false
Catalog
EP 101 2020: A Virtual Program for Incoming EP Fel ...
Ventricular Tachycardia in Ischemic and Non-Ischem ...
Ventricular Tachycardia in Ischemic and Non-Ischemic Cardiomyopathy
Back to course
[Please upgrade your browser to play this video content]
Video Transcription
Great. So we're going to go into a video for Wendy Zou talking about ventricular tachycardia, and she is a fixture every year and just gives a wonderful lecture. One of my trainees, one of my fellows when I was at Penn, now out heading up the program in Denver, University of Colorado. And without further ado, let me introduce her talk, and then Dr. Zou will join us for Q&A right afterward. It's my pleasure to talk about this topic, VT and ischemic and non-ischemic cardiomyopathy, and Dr. Kim's talk led into details that I'll discuss here very nicely. So you've heard over the course of the last couple of days many great lectures about mechanisms of arrhythmias, and each of these mechanisms may be involved in VTs. When we're talking about VT in the setting of structural heart disease, because that's what we mean by ischemic or non-ischemic cardiomyopathy, we're typically talking about re-entry as the predominant mechanism. And this is mostly due to the presence of scar. This very well-circulated photograph, for which Bill Stevenson should get a lot of royalties, demonstrates the concept nicely. See, remember that the scar is not itself arrhythmogenic. It's the mixing of scar tissue and normal tissue that alters the conduction properties of that tissue, to which it is adjacently attached in series to like normally conducting tissue. So that's the perfect milieu for re-entry to occur. It allows for unidirectional block to occur in induced fashion if it's not present at baseline, and it also offers a region of slow conduction through which circus movement activity can propagate once it initiates. The importance of scar historically has been known for some time. The early experience was in the OR among patients with completed myocardial infarctions in an era that predated early revascularization. Aneurysms were very common in these patients. It was common thinking because a lot of these patients had ventricular arrhythmias too. Well, if we get rid of the aneurysm, then maybe that'll take care of things. So that was the initial approach, was to do that and then to revascularize these border zone tissues that still had scar tissue, but some living tissue that they thought could eventually wake up and contribute to contractility. Unfortunately, that wasn't the most effective way to control VT, and it was a pretty high mortality surgery as well. It was with the work of Dr. Mark Josephson, truly one of the luminaries of EP, and his colleagues who were at the time embarking on this process of doing invasive electrophysiologic studies, a really revolutionary concept at the time, that they identified a couple of things, several things. One, that truly the mechanism accounting for VT in most of these patients was re-entry. Two, that they could identify critical circuitry elements, and most of those elements lay closer to the endocardial surface than beyond it. In fact, they could do a study, identify a really arrhythmogenic region on the EP study, and then direct the surgeons to resect just a portion, just a little peel. It became known as the Pennsylvania peel. Two to three millimeters of endocardium to really get at the most critical elements, and those patients in general did very well if they survived the surgery. But this is an example of the data that sort of was understood, collected, and really provides the foundation for how we treat people invasively for VT ablation. So a cartoon of the left ventricle with the apex incised, and you see this electrode covered plaque placed against the abnormal septum both before and after subendocardial resection. EP studies done before and after show very nice concepts. So on the left panel here, you see VT is easily induced before resection, and you see here all sorts of signals recorded from this active, that from this site here where the plaque is located. And what you can make out is there are these discrete signals that line up with the QRS complex. So that represents probably far-field activity that lines up with the global activation of the ventricular myocardium. But what is all this other stuff in between? I mean, it's ventricular activity, but it's clearly abnormal. It's not lining up with normal activity at all, which is what you see with these other electrograms that line up with the QRS. You also see these, so these in VT are known as mid-diastolic potentials because it's activation of the ventricle, but not at any other time that the rest of the major brunt of the ventricle is being activated. You can see signs of it, signs of disease and abnormal conduction in just sinus rhythm, and you can see them here, evident as late potentials. So late potentials in sinus rhythm, mid-diastolic potentials in VT. It's terminology difference, but it's kind of an important one to remember. But you see here the activation of the rest of the myocardium lining up with the QRS activation, and then discrete activity occurring late after that. After resection, you see disappearance of the late potentials, a greater, more near-field appearance of the more normal tissue beneath the scar, and importantly, lack of VT-inducibility and very good VT-free survival outcomes, assuming, of course, that you survived the surgery. Now, advance a couple decades later, and the whole concept and technique and technology of radiofrequency ablation was introduced. So the idea that you could deliver therapy through catheter-based intervention that was invasive, but not surgical, opened up the door and the opportunity to offer this kind of treatment for patients, a wider population of patients. And again, to utilize this drawing, which is just so nicely illustrative, you can imagine lots of potential different circuits that could form here, because there's lots of viable tissue that is interwoven within scar tissue. In fact, this patient with a history of transmural infarction really only has one segment here where there's truly transmural scar. Everywhere else is free game as far as an arrhythmia is concerned. And in fact, so let's just draw a little path through here because it's the easiest one to see. As the wavefront of propagation goes through this critical isthmus, you don't see that part on the EKG. That's what's showing up in mid-diastole, when you have a catheter or a plaque with electrodes at a given location of interest. What you see on the EKG is the exit, the exit that then activates the rest of the ventricle. So that's what produces the QRS complex. So when we talk about where is this VT coming from, it's not so much the true origin of the VT, but where it's exiting. We don't care that much about the exit, but we do care about the critical components to which it's attached and the components that are most easily addressed with ablation. So it kind of comes around and moves around. And then if you happen to have a catheter surreptitiously placed in the right location, not so surreptitiously, actually, if you've used the logic that Dr. Kim has already described in getting a 12-liter of the VT, figuring out where you had to go, you know right away where you should at least aim to get the catheter in the beginning. And if you're lucky and very excited, you'll see a potential like this at an isthmus site, a potential more like this at an entrance. Because remember, it takes time. If the activity recorded from here, it takes time for it to get from here out to here where the surface QRS is activated. So again, this is the signal, the local EGM that is attached to then a subsequent QRS complex. And then a different timing, if you're