Cardiovascular Diseases and Cardiac Surgery

Overview

In this Heart Rhythm Challenging Case Discussion, Mayo Clinic cardiology experts Samuel J. Asirvatham, M.D., Abhishek J. Deshmukh, M.B.B.S., Christopher V. DeSimone, M.D., Ph.D., and Siva K. Mulpuru, M.D., M.P.H., discuss structural ventricular tachycardia (VT).

For more information or to request an appointment, click here.

SAMUEL J. ASIRVATHAM: Thanks, everyone, for joining in. And remember, we want to discuss your questions, your cases, anything that you'd like to discuss. Today, we just have a focus just as discussion points about ventricular tachycardia, ablation in structural heart disease.

So for those of you who sent in cases that you had questions on, most of you, either me or one of my colleagues, has responded to. For those of you who'd like to present a slide or a case or come up as a panelist and discuss your question, please just let us know and either use the Q&A or just come on the chat and let us know you'd come-- you'd like to have a question discussed. And we can bring you up here.

There's a few of you who had a case that maybe fits in with one of the other sessions that we'll have as this progresses, the series progresses. And we'll make sure you're available on that day to discuss that. So maybe I'll just get started, and then we'll have our panelists and any of you who'd like to bring up specific questions. We've, for those of you who emailed questions, have taken the four or five most common questions we've got, and we'll make sure that we cover those as well.

So just a couple of things on where we see the bulk of questions, either that we had as we developed in our careers or what you have when you come up with cases. And I'll just highlight maybe five or six themes that we can use as a discussion point.

So one of them has to do with signals. So we have a lot of different types of terms for signals, pre-systolic, late systolic, mid diastolic, near field, far field. Is that more important? Is the voltage itself that we pick more important? And to have some kind of mental construct, electrophysiologically, what are the signals, and how do they tie in to these circuits for ventricular tachycardia that patients have? We'll discuss this and use some examples of how to analyze in the context of a given case.

As background, I think this picture goes over really up on what we are up against, and very useful for trainees especially to keep in your mind of what it is that we are recording with those signals. And I'll make a few points from this. So first one that you can recognize is there is really the definition of what is scar versus unhealthy or normal tissue is a continuum. They closely coexist with each other.

And places where the myocardium that's viable dominates has a different type of signal. Those where the fibrosed tissue dominate have a different type of signal. And to be able to interpret all the ways that we map-- entrainment mapping, mapping of substrate, activation mapping-- looking at the signal that we see and correlating in our mind's eye what the picture would look like is one of the great challenges and most useful skills that an electrophysiologist has.

You can also see a couple of other things that the substrate in the same patient in some areas is predominantly epicardial. In some areas it's predominantly endocardial. So it's the same patient, but we just have to be prepared rather than going in with the idea of this is an epicardial substrate versus this patient has an endocardial substrate. On a given day, the relevant circuit in the same patient could vary from one to another.

You can also see in this picture another common issue is there's structures within the cavity of the heart. So for example, cut section of the papillary muscle, that may have a superficial substrate. In another situation, it might be just hiding a deeper substrate where it's very difficult to flatten out to a two dimensional image and create a map that says, here's what we saw, because the catheter that's here is actually mapping three or maybe even four dimensions of normal and unhealthy tissue.

And final point I think that's good to look at is the role of the septum. So the septum is one step more complex than just the endo epicardial substrate issues or endocavitary structures. It's all of the above, plus we have this delineation for right versus left ventricle. You'll notice, for example, the myocardium of the right ventricle, it just tracks differently from the left ventricle.

But substrate here with an exit somewhere here would be an LVVT, but that substrate might be easier to reach from a catheter in the right ventricle. So issues related to vantage point, access, adequacy of mapping, and how we handle complex substrate are all just essential prerequisites to picture to even get close to success in structural heart VT.

Another level in our mind's eye just looks at the histological equivalent of what we saw there with the naked eye. And that is that we have tissue that survives even what we call as a healed infarct, a transmural infarct. We routinely will have tissue that survives. And that surviving tissue is highly arrhythmogenic.

And where we see that surviving tissue is typically at the edges and on the endocardial surface. And the reason for the endocardial surface being something that survives is the Purkinje network with its own kind of glycogen and energy stores tends to survive even a lot of ischemic infarcts or other types of injury to the myocardium and often is what's the surviving strands that we map endocardially.

