Good morning, everyone. My name is Julie Mobate. I'm the product manager for the LAO Registry. And today, this morning, we're welcoming Dr. Homsey. He is from the Ohio State University, and he is going to talk to us about residual leaks post-LAO imaging. So, thank you. Oh, one more thing. If you have questions, just put them into the app, and we'll discuss later. Good morning, everybody. This is actually a very interesting topic, and it's important for NCDR to report it. If you remember a long time ago when we have the appendage clipped by the surgeon, and we said any communication is like should be treated, and we should put people back on or into calculation. But what we found, if you remember the first study we did, which was protect AF, we said if you have less than five millimeter gap or per device leak, that means it's closed. That's how we define it. And we start taking people off the medicine, but then we look at through some of the slides that we start seeing some problems. So, really, I think any leak is not a good leak, you know. So, we'll discuss that in detail. And how we can define the leak, you know. I don't have any disclosure. So, I'm hoping in the next 45 minutes to cover all this objective, we'll focus on various modality available for post-procedure imaging. Is it T, is it CT, or MRI? You remember about two years ago we have deficiency, and we have shortage on the contrast. And we did, we published our data about MRI, enhanced MRI, showed some reasonable data, you know. But that was pre-implantation, not post-implantation. So, the, what we're trying to figure out and try to understand what kind of imaging modality we can use to determine residual leak. And if we find, identify the different modality to define the leak around the device, or incomplete closure of the device, how significant is? Is it one millimeter significant, three millimeter, four millimeter, five millimeter? All these numbers like everywhere. And where this number come from? So, we'll review some data and go through this. And try to connect this data to the outcome, you know. Then, if we have a leak, how we deal with the leak? You know, everybody tell you when the left anterior appendage occlusion device goes in, it doesn't come out, except surgically, you know. So, you have to deal with that, you know. Because the way the device is designed, it's designed to have anchors. And once the anchor is engaged, really cannot take it out, you know. But there are some good, some promising and good therapeutic approach to peridevice leak. So, let's focus on the first objective, identify different modality available for post-procedure imaging. First, let's define what is peridevice leak. It's a residual communication between the left atrium cavity and left atrium appendage. And that is being noticed on the CT or TE. Now, this could be intraoperative, while you're doing the procedure, could be post-operative. You should, all implanters, they know that, when they have this, for example, this TE, let's say I'm implanting a procedure. Hopefully, we still have connecting, still connected to the delivery system. And we see this when they said we're not satisfied with this. So, what we do, we capture the device and maybe try to put a different trajectory or maybe upsize, go a little bit bigger. But this, it happened to be, this is a 45-days follow-up. And now, what we see here, this is on the TE. This is where we see the left atrium proper. And this big globe, uncovered by the Washman device. And we have this connection here. Very big globe here, slow, sluggish flow. And that's big concern, all right? Now, on the CT here, which is the other device, this is the disc and lobe. And you have different level of cross-section from the CT. And you can see there is a gap here. No matter where you look, from the surface of the disc, the proximal, distal, and mid-level of the lobe, we have gap. And that gap is concerning. And the reason we can talk about that more in the CT, when you see the contrast in the left atrium, exists beyond the device, it's that the contrast is making it all the way in the back. So there's communication here. So that is what we're concerned. That's something we don't want to see. But if you see, you have to deal with that, you know? Now it's 45 days, and the patient is looking for what? Stop in the oral articulation. And we say, no, not yet. Let's do something else. So post-left atrium appendage occlusion imaging evaluation is beyond that. It look multiple, for multiple evaluation of the left atrium appendage occlusion device. First, it's going to look at, where is the device in proposed or defined left atrium appendage landing zone? Is it at the zone? Is it deeper? Or it has a shoulder, like a little bit high? It look at what we discussed yesterday, last night, presence and absence of device-related thrombus. And also, it looks all presence and absence of peri-device leak, and define the mechanism, how it look like. Where is the leak? And does it uncover LA lobe? And we'll talk about, also, it looks at possible complication. You can see pericardial fusion, you know? And we can