Good morning, everyone. It's a pleasure for me to be here in this joint session with the Haring Society. It's a pleasure to introduce Dr. Jodi Hurwitz. She's a Haring Society current president. She works in Texas Heart Center in the U.S., and now she will be my co-chair. Jodi? Thank you very much. I'm delighted to be here. This is a great session. This is wonderful to be part of, Lars, and it's a true honor to have my co-chair here sitting next to me, and I appreciate this wonderful meeting. So our first talk is going to be by Dr. Rod Tung. He's a cardiovascular director of clinical research in Phoenix, Arizona, and he's going to be talking about how should we approach CSP in narrow QRS patients with AV block and normal LVEF, a nice long title. Thank you very much for the opportunity to speak with you this morning. I think that pacing affects every electrophysiologist, and I'm delighted to talk about how we should be choosing or thinking about conduction system pacing. This is a beautiful, beautiful picture of the HisPurkinje system that is in all humans, and it's a shame not to use the HisPurkinje system. It is like the high-speed internet, and everything short of that is not really having access to a high-speed mechanism. Now, it's very intricate. It's very redundant as well, and this is all by design. It's important to remember that even though the EKG is very primitive, it still is informing us to make decisions based on three 100ths of a second. We understand that someone with a QRS of 150 is very different from 120. And it's also important to remind everyone that the normal physiology on the left septum goes down the tri-facicular nature as first described by Dirk Durer, and when we do CRT, even though it is resynchronization and a relatively narrow QRS, it is anything but physiologic. RV apical pacing fusing with an LV epicardial wavefront to meet somewhere in the middle and give a reasonable QRS. So when you talk about a 57-year-old woman with a narrow QRS that presents with fatigue and shortness of breath and has 2 to 1 AV block, then we always have to think about what the options are for pacing. And in this case, this patient was at University of Chicago and received a standard RV apical pacemaker, and this is the activation of RV apical pace, which is right ventricle first, and then secondary transeptal activation to the left ventricle, which takes 30 or 40 milliseconds, and then about 140 or 150 milliseconds across the lateral wall. So this is what we do to every patient with RV apical pacing. And we understand that this is what we try to undo with CRT. So is this how you want your own ventricle to activate if you needed a pacemaker? Well, three years later, she comes back with shortness of breath and congestive heart failure. And here you can see the ejection fraction may be around 30% at best, and the RV pacing increases over time over 40%. And this is PIC, pacing-induced cardiomyopathy. How frequent is PIC? Well, if you read, it may be as low as 10%, but it may be as high as 60%, but on average, a very reasonable estimate may be around 20% to 30%. If that is the case, then for every three pacemakers that go in, one person may have an adverse outcome or create heart failure, and that is quite significant. So much of the question to answer, should we be doing conduction system pacing for patients with AV block, is all based on what the true incidence and prevalence of PIC is. And I'm not certain that we know all the right answers of this, because you have a very wide distribution. The upgrade menu consists of coronary sinus cannulation, an epicardial lead by the surgeons, experimental LV endocardial pacing, which can also be done leadlessly now, and then his bundle pacing and obviously left septal pacing, which is left bundle area pacing. And in this case, we gave the woman a his bundle pacing upgrade, and in only 45 days, the heart function is back to normal. That's amazing because that's earlier than we think about for remodeling. So a lot of this is purely electrical rather than structural. For the fellows in the audience that are still learning EP, one pearl is that when you have two to one block with normal PR and narrow QRS, the site of block is not infrahisian, it is not avianodal, it is intrahisian. And here's a nice example of the fact that we even mapped this woman, and we were able to demonstrate intrahisian disease, H1, H2, and the block is between H1 and H2. So this is very nice physiology. So the preservation of conduction is first described 23 years ago by Pramod Deshmukh, who was a private practice physician, which goes to show that there is no such thing as academic medicine. Academic medicine is what is in your heart. And here she decided to stylet shape after avianodal ablation to be able to maintain QRS synchronization after AB block. Now it takes almost 20 years later when Dr. Vijay Raman, who was one of the new pioneers of his bundle pacing, connections and pacing, to look at the comparison of his group at Geisinger. Most of the time, if your group does not behave and they do things differently, we get sometimes upset. But sometimes it's great because it provides a comparison that's not randomized, but one part of the group was doing his bundle pacing, and one part of the group was doing RV apical pacing. And they looked at the differences between the two, and they saw that there was a very significant reduction in death, heart failure, and the need to upgrade. And what's more important is they almost saw a reduction in mortality for all comers, for those receiving a pacemaker, in favor of his, and heart failure hospitalizations were significantly reduced. So his pacing clearly can prevent the development of pacing-induced cardiomyopathy. So why don't we do this in every patient? Well, how many times do you need to do this to save one person? We need to look at the statistics, and you say, well, there's a 5% difference, which is a number needed to treat of 20. So for those that are not comfortable with saying, I need to do connections and pacing for everyone, they are correct because you have to do this in 20 patients, and 19 may not benefit, and one, you could prevent a heart failure. But if the pacing is over 20%, then the number needed to treat is only 12. And my assignment is for every patient with AV block, and that is going to be higher than 20%. So the number needed to treat of 12 is quite good in favor of preserving normal synchronization. But why do we not like to use his bundle pacing? Because the problem with long-term outcomes is that there is a need to