Just to show some other challenges here is just the thickness of the myocardium, especially like in hypertrophic cardiomyopathy. We may have patchy scar circuits that join. But to be able to reach and going through and trying to spare as much as we can of normal myocardium is a challenge. And we should always [INAUDIBLE] not only the challenges that the substrate gives us but a thorough understanding of the limitations of our own mapping and recording.

So a simple thing like catheter orientation where only the distal electrode is making contact with tissue versus having a distal and ring electrode making contact with tissue can give us a completely different picture of what we're analyzing from that substrate. And I know several of you had questions about what is the best target? I believe the question was fragmented or late signals. And we'll address this issue of catheter orientation when we talk about that.

Another concept to keep in mind is near field and far field has a very nuanced meaning in structural heart VT. It's not like what's close to our contact point versus what's further away. It's that plus we have to look at the health of the tissue that could make something that we're making contact with but not look like a near field signal. We have highly fragmented signals. And then we have epicardial fat that can separate the closest tissue that we have to the electrode but that separation gives this far field flavor. So another challenge important to understand straight off the bat.

And I believe one of you wanted to come up and ask a question very specifically about the role of conduction tissue arrhythmia as how would we map with point-to-point and multi electrode catheters. And we're happy to discuss that as well. And we'll go through each of these and concepts or things that are important for substrate mapping.

Now, we could maybe get started with some of the questions. Please feel free to send them in live as well. I'll take the questions as we come up. And I would like if maybe we could bring up Dr. Friedman as a-- up to the panel here. And I will use his experience to discuss the first question. And then we'll go to the other questions as we keep going.

So one of the very common questions that's a theme that comes up is when we say we're mapping substrate what exactly are we primarily mapping? So some of the questions were related to use voltage pros and cons. Others fragmented signals are the key. And others where like late potentials. Should we be ablating the late potentials primarily?

So a lot of things here are based on judgment, but we always fall back to our basic constructs and the type of arrhythmia we're ablating. So maybe I'll ask Dr. Friedman here to maybe just give from experience what are the really tough challenges for structural heart VT and maybe get us started on this discussion on fragmented versus late potentials.

PAUL A. FRIEDMAN: Sure. Thank you, Sam. Structural heart disease is a broad category, of course. And I think one of the first things we think about even before we put catheters in the body is what available information do we have in terms of surface ECGs, ventricular tachycardia, and knowledge of the disease to suggest where we might find abnormalities.

And so one example would be for arrhythmogenic right ventricular dysplasia, or dilated cardiomyopathy, where we'd be thinking more about how we might approach the rhythm meaning epicardial versus endocardial. And then you nicely showed the structural abnormalities. And I'll just make one or two brief comments. And it looked like maybe you had some additional examples to show. And that is--

SAMUEL J. ASIRVATHAM: [INAUDIBLE] I just wanted to emphasize something you said.

PAUL A. FRIEDMAN: Oh, sure. Please go ahead.

SAMUEL J. ASIRVATHAM: So the EKG, so especially for trainees or newer electrophysiologists, EKG is what we really grow up on for as electrophysiologists to help us to know where to ablate. But with structural heart VT, with reentrant VT, the ECG rarely will point us to where we need to ablate. And the reason for this is it is a circuit. And this circuit where it transitions from abnormal tissue to normal tissue produces this concept of the exit.

So the exit is where that tissue is becoming relatively healthy tissue. So wherever the circuit is held, where it exits out to normal tissue defines the ECG. So I think the point that you are bringing up as we study the ECG not to know where to ablate but to give us an idea of that transition site. So using that and then looking at voltage and other aspects to tell us where compared to the exit is the abnormal tissue, where we can focus our ablation in.

PAUL A. FRIEDMAN: Which brings up the point you said, there's bad tissue and then just an exit site. And what we really want to know is, is our catheter next to bad tissue, number one? And number two, is that bad tissue part of what's causing the arrhythmia? And if we simply start by saying, how do we recognize bad tissue-- and you brought up several different points. One is voltage size. One is late potentials, and one is fractionation. And I think they're really important considerations.

When we look at voltage, the basic idea is that if healthy myocardium is replaced with scar, it won't be able to generate as much electricity. So the voltage may be smaller. But the caveat there is that there are other things that could make the voltage smaller that we have to be mindful of. And one is, are we in good contact? So in the endocardium, there are many structures we don't typically see. And if the catheter isn't making good contact, that could give you a false low voltage.

The bigger issue is actually in an epicardial position, where often there may be intervening tissues between active myocytes and the recording electrodes. And that could be fat, for example. And so you have to be particularly careful to say it's scar based on voltage alone when you don't have good contact or when you're in the epicardium, and there may be fat between your catheter and working myocytes.