see, also, sometimes we see the define of the leak, if the leak complete or incomplete. And we'll come to that in a second. So we'll start with the case here. This is a 65-year-old gentleman with a history of coronary disease, cardiac heart failure, persistent atrial fibrillation, staspos left atrium appendage occlusion device, six months prior to his visit to the emergency room with atrial flutter. I'm an electrophysiologist. And usually, if you're a patient in atrial fibrillation or atrial flutter, you need to know how long he's been in atrial flutter or atrial fibrillation. It's more than 48 hours. You start anticoagulation and do imaging procedure to define if patient has left atrial thrombus. People said, you know, he had the TE, which was done here two months after the procedure. And it was reported that no leak, no thrombus, and well-sealed device. Now, I don't look at the procedure. But this, my colleague said at Ohio State University, that actually, this was outside the two months, because we do 45 days. And the patient, actually, people argue with me. He has everything was reported. And I said, I want to see that with my eyes. I'm concerned. Usually, in a person who has risk of CHAD-VASc 3 or 4 or above, for me, if a patient has a stroke, no matter how old the left atrial appendage occlusion device, I will do TE all the time. That's me, not in the guideline, because a patient has stroke. And we have to prevent patient from stroke. So again, a lot of unknown in this area. But let's say, in six months, it would look good. So I said, you know, I want the TE. So I put them on heparin and sent them to the lab for TE. And this is a TE. Let's make it look here. So the TE, if you see here, this is the left atrial appendage watchman device, occlusion device here. This is the left pulmonary vein. And this is called left upper pulmonary vein bridge. And I'm going to play it one more time here. You can see this called color doppler jet. You see something going here in this area. And usually, if the angle is right, how would they put the angle of the probe? And if you have a communication, usually this should go past the device. But they tried different view, and they didn't see any good communication. And they didn't call this leak at all. And if you call the leak, you have to tell how much leak. So they couldn't call this leak. They were concerned about this area here. We're concerned because they were thinking about this is, you know, it looks like thrombus. Thank god they were concerned about that. Because they said, you know, if this is the problem, we are happy with the device. Looks seated well. And we're concerned about this is a thrombus here. Where is the right arrow? And we don't see any leak. So that's at Ohio State University. We're doing this. That's OK, right? You look for thrombus. Can you define the thrombus? So let's do it again and see if we can. So they did the CT. Let me orient on the CT here. This is the device sitting here. And this is going to go from anterior to posterior, scanning the device. And we're going to just focus in this area here. Yeah. It went fast, but let's try it again. So let's focus in this area here. So device sitting here. And you see something here. You see that? You see this one here? So this should be the landing zone here. And you see this stuff going this way. We have a good area of uncover, devices uncover. And it's getting bigger. All right. So report came back, no thrombus, no device-related thrombus. But we have 10 millimeter peri-device leak and malposition. And this should be the landing zone. Usually, you have here the circumflex. And you have this ridge between the left upper perineal vein and left at the appendage. And the device has to be here. And the device is canted, is tilted. And if you look at this, this is where the face of the device, you have a tilted and malposition. And now we have to deal with this. Notice something interesting. You see this dark area here? This thrombus. But we still have some leak here. And that is concerning. Of course, what we did was surface-to-surface anticalculation. And we did the ablation. And again, what they did actually before that, they went back to the TE. And they tried to look at the front view and said, oh, yeah, we see some leak here. So what you get out of this, that means the CT is more sensitive of defining the leak, more sensitive than the TE. Interesting, when we have the first few studies, if you look here, the first two that studied protect AF and prevail, both of them used TE before and TE after. We did not use a CT, even in PROC and ASAP2. So most of the study, they did not use CT. So we used just a TE. And usually when they designed the study here, in 45 days, we discuss about anticalculation. If it looked good, it looked fine, it's well-sealed, no device-related thrombus, you stop the anticalculation and you proceed with DAPT, doing antiplatelet therapy after six months, then after just aspirin by itself, and they do another TE. That was a design study. Most people, they do this. I saw one of poster from Montana. Actually, they have it. They serve a large radius area, about 