talking about something that's recorded closer to the exit. Now, of course, you don't know for sure that that's what those signals are. They're just suggestive based on timing. You still have to do the appropriate maneuvers that Dr. Winterfield very nicely described in order to confirm whether or not these are truly important sites or if they're just passively activated. But there are limits of entrainment as well as activation mapping, as satisfying as it may be electrophysiologically. One assumption, especially with entrainment mapping, is that the conduction velocity through the reentrance circuit does not change with pacing, which is not always true. And that's why it's very important to pace just slightly faster than the tachycardia cycle length, no more than 30 milliseconds. You want to capture, but you don't want to disrupt the circuit. Multifractionated signals can often confound interpretation both for activation as well as entrainment mapping. And most importantly, less than a third of patients who are referred for VT ablation have human dynamically tolerated VTs that can be mapped like in this fashion extensively. So knowing these limitations, knowing that VT is a big problem in people who have structural heart disease, a concept arose. Okay, so we know that SCAR is involved in post-infarction VT. That's pretty clear. It also became known, not immediately known, but later has become known, that SCAR is also responsible for non-ischemic etiologies of VT. It's in different locations and there are other different characteristics, but they have that commonality. And so with that in mind, a group from Penn and probably some others, but Penn gets the credit, thought that if you could see it in the OR, there must be some surrogate that you could use electrically to kind of see it in the EP lab. And that's exactly the approach that was kind of undertaken in this really landmark study, which if you have not read it yet, you need to, because this provides the foundation for what we do in terms of substrate-based mapping, at least for the better part of the last two decades. So the concept here was that we could potentially identify SCAR just based on voltage, not just voltage, but voltage was the biggest part. So low amplitude signals could direct us to areas of SCAR. And if we could couple that information with limited activation, limited entrainment, and pace mapping, then we could identify sites where ablation might feasibly cut across some circuits. And we do more extensive ablation than just focused ablation targeting termination of VT, or just based on simple activation. Analogous to the surgical approach of submittal cardioresection, but just with ablation lesions. Now how was SCAR defined? I bet none of you knows that answer, maybe. So it was based on statistics. So they mapped several patients' normal RVs, normal ventricles, and basically what they found was a threshold above which 95% of the recorded signal amplitudes from the endocardium of both of these chambers were included. And from the RV, it was 1.44 millivolts, and from the LV, it was 1.55 millivolts. To kind of take an average and make it easy to remember, they call it 1.5. And remember it, we absolutely have. I mean, this has, you know, kind of been the definition of abnormal when we talk about abnormal voltage and defining SCAR on electroanatomic mapping, again, for more than the last 15 years. And this is kind of the electroanatomic map that has been created when you couple it with an electroanatomic mapping system. So red correlates generally with dense SCAR. And again, you can alter these color bars to whatever you want. But these are the traditional settings. Less than 0.5 was determined to be dense SCAR based on the OR studies. And again, this greater than 1.5, the lower threshold of normal of 1.5 millivolts determined from the mapping studies. But more and more, especially as these mapping systems become more sophisticated and you acquire more information, more and more and more, you have to be really cautious that what is being collected is true to what is the true state of things. You can get red by just not being in contact. Low amplitude signals will all produce red, but that doesn't mean that they're all important. Signals that you want to see look not only low amplitude, but they're abnormal in character like this. So they're highly fractionated, long in duration. There are late potentials incorporated within the signals. Those are the ones that are most commonly found in central isthmus sites, which include the central isthmus itself, but also the entrance and the exit. The next thing you want to do is you want to have some idea, because this is a fairly large area of SCAR, where to start. Where do you want to concentrate your efforts with respect to ablation? And that's where having the clinical VT is very helpful. And Dr. Makati said this in his talk. Dr. Kim emphasized it in hers. And I want to emphasize it again, that by the time you are done with your fellowship, hopefully actually within the first six months of the end of your first year, you'll be able to look at a 12 lead like this, a VT, and be able to tell immediately where you're going to need to start your work once you get into the lab. Because talking about it is very leisurely and academic. Being there in the lab when a live VT case is going on can be rather stressful. And so the quicker you are with your brain and your hands, the smoother that things will typically go. So the algorithm that EJ went through was a very good one, and I have a very similar one. So you imagine the heart with respect to the limb leads is kind of viewed in cross-section. And then the pericordial leads are as they are, kind of from base to apex. And so I usually start with looking at lead V1. And if it's predominantly positive, we call it a right bundle morphology, indicating that it is likely exiting from the LV. If it is predominantly negative, it's a left bundle morphology exiting from either the septal aspect of the LV or the RV. I next look at the limb lead axis starting with lead 1. If it is predominantly positive, it's moving with the vector of lead 1. Therefore, it's more likely to be septal or RV exit. Opposite, if it is negative in lead 1. So we call that, say that in this particular patient's case, it was a right bundle, right axis. And then the inferior leads are telling. So if they are superiorly directed, that means that vector of activation is north, and that means the QRS complex in 2, 3, and AVF is negative, then that means it's most likely coming from the inferior aspect. Opposite, if it's coming from the anterior aspect, in which case we say it's inferiorly directed. So several just short phrases coupled together, and I know exactly what you're talking about when you talk to me about a VT. Right bundle, right inferior. OK, got it. And then the pericordial transition also gives us information about whether or not the activation is more likely to be coming from the base, the middle of the ventricle, or the apex. So going back to this case, this right bundle, right inferior axis VT with a relatively early transition can really only come from the anterolateral aspect of the LV. Now, why is the transition so early when the apex is out here, and this is kind of where the VT might be exiting from? Well, you have to also think that in this patient who's had a fairly extensive and dense apical myocardial infarction, that this territory for the most part doesn't activate. So it won't contribute to the surface QRS. So it's one of the caveats that you have to think about when you think about ECG interpretation in the setting of prior infarction. But you have that information, so you go to that area that has the very interesting baseline sinus rhythm abnormalities. And if you're very, very