In contrast to that would be-- and I'll pause in that in a second-- would be fractionation or late potentials, which suggests that there's abnormality in how electricity is passing through myocytes. And it gets back to that beautiful and a picture you showed where you picture water flowing down a stream very smoothly. And then all of a sudden, it hits pebbles or the scar that you showed us.

And we start getting small chaotic signals that don't have as much myocardium behind them. And they may have varying frequency content and then would give rise to those abnormal either fractionated or very late signals, because they just have to work their way through a lot of tissue. But yet, they haven't gone far from the catheter, because we're still recording them, just much later in time.

SAMUEL J. ASIRVATHAM: Thanks, Paul. So maybe, Siva, if you want to pull up an example of an abnormal signal. I'll just tabulate some of these terms and then see what it is that we're trying to target here. So we heard here just low voltage. Then we heard also about fractionated or fragmented signals. And we heard about like mid diastolic signals, specifically. Then we also hear about late potentials, or late signals.

And a very valid and frequent question that comes up is, are any one of these fundamentally better to be sure that we are categorizing well and can target for ablation? So maybe if we just try to think about each of these as pros and cons. So maybe I'll ask Dr. Deshmukh. Between each of these, do you prioritize in your mind like if I see this, I need to specifically think about targeting it?

ABHISHEK J. DESHMUKH: So if you are mapping in sinus rhythm, then we are hoping that we might see the fragmented signal, maybe mid diastolic signal or late signal. But none of them are mutually going to be exclusive as sites for ablation. And lot of times, you have to mutually include a lot of them if you are in the right area where the abnormal scar may be.

And for example, it also depends, as you showed, based on electrogram and the electrode orientation. So you might be at the level of a scar boundary, and then you might be picking up a signal across it on the other side of it. And that may certainly look like a late signal. Whether that late signal is truly arrhythmogenic, that's unclear. Similarly--

SAMUEL J. ASIRVATHAM: So maybe, Abhishek, I'll just take your point there, and I'll just draw a figure to say. So let's say we have some scar tissue here. We have some scar tissue here, maybe something else around here. And then, we have relatively normal tissue, maybe around here and here.

One way to picture this, especially for someone who is new to EP, is if we have an area that's just struggling to come through, so taking a long time to conduct through a site, and then comes out to the rest of the myocardium, just in sinus rhythm, then this area, because it's slow conduction, and we have an electrode that's just interfacing with this area right here, we show up with a multi component signal, because just in that small timestamp in that small area, we have in that area signals that span a significant portion of the QRS or inter QRS or during tachycardia, the cycle length of the tachycardia.

So this fragmented signal, if we have good contact and we're able to see these signals all within one interspace of an electrode, it suggests there's really slow conduction there. But the consequence of that slow conduction is that once we're done getting through that conduction, and then we finally do exit, that's going to be an area that may be a small amount of myocardium but is likely to be late after the QRS. In a way, the late potentials are like the consequence of slow conduction. But the slow conduction itself is what's making late potentials and is really the substrate for initiating and maintaining VT.

So one of the specific questions that's asked is, if you ablate, do you change the substrate at sites different from where you're ablating? And I think if we keep this construct that the answer is yes, because if we do ablate the slow conducting areas, then we don't have that delayed conduction that gives the late potential. So it's possible you start ablating and remap the same place. So you tagged, for example, in a map fragmented site and a late potential site that you ablate the fragmented site, and then we no longer see the abnormal signal that's a late potential.

Now, another way of also asking that question is, is it better to ablate all the sites that can just leak out? So in other words, we can't really find where the slow conduction is. But if we actually plug or ablate all the late potentials, then aren't we getting rid of all the sites where a potential circuit could come out? And I think that's absolutely true.

But it would have to be perfect. It would have to be complete to be able to get all those locations like that. But that may be the only way that we can ablate if, say, it's a deep septal substrate or a epicardial substrate near a coronary artery that we can't get to. So I think part of our skill in substrate mapping is to tag each of these. And wherever we see them, we ablate, either because they're cause or enough of an effect that we can [INAUDIBLE].

Maybe we could get-- maybe, Siva, if you have a slide to share with us of a unique one, this one that we haven't discussed here, of what is this mid diastolic signal, and why does-- does it have significance? So one of the-- Can we make that a large screen, Siva?

SIVA K. MULPURU: Sure.