replace the generator 9% of the time versus 1% of the time, and 7% for lead revisions versus 3, and that is a number needed to harm of 13. So for every 13 to 20 you put in, you will have an early battery replacement problem because the thresholds tend to be higher than myocardial thresholds, and sometimes the leads are not as stable. But nevertheless, his pacing is still in the guidelines, and with block HF criteria, if you have an EF of less than 50% that was shown by Ann Curtis's paper in New England Journal, and you expect to pace more than 40% of the time, and that is the number needed by given by David trial, that CRT or his bundle pacing is recommended as a class 2A. It is important to remind everyone that there was no randomized trial with his pacing. This goes to show how compelling physiologic pacing is with a narrow QRS that it makes the guidelines without any randomized trial. I was surprised, but remember this is level of evidence B, and even class 2B in patients with AV block at the level of the node with permanent pacing. So this is now currently in the guidelines, but the way that you can overcome the high thresholds with his is the beautiful story of left bundle area pacing, and I remember when Dr. Weijun Wang from Wenzhou, China was drawing on a napkin for me that he felt like he could do intraceptal pacing and engage the left side of the bifurcation, and this in fact has really swept the country not only in China, but the countries in Europe, in Asia, and the current countries in North America, and this is now the real mainstay of conduction system pacing. Why? Why is it the mainstay? Well, when you do his bundle pacing, it is very specific, the anatomic location. It is always at the tricuspid angulus between the left, the non, and the right coronary cusp, but when you do left bundle area pacing, it is a wide area that is shown by the left conduction system, which includes the posterior fascicle, the anterior fascicle, the septal fascicle that you engage, and like I showed, if you can get anywhere with high speed internet access, then you will have very rapid conduction with the left ventricular activation. These are some of my own tips for left bundle area pacing. In RAO, the catheter should be pointing into the screen, but LAO, we like to use a little bit of contrast to show that you are perpendicular and the sheath is well positioned here. The sheath positioning is the most important thing, just like transeptal puncture for atrial fibrillation, and here you can see the shape of this, and the orientation will make it much easier to penetrate the RV endocardium, and here is a nice image that we did with contrast injection and simultaneous rapid rotation. You can see the lead right there jumps into the septum, so we were showing contrast and the ability to be able, and I'll play that one more time, to penetrate the septum, so if it does not go in in the beginning, then you want to reposition the sheath, but this is here real-time illustration of intraceptal fixation to be able to pierce the RV endocardium. Once you get into the septum, it's much easier to advance, follow impedance parameters, and engage the left side of the septum. The Chinese have demonstrated a nine-segment model. We like to target segments four and five, and I just caution that if you are targeting segments one, two, and three, this is much more on the superior fascicle or the anterior fascicle. Seven, eight, nine is targeting more of the posterior fascicle. I personally think that it's probably best to try to get the common regions, which are the septal and the common left bundle, and it's very important to look in RAO and LAO. This is a case report that we hope we never have to see again because this is a lead in the left anterior descending artery. The problem with this is they did not look at RAO and look to see that they are in the anterior border. They were only looking in LAO and fixating this, so it's important to look at two views because this should not be a risk of left bundle area pacing. This is only because they're not looking at RAO and almost exiting the anterior superior border on RAO. The criteria for left conduction system pacing are quite complicated. What's important is looking at the overall activation of the left ventricle, which is called the LVAT or the peaked R-wave time, and in general if the LVAT is less than 75 or 85, which is the stem to the R-wave in V5 or V6, this has been correlated with conduction system capture in patients that have left bundle area pacing. This has really gained a lot of popularity in Europe, and I know it is gaining popularity in Latin America and South America because I have traveled to Puerto Rico, Dominican Republic, and seen a lot of this in Colombia. This is 2,500 patients in 14 European centers published in European Heart Journal called the MILO study, showing that there is a 93% success rate with bradycardia indication. So this is highly successful in being able to achieve normal synchronization of the ventricle. And lastly, I think that electrophysiologists have a lot to learn with clinical trials. We can learn from our heart failure doctors. How many patients were needed to be able to show improvement with SGLT2 inhibitor? Well, in these situations, you have over 3,500 patients to be able to show that you can reduce heart failure hospitalization. 3,500 patients. We rarely have this many in electrophysiology. And for an SGLT2 inhibitor to prevent heart failure or to reverse heart failure, the number needed to treat is 20. But we use guideline-directed medical therapy and SGLT2 inhibitors for all patients without asking any questions. But as I have shown you, with AV block, the number needed to treat is only 12 to prevent heart failure. And they will never need guideline-directed medical therapy. So I think it's important for us to look at heart failure outcomes as well. And when we look at ARNI or the Entresto or the neprilysin inhibitors, they needed 8,000 patients to be able to show 3% difference, which is a much higher number needed to treat. So in summary, ladies and gentlemen, RV apical pacing, I think, is dead. We are 100% conduction system pacing in phoenix because you will never get diaphragmatic stimulation. You will never get a late RV apical perforation. And you don't get as much dyssynchrony. If pacing-induced cardiomyopathy is maybe 30%, then left bone marrow pacing will prevent harm in many cases, maybe in one out of 11 cases, and reduce the incidence of heart failure. So we don't need heart failure medications. Thank you very much. Thank you