lucky, you'll get a perfect pace map. I'll tell you that doesn't happen most of the time, but it's still good to see it. And so the general approach for traditional substrate-based abrasion, we start either with a voltage map or we start with identifying the clinical VTs. If we don't have a 12 lead of a spontaneously occurring VT, then we try to induce in the lab. And that is important for a couple reasons. One, you know, we're getting into this era where we're talking about blading, ablating, and ablating a lot more substrate. We want to make sure that that's the relevant thing to ablate. It is feasible, for instance, to have bundle branch re-entry outside of a large scar that is actually the patient's problem. It would be silly to miss that by just targeting substrate that has nothing to do with the problem that the patient presented with. And then based on information from activation, entrainment, pace mapping, we have also a color coding system, although it's not nearly as consistent as Dr. Michaud's, but kind of identifying all of these interesting signals or sites based on all of those criteria along the septum and the apex. And that's kind of how the lesion set was designed. This was kind of the initial approach. Again, when you look at this, it's not that much ablation, but it was certainly more than what was being done up until that point in time. With that, it was effective. So this is a mixture of both ischemic and non-ischemic cardiomyopathy patients. And you can see that acutely, most of them did very well from an overall VT burden reduction standpoint in intermediate term follow-up. When we look at the clinical trials, there's an improvement with how patients do. And granted, these are all post-infarction VT patients. You can see that compared to standard medical therapy or escalated medical therapy, in most of the trials except for this most recent one, which included patients with only unstable VT from Germany, you can see that there is an improvement in the ablation arm compared to the other arm. It's disappointing though, right? I mean, you look at these curves out through time and you think, gosh, why are so many people recurring with VT? Because, you know, ostensibly, a lot of work was done in each of these procedures. Important caveat, and this is kind of a tangent, but one of my big beefs with looking just at the Kaplan-Meier curves is, you know, it's very nice they kind of relay a simple message. But if you don't understand what went into the picture, then you really might lose the information that is trying to be translated. It's important to note that although there wasn't a difference in VT-free survival, recall that any VT that recurred, whether treated with ATP asymptomatically or treated with multiple ICD shocks, were counted the same, even though clinically we know that that's not the same. And as well, oh sorry, the patient populations were probably rather different in each of these circumstances. That being said, so one of the potential limitations that might have occurred in these trials, as well as in the way that we've ablated VT to date, is this reliance on, you know, very strict characteristics for pace mapping. This is relatively old work from Bill Stevenson, but you can see here the concept is demonstrated nicely. So you have scar tissue shown in gray here, a VT circuit propagating through all of these different channels, and this is an isthmus site. And so pacing during VT, because of the fact that a wavefront of propagation is coming in this direction in general, any wavefront going in this direction induced by pacing will be extinguished by the wavefront coming in the opposite direction. Thus the only effective wavefront will exit out where the exit of the VT circuit is during VT, and you'll get concealed fusion. You won't see such a good pace map necessarily at that same site in sinus rhythm because you don't have the VT activation wavefront interfering with conduction in these other channels as you would normally. In fact, this study found that less than 30 percent of perfect pace maps, and we, you know, talk about 12 of 12 matches, etc. Less than 30 percent of the perfect pace maps were found at lightly circuit sites, and only 9 percent of them were perfect at RF termination sites. This is more recent data from Christian Deshalieux, who has done some work with pace map mapping. So what you can see here is an isochronal activation map during VT showing this nice figure of eight circuit in a patient with ischemic VT, and the correlating in sinus rhythm pace map map. So what he did was he paced all along the corridor here in regions that he knew based on entrainment were found to be important and found drastically different pace maps all throughout here. In fact, a criteria that is probably a little bit more specific is observing a long stimulus to QRS time. So that means the time that it takes if you're pacing in a diseased spot or a spot that's relatively insulated, you pace here, it will take time for it to get out and then activate the QRS. So that's a long stem to QRS. Greater than 40 milliseconds is typically what we call that, and it was observed in this work in the majority of re-entry sites. Another important clue that can be gleaned from pace mapping is this concept of alternating exits, just pacing at a rate of usually 600 milliseconds. So if you have such diseased tissue and you're pacing at a site that you're variably inducing unidirectional block in various directions, thus producing different wave forms, that's also an important clue, especially if there's a long stem to QRS. An example of one of our patients, you can see here, pacing in the middle of this dense scar produced two totally different morphologies that happened to kind of match. And this is usually what we get, by the way. It's a kind of match of the clinical VT morphologies. And in an area that is this big, especially if you're talking about repeated ablations, sometimes more ablation is necessary. The other big, big problem with the way that we approach this is what we determine our endpoint to be. Usually, it's inducible VT. EJ, as well as I, have already alluded to clinical VT, as though we know what that is. Really, what we mean, I think, is what VT did the patient have spontaneously. So either captured on a 12-lead or captured by ICD electrograms. That's kind of what we term the clinical VT. However, so take this patient, this very frequently occurring spontaneous VT. These other morphologies were very easily induced, singles or doubles, in the same ablation setting. And this is kind of the substrate that we identified in the left ventricle towards the apex. Pretty large portion. Now, if you look just at this one, you could imagine, based on the morphology, this right bundle right kind of indeterminate axis, it's kind of coming from this mid-lateral portion of that apical scar. But if you look at all of these others closely, you can see that they probably exit from other portions of the same scar. So I'm not sure that I'd be high-fiving myself if I get rid of this one, but I can still induce these ones very easily. We've technically eliminated the clinical VT, but we haven't necessarily, in my opinion, sorry, eliminated a clinically relevant VT. And that patient, if that is all that is done, typically would be back out for a repeated effort. So some contemporary substrate-based approaches have evolved to try to kind of overcome some of these endpoints or endpoint problems. This one is probably the most straightforward that most of you are familiar with. So this is the substrate homogenization effort, looking at the entire area with abnormal electrograms and basically ablating through the whole area. It's a very pragmatic approach. It's a lot of ablation, but it does probably get to all of the