SAMUEL J. ASIRVATHAM: So this is a unique slide where you have VT on the right part of your screen. And we have some signals that are actually within the QRS and maybe some signals that go outside the QRS as well. So maybe I'll come back. I'll come back to here and just give us-- I'll just run this to see what we mean by this.

So this is like in tachycardia what we're trying to look through. And Siva, you know this figure well. But if we imagine that there is a circuit that has some boundaries of normal tissue, fiber orientation, scar, depending on the nature of the pathology, and then we have to have some place where you get to relatively normal myocardium, and there has to be enough of that myocardium for us to be able to generate a QRS.

So when we talk about mid diastolic versus within the QRS, is there a fundamental pathophysiological difference? So the QRS is being made once we get out from this exit site. There's always going to be electrical activity that's going on within the circuit. So in other words, there's going to be from beginning of one QRS to the beginning of the next QRS some electrical signals somewhere that are in that circuit during ventricular tachycardia.

So during VT, whether we see the signals within the QRS or not really has very little difference in pathology or pathophysiology of the circuit. However, any site away from the circuit, you will get signals close to the QRS, but you're unlikely to see anything in diastole. So in the circuit, it's true. We see signals that span the entire cycle length. But outside the circuit, it's going to be incredibly unusual to see things distant from the QRS. So when we do see mid diastolic signals, we get really suspicious that we probably are in an area that's either close to or housing part of the circuit.

Now, it's not impossible to get late signals or mid diastolic signals, even when we're not in the circuit. And I'll come back here. For example, if this is abnormal tissue, and all of this is normal, and we just have a bystander area of abnormal tissue, VT could be here, gets to the normal tissue, creates the QRS, and then starts meandering into this abnormal tissue.

So a signal that you're going to find here will also be well after the QRS, maybe even in the middle of diastole. And that's why you heard that point from Dr. Friedman that we'll have to correlate the signals that we see with some kind of dynamic maneuver. Sometimes, that's entrainment. Other times it would be like when we actually see the tissue.

Now, I'll just play back this screen here. I think maybe Siva's screen was on at that time. Everyone able to see this? So this is what I was referring to as trying in the actual circuit. We will have signals that span the entire cycle length. So there's nothing inherently better or worse with a mid diastolic, pre-systolic, late systolic, or within the QRS signal. But outside of this area in the normal zone, it's going to be exceedingly uncommon to see a signal that's in the middle of diastole, except in these rare situations that we went over here.

SIVA K. MULPURU: And one of the--

SAMUEL J. ASIRVATHAM: Yes, go ahead.

SIVA K. MULPURU: One of the questions there is, what type of filters, if we are trying to see more of these late potentials or fractionated signals? Do we manipulate any of the filters, or do we take the filters that are already there in the electroanatomic mapping systems?

SAMUEL J. ASIRVATHAM: Yes. So that's an excellent question about changing filter settings both in conventional mapping systems and some of the newer mapping systems. So I would say right now-- and please feel free to chime in, anyone here. But I would say right now there's very little value in changing filters, except you should be aware of what's the likely effect going to be.

So for example, if you have a filter for electrical noise as part of that system, you're going to do a very good job of taking away that 50, 60 hertz noise. But some signals, especially fragmented signals when the channel is very small, like highly arrhythmogenic signals, or conduction tissue can have a very similar frequency spectrum. So if we take that out, we may filter that away.

The other aspect of filters we should also keep in mind that when we over try to filter we can introduce signals like ringing or introduction of signals that don't even exist. So we may wind up targeting a signal that actually is just an artifact of our filtering. So a couple of ways around it when we have a doubt on, is this a signal, am I missing something, is it real? And we run into this a lot for electrical noise, artifacts as a result of filtering, or even mechanical artifacts like valve closure.

But we have some nice tools to try to work around this. One very important one is pacing. So if we have a doubt about a signal that doesn't look normal, but not sure is it noise, is it some other, its timing is way off from the QRS, you pace and see if you capture. If it's really contact and a signal there from the tissue you're making contact with, you'll capture. So first thing you'll realize is that signal disappears. It comes very close to the pacing artifact.

Second is that capture could give rise to a global QRS. And if that looks similar to the VT that-- or one of the VTs that you had clinical or induced in the lab, that's also really good proof that that is an abnormal signal, and you have it underneath your catheter.