500 mile. And they have a lot of people, they don't come back for the 45 days. Makes sense sometimes because so big. But they find a lot of problem because people feel fine. And guess what? People stop the blood turn on their own. So it's concerning what happened in that. But again, that building's Montana. And they're trying to work hard. And they actually start to reach to the patient. And if you see that, it's really good poster. If you look at that poster on the LAA closure, they make a lot of effort to patient and brought their TE higher to show up with a higher percentage of follow-up and doing the TE at 45 days. Multiple study, again, all of them, either 45 days, two months, three months. And we talk about this. If you have three months, usually a patient have stroke. I do three months. And if a patient have no stroke or low CHAD-VASc 3, we go 45 days. So it's really variable. And the orange, actually, it depends on if you see device-related thrombus. So you have the follow-up. Or if you have leak. At The Ohio State University, since 2016, April 2016 until June 2023, we look at the imaging modality at 45 days. No surprise, because we are following the studies, about 75%, 77% almost TE at 45 days, 6.6% at CT. Usually, for people cannot tolerate the TE, if they have history of esophageal varices, or they have other issue, not because of patient selection, just because they didn't get the TE done there. And at six months, if we have about 10.8, again, Ohio State, as a Frank County, we serve a big radius area, almost like four county, northwest and west, as well as south. So sometime, the follow-up will happen at the outreach hospital. So we don't have a lot of data, but everything report to us. So nothing happened in Ohio State, all of them, but we have some data come from our site. So let's go ahead and look at how the cardiologists will report the device leak. And this is, again, to orient you here. This is a transvaginal echo. And the sonar beam is coming from the esophagus, looking at the cavity of left atrium. Here, you see the left atrium appendage closure device, which is Watchman. And you see this big color flow going through. And you see this green color is called vena contracta. When you have something tight, you have this kind of like a jet coming out. And that's what you have to measure, the widest area here, and see what is the peri-device leak. And this is estimated about to be 6.7 millimeter, above 5 millimeter. So you see a visible gap, and you see the communication. You put the color Doppler, and then you determine the vena contracta, the highest one, and you measure the maximum area of peri-device leak. So the advantage of transvaginal echocardiogram, again, is you don't have to worry about contrasts. You don't have to worry about affecting the kidney disease. If the patient has mild kidney disease, you don't have to do it. And most of the study, again, was done by TE. And most people, actually, they don't want to try different modality. The disadvantage of that is less sensitive, and large leak could be missed. Sometime you don't see it, and they have the risk of procedure. In a good hand, you don't have problem. But in some places, you might have the high volume. If you have a high volume center, you're OK. Patient comfort. And sometimes they have, even they use a 3D assessment. It might help you see other structure, but it's not optimized. And the concern was, during COVID, was aerosolization. So people avoided that. We did more CT at that time compared to the TE. And it can miss low and large lobe, as we showed. If you look at the CT, of course, the CT is more robust on that. It can give you the device position. Hopefully, you don't look at the CT and you don't find the device. Yeah, early stage, we see some embolization. People look at device. There is no device. It's somewhere. But we don't see it on the new devices. All you see of the devices is canted or have a shoulder. You look at the leak, and also you look at the device-related thrombus. We talk about hypothyroid, and it's called HAT. And define it on to define if this is a thrombus or normal healing yesterday. And look at the surrounding structure. You look at the impact of the device pushing on the left upper vein, mitral valve, and pulmonary artery. And also look at plasma complication like pericardial fusion. So same like the TE, but it's higher and better resolution. When you look at the TE, for example, here, you have Watchman device at a CT. And what you see, you look at the contrast in the left atrium, which is here, and beyond the border of the face of the device. And you see the contrast is going through to the distal area of the device. So if you see a pacification behind the left atrium appendage, you have a leak. You have incomplete seal of the device. And then you have that. And then you have to determine the largest area. And you can have 360 degree and define the largest area of leak. And then you can also define the mechanism of leak. This is the first one I showed to you earlier here, first slide. And really wanted to look at a couple of things here, which is really