potential circuits a little more efficiently. Other approaches that have evolved have really focused on kind of activation timing. And I'll get into that in a little bit, but if you look at this, so this is an approach whereby all of the late potentials are eliminated. Now to orient you as to what's shown here. So shown on the far left here is a standard voltage map from the endocardium and the epicardium. So red here being scar, purple being normal, same on the epicardial surface of the LV. Now this is what they call a late potential map. So you've heard about activation maps during tachycardia. You can do activation maps during sinus rhythm as well. And it was revealed to me during one of our morning conferences this week that my fellows didn't quite understand what I was talking about. So basically, it's the same concept. You pick a fiduciary point, a reference point in the QRS, you can anywhere, that's reproducible. And then what you're looking at is the timing after that. So you time to the late potential. So the earlier the late potential is, it'll be red. The later it is, the more purple it will be and everywhere in between. So then the purple areas here is highlighting as late potential areas, which you'll note are relatively small in area compared to the very large area of scar that's seen on the endocardium. So the approach using this method is to completely ablate and kind of abolish at least entrance conduction, eliminate the late potentials in those regions, which is demonstrated on remapping. Similar concept, but using this technique to provoke late potentials if they're not immediately obvious, sometimes with pacing maneuvers, with so-called local abnormal ventricular activities. But similar kind of color coding scheme, similar sort of endpoint. And then this one is kind of targeting the timing of entrance signals into the scar and peripherally trying to shut off all the entrances into the middle part of the region. You'll notice that although all of these things, all of these techniques utilize slightly different terms and methods for identifying regions that they're gonna target with ablation, they all mostly target the same areas of ablation, right? We came up with approach as well that is not that different than that, but adds a slightly extra component. So called core isolation. So this is an electroanatomic map where we've altered the color scale. So the upper limit of abnormal, lower limit of normal is one instead of 1.5. That kind of shows this area here, which was red when we used the standard definition or the standard criteria that was rich with late potentials. So we paced at all these areas to see what captures. This area was truly inexcitable, didn't capture at high output pacing. This area all captured, all had interesting looking signals and reasonably good looking pace maps to the clinical VT. So the strategy was to incorporate this whole region and use this kind of as a buffer for our ablation lesions. This is a nice example of the electrophysiology that can be observed if you are looking for it. So in this case, a multipolar catheter placed in the center as the ablation lesion is being completed, completed right here. And you can see here, this late potential that follows each of these first two beats is no longer there. So entrance block. And then not too long after, we actually see this phenomenon of dissociated firing. It's almost like we're doing a PVI here, but it's way more fun. And then you pace and you capture the signal and lo and behold, you have exit block. So that's what we call core isolation. Nothing really rocket science about it, that's just another endpoint to assess. And we found that there was actually incremental benefit in doing so. In addition to the gold standard of VT non-inducibility, which is still the gold standard. But clearly those patients in whom VT was still inducible, we couldn't achieve core isolation, VT free survival was very poor. Interestingly, in those in whom any VT was inducible, but we achieved core isolation, the VT free survival over a 16 month median period was rather good. And of course the best outcomes were those in which we could achieve both. And that was sort of demonstrated here in the Katherine Meyer analysis. And then a recent meta-analysis done by a fellow who was a frequent attendant of this conference actually, so you should aspire to this level. But demonstrating that the more comprehensive approaches that were described in each of these individual studies when combined together, kind of favors more comprehensive ablation, more comprehensive targeting of substrate at the time of a single ablation in order to achieve more durable arrhythmia control. Now this included non-ischemics and ischemics, but when you look at just non-ischemics, unfortunately, consistently the ablation outcomes are worse in those patients. Repeated ablations are often necessary. So this is data from the International VT Center Collaborative Group, which was a conglomeration of 12 international VT centers from which, at which more than 2000 patients with structural heart disease were ablated. So we looked at data from patients who had information about first ablations at the time of their entry into the registry, or a repeated ablation attempt at the time of entry. Non-ischemic cardiomyopathy, not that surprisingly, was more common in the repeat ablation group. And the procedures were more complex, more epicardial access, only unmappable VT, and VTs still frustratingly inducible after a lot of ablation, and more often than the repeat ablation group. Now why is that? I mentioned that scar is the common endpoint for both of these substrates with respect, or maybe the starting point, the common starting point for the arrhythmias in these patients. But that's where the similarity ends. In the post-infarction substrate, it's much more two-dimensional, localized towards the endocardial surface. Most of the time, we can just get away with ablating from the endocardium. Non-ischemic's much more heterogeneous, far more three-dimensional, doesn't always involve the endocardium, or at least you can't see that it involves the endocardium. Often involves the mid-myocardium and the epicardium. This is a common example, so normal bipolar endocardial voltage map. You only see an abnormality if you happen to have an ice catheter introduced, which we always do for our VT cases. But here you see this stripe of hyperechogenicity, which is confirmed to be abnormal when we go to the epicardium and map there. Not just low amplitude because of fat surrounding vessels, but look at the very abnormal characteristics of these potentials that were uncovered there. An emerging kind of substrate is this mixed cardiomyopathy group of patients who have had a history of coronary disease, even have had an infarct, but they don't have any sign of post-infarction substrate on their mapping. And in fact, this was a patient who had non-ischemic substrate, had to go to the epicardium, absolutely no scar on the inside of the LV despite having had early revascularization. So clues to this are VT morphology that's inconsistent with the coronary disease distribution, including an epicardial exit, or an imaging abnormality, pre-procedurally, peri-procedurally, that is inconsistent with or out of proportion to the distribution of coronary disease. This is one example that occurred a couple years ago in our labs. You can see this patient came with a history of RCA infarct, multiple stents, had had multiple ablations. That's always a clue. If they've had lots of ablation of the same area and they still continue to have VT, that's also suspicious. But you see here this distribution, completely inconsistent with post-infarction substrate. The way that coronary vasculature supplies blood to the heart is from the