Another thing that's useful when we have a doubt, is this real or not, you can just perturb by pacing not from your catheter but from somewhere else. So if you have something you worry, is it a valve artifact, the timing of that valve closure is going to change between sinus and just a PVC and a post PVC beat. So if that timing changes and in a way you predict that, well, movement is going to change, then that gives us some information.

The other way is you could filter not so much to see your signal clearer but by changing the filters to see if some signal completely disappears and yet it's not in the range of what you're filtering. That would suggest this was an artifact from the filtering itself that you used. There's actually a very nice review of how filtering can affect the signals and introduce signals. And maybe I'll ask Chris to find that paper. Dr. [INAUDIBLE], my colleague from Jacksonville first thought we could send it to anyone who had that type of question. Anything you want to add to that, Siva?

SIVA K. MULPURU: Yeah. Sticking to conventional filters for the unipolar, if you're relying on unipolar signals, in structural heart VT, sometimes narrowing the bandwidth of the filters may help you when you're relying on the unipolar signals. Whereas, bipolar, I usually keep the same. I do not really manipulate my bipolar bandwidth of the filters.

SAMUEL J. ASIRVATHAM: OK. Great. So then, the next question that I have here is a little bit more complicated one, but it's-- and we can certainly explore it with cases as our sessions go by. But substrate mapping versus entrainment versus activation mapping. What do we actually rely on? That's basically the question. And we can certainly explore each of these in more detail. But what do we rely on, and what are the pros and cons of each? Chris, do we-- do-- would you like to take a shot at that? And then, maybe I'll ask Dr. [INAUDIBLE] and Dr. Cannon to come up here and give us some thoughts as well.

CHRISTOPHER V. DESIMONE: Can you repeat that? I was just looking up your signal processing.

SAMUEL J. ASIRVATHAM: Yeah. Maybe I'll draw. I'll just draw it out here. The question is not so much about substrate mapping, what do we target? And I just see another question also related to substrate mapping, specifically about [INAUDIBLE] and what that is. Maybe Chris, you can prepare something on that. But this was more as do we rely on activation mapping, entrainment mapping, or substrate mapping? Not so much as theoretically, what's better. But what do we actually focus on when we are doing a structural heart VT case? So no right--

CHRISTOPHER V. DESIMONE: I'd like to-- Sorry.

SAMUEL J. ASIRVATHAM: Yeah. No right answer on that one. But maybe we'll share your thoughts.

CHRISTOPHER V. DESIMONE: Sure. Well, I like to use all three if I can. Usually, it depends on what the patient will allow me to do, hemodynamic stability or not. But I always usually start off at the beginning with the substrate map to get a sense of the voltage, get a sense of these abnormal signals.

And then once I've honed on to where the abnormal signals are, where there's some scar, where there's some area of these late or fractionated potentials, as we mentioned, then I'll use that time to, after I've gotten the patient hemodynamically stable or at least the blood pressure up, then try to induce VT. And that's when I try to do my activation mapping, my entrainment mapping. I don't think any of them is better than the other. I think they're all complementary.

SAMUEL J. ASIRVATHAM: OK. So a luxury would be a case where we really have all three. We have a full activation map. We are able to entrain, identify different parts of the circuit. We have substrate. But if we had to favor one or the other for the main negatives, the main problems, that we run into while doing this, I think each have their own unique problems. You mentioned briefly that entrainment, we have an issue that we could change the circuit. We could change to another tachycardia. It's hard to know which signal we're actually capturing to make the measurements that we look at.

But I think it's fair to say activation mapping also has its own pitfalls. And it's useful to think about this. And I think that's what the doctor who was asking this question was talking about. Maybe I'll ask Siva. Do you want to give us some thought on where you can think of a case where the activation map failed you? And we'll try to troubleshoot to help answer this question. And then, Jeff, if you could also get Dr. [INAUDIBLE] here, we'll also get some thoughts from her.

SIVA K. MULPURU: So one situation when activation mapping-- So for example, if you're mapping a chamber or a vicinity of a chamber that is relatively remote, like an epicardial substrate, and you're doing a good endocardial map, you get some clues. But the activation map-- the targets on the activation map may not be successful targets for ablation. But the clues there are if it is a very late, a broad area of early activation, it clues you into do more activation mapping before coming on ablation.

SAMUEL J. ASIRVATHAM: So maybe I'll just outline that a little bit more, Siva. I'll just go here. And let me know if you have any trouble seeing my slides. Able to see this OK?

SIVA K. MULPURU: Mm-hmm.