interesting here to look at the Hounsfield unit. So if you have Hounsfield unit measured here like 558, and it's less by 100 or more in the area of distal left atrium appendage, that mean you have a leak. Or the percentage of this number to this number more than 0.25. So first, you see the pacification. You have a pacification, you have a leak. Now we have to determine where the leak comes from. Sometimes you see it, sometimes you don't see it. And you can, again, you can get cross-sectional from long axis to short axis. And you can define the largest leak, which happened to be in this area here around the proximal face of the loop. Here, we see something really interesting here. If you remember, the watchman device has this fabric. And the shoulder was so high and the contrast is going under the fabric from this side and going from the top as well. So you have double leak on this side here. So again, when you look at the device leak on the cardiac CT angiogram, really what you want to see, you want to see this. You want to see the lead again. You want to see no contrast beyond the face of the device. And the device is dark. You know, usually you're going to have a clot. And you remember yesterday we talked about hat zero. That means because of the design, the watchman, the fabric will be in this area and you see some clot underneath. And this is under the fabric, not on the face. So this is called hat one. You don't need to do anything here. The concern is here, watch and B. You can see the scaffold and the outline of the device looks really opposed to the wall, but you see the contrast here, same contrast. You see it passes here and you don't see a lot of darkness here. There is some leak. Is that leak going through the face of the device because you have lower endothelialization? Is it when they move the rod or the delivery rod, it injured the face? Because sometimes, you know, they don't pull it back and it might have damage to the face of the device. Is it a micro leak around the device? However, all this gets concerning because if you have incomplete endothelialization, do we decide to keep patient on blood thinner a long time? And that is a question we ask our CT, CT endogram specialists at OSU to define if the face will have a good endothelialization or not. They don't have a tool to tell us if this is endothelialization or not. So it will be a shared decision to discuss the patient. And it's okay to take patient off the anticoagulation and continue on DAPT, dual endothelial therapy, or you can lower the dose of DOAC to half. Or if patient is in agreement, you can stop it and in three months repeat it. Hopefully you don't see anything on the face of the device. So really we are not sure yet if we have a mechanism to tell us this is the normal healing, don't worry about it. It's micro leak and micro leak will not cause any clot release from that area. The concern is when you have a visualized leak or big leak here. And then you have to define how we deal with that. Now when you look at the mechanisms, which is really interesting here, if you look at patient on the first row here, you can see this device actually, it uncovered the whole lobe, big one, the same one we showed earlier. And you see still some opacification and some partial clotting inside the device. It's okay to have a clot in the device because this is what's going to happen in the end. You're going to have a clot and nothing is going to move. But other thing here you can see also, you can see the same thing in this patient and this patient, in the amulet and the watchman. Uncover approximately tissue, you can see that one and it's a little bit, they have this in this area here. Again, you see some opacification distally, but you see the channel. Same thing in 2B and in the amulet. If you have, again, if you have long, high shoulder, more than half, you're going to have leak around the fabric and that would kind of pacify. So you try to avoid the shoulder. Sometimes it's difficult because some of the appendage are shallow, so we recommend to go to use the amulet instead of watchman. And here is canted and tilted device. So I tell you also about the mechanism. So CT is much more, it's much more better and could be the mechanism of peri-device leak. So the advantage again is non-invasive procedure, no sedation, 3D assessment for the device and surrounding and adjacent structure and increased sensitivity, especially when you have a delayed phase. The problem with this is contract, especially when the patient has kidney disease or have any other clinical irrelevant leaks. Now, we don't have data head-to-head cardiac CT and TE, like compare head-to-head. We have some data here that CT and TE were done not the same time, close time, but you can see from all this number, I highlighted the green for TE and the yellow for the CT and you can see either equally sensitive or much more sensitive to detect the PDL. So really we don't have a good study, we don't have a good comparison and they are different way to look at it because TE is a procedure of sedation, CT you don't get it, you don't, you have different