epicardium to the endocardium. So when there was an infarct, the tissue that sacrificed the most is the endocardial surface. You might have a little bit of that towards the base, but certainly not the majority of the substrate that you see there is due to post-infarction substrate. So when we have to go to the pericardium, we use this technique that was described some time ago by Dr. Sosa, Skanovaca, and their colleagues in Brazil. It's this percutaneous access with a tui needle, typically, and you frequently inject dye along the way in order to sort of just see where you are. Usually there's a perceptible, tangible pop as you go through like the parietal pericardium. Once you're there, you inject dye or insert a wire and you have to make sure that the wire crosses both planes and hugs the cardiac silhouette and LAO view in order to confirm that you're in the right spot. The other thing you have to do is be in a center that is used to doing these and that importantly has CT surgical backup. Because this is a very difficult thing to recover from if you need CT surgery and you don't have it. Okay, so other clues that can be derived from mapping on the endocardial surface is using unipolar mapping data instead of bipolar. So with the bipolar electrode or the bipolar pair, it provides a very limited field of view compared to what you might see if you took the anode out to a farther location, usually the Wilson central terminus. In doing so, you lose detail in the electrograms, but you see a lot more about voltage abnormalities that are beyond where the catheter is directly contacting. And so this is an example, again, work from Penn. So normal LV on the top, non ischemic and VT on the bottom, normal endocardial bipolar voltage, unipolar voltage abnormality is brought out when you use this threshold of 8.3 millivolts as the lower limit of normal. And that is corroborated by mapping on the epicardium. Similar concept demonstrated on the RV but with a slightly lower threshold, 5.5 millivolts, given the thinner RV wall. And you can see here that this ARVC patient who at first seems like doesn't have much actually has a lot going on, which you might expect anyway because of the ARVC substrate. The most vexing kind of substrate, of course, is mid myocardial. And if you happen to have the luxury of having a cardiac MRI of this quality before you go into a case, you at least are prepared. But many times you don't. And all you have is this. You map the inside of the LV, you see nothing. You might get some superiorly directed VTs that make you think, okay, well, we should map the epicardium. But guess what? The epicardium won't show much either. All of these areas that look kind of red, those are low amplitude areas because of fat, not truly abnormal. And the impact of this can't be understated. I mean, even if you know it's there, like if you have that imaging, you have abnormal unipolar voltage criteria, you have ice to guide you, to show you that there's a mid myocardial substrate. Then all you're left with is the knowledge, but maybe not the tools to be able to penetrate to the area that you really need to. In this large series of data, nearly half of the acute ablation failures were accounted for by presence of mid myocardial substrate. So in that kind of patient, and non ischemics comprise a big portion of that, you have to rely more on mapping during VT, a little bit more uncomfortably so than you would otherwise. Because this is particularly true when you can't really see the substrate, or the substrate is underneath a layer of healthy tissue that isn't completely dysfunctional. And you wanna minimize the amount of collateral damage that's done. So this is an example here of some endocardial maps from three different patients here. You see this far field activity that's highlighted here. You can't capture it, you're not sure if it's just this in particular, not sure if that's artifact or something significant. And lo and behold, go to the opposite side and you see something sharp, which is what was being picked up here on the endocardial map. In some cases, although mechanical hemodynamic support has never been shown to improve outcomes with respect to VT ablation, it can be of utility in this population of patients where you have to rely more heavily on mapping, and you are suspicious that the patient won't be able to tolerate it hemodynamically. So various products are on the market. TandemHeart is this left atrial to right iliac artery bypass centrifugal pump here mounted on the leg. It's a relatively large bore access, both at the arterial side as well as the transeptal. And most people don't really use that technique. Impella is used a lot more because of the ease and familiarity of insertion. So it's basically this miniaturized axial flow pump that's mounted on a pigtail catheter, and it pulls blood out of the left ventricle and ejects it right out into the aortic root, so direct LV offloading. And then ECMO is always an option, assuming that you have the surgical support as well as perfusionist and perfusion machine. More and more, I think, as we're getting into defining substrate, we're finding that this fixed definition that we've been using is probably too simple. This is work from Rod Tung's lab, but the concept has actually existed for quite some time. I remember reading a paper from Andy Witt from a couple of decades ago that demonstrated this same concept of identifying late activation in sinus rhythm and identifying and categorizing these into isochrones of activation. And so this is, again, that concept of activation mapping during sinus rhythm. You're mapping the latest activity. So in this case, he has mapped to the latest offset just to standardize it. And then he created partitions. ROYGBIV is what he says. The purple is the latest, red is the, well, white is the earliest. Red is the early in the scheme. And basically, what he has found, actually, so you would think that if you just targeted this region here, that maybe you could limit the amount of ablation in this tremendously large portion of scar. In fact, you wonder how the heart is even beating with all the scar that they found there. But in this case, putting a catheter there during VT induction, that's the electrogram that you see there. It's nothing. It's not even passively activated. There's functional block there. It's actually in this region here that doesn't look that interesting in terms of lateness of potentials, but where you see the mid-diastolic potential that's of interest and that actually leads to VT termination. And so it's not so much in this algorithm or in this schema, the latest of late potentials that are important, but the crowding of isochrones that might occur that indicate deceleration of conduction. And high-density mapping has certainly changed the landscape in terms of how we can effectively map VT in patients like this. Not only increasing the speed, but the density with which we can acquire high-fidelity signals. This is a recent publication. If you haven't seen it yet, you should. It's in Jackie P. But looking at the Orion system and their newest algorithm called LumaPoint, you can see here, so it's a couple things. One, you can see very nicely this kind of figure of eight classic VT circuit. But if it weren't that obvious, what you could do here is use this kind of skyline feature that they have built into the program, whereby it focuses, it's like a histogram of ventricular activity. So you wanna focus on the areas where not that much of the ventricle is being activated. And if that's during the diastolic period, then that just kind of confirms that that's the area of interest. So that is an example of how a map could be acquired relatively quickly with efficient information gathered. This is another example. So a