SAMUEL J. ASIRVATHAM: Maybe I'll use this atrial map as an example here. So here, I think this illustrates like the issues with activation mapping. So one of the things we stress to trainees is never construct an activation map after you've already decided what the circuit is-- so if you're doing an atrial arrhythmia, and you just say this is flutter and a typical flutter, and I do a map for VT that you know from where the exit is, that's where it is, and try to do a map.

You have to really let the map speak to you, and make your own determination from what that information is. That's probably one really important pitfall is really look at the map. Don't make the map what you feel it already-- you think it is where you need to ablate.

The second thing is this importance of circuit length. And I think this is where we get most questions related to this. Everyone has heard this concept that we have to map the whole circuit length to know that we've mapped the VT. And the construct is really simple. If you have two chambers, and if you have circuit here in one chamber, and you're mapping in this chamber, this is all bystander. So this is going to be a very small part of the cycle length of the tachycardia.

So tachycardia cycle length is 300 milliseconds. And you've only mapped 80 milliseconds. And you did pretty much a whole chamber. Then it's unlikely that that circuit is in that chamber. You still could argue you missed some spots. You misinterpreted some signals as noise. You have a primarily epicardial substrate that you just haven't mapped, but it's really suspicious when it's this much less that you just have to go somewhere else and map. So that's the construct that we do need to be able to get the whole cycle length. But there's some caveats there too that make it not easy to know what that means.

For example, we always have to have a reference, say the QRS we take or a particular signal, say a coronary sinus electrode. And if we have a cycle length in tachycardia of 300 milliseconds, and by chance, by chance, we have conduction in a remote area from the circuit, that based on what we took as a reference, that wavefront just got there, say 150 milliseconds after a reference.

And the very next signal may be, say, 148 milliseconds before the reference, and one here is, say, 152 milliseconds after the reference. Then, if we took the windows for our map as 150-- we just split that 300 milliseconds, 150 minus, 150 plus-- you would have had one signal that you've recorded at 149 and another that you've recorded at 151. So you add that up, and you get-- I've got 300 milliseconds. And it has really nothing to do with the tachycardia.

So a very important corollary to the general concept that you need to get the circuit length as yes, if you don't have the circuit length, you don't have the circuit. But even if you have the circuit length, you have to have all the components that span the circuit length and all contiguous with each other. So it's having circuit length is a necessary but not sufficient parameter for you to say I've got the circuit.

You also have to be able to find each of these steps in between. And that can be difficult, very difficult to do or visualize when we're actually looking into this map. That's one thing to keep in mind when we are looking at a tachycardia. So here's an example. In this one, this is a patient with multiple congenital heart surgeries, had a cycle length of, say, x milliseconds. We mapped, and we got-- we couldn't get the circuit length in that chamber.

So this particular patient had a type of Fontan where we had some of the atrial tissue now part of the left atrial circulation, a neo-left atrium. You need to map that as well. And then you put it together. You're not only able to span the circuit length, but you also see this contiguous pattern that you can legitimately visualize the circuit.

So activation mapping for focal source tachycardia is much, much easier than activation mapping when we're looking at the circuit. And it's good for anyone in EP, anyone tackling structural heart VT, to just be prepared for those. It's not going to be apparent. It's a hunt in the picture we create with whatever mapping system we use.

Whether it's something that's showing you a propagation or just showing you the timing of the signals at different sites, we'll have to see, where do you have the circuit? Where all do you have the circuit? And where all can you actually trace this? Then you compare that with the substrate to see where it makes sense. And then, the ECG for the exit to say, does that exit make sense from that site? And we're able to get that kind of information from the activation map. Anything you'd like to add to that, Abhishek?

ABHISHEK J. DESHMUKH: I think this is a very important point [INAUDIBLE] was fantastic. There was one question that how do you assess the end point of VT ablation, seeing the non-ischemic cardiomyopathy? Is it like taking care of the clinical VT or keep inducing other VTs, what we could induce, and then map and ablate?

SAMUEL J. ASIRVATHAM: [INAUDIBLE] Ammar, do you want to share your thoughts on that? And then, if we have Dr. Cannon still available, we can get his thoughts on that as well, particularly the unique end points for ablation in patients with VT in the context of congenital heart disease. So Ammar, end point for ablation. And then maybe I'll add something about the future of what those end points might look like as well.

AMMAR M. KILLU: Sure. Sounds good. Can you hear me OK?

SAMUEL J. ASIRVATHAM: Yes.