environment, do you do the CT first and TE first. We are in process of doing a study, actually a multi-center study, including Ohio State, to look at after procedure CT as well as TE to compare. But again, this will tell you they are much more sensitive than the TE in detecting PDL. So in comparison, TE versus cardiac CT was left-handed appendage occlusion device. Cardiac CT use has increased over time because we think we need to use it more. Anything questionable, like in the case I see a jet and I can't understand why the color jet is coming, I will order CT. The advantage of TE include no renal injury and less sensitive. But sometimes if you use a contrast, you know, some can, in the TE, like we use it to look for PFO, sometimes you can see if there is a leak around the device and that sometimes we've done it and we have some benefit on that. Again, this is a study we're doing now, it's called the CT versus TE post-let-handed appendage occlusion device. We have four centers, Ohio State University, University Hospital, St. Francis in New York, and North Carolina and one in California. So we'll let you know, we're going to start the studies soon and then hopefully we'll have the result to be reported. I didn't find a degree of pre-device leak. If you look at the studies, you know, from the earlier studies, the PROTECT-AF, you know, they have leak, but again, we say it's less than five, you stop them, but they have leak. When they report all the leak, they are really up to 40%. The amulet, when they look at the amulet, the IDE study, they compare Watchman to the amulet and they found that more leak in the Watchman compared to the amulet. But again, this was the first generation. So it goes as high as 55%, you know, 54%. Evolution actually, it's lower. So it's all over the place. It depends on how you define the leak. Again at OSU actually, we look at our data, again from 856 patients from 2016 to 2023, we have 35.2 leak. It's not as much as the NCDR, we'll show it to you in a second because it's lower at that time, which is include 51,000 plus patient. But if you break these group to two group, less than three millimeter and more than three millimeter, you find it's almost like a little bit higher and more than three meter. And why is this important? We'll show in a few slides because usually this will regress over time and they close up. If you have more than three millimeter, it's either stay the same or progress. This again, six months imaging at 45 days at OSU, we don't have a lot because everything could happen on the outlier, none done at OSU, but we have some people as more than three millimeter and have less than three millimeter. So this is the registry from 2016 to 2019, NCDR registry. This is really eloquent paper by Al Cooley from the Mayo. He put together the data and it's really impressive when you look at this. If you have no leak on the Y-axis here, you look at this stroke and ischemic or TIA or embolization incidence. So if you have no leak, people, which is about 73.4%. And if you have a leak, it's the rest of this patient. And if you look between zero to five, there are 25.8%. Let's focus on this group, it's called small leak. We think it's small. And when they compare small to no leak, they have high, a little bit higher hazard ratio, about 1.1 for stroke, TIA and systemic embolization. And also they have more bleeding because we're going to put patient back on anticoagulation and major adverse event. So really what you need to do, what to take home message from this, want to be here as much as you can. And if you look even further beyond this, this is the NCDR we just talked about. And we look at the amyloid IDE. Again, the amyloid look at two devices and they use donor-integrated therapy. And they have leak also, but it's lower and follow-up is 1.5. This is the earlier study of the TACT-AF. The TACT-AF prevail and continuous access about more than a thousand. They follow a patient up to five years. And what they found that at 45 days, they have leak is higher, but when you follow up them at one year, it gets lower. If you have leak, it gets higher. If you have no leak, you have less stroke. So really any leak, it will predict stroke. So we have to close the leak. So when this came out, it was a really awakening point for us about how we deal with this. Do we screen people all the time? Do we try to close this leak and how we want to deal with that? This is the same what we did about the five-year follow-up. And if you look at the five-year follow-up from this thousand patient, just spread it over time, and you can see about, again, it's almost close to the NCDR. It's about 71.6%. They have no leak. Zero to three is about 15%. And more than three to five is about 12. And about 0.7% more than five millimeter. And again, it predicts higher. If you have leak all the time, you can predict more higher non-disabling stroke and cardiovascular disease. Now, overall, cause mortality did not show any difference. Really, all of them didn't show any difference. The only difference I showed here is non-disabling stroke, but it's still a stroke. So Dr. Afzal, actually, my