voltage map was created, activation mapping on the endocardial side, the epicardial side. VT is tolerated, but only briefly. But because we had looked at the surface 12 lead, we knew exactly where to place that electrode or that recording catheter on the epicardial surface as VT was induced. And look at all of the signals cumulatively and how much of the mid-diastole it kind of encompasses. Very interesting, much harder to do this with single catheters. If you can imagine just moving around in this space with just a single catheter, how much more time and stress that might involve. And in fact, just getting a couple beats of VT on the epicardium, you can demonstrate in fact, if you get it on the epicardium as well as the endocardium, this transmurality of circuit. You see ROYGBIV here, we're missing purple here. So purple's somewhere in here. So it like activates here, breaks through, gets the right, the endocardium then comes back out again on the epicardium. Pretty cool. Hard to do with like just single catheter mapping. Okay, so spend a lot of time about mapping because I think that that's really, really important. We do have some adjunctive ablation techniques to kind of augment ablation efficacy. I won't talk about all of them, but the ones to highlight include those that can be used with tools that we kind of already have that augment radiofrequency energy. And that includes bipolar ablation, use of half normal saline irrigant, just altering the ionic concentration. The needle catheter, although it's also experimental, probably at least for the foreseeable future, not gonna be usable. The hybrid surgical technique using cryoablation and the OR and then use of selectively myotoxic substances that are not radiofrequency energy or cryo. So bipolar ablation, unlike with mapping, the usual ablation that we do is unipolar ablation. So the circuit is completed from this active electrode to a large surface area impedance patch that's placed near the heart and that forms the circuit. And so you get resistive heating and then conductive heating subsequent to that. If you actually change out the patch for a small surface area other electrode and place it in close enough proximity on the opposite side of an active electrode, it will function like an antenna. You'll actually get heating from both that synergistic and create synergistic conductive heating, which can produce in the right context, transmural lesions. The lab at Mount Sinai as well as ours has done a fair amount of research looking at this and found that consistently we would get larger transmural lesions with bipolar ablation usually using open irrigated catheters compared to sequential unipolar and even compared to simultaneous unipolar. Another technique that's actually easier than bipolar, because you don't have to construct any weird cables, is that you can change the ionic concentration of the irrigant. So you had a very excellent biophysics talk by Ed Gerstenfeld, so I won't belabor too many of the details. But we know that limitations in RF lesion are several fold. So at first, you have energy that goes into the tissue, it's heating, and it creates this resistive heating core, which then produces thermal conduction, and you get the lesion. In the circulating blood, you have high-tip tissue temperatures that can form at the interface. That can break down blood cells and cause char formation, which then limits the size of the electrode. You also have convective cooling from the blood going by, which pulls heat away, actually. And so then you end up with a smaller lesion than you would otherwise be able to achieve. By actively cooling the tip, you prevent those high-tip tissue temperature interfaces from occurring, and then you get more energy directed into tissue, creating larger lesions. The issue is that most of the time we use normal saline. That's what is FDA labeled. But it has an ionic charge. Ionic charge substances can conduct electrical current. And remember that RF energy is electrical current. So you get energy shunted away in this process, actually. By reducing the ionic charge, for instance, with using half normal saline, we direct more of the energy into the tissue and thus create a larger lesion. In our preclinical studies, we have demonstrated this consistently in both of the ventricles compared to normal saline, both in unipolar as well as bipolar configurations. We have recently collected our cumulative experiences with other friends who have caught word of this technique across the country, and the world, actually, and kind of published our efforts. It initially failed attempts with ablation with normal saline, then subsequently ablated with half normal. All the usual suspects for causing headaches are involved here. And what we were able to do was achieve acute success more often with use of half normal saline and urigine compared to without, and that usually translated into better arrhythmia-free survival in those patients. The final approach, which still can be done, we can go to the OR if you really have to, for selected cases. But it's possible to do that with contemporary mapping tools. So in this case, we've gotten lateral thoracotomy access. We're mapping on the epicardial surface of the heart. You can see here this catheter can be moved relatively easily in the epicardial space. Or yeah, the epicardial space, sorry. We can induce, if the patient happens to be on bypass, which this patient was, and you can map for a while in VT, you don't want to obviously take too long doing it, because being on bypass for too long isn't great. In this case, this area of interest correlates well with an area of voltage abnormality that was seen on the map. We can entrain there, confirm that it's a site of interest by jerry-rigging some materials from the EP lab. And then we apply cryoenergy there. And you see you get delay and then termination. Really, tears of joy flowing throughout the room. But we're not done there, obviously. You can't just go for VT termination as an endpoint. So we kind of consolidate that region pretty heavily and demonstrate non-inducibility. So looking at populations of non-ischemic VT, this is data from Penn. You can see that although the VT free survival as an absolute is compromised many times in these patients, the overall VT burden and quality of life is improved. But sometimes it takes multiple procedures. Note that this is a Kaplan-Meier curve showing VT free survival after the last procedure. So that means multiple procedures. In fact, more than one procedure in more than a third of these patients. But the message is true that many times it just takes more attempts or multiple attempts in these patients, more effort. This was data from our institution looking at patients with mid-myocardial substrate already undergoing a repeated ablation. And just on their first repeat, do worse. Subsequent repeats, once we kind of were satisfied with how everything was going in clinical follow-up. And you can see that you actually can achieve the same level of VT for free survival in those patients compared to those who don't have that very vexing substrate. But importantly, we had to be creative in about a third of these patients. So that was a lot. I'm going to conclude by saying that reentry is responsible for most VT in structural heart disease. The ablation procedure, as well as mapping techniques, have evolved. And overall, we've improved in outcomes through a series of multicenter collaborations, as well as multidisciplinary efforts. Challenges remain in controlling the ventricular arrhythmias involved in mid-myocardial substrate. And advances continue to be made. And judging by how completely full and full of excited and engaged faces that I see in this room every year that I'm here, I know that the future is bright for