AMMAR M. KILLU: Perfect. So I think a lot of the time we go after what we call the clinical VT. And that is typically the one that we have seen on a 12 lead ECG, or holter, or something like that. Sometimes, we don't have that information, despite best efforts. And we just go based on the ICD electrograms and try to correlate what we have in the case. I think at the very least, we should try and eliminate the clinical VT, if we see it. But I favor trying to eliminate as much of the abnormal substrate and as many of the inducible VTs as possible.

I think if you don't, invariably, it will become a problem at some point. Now, you may be able to do drugs or something like that and keep it at bay. But if I can eliminate as many of the VTs as possible, I do try to do that. So I favor complete substrate elimination, personally, as an end point. But it's not always possible. Many different things can make that the case-- patient tolerance. Sometimes, you're just unable to. Sometimes it's too much, so on, and so forth.

SAMUEL J. ASIRVATHAM: Yeah. I think this is a very practical point, Ammar. There's no easy way to do structural heart VT or substrate ablation. So sometimes, when you hear someone did the structural heart VT in one and a half hours, and it's either someone's really lucky or it's just upfront saying, that's all I could do, and we need to do something more later.

It is very challenging and at the same time extremely rewarding. We've all seen patients who have frequent shocks, running out of options, not good candidates for transplant. And doing a really thorough good substrate ablation is life changing. So I think that those are things to keep in mind that don't set yourself up like I have an easy fix. But a few things maybe in the future that will be useful for us to understand end points.

Specifically, I believe the question was even not only clinical but just substrate. How do you know you're done ablating substrate? How do we know? It's like we saw. Am I finished? Is there more substrate that I need to do? Just a few things for the future that might be very useful in understanding this.

One is an approach that looks like real time display of late potentials, very similar to the signal average ECG. So just like we have the signal average ECG as one thing that we can get for patients before ablation and display the late potentials, some signal averaged, some mapping systems are trying to incorporate signal averaging in real time. So you'll have the signal average late components. And as you chip away, you can know percentage wise how close are you to having ablated these late sites. So that's one.

Another also unique approach, one that we are involved in, is even without signal averaging, taking just a few very small number of QRSs, VT, and then the resting ECG, but using machine learning from models that were already created, not dissimilar to models like for entrainment that we borrow on today for all the entrainment mapping we do, but just looking at that ECG to not only tell us the degree of late signals that we have but a vector for those late signals, so that we'll know where in the heart they're likely to be found and may serve as end points in the future to know, OK, here, we are actually done.

But practically, as Dr. Killu pointed out, just given the complexity of these cases, many times we're done just basically when we just feel either it's been a long case, we're not able to do this any further, or it's this is as much as the patient will tolerate. And we consult patients saying that we may need to come back to complete it and so on.

Do we have Dr. Cannon up here? And then, Siva, I believe someone sent you a case now that you'd like to-- we can use to discuss as well. Bryan, any thoughts to share about VT structural heart in the pediatric and congenital heart population? What do you use realistically as an end point to say, I know you have a lot more energy than we do, so you'll go a lot longer? What do you keep internally?

BRYAN C. CANNON: We just have smaller patients, so we can go longer than [INAUDIBLE]. But for the adult congenital population, I think one of the most important things is understanding the anatomy and understanding the potential circuits that are likely to be important. And as Dr. Killu said, these can sometimes be very challenging. You may have five or six circuits that you can induce. Ideally, you should try to get rid of all of them. But if that's not possible, try to figure out what your predominant circuit is, and then attempt to get rid of it.

In addition, you also have to think if by avoiding this, am I truly getting rid of the underlying risk for VT? Or do we need to consider an ICD placement in addition to this, because not all of these patients will have ICD? So it can be a very challenging thing. But I think if you can get rid of the clinical tachycardia, that's an end point. But I also agree that if there are other substrates that are potentially viable to ablate, that's acceptable. But then again, you don't want to be spending 12 hours putting 90 or 100 lesions in there that may potentially cause damage in the future or AV block. So I think it's a difficult decision and difficult question to answer.

SAMUEL J. ASIRVATHAM: Thanks a lot, Bryan. But one thing I would say to-- so maybe we can keep that case also for next time, Siva, so since we're getting close to time. But one thing I'll say to add, especially for trainees, folks who don't do a lot of VT or haven't done a lot of VT ablation in structural heart disease, if you do get a patient who's tolerating the procedure, and it clinically makes sense to do whatever you can to eliminate the VTs-- so they've had shocks. They've had multiple morphologies. And they're good on that day.