very much. For time's sake, we'll go on. We'll see if we have any time for questions at the end. Next, I'd like to invite Dr. Brigada. He is Cardiology Chief of the Arrhythmia Department in Barcelona. And he's going to talk about Brigada syndrome after 30 years of the first description. Welcome. Thank you very much, Jody. Thank you, Ulysses. It's a pleasure to be here and to present to you a little bit of the history of the Brugada syndrome and what has happened during these 30 years. We say 30 years later and we keep learning. Let's remember a little bit of history. The first patient we ever saw with Brugada syndrome was this very young boy, two years of age, who was recovered from sudden cardiac death. We learned that his sister died suddenly at age two and we could also recover some diseases from the sister because she had some episodes of syncope before dying. This was in 1986. It was not until 1991 that we presented these two patients plus two other patients that we identified after sudden cardiac death and had similar abnormal electrocardiogram. We presented these in Washington in 1991 and then four other patients came to our attention. Finally, the paper published in 1992 with these eight patients, all of them recovered from sudden cardiac death. One of them died, the sister, and all of them presenting, which was an abnormal, strange ECG. The ECG looks like that with this ST segment elevation in right precordial lids with the right pundal branch block pattern in the majority of cases with a slightly prolonged PR interval and this terminal negative T wave in right precordial lids. Obviously, this was not due to any coronary syndrome and it was a persistent ST segment elevation at that time. We realized that when you had this abnormal ECG, some of the patients were presenting with this ventricular arrhythmias, polymorphic ventricular tachycardia, and sudden death. Then, as I said, we published these eight patients in this initial paper in 1992 where we were describing right pundal branch block, persistent ST segment elevation, and sudden death. The history has shown that not every patient has right pundal branch block, that almost no one has permanent, persistent ST segment elevation, and fortunately, only a minority have sudden cardiac death. This is part of the thing that we have learned during these 30 years. It was in 1996 that Miyazaki, for the first time, named the syndrome Brugada syndrome in this article. Since then, this is how it is known in the literature and in the textbooks. The very funny story about Brugada syndrome is that 30 years later, and the number of papers published that have in the title Brugada syndrome keeps growing, and the interest is still intact, and the number of centers around the world that are studying Brugada syndrome keeps growing and growing and growing, and the information we have about it keeps growing. Why? Probably because at this time that we describe it, it represents a completely new way of looking at sudden death and arrhythmias. And the introduction of genetics in cardiac arrhythmias, the introduction of the notion of channelopathies, and all these make the interest on the Brugada syndrome extremely high, but 30 years later it remains intact, and the people still publish a lot on Brugada syndrome, which is good because we keep learning. We know that as any genetic disease, and we will talk about that in one minute, there are different prevalences around the world, and the highest prevalence in the whole world is located in Southeast Asia, Philippines, Thailand, Cambodia, south of China, Japan, et cetera. And for the rest of the world there are different prevalences that have been described. So the most important issue is that this is a fact, is a disease that is based on an ECG, and of course looking at an ECG is a very subjective matter because I can see the ECG in one way and my neighbor can see it in a completely different way. So that's why it was important to try to clarify what is the abnormal diagnostic ECG, and it's what we call the type 1 ECG. So you can only make the diagnosis of Brugada syndrome if you have a type 1 ECG. The type 2 and type 3 are suspicious ECGs but are not diagnostic ECGs. They can be due to many other causes, and you have to prove that the patient has a type 1 ECG before you put the diagnosis or you label the patient as Brugada syndrome. And we know that, and this comes from the notion that at the beginning it was the persistent ST segment elevation, but we realize that this is not true, that the ECG normalizes in many, many patients during the follow-up. And also we know that because this is related to a fact that there is some defect or some imbalance of currents in the right ventricular outflow tract, it will depend on the location of the outflow tract in relation to the thorax of the patient that we will record the abnormal ECG at different levels in the thorax. That's why it is so important to record in second, third, and fourth intercostal space the right precordial leach, because you might miss it if you only record it in the fourth intercostal space, as is shown in this slide. So as I was telling, the ECG normalizes over time, and this is a classical picture that I'm showing since many, many years with one of our first patients that we are following leach V1 to V3 during a 2-year period. And you see that the ECG every time is completely different. There are not two identical ECGs, but even, and you see the red arrow, you have at one time the ECG is completely normal. If we have seen the ECG only, this ECG only at that time, we had never made the diagnosis of Brugada syndrome. So this, of course, was complicating everything in relation to diagnosis because we might have normalization of the ECG in patients like this one that you see had two episodes of ventricular fibrillation before and after recording this normal ECG. So a very high-risk patient in this case. So that's why Miyazaki and others described how we can manipulate the ECG by using antiarrhythmic drugs. And very strange in the field of electrophysiology, in Brugada normally in the field of electrophysiology you use antiarrhythmic drugs to treat arrhythmias and you use isoproteranol to induce arrhythmias. In Brugada syndrome you do the contrary. You use antiarrhythmic drugs to induce more the disease, like class I antiarrhythmic drugs is the diagnostic test to unmask the abnormality, and you use isoproteranol to normalize the ECG. So from the physiopathological point of view this is extremely interesting because it's a completely abnormal behavior in the field