this. So thank you. Thank you. With Dr. Zou, thanks for that great presentation that we recorded. And now we have a lot of questions coming in for you on this complex topic, which is broad. We'll try to cover as many as we can. One of the first questions that came up has to do with core isolation. And the question was, if you're going to try to isolate a whole area of ventricular myocardium, are you not also getting rid of normal tissue? And what happens to the EF, or ventricular contraction, if you're also eliminating areas of heart that may be normal? That's a good question. So when we talk about dense scar, scar has many different varieties, right? So the densest scar is the area of tissue that's hardest to excite, but may still have living, viable channels within it. There is border zone tissue, which, strictly speaking, although there's still viable myocardium within it, typically it is not as normally functioning as completely normal myocardium is. So more so in the cases of non-ischemic cardiomyopathy, where you truly have greater heterogeneity in the fact that there's perhaps better functioning tissue mixed with denser scar that's more patchy, in the cases where you already have a fair amount of dense scar and border zone tissue associated with it, normally that tissue that is associated with the dense scar is still not normally functioning either. And for the most part, especially in people who have ischemic cardiomyopathy, when we're ablating in those places, we have not seen a change in the EF, or a change in the focal wall motion abnormality in that region. Even in non-ischemics, where there's mid-myocardial or mostly epicardial substrate, commonly that wall motion there is not normal either. And so we haven't observed to date any significant change in the overall function. Now, that being said, there does, as you've sort of suggested, have to be cautioned exercise in this. Because especially as we've gotten more comfortable with ablating more and more, we really, really do have to be mindful about not affecting things adversely. The early surgical literature sort of taught us that as well. Alongside the sort of Pennsylvania Peel approach was this concept of just doing a full thickness ventriculotomy of areas that had scar, as well as border zone tissue. Those patients did OK acutely, the ones that survived. But they tended to do much worse in longer term follow-up because of heart failure. And so those are important lessons historically that we always have to remember as we move forward in this. A little bit of a corollary to that is if you're doing core isolation, but your lesions are not fully transmural, how is it that signals inside the core are not able to escape out to the rest of the myocardium in the deeper parts of the tissue? That's a good question. Many times what we observe with scar, and you can kind of see that in that autopsy sample that Bill Stevenson has shared with many, you have layering of scar within there. And there's some relative compartmentalization. And so luckily for us, the lesions don't always have to be fully transmural in order to be effective because you just have to penetrate to that border of inexcitability and kind of link those areas up. And then you can kind of, relatively speaking, contain whatever excitement is going on in the middle because it has no place to go, theoretically. A question came up on epicardial mapping. How do you distinguish low voltage that's fat from low voltage that's scar? That's a great question, and one that's really, really important because I maybe didn't emphasize that well enough in the discussion. But low amplitude signals will all look red. Same with low amplitude signals due to poor contact. Poor contact can be induced because of some tissue that's not myocardial intervening, and that's fat. So when you only get a low amplitude signal that kind of looks a little far field, but other than that is not particularly abnormal in terms of other characteristics. So there's no evidence for local conduction delay. The signal isn't fractionated, it's not long in duration, and there aren't late potentials associated with it. That's more likely to be truly abnormal. If all you have is low amplitude signal, and it's in a region that you would suspect epicardial fat to exist because of coronary anatomy, then it's much less likely to be a place that you would want to target. In fact, the coronary vasculature being there, especially if it's at the more proximal portions, you should absolutely not ablate unless they've got a known infarction in that territory already, or else you could cause more trouble. Yeah, those questions, by the way, are from Dr. Sterber and Dr. Premji. A question came up from Dr. Nino Pulido regarding MRI. What is the importance of planning with an MRI? And I'll add, do you get it in every single patient, or just on ischemics? Or how does that play a role in planning for VT ablation? That's a great question. I mean, academically, I would love to say that we get one for every case before ablation. And certainly, in some institutions, that's what's done. At ours, it's not feasible. For one, many times they're coming in with storm, and there's not the time to slot that in easily before we can get the patient to the lab. And as well, the presence of devices in most of these patients makes the imaging quality less ideal. We don't, at our center, have the kind of system where you can do all the fancy subtraction and imaging processing that makes scar more apparent. And so I would say that when you can get it, it can be helpful, especially if you're talking about someone in whom you're talking about doing a redo procedure, or you want to get some idea of what the substrate might be. If you question the report that it's a post-infarction VT, for instance, it can be helpful in those circumstances. But it's certainly not essential. A couple of questions came up on how to do things, or can you do things. Dr. Rashid is asking about the CARDO3 system. And can you use a pentaarray catheter to do unipolar endocardial mapping, and also to do late activation mapping? That is a good question, too. The multipolar catheters can all be used for unipolar mapping. And that can be in any system, really. Because really, all you're doing is you're taking whatever pole that's on the catheter, using that as the cathode. And then you're using this indifferent electrode, either Wilson Central Terminus. Some people place an indifferent electrode within the blood stream and use that. But the simple answer is yes. And all of the systems can portray that information. The late activation mapping within CARDO is not something that's programmed in. It can be done manually. But you kind of have to walk your mapper through it with respect to how to time during the map, doing the map real time. Yeah, so I'll leave it at that. Yeah, it's a good point. We can sometimes sort of jury rig these mapping systems to give us information that they're not originally programmed to do, just by simply having the mapper put the timing icon, a timing point, at different places, whether it's an electrogram-based or simply a time-based positioning of that icon. And then you can create different zones of color based on the parameter that you're looking to measure at those points and visually represent it in sort of a hacking the map kind of way. A question came up. And this is going to be hard without extra slides. But just basic principles, when you're doing entrainment of VT with multiple electrograms on that channel, how do you recognize which were the ones that you captured so that you know how to measure the last entrained beat to the relevant EGM? Yeah, that is a hard question to answer without some illustration. So I actually drew out a little cartoon. I don't know if it's going to