colleague at Ohio State University, he had multi-center retrospective analysis of, and then I said about all patient has Watchman implant. And he included 1,039 patients without device leak at the implantation. The report showed no leak. And they have a follow-up, about one-year follow-up, transvaginal echo to look for PDAL surveillance. Again, we didn't have any, but they took a patient who had this follow-up. And they look at combined primary outcome include failure to stop or decagulation, because there is a leak. We kept the patient on leak at 45 days or 90 days. TIA stroke, device-related thrombi, and need for peri-device leak closure. This is a primary outcome, big, ambitious evaluation. And they look at new PDAL identified 45 days, 90 days, using transvaginal echo. Again, this is TE era, noted, and they found it to be about 10.5%. Again, this patient, they have complete closure of the device. We're happy with that when they left the lab. And they have no leak. All right, let's show you an example. So this first patient, you can see, this is when he was in the lab. So we have zero degree. If somebody is familiar, they are different degree to look at different angle of the device. So usually, zero degree and 30 degrees show, usually you don't see any leak. You see the leak on 131 and 90. So zero, 30, again, this is the left anterior appendage. This is the device. Look good. The color, I don't see anything coming out on the device. At 30 to 45 degree, again, this is the aorta here, and you don't see anything. The aortic valve here, but you don't see any leak around it. About one third of the device has a shoulder here. So the fabric is in this area, close, but we don't see any leak. This is on the day of implantation. At 131 or 135, we didn't see any leak around the device. Now this is 45 days, same patient, trying to get the same angle close. So here, we don't see anything. Again, zero, 45, you don't see a lot. But when you start going to 90 and 131, you start seeing this. Start evolving. And it said, what happened here? And they came up with statistical analysis to determine if it's three millimeter or less than three millimeter. And when they look at that at 45 days, nine days, and one year or longer time to see how the trend is going to go. When they came up with the statistical analysis to use a three millimeter cutoff, that mean if patient has less than three millimeters, usually it regress over time. So you go from 2.2 plus 1.8 to 1.6, significantly noted here. If you have above three millimeter, it does get lower sometime, but it's not significant statistical. So they use now three millimeter because of that to be the cutoff. So if you have a patient on three millimeter, you can continue on anticalculation and repeat it in three months and see if it closed up. So what they conclude that the primary outcome which we talk about like not to stop or anticalculation or stroke or TIA or devastated thrombus or need for closure was higher if you have more than three millimeter. So it was 69 percent compared to three or four percent, which is significantly higher. And the incidence of the stroke was also significantly higher than patient without EDM. So they came up with this flowchart. So this is the patient is getting the device in the lab. And hopefully everybody leave with no leak. And hopefully if somebody has leak, they did something, recapture the device, redeploy, upsize it. We ended here. So hopefully everybody has this. And I think all of us, we do that. We be sure that the device is not leaking. If still leak and have problem, we take it out. But after six weeks, which is 45 days, we have less than three millimeter or more than three millimeter. If you have less than three millimeter, the authors, you know, they suggested you can stop anticalculation and start dual antiplate therapy and repeat in six to 12 weeks. And see if you have less than three millimeter, you don't need any follow-up. If it's more than three millimeter, you do the same thing. If it's less than three millimeter, you can continue the screening. If you have more than three millimeter and be persistent, then you try to close that gap. Our institution actually just published this data on WatchmanFlex. And they include 355 patients, again, without PDL. And they look at the evolution of device leak. And the primary endpoint was, is there any PDL? And the composite secondary endpoint is the continual anticalculation, device-related thrombus, TIA stroke, same what we've seen in the first study. But this is more observational study. And they look at the surveillance of the imaging to see if we have PD device leak. And what we found out, out of 355 patients, now, if you remember the pinnacle study, they have 400 patients. They have no leak. And if you remember our comment yesterday, this was restricted to a skilled physician who did really good. And they did, found like, almost like no leak. But here, in real life, we have 139 out of 355, which is almost 40%. They have leak at 47 days. And the average leak was 3.2 plus minus 1.4. And then the composite, again, here, because you have leak is about 