of electrophysiology. Of course you have to exclude other causes that might produce a similar ECG. That has been called phenocopies of Brugada syndrome because we know that there are situations that produce an ECG that might mimic a Brugada ECG, but when the situation is over the ECG normalizes. This is the case in a patient with a pectus excavatum, for instance, or a mediastinal tumor. These are just two of the examples of phenocopies, but you have dozens of different causes that might produce a similar ECG and they are not related to Brugada syndrome, but the cause that produces the abnormal ECG. We learned, and we knew it since the very beginning, that it was a familial aspect in the disease, it was after we found this family in Italy where we could analyze the family with nine members affected and we could make the genetic testing in this family and after the genetic testing we described for the first time the mutation in the sodium channel as the cause of the disease in Brugada syndrome. Since then more than 300 mutations have been described in the sodium channel. All of them have the same pattern of action. It is to premature closing the sodium channel. And by premature closing the sodium channel you create an imbalance of currents in phase one of the action potential where the ITO and INA are competing because the INA is no longer there. ITO has no competing current and everything creates an imbalance of currents and a gradient, an electrical gradient, between different parts of the heart, basically in the right ventricular epicaldium. That's where ITO is predominant in the human heart. But, of course, this is a simplification of the situation. We like to say one mutation, one disease, and this is true for Brugada, for Long QT, for Short QT, whatever, but the genetics is probably much more complex than that. Let me show you this example. This is a patient from Barcelona. We saw this patient and he had two mutations in the sodium channel. His father had one mutation, his mother had the other mutation, and he had two daughters. One daughter inherited one of the mutations and the other one the second mutation. And when you look at the ECG, it is very funny because the patient has a prolonged PR interval and a very nice type 1 ECG. In fact, he was recovered from sudden cardiac death and he has an implanted ICD. The father and the mother had no Brugada ECG, but the mother had a prolonged PR interval. And the two daughters, none of them had the Brugada ECG. One had a completely normal ECG. The other one had a prolonged PR interval. So this is a kind of demonstration that putting together the two mutations was producing Brugada syndrome in this family, but each one of the mutations separately was not. So we still have a lot to understand. We still have a lot to learn about genetics. We learned that the mutation producing Brugada syndrome is temperature-dependent. That means that the higher temperatures produce more defect in the channel that has the mutation. And this is very interesting because in the very first patients, the eight initial patients, two of them had ventricular fibrillation coincident with febrile episodes. So now we know that the febrile fever is also a very good diagnostic test for our patients, especially for children. And when we have a child in a family with Brugada, we always ask the parents to bring him to the emergency room before giving anything to check for an ECG when he has fever because if the ECG is negative during fever, you can be sure that he has not the mutation and that he will have a pharmacological test that will be negative and probably is not necessary to do the pharmacological test in the child. When we look in this study that was a worldwide study, it was shown that the episodes, the severe episodes like silent death and ventricular fibrillation related to fever were concentrated in the very young patients, in the younger age. So that's where we have to check very carefully the influence of fever in the very young patients or family members of patients with Brugada syndrome. We understood that females were somewhat protected during Holowat about the incidence of silent death and ventricular fibrillation, not on the incidence of Brugada syndrome because in a given family, males and females do have the same percentage of affectation. But during Holowat, females are somewhat protected and there are several theories about that, hormonal, et cetera, et cetera. We are still working on it. We also know that the patient that has had one episode of ventricular fibrillation or silent death is very prone to have a second one. And this is a patient at an extremely high risk. Nobody discussed that. Also, the patient that had a cardiogenic syncope has a high risk of ventricular fibrillation. And there is some discussion about the patients that are more and more diagnosed as asymptomatic because we identify an abnormal ECG in a routine checkup or because of family screening. One of the things that has been over discussion for 30 years is the role of programmed ventricular stimulation. You know that we have always been defending that this is extremely important for risk stratification. And, in fact, again, just how it reacts to the different drugs is something that should attire our attention. Because, again, when we have a patient that is inducible, we give an antiarrhythmic drug to make him non-inducible. And if he is non-inducible, we put isopreterinol to make him inducible. In Brugada, it's the contrary. You use antiarrhythmic drugs and the patient remains inducible. You give isopreterinol and you cannot induce the patient. So, simply from the physiopathological point of view, the role of programmed electrical stimulation is extremely interesting and we should not miss that. But also, when you look at the follow-up, and now even those that were saying that programmed electrical stimulation is not useful, they even are now accepting. And they published a circulation three years ago where they say that, yes, finally, programmed electrical stimulation is good in risk stratifying patients with Brugada syndrome. And our data are very clear on that. So, for us, being a male, having hypersyncope, sudden death, or being inducible in a patient with a type 1 ECG are the characteristics that identify a patient at risk of sudden death. And these are the characteristics that should be protected. So, when you look at the male that is inducible and had a positive ECG, you see that his likelihood of having sudden death during his whole life is almost 50%. If you are a female, non-inducible, and