work. But OK, so if you imagine, like maybe, can you see this? So this is the VT. This is the electrogram that's seen locally. And what you're talking about are these very exciting-looking site, right? Lots of mid-diastolic potentials. So then when you overdrive pace, you capture something in here, because you have advanced the tachycardia. This is an exaggeration. You never want to pace this much faster during entrainment. But just to illustrate the point that this is the first return, it looks, let's just say it looks concealed. But so this is the pacing stimulus. And then what do you measure to out here once the tachycardia resumes? So one way to do it would be to kind of, first of all, you can see clues, right? So in here, you see this in this portion here. During VT, there's this electrogram right here. That's not captured, right? Because you can still see it activating before the pacing has occurred. And then there's something after it that may be of potential interest. And then there's this little signal here that comes in right before you start pacing again. So one way to tell what you might have captured is to measure from, because what you've captured should be pulled in earliest. So you measure from the pacing stimulus to, say, a fiducial point within the QRS. When it's concealed, it's easier, because then you can kind of go to the same area. So that stim to QRS time, if you measure that timing out, let's say it's 80 milliseconds. You go out to this first return here. 80 milliseconds back would put you about right here. So that should conceivably be what you actually captured during the maneuver. That should be what you measure to when you measure the PPI. Yeah, that was magical, what you just did there. I don't know how you prepared that so quickly. But I guess to reiterate your point is that you definitely do not want to measure to electrograms that you still see during pacing, because clearly you're not capturing those. That's not the immediate local potential from underneath your catheter. And then there are some other tricks that you can use of the ones that you hadn't seen that then reemerged. That was an amazing summary with great visuals. A question came up about the QRS morphology and localizing VT and how you factor in known areas of scar and how that confounds your ability to identify the origin of VT from the EKG, from the 12-lead. So that's one little semantic change that I've tried to emphasize. We always used to talk about where is this VT coming from. It's not really where it's coming from. It's coming from usually a large area of scar. The QRS tells us where it is exiting from. And so the critical circuitry, the central isthmus portion, is probably somewhere distant to the exit, but it's attached. And so the exit provides information about at least what a viable circuit is within the area of scar, and specifically the exit portion. Why we end up doing more ablation now than we used to is because it's recognized that there can be multiple exits. And attached to a single exit could be multiple other isthmuses. And so gathering information about the scar is helpful for anatomic localization of where we're going to likely ablate. The exit is important because we want to make sure that we really reinforce that side, because that's what the patient has had spontaneously. And so we kind of marry that information together to sort of figure out a strategy. I would say that most times, more often than not, we're ablating more areas of that scar than just the exit would represent or where the VT is coming from, just on the surface QRS. Another practical question that came up, you showed ablation in the OR and doing mapping. And the question is, how do you pull that off? Are you having your surgeons come to the EP lab to do their surgery? Or are you somehow transporting your mapping system into the OR so that the surgeon is at home? That's an excellent question. So we coordinate this very carefully and closely, and it's not easy, with our cardiothoracic surgeons. So they're an active part of the procedure. The case that I showed was done in the OR, in the hybrid room, where you could kind of see the electroanatomic mapping system information projected on the screen within the OR. There's fluoroscopy if needed. In general, I haven't pulled it in because of the logistical complications associated with doing so. But we have also done it in the EP lab. It's much easier to do in the OR for the surgeons. And honestly, because we're involving them, their team, the perfusionists, et cetera, as well as the OR time, I kind of yield to whatever they'll accept doing. And most of the time, they're most comfortable doing it down there. Now, on occasion, we've had the situation where we've had to do it in the EP lab because we couldn't get epicardial access, for instance, and we needed the surgeon's help with that. And so they have been, we've been lucky to have a collaborative enough relationship with our surgeons to allow that to happen, but that probably can't happen in a lot of places, or in every place, anyway. Maybe one last question before we go to our next video. It deals with epicardial access and anticoagulation. What's your workflow? And do you puncture while you're anticoagulated, or do you have to factor that in? Do you start epicardial? What's your sequence? That's an excellent question. If I'm not sure if I'm gonna go epicardial, I will do the endocardial stuff as planned on anticoagulation, and then I reverse before going epicardial. There's some data to suggest that you can do it safely on anticoagulation, but frankly, I haven't wanted to take the risk. If there's time and it's not an emergency to get epicardial access, then I take the time to allow the patient to become subtherapeutic. If I know at the outset, so for instance, an ARVC patient, or if it's a redo patient where clearly the epicardium was needed to be targeted, then I'll obtain that access up front before anticoagulation has been initiated. Fabulous. Well, I think we're going to end the Q&A to stay on time there. Thank you so much, Dr. Zou, for joining us. Thank you.
Video Summary
Dr. Wendy Zou discusses ventricular tachycardia (VT) in the setting of structural heart disease, specifically ischemic or non-ischemic cardiomyopathy. She explains that re-entry is the predominant mechanism for VT in these patients, usually due to the presence of scar tissue. Dr. Zou emphasizes the importance of scar tissue and its role in altering conduction properties and allowing for re-entry to occur. She also discusses the historical approaches to treating VT, such as surgically removing scar tissue and revascularization procedures. Dr. Zou then describes the advancements in VT ablation techniques, such as substrate-based mapping and ablation, as well as the use of high-density mapping, bipolar ablation, and altered irrigants to enhance ablation efficacy. She also addresses the challenges of mapping and ablating mid-myocardial substrate, as well as the need for multiple procedures in some patients. Dr. Zou provides examples of mapping techniques and the interpretation of electrograms to guide ablation. She concludes by highlighting the importance of ongoing advancements and collaborations in the field of VT ablation.
Asset Subtitle
Wendy Tzou, MD
Keywords
Ventricular tachycardia
Structural heart disease
Ischemic cardiomyopathy
Non-ischemic cardiomyopathy
Re-entry mechanism
Scar tissue
Conduction properties
VT ablation techniques
Substrate-based mapping
Heart Rhythm Society
1325 G Street NW, Suite 500
Washington, DC 20005
P: 202-464-3400 F: 202-464-3401
E: questions@heartrhythm365.org
© Heart Rhythm Society
Privacy Policy
|
Cookie Declaration
|
Linking Policy
|
Patient Education Disclaimer
|
State Nonprofit Disclosures
|
FAQ
×
Please select your language
1
English