unmet needs are just looking at this figure itself. And then you throw in the myofibrillar pattern of an MRI out there, and it tells you how complex the heart is. And into that myofibrillar pattern, throw in a little patch of fibrosis, throw a little bit of scar in, throw a little bit of dysrhythmia in, you then understand that this heart is really more complex. And the simple CRT strategies, even though they work really well, and this is the MATED-CRT, the reverse HF and RAF studies, which have really changed the paradigm of how we actually treat patients with heart failure with a wide QRS and low ejection fraction, do really well, but you know that it can still do much better. And that's where I think the conversation needs to go, is what can we do better? Because there are still patients who don't respond. There are a substantial number of patients who respond, and we really hang our hat on that. But how can we convert those non-responders into responders? And how can we convert the level of response in responders into responding much better? And why is that important? It's important because if you don't respond well, you have more propensity for having sudden arrhythmic death. You have a higher incidence of shocks, and this is data from the MATED-CRT, which actually shows that patients who did not have a good response actually had a higher arrhythmic event out there. And this is a study which said, OK, resynchronization therapy is trying to treat the pathophysiology of an abnormal electrical activation. But what does that abnormal electrical activation lead to? It leads to mechanical dyssynchrony. So why don't we just target mechanical dyssynchrony itself and not really worry about the electricity within the heart? So there was the ECHO-CRT study, which everybody knows about. And when we actually went in to look at that, we found that, you know what? It didn't work. We actually landed up killing people. That study had to be prematurely terminated because it was neutral to negative. And that, again, indicates that there is a lot of variability out there at the individual level, but the mechanical dyssynchrony level is not measured very well and very reproducibly, and as a result of which, we still need to focus on the electricity. Now, why the electricity? Because if you look at the electrical activation pattern in a failing heart, and we look at the wide QRS, we can say, OK, this patient has a left bundle. This patient has a non-left bundle. This non-left bundle is a right bundle branch block, or it could be a right bundle branch block with the left hemp or hemiblock. And then we wonder, why do all our patients don't respond to CRT? And they don't respond to CRT because our strategy has always been a one-size-fit-all strategy. You put the lead in the lateral wall, and you hope it does the same effect for every different electrical activation sequence. Now, it works quite well, and it works really well, and it has transformed the way of life of a lot of patients, but it doesn't work uniformly well because there is so much of individual variability in the electrical activation sequence, which, again, is an unmet need of how we need to approach that and really try to develop strategies that will make these patients feel better. Now, this is just to show you that ischemic substrates behave differently from non-ischemic substrates. And this is the study that I think most people know about. This was really looking at non-ischemic patients and found that these patients actually did pretty well with just CRT and didn't necessarily need an ICD, although there is a preponderance of sudden arrhythmic death in patients who didn't get an ICD. But overall, resynchronization therapy actually works quite well in non-ischemic to the extent that when we look at trying to understand why the pathophysiological response within certain patients of heart failure varies, it's because they have a different substrate, they have a different electrical activation sequence, and all of those things actually impact the outcomes. So if you look at it out here, you can see that patients who have a wider QRS, who have a left bundle branch block, and who are non-ischemic do pretty well to resynchronization therapy. But the moment you're ischemic, you have a narrower QRS, and you have a non-left bundle branch block, you don't do as well. And it's also important to recognize that there is a wide variability in the level of response. You can have a super responder, you can have a responder, you can have a non-responder. So it's not a dichotomy. It's not that you responded or you didn't respond. It is a continuous feature. And I think as we work on our devices, I think it is so important to have sensors within our devices that can quantitate, quantify response, and individualize therapy so as to use the sensor-aided strategies to further facilitate response in our patients with automatic closed loop strategies that will adapt adaptive pacing strategies within the patients to enhance response. I'm going to quickly shift gears on the pathophysiological component of the autonomic nervous system. I think it's really important to recognize this is very simplistic, that there's a cardio-acceleratory limb and there's a cardio-inhibitory limb where our heart rate is, where our blood pressure is, is a constant push and pull balance between the sympathetic nervous system and the parasympathetic nervous system. And it's important to recognize that patients who have an abnormal sympathovagal balance, patients who have a higher sympathetic tone and a lower parasympathetic tone have a higher predisposition for having abnormal cytokines, have a higher predisposition for having prorythmias, have a higher predisposition for actually having a compromised LV structure and not remodeling as well. That's why beta blockers help. That's why modulating the autonomic tone helps. That's why modulating the sympathovagal balance helps in these patients. Now, I showed you a very simplistic strategy. But if you look at the heart, it's actually very complex. It is made up of billions of ganglions on the heart, millions of ganglions on the heart. And there is actually something called the small brain. And that small brain is on the heart. And it's constantly modulating itself based on the input it gets. So it's not static. And if you look at the little cross-sectional histopath out there, it just shows you a 0.1 millimeter little excerpt from the heart, which shows you how interconnected it is with the autonomic nervous system. Now, why am I showing you this? I think, again, we'll talk about this in newer trials. But I think it's important to recognize that there are many arcs in the autonomic nervous system that can be used to prevent heart failure. It could be stimulating the carotid sinus. It could be vagal nerve stimulation. It could be spinal cord stimulation. It could be renal denervation. It could be stimulating the aortic arch, where there are baroreceptors out there. It could be stimulating the pulmonary arteries to enhance cardiac contractility. It could involve denervating the pulmonary arteries for pulmonary hypertension. And at the same time, there are many other plexi that we have not even talked about. So there's a whole unmet need in extracardiac stimulation and extracardiac neural stimulation that we can actually use. So again, just going to leave you with this. This is my last slide, that there are many pacing strategies there. We don't know which one works well, but there is. All of them work, but we need to be able to phenotype our patients a little better, both for their myocardial function as well as their electrical phenotype. It could be the His-Bundle strategy. It could be the Cori-Sinus pacing strategy. It could be the Left-Bundle Branch Block strategy. It could be multi-point pacing strategy, whether it's sequential or simultaneous. It could be LV endocardial pacing strategies, whether they're single site, multi-site, or leadless strategies that are coming around the pike, which I'll talk about later. And it could be multi-site strategies, where two or three different, in combination, you can have RV and LV pacing. And then a vast number of extracardiac stimulation, such as neuromodulation. So with that, I think understanding the pathophysiology of the heart really is important, because it helps us separate between HF-PEF and HF-REF. It helps us understand that conduction defects are really important to understand as determinants of response to therapy. We recognize that pacing strategies actually can prevent progression of heart failure and reduce the incidence of sudden cardiac death. And obviously, it allows us to individualize electrical therapies. Thank you.

heart failure

conduction defects

pacing strategies

sudden cardiac death

individualizing electrical therapies

LVADs