based on normal ECG, only abnormal after drugs, your likelihood of having sudden death during your whole life is only 3%. So, when we have a patient with Brugada, we have learned that the Brugada behaves in a very funny way. The patient is completely asymptomatic for years, and then suddenly he has an electrical storm. And he has 8, 10, 12, 15 shocks in 1 or 2 or 3 days, and then remains asymptomatic for another 3 years. So, of course, when you have these electrical storms, you have to behave, you have to do things. One of the things is using isoproteranol. So, isoproteranol is a very good therapy to stop the electrical storms in patients with Brugada because it normalizes ECG, normalizes the disability of arrhythmias, and normalizes electrical storms. And this is something that should be considered as the first-line therapy when you have a Brugada electrical storm. But you can also use quinidine, which has been used in many patients, and this is a study done in Spain some years ago, where it showed that two-thirds of the patients with quinidine during multiple shocks were normalizing the number of shocks. But it was the group of Dr. Zeib that, some years ago, started studying a model of Brugada syndrome, and they realized that doing ablation in the epicardium of his model they were able to avoid inducibility of arrhythmias. This was used by Nat Demany some years ago, some ten years ago, in nine patients. He did ablation of the epicardium of the right ventricle and he found that he was normalizing the arrhythmias in these patients. And we also did that together with Dr. Pappone in Milano in 135 patients. And we realized that when you were mapping the patient, the endocardium, the epicardium of the patient with Brugada syndrome, at the moment he had the normal ECG, the mapping was completely normal in the epicardium and endocardium. And only became abnormal when you gave antiarrhythmic drugs to mimic the ECG, not to mimic, to unmask the ECG in the Brugada patient. And we had these very funny and strange signals in the epicardium, very long signals, very different from the, let's say, fractionation or slow conduction that we see in structural heart disease and that were only present when ST-segment was elevated, not when the ST-segment was normal. So it has nothing to do with structural disease. It was absolutely functional and, to our view, something that relates clearly to the repolarization. And this is the ECG during the radiofrequency start. You see a very nice type 1. And then you see how the morphology of the ST-segment elevation keeps changing while we are eliminating one layer of the epicardium. And finally, this is the result. So you have a patient with a type 1 ECG. After radiofrequency, the type 1 has disappeared. You have an ascending ST-segment elevation. And you give flecainide at full dose and nothing happens. So we have normalized ECG. So this is the epicardium in the right ventricle in basal with the ST-segment elevation. You have the abnormal signals located there. When you give flecainide, you get more ST elevation. You get more abnormal signals. The area is bigger, and the duration of the signals is bigger. And after the ablation, you see here the voltage before and after. In the epicardium, of course, you reduce the voltage of the epicardium. You normalize the duration of the signals, and the activation map does not change because you are doing an epicardial ablation or a transmural one. So the activation of the right ventricle remains unchanged on the endocardium. So this is the ECG with isomalin before the ablation. And six months after the ablation, there is nothing. No change at all in the ECG with the isomalin. So the pharmacological testing has become negative. This is the inducibility before and after. The patient remains non-inducible six months after the ablation. And probably this is, and we published that in circulation some years ago, with this 135 patients, showing that probably what we are doing with the ablation is normalizing the phenotype of the disease. We are not changing the disease. We are not curing the disease because it is a genetic disease, but we are modifying the phenotypic anomalies related to the disease. We normalize the ECG. We normalize the pharmacological testing. We normalize inducibility of arrhythmias. And we hope that this will result in non-arrhythmias during follow-up and probably will be an alternative therapy for the future in low-risk patients. And, of course, this is my last slide. This is the Brugada family. Pedro saw the first patients in 1986. Ramon, who did the genetics while he was working in the States, and now he's in Girona, chief of cardiology. Georgia, which is our niece, that is doing the pediatric patients with channelopathies and myself. Thank you very much. Thank you, Dr. Brugada, for your excellent presentation. The next speaker is Dr. Judy Hurwitz. She will talk about anticoagulation for AF, what's new, and where we are going. Thank you very much. So I'm delighted to be here. I was asked to talk about anticoagulation for atrial fibrillation, what's new, and where we're going. And really what I want to do is talk about stroke prevention in patients with atrial fibrillation and really about the duration of atrial fibrillation and what's important and what's not. So I think it's safe to say that if someone documents 48 hours of atrial fibrillation, we know what to do. We go with the guidelines in terms of anticoagulation. I think for most of us in terms of patients who have 24 to 48 hours of atrial fibrillation, we also know what to do, and we go with the guidelines. But as you can see all the way over on the right, with atrial fibrillation of less than 24 hours, we still have questions about this. How short is too short to cause a stroke? How long is too long not to be treated to prevent a stroke? So what we're really talking about is atrial fibrillation associated with thrombus formation. As you can see here, in 1942, Hay and Levine commented that auricular fibrillation increases the incidence of auricular thrombosis. And what's the temporal relationship between atrial fibrillation and thrombus formation? On the right, you can see Virchow's triad as far back as in the 1800s that you can see that you need stasis, you need vessel wall injury, and you need hypercoagulability really to form thrombus. So where does that lead us? Again, the real question is, does atrial fibrillation cause a stroke, or is this just a marker for stroke? As you can see here, an example from a device, there are three different episodes of atrial