You're able to tolerate it. They're able to tolerate it. You're not putting them at undue risk. It's a tremendous experience as an electrophysiologist that you have 17 VTs. You have relatively limited substrate. And you do a thorough substrate ablation along with activation mapping when appropriate, entrainment when appropriate. And you don't have anything inducible at the end of it. It's not every case. I can say just in my own career it's a minority of cases, maybe 25, 30%.

But when that happens, and the patient allows you, the arrhythmia allows you to do it, there's no more powerful learning experience as an electrophysiologist that substrate is a real thing. So otherwise, it's always pictures. It's maneuvers you see. And it should never be just the lack of belief that this will help the patient that prevents you from going further. It should be something that's organic, like hemodynamic intolerance. You just need to do something more for the next day.

Now, I see a question here. And we'll take that question, and then maybe I'll just ask Siva to make a point about this. So we have one. Do we use anti-arrhythmics intraprocedurally? Do you want to take that one, Abhishek?

ABHISHEK J. DESHMUKH: Yes. So I think using anti-arrhythmics intraprocedurally becomes important, especially, at least I-- whenever I'm doing a PVC triggered VF case. And if we are inducing a lot of VT, then at least I have our anesthesia colleagues hold lidocaine in their hand ready to inject, if we are not able to defibrillate the patient from VF. That would be one. But generally, I personally haven't used a whole lot of IV amiodarone or procainamide during a case as we are mapping and ablating, but certainly lidocaine if we are having or inducing a lot of ventricular fibrillation.

SAMUEL J. ASIRVATHAM: Yeah. I think the time when it really becomes valuable is first thing is clinically when you're deciding who should you take off anti-arrhythmics. So if you have a patient who's having the VT that you want to get rid of, and they're having it on the anti-arrhythmic, there's very little value to stop the anti-arrhythmic, because the thing that could happen is you have multiple other VTs, induction of [INAUDIBLE], shocks. So if they're getting it, maybe it's slowing down the VT. They're able to tolerate it. You'll be able to do the ablation.

Intraprocedural procainamide is tough to use once you have hemodynamic consequences. So if you're already in a shocky state, it's a tough one to use. So you have to make this decision early to say, OK, I know this is a rapid VT. I'm going to give some procainamide, wait till the pressures stabilize well beyond the procainamide vasodilation, the procainamide gangliocytal blockade. And then you induce the VT and see, are you getting the best of both worlds, a slower VT that you can map or ablate?

Now, I have maybe one question we'll try to get to. And we will get to the other questions that you had, both email and cases, in a section that we'll record and post on YouTube as well. But maybe, Siva, a quick one here. This is a common question. How do you use the device electrograms in your VT ablation procedures?

SIVA K. MULPURU: OK. So device electrograms will give us the cycle length of the tachycardia. So one way, if it is really fast, and the patient was hemodynamically unstable even before coming to the lab, that patient, I may pursue more of a substrate modification approach and recognizing that he may be unstable with that fast VT. The second thing is looking at the morphology of the electrograms we can use a far field channel to see that morphology and compare to the VT induced in the lab. So if we have a reasonable match, we can assume it to be like a clinical VT and at least eliminate that particular VT.

SAMUEL J. ASIRVATHAM: Great. And I think also maybe I'll just have an extension to that is if you have one VT that a patient is getting shocked for, or one PVC that's triggering VF, but in the EP lab, we are just seeing a whole bunch of stuff, we can, program sensing with the device, run the strips and see which PVC or which VT mapped the intracardiac sequence, so far field versus near field on the device electrograms, and also the morphology. So if those look very similar, that might also be a clue, like here. This is probably the main substrate that-- or the main target that I want to go to.

ABHISHEK J. DESHMUKH: Sometimes, [INAUDIBLE] tachycardia pacing is also important, because you can look at from where the ATP is being done, say RV lead, and then looking at the post pacing interval, how close or far you could be from the [INAUDIBLE] exit site of the VT.

SAMUEL J. ASIRVATHAM: That's a fantastic point. And maybe next time, we can start with an example of that, Abhishek, if you want to put that together. So we're at 8 o'clock. And just want to thank everyone for being here for part of the evening here.

And I can see we didn't get to about five questions. Between today and the next session, we'll make sure that we get that covered. And then I know, Siva, you had a case that someone had brought up. We will respond if it's a clinical question back to you. And then we'll try to bring up that case and perhaps have the person who sent it to you, Siva, to present the case to us as well. Thanks, everyone, for this. Thank you, everybody on the panel as well.