30%, any PDL, like for the first one, you have about 40%. And looking at the TIA or need for closure device and intervention was in 30%, so 42 patients. Now, the PDL was noted in 33%. They have no PDL, you know. I'm sorry. The composite, when they look at patient has PDL at 42%, and which is 30%, and 15% without leak. So, even has some stroke, even they didn't have leak. But most people, if you have a leak, you have higher chance of having stroke. And out of this 43 patient, you know, and 139 who has a leak, you know, they have a follow-up on their leak and see if it's regressed over time or not. And what they found, that the leak will regress in 77% and stay stable in 4% and 9% and progress, get worse in 14%. Now, this is a WatchmanFlex, not WatchmanFlex Pro. And this is how they look at the baseline at the leak. You know, if you detect a leak at 47 days, that was the mean. And it's about 3.75 in the upper end and 3.4. And over time, overall, they have some regression. But if you plot all the patient, and you see the blue, this blue is regressing. This is interesting. You see 9 millimeter, it regressed to 6. But what we did, actually, we don't like that, so we close it. The purple mean we did a closure device or coil. Then you see red, the one that got worse, and we have to close one of them here. But most of the patient, you know, they regress over time. So, their strategy, if you are less than 3 millimeter, again, you can wait and see if it can regress. But again, we don't understand why this regressed like this. And if you waited longer, maybe it will regress more. But we have tendency, if it's, you have more leak to, and it's higher than 5 millimeter, to schedule for closure device. This patient, actually, in particular, patient has, when C9 millimeter, we continue that calculation. We scheduled for the closure device. On the day of closure device was six meter. However, we proceeded with the planned device to clear and close that leak. The author actually has predictive markers about who is going to have PDL, or Pre-Device Leak. If you have difficulty position the device, is it because it's shallow, or it has different flow, and it cannot do it. So multiple attempt increase your risk up to 2.7 times. So if you have a problem with that, stop there. Use AMULET. Use different strategy. If you have larger device, larger device, sometime, if you remember, when we put the device and there are criteria for putting the device, called pass criteria. One of them is how much compression the device has. And you have to be sure that it's about at least 8% to 10%. We didn't have large device in some cases. Like the largest one, we have 35. Now they have 40 millimeter. So large device sometime might have higher risk of leak. Usually what we do at the end of procedure at the high state, we do a picture of venogram. So we put the sheet close to the face, and we'll get pictures, contrast, and see if there's a leak. And if you use the contrast, you have better prediction that this will not have a leak in the future. Some institution, they don't use contrast at all. Now we look at the factor. What is really, what the cause of this leak? And there are really multiple proposal or concern. One of them is how is the, if the patient is persisting AFib, is the patient persisting AFib, is the patient has remodeling, is the patient has more fibrosis, or the orifice or landing zone, is it eccentric or it's irregular? And if you did the dotted spot, this is where you put the device, you're gonna have leak. And it's the ideal position would be the red. But if you're not aware about all this irregular area, you will choose this area. If you measure from here to here, and you put the device there, you're gonna have multiple area of leak. The other thing, if you look at the side profile, if you have also high point and lower point, and if the device is sitting here and it has a little shoulder, you might get some leak under the fabric. So you have to really have a good look at this. And that would make CAT scan and computational simulation much more important because you can look at different angle. And again, how much the compression of the device, how much diameter of the device, and how much coaxial alignment. Here, one of the computational simulation, I use CT before and we look at different view. This is long axis, short axis, and we can get the perimeter here and detect the shape of the device. And you'd be surprised. We've seen some shape like this. And if I measure this way, definitely I'm gonna have a small device here. And you might see it like tear drop, triangular, elliptical, you name it. So once you have perimeters, you can predict the average diameter and then you have to oversize. And then this simulation for watchman, they can actually simulate before you put the device in, the size of the device. So if you put like 27, you're gonna have 10%. If you have a 31, you're gonna have 25%, you're gonna have shoulder. The CT also give you a good prediction of what area you have to go through the septum. So it has multi computational simulation become part of pre-device evaluation. I switched totally