fibrillation. And in the middle section, right after those episodes of atrial fibrillation, that's really when you would think that there's the highest risk of stroke occurring. But as you can see all the way over on the right-hand side, the blue line, that's when the stroke occurred, several days after the episode of atrial fibrillation. So again, looking at studies from ICDs and pacemaker patients and the timing between atrial fibrillation and the stroke, is there really an association, or is there really a dissociation? These are three somewhat different studies. Both have about 2,500 patients in them. And they all had devices, and they looked at the timing of stroke, and they looked at the timing of atrial fibrillation. So the red lines in each of these are the incident is the episode of stroke. The black lines here are the incidents of atrial fibrillation. And you can see that there are times when the stroke happened at the very beginning of an episode of atrial fibrillation. There are times when a stroke happened well during an episode of atrial fibrillation. And there are plenty of times here when strokes occurred and there was no atrial fibrillation, or the atrial fibrillation happened much later than the actual stroke. These are just a little bit more of a deep dive into a couple of studies. Again, looking at the incidents of silent atrial fibrillation in relation to thromboembolic events. And you can see kind of over on the right-hand side, again, the vast majority of these patients did not have episodes of atrial fibrillation within the 30 days before a stroke. There are a variety of different numbers of patients in these studies. But what I really want to draw your attention to is, I think it's the fourth column, the definition of an episode of atrial fibrillation. This varies tremendously, and it has in all of these studies, from anywhere from 5 minutes to 5.5 hours. And I think that's one of the big issues that we're finding now with a lot of these studies is the definition of atrial fibrillation. So it goes back to the very first question that I was asking, what duration of atrial fibrillation really warrants anticoagulation? And these studies have really looked at different timing, 6 minutes, 6 hours, 12 hours. Again, I think we know for 24 hours what to do. This is a really interesting and nice study that I think reminds us that we're not just treating an arrhythmia, we're actually treating patients. So Kaplan in 2019 looked at about 21,000 patients, looked at electronic health records and care link data, but combined the timing of episodes of atrial fibrillation to Chad's VASC score. And again, Chad's VASC scores remind us that we're dealing with patients who have multiple morbidities. And I think it's interesting to see if you look all the way over to the right that if you have a Chad's VASC score of 5, even with no atrial fibrillation, the risk of a systemic embolism or stroke is higher than if you have 24 hours of atrial fibrillation but a Chad's VASC score of 0. So we really need a lot more information before we can answer some of these questions. NOAAFNet, which is a study that included patients who had documented atrial high-rate episodes of at least 6 minutes, patients who were at least 65 and had one Chad's VASC factor, were randomized to treatment with endoxaban or placebo. And this data was published simultaneously at ESC and in the New England Journal last month. And what they found was that there was no difference in stroke, systemic embolism, or death from cardiovascular causes between the two groups. And in fact, there was an increase in the major bleeding or death from any cause in the endoxaban group. Another study, Atresia, which is looking at sort of similar patients, so subclinical atrial fibrillation patients who have devices that range between 6 minutes and less than 24 hours, who have a Chad's VASC score of at least 3, were randomized or have been randomized to placebo or apixaban. And the information from this study will be presented next month at the American Heart Association. We're all looking forward to this data. It's interesting that the previous study was stopped early for futility, and this is not, but we don't know what the exact results will be. The Danish Loop Study also I thought was very interesting. It looked at 6,000 patients who were at least 70 years old and had either diabetes, hypertension, congestive heart failure, or prior stroke, and they randomized these patients in a one-to-three fashion to implantable loop or standard of care. And the idea was to look for ischemic stroke or systemic embolism. It's really no surprise that there was a higher incidence of atrial fibrillation detection in the loop group compared to the control group, and that translated to a higher incidence of oral anticoagulation. Interestingly, there was no difference between major bleeding between the two groups. But really what was interesting is that in terms of 6-year follow-up, there was no statistically significant difference between stroke, systemic arterial embolism, ischemic stroke, and as you can see on the two slides here. And so really what it showed, of course, was that even though there was a screening that showed three times increased incidence of episodes of atrial fibrillation, again, which translated to anticoagulation initiation, there was no significant reduction in risk of stroke or death. So really, is the question not all atrial fibrillation is worth screening for or not all atrial fibrillation merits anticoagulation? But again, perhaps we're looking at atrial fibrillation the wrong way, and it's not really duration, it's burden. And interestingly enough, in this study, more than half of the patients had a reduced atrial fibrillation burden in the last half of this study. So maybe really we're looking at this the wrong way, but we don't have that information yet. So what I want to sort of end with is two interesting studies or ways of looking at this that may change what we're going to do. The first is the REACT study, which is the Rhythm Evaluation for Anticoagulation Therapy for Atrial Fibrillation. It's an NIH-sponsored study, and it's a one-to-one randomized trial of chronic NOAC therapy or watch-guided targeted NOAC therapy. So what the watch-guided targeted NOAC therapy group is going to have, they're going to have a special Apple Watch, not available except for patients in this study, and they are going to self-administer one month of anticoagulation if the watch tells them they have episodes of atrial fibrillation. If they have no atrial fibrillation after their 30 days, they stop it. And the idea is looking for ischemic stroke, arterial embolism, or oral cause mortality. And the study is just getting started. So that will help us a little bit more in terms of really how to recommend anticoagulation and what's significant and what's not. And then, again, I can't leave this topic without talking about the factor XIa inhibitors. So this is totally different. These are decreased pathologic thermo and preserve the ability to form clots in response to bleeding. So on the left-hand side, you have the intrinsic pathway, and you can see where factor XIa inhibitors would work. But it doesn't affect the extrinsic pathway. The Pacific trial was published and showed that in patients who received apixaban, they had three times the bleeding risk of those who took a factor XIa inhibitor. The Oceanic trial is ongoing right now. There's no information about it whatsoever yet out in the literature, but it's going to be looking at these drugs. And I think really this could be a true game changer in terms of how we treat patients, their risks, et cetera. There are this coming up in the next six months. There's going to be two AF guidelines that are going to be published, and we're all looking forward to that information. So that may also change a little bit of what we do, especially looking at burden rather than duration. So in conclusion, atrial fibrillation of 24 hours, well, we know what to do. We anticoagulate. Atrial fibrillation that's less than 24 hours, I think we have to remember we're dealing with patients again, and we need to look at the risk factors. We don't have a magic number of what the duration should be. And, again, the question is, is this duration that we're looking at or is this burden? And we don't have that information yet. In terms of atrial high rate episodes, at least from no AF net, we know don't treat them. That's not going to help. We'll see what, again, atresia shows us coming up in the next few weeks. And, again, we need to continue to wait for randomized clinical trials and the AF guidelines, which should be very interesting. So thank you very much. Thank you so much, Dr. Judy Hurwitz, for your excellent presentation. The next speaker is Dr. Luis Aguinaga. He is cardiologist, electrophiologist at the Integral Arrhythmia Center in Tucumán, Argentina. He will talk about update on risk stratification for sudden cardiac death. Good morning. Thank you, Dr. Roque. Thank you, Dr. Hurwitz. It's my pleasure to be here. First of all, I would like to thank the organization committee for inviting me to take part in this very important joint session. The topic is risk stratification of sudden death. I will try to give a challenging presentation. First of all, I will talk about ischemic heart disease due to its high prevalence around the world. As you know, in ischemic heart disease, sudden death is related to two clinical scenarios. The plaque substrate that occurs during acute coronary syndrome with thrombosis and the post-myocardial infarctor scenario related to fibrosis. But in both scenarios, ventricular arrhythmias or sudden death can occur. As you know, current guidelines recommend the ICD therapy implantation in patients with coronary artery disease with ejection fraction less than 35%, functional class 2-3 or heart failure, despite of three months of optimal medical therapy. But these recommendations are based on old trials, on some old evidence, mainly due to a tool that was published 20 years ago. But patients, therapy and evidence have changed in this time. Due to that, I will analyze every tool or every element of these recommendations and maybe its limitations. For example, ejection fraction. Ejection fraction is not a sufficient criterion for re-stratification of sudden death. For example, in patients with ejection fraction less than 35%, there are a lot of patients with high risk of non-arrhythmic death, therefore less ICD benefit. Or, on the contrary, patients with ejection fraction more than 35%, a lot of patients or many patients can suffer sudden death. Additionally, some studies, as this study published two years ago, this European study based on clinical effectiveness of primary prevention ICD, showed that only 7% of patients received at least one appropriate ICD shot in a follow-up of 2.5 years. This is a little percentage or a very low percentage of therapies in this group of patients. And patients older than 75 years and diabetic have less benefit. And optical medical therapy. As you know, in the last 15-20 years, there was a huge evolution in medical treatment on heart failure. For example, since 1995 to 2014, there was a 44% reduction of sudden death across heart failure trials due to the incorporation of agiotensin inhibitors, beta-blockers and mineralocorticoid antagonists. Maybe we need to start new trials including these drugs. And lastly, drugs like saccubitril and glyphosate have shown a reduction of 20% in ventricular arrhythmias and sudden death. And with each drug added in patients with ICD implanted, it was observed a 34% reduction of death. Regarding the time of optimal medical therapy, three months could be premature. For example, with saccubitril, recent trials have shown at 6 months, 32% of ICD-eligible patients become ineligible due to the increase of ejection fraction. And at 12 months, 62% of patients become ineligible. And also, with glyphosate, at 12 months decreased the risk of ventricular arrhythmias or sudden death. Again, maybe we need trials using these new pharmacological treatments. Regarding to the functional class, we can say that, as you know, guidelines say functional class I do not have benefit from final prevention of ICD. And again, these major trials like EMPEROR and PARADIGM in heart failure using saccubitril and glyphosate have shown that they can improve functional class. And 70% of patients had functional class II before the treatment in these trials. Due to that, a lot of patients do not have meeting indication of ICD in primary prevention of sudden death. Due to that, we can say that delaying ICD implantation for 3 months may be earned. We can expect more benefit with medical treatment. And delaying the ICD implantation, we can decrease the number of unnecessary ICD implants. We can reduce the risk of ICD implantation, infection, inappropriate shocks, and cost. A topic very important in our Latin American countries or developing countries. But we know the concept argument that some patients may have sudden death awaiting the improvement of ejection of functional class. And this could be minimized with a personalized approach and with new tools for risk stratification. Let me show you this study that we published 8 years ago in Haridu. We included patients with ejection fraction, left ventricular dysfunction, 28%, with criteria for primary prevention of ICD, with high burden of PVC, with performed ablation. And after that, 64% of patients went off of the indication. As you can see here in red, patients with no ischemic heart disease from an initial number of 55, just 23 remained with indication of ICD at 12 months. And for initial number of 11 patients with ischemic heart disease, just one remained with indication. This study showed that waiting for the recovery of ejection fraction could be safe. Any patient suffers sudden death in this time. And it's important to remember that the majority of events post myocardial infarction of sudden death occurred in the first 3 to 6 months in different levels of ejection fraction. Due to that, the new directions in sudden death should be We need to improve the risk stratification to select patients with more benefits of ICD, to select patients with less benefits to decrease costs and to decrease complications of ICD. How can we do that? With two specific elements. The first is using the risk scores. And the second is using the new tools for risk stratification. The risk scores, I will show to this one. The risk score obtained from a made-up study. They compare the ICD benefit in a personalized evaluation and compare the risk of arrhythmic versus non-arrhythmic mortality. There is a calculator to compare the benefits in patients. For example, older patients and diabetics with ejection fraction less than 25% have less benefit with ICD therapy. And regarding to the new tools for risk stratification, I would like to mention the electrocardiogram and specifically this study from Finland that the authors found that with this electrocardiogram abnormalities in general population, if you find more than three abnormalities, these patients or these people have 10 fold risk more than the general population risk of sudden death. Additionally, this year was published this Japanese study that included patients with guideline-based ICD indication due to the ejection fraction and functional class. They improved the sensitivity and specificity using artificial intelligence in the electrocardiogram. It's a very important tool. As you can see in the raw code, they improved both sensitivity and specificity. As you know, maybe the most important tool is cardiac MRI. With cardiac MRI, we can evaluate myocardial fibrosis, the burden size, localization, characteristics, and it's the software for ventricular arrhythmias. Let me show you two studies. The first one is from Leiva, from United Kingdom. The authors studied patients with ICD implanted. They found that, for example, patients with myocardial fibrosis, visual assessment, and total fibrosis and Grayson fibrosis were predictors of sudden death. They proposed this level of risk. In the first group, no myocardial fibrosis visual assessment is a group with low risk. The percentage or the probability of sudden death is 0%. The negative predictive value is so important. On the other hand, in the other group, high risk myocardial fibrosis visual assessment and Grayson fibrosis positive select the high risk group. And most important, this study published last year in patients with ejection fraction more than 35%. They compared Grayson, total fibrosis, and ejection fraction. As you can see in the raw cures here, the Grayson fibrosis was more important than total fibrosis and also than ejection fraction to predict events of sudden death. But not all images in resonance are equal. As you can see, there are a lot of parameters that we can evaluate in MRI study. The transmorality, entropy, radiality, interface, local patterns, etc. And we don't have much in learning to help us in the analysis of these patterns. For example, in the left, you can see a regular pattern of fibrosis in a patient with no events of ventricular arrhythmias. In the right, irregular patterns you can see over there. These patients have events of ventricular arrhythmias. And there is an ongoing trial using MRI in patients with ejection fraction between 36% to 50% to evaluate the value of MRI. Another important tool is the biomarkers in patients with heart failure and specifically the NT-PRO-BNP. In recent trials of heart failure, for example, DAPA heart failure trial, NT-PRO-BNP was the largest predictor of ventricular arrhythmias of sudden death. Or as it was shown in this very important meta-analysis published this year, evaluating NT-PRO-BNP in patients with ICD, they showed that it was a strong predictor of all-cause mortality and also was a predictor of ICD shocks. And finally, the polygenic risk. As you know, genetics is a strong predictor of risk stratification of sudden death. I would like to show this study published last year. They included patients with ejection fraction between 30% to 35% coronary artery disease and they performed a genome-wide polygenic score and they identified patients with high risk of sudden death, specifically in a very large group of almost 50,000 patients. But all these new tools should be evaluated in the concept of shared decision-making process. We must evaluate the risks or explain to the patient the risks, the benefits and the expectations of the therapy, ICD or no ICD. In conclusion, risk stratification of sudden death should be improved. Current guidelines are based on all evidence. 20 years ago, the major evidence was published for the actual recommendations. And there are new elements to identify patients with more benefits. Electrocardiogram, MRI, genetics and biomarkers. And we should use a personalized approach in the shared decision-making processes. And finally, I would like to show this picture. This is a historical picture and this is a joint session. This picture was taken 10 years ago. And this is our first agreement between HRS and SOLA-S. Dr. Kolkis, Dr. Fogel, Dr. Anguillez and J.S. Jamblon. We signed the first agreement between HRS and SOLA-S, now Latin American Heart Research Society. And thank you for your attention. Thank you very much. This was a great session. And I'm afraid we don't have any time for questions. We need to go on to the next session. So, enjoy and thank you.

anticoagulation

atrial fibrillation

stroke prevention

risk stratification

ventricular arrhythmias

pharmacological treatments

NOAC therapy

sudden cardiac death

ischemic heart disease

personalized approach