to CT prior to this, and we use MRI just short period of time because we were out of contrast. Let's talk about how we deal with the leak. So first we have to understand the mechanism of leak. Do we have the edge leak here? Do we have uncover loop? Or we have canted device, it's already tilted in. Or the leak going through the fabric. And the same thing for the same, both devices. We showed this picture earlier. So there are three way we do it. We can either use a plug, and that plug actually or duct occluder, and you need a CAT scan. And this is, you don't see that frequently that let's say you have a watchman in and you fit the whole loop and sometimes you can fit another occluder there. Uncommon, the common one you see this big leak here and long leak. You have to be careful with this plug. This plug doesn't have anchors. So you have to have the end of this device is being anchored either behind the device or sometime you go through the struts. You have to keep it there, otherwise you have imbalization and has to be compressed 50%. Here an example of TE, we showed the leak in this area at 94 degree and you see the color and it's going through this area here. And this is the 3D we're showing at 120 degree. And we did the multipurpose catheter through the agilis sheet through the transceptals and we find our channel with the contrast and we find the area of leak. Then we'll put the device 12 millimeter implants and the final, you can see it is gonna push the device a little bit, but you can see the face of the device. Now we have this device here and the end of this is anchored in the back. We have some cases sometime really cannot get, you have to push this inside the device here because you don't wanna have this plug imbalize. Now people say, what you do after this? Do you give anticoagulant and aspirin because you have another device? Usually that's what I do. Some people they do do an anticoagulant therapy and then you do another TE in about three months or CT. Endocardiac coil, this coil actually has been utilized for cerebral aneurysm, AV malformation in the guts, people who bleed and the intervention goes in and try to coil it and close it or you have shunt either congenital or iatrogenic. Usually you use it if you have shallow and narrow landing zone. When you use this coil, you have to be very careful that not to put it, you have to put it really deep beyond the device because if you put it in this area here alone, it might imbalize. So you have to be careful. You have to define physical boundary to ensure safe deployment and avoid imbalization. And we put a lot of coil. The one we use a lot is called Azure Thermal System. Maybe I'm not saying it right. What we do here, if you see this case here, we have this Watchman device and again, we see all this area pacified. So we're gonna fill all this area. So first we put the framing coil and if you see this framing coil, it's really outline the area of contrast here. Then we'll put this platinum coil with hydrogel coil. What it does is expand about four or five times inside that area. And at the end, we see a little leak here. We add more, but in the end, when you do pictures from the front, we don't see any more leak. So it has to be, you have to put it behind the device. Otherwise you imbalize. We are switching a lot to this coil nowadays. They are soft and they are really, we're feeling much comfortable with this than others. Radio frequency ablation. Think about it, that if you have a little bit, if you have a small leak here, really how are you gonna put this plug? You can't do it. How are you gonna put the coil? You can't do it. It's so difficult to get in. So if it's small, what you could do, you can ablate on the area side. Do not go inside and ablate. Because if you ablate here, you have what's called steam pop. Sorry, it's pop, not pus. And what steam pop mean, it's like rupture, disruption of the tissue and cause pericardial fusion. So you need to make ablation in this area and then have multiple ablation. And then when you have the multiple ablation in that area, you have not to do one or two, and you have a short burst of ablation, you know, 15, 20 second and watch the impedance. If the impedance drop too much, you stop and you watch the temperature sometimes, if it's temperature driven ablation. Again, avoid wedging. You have to be careful not to wedge it inside in the channel here because it could cause rupture and pericardial fusion and perfusion, perforation. Here's a case. Again, we look at the, there's a leak. So this is a device sitting a little bit in this position here. And you see the left upper main vein and you have the leak here. The three dimension, it's at 12 o'clock between 11 to one o'clock in this area here. And you can see in this area here, there's a white color. This is the edema from the ablation. The concern of this approach, that if this is edema and the edema is solved, could that have recurrence? It could. So that the reason we do multiple lesion to enforce this area of that to be sustainable. This is the least, it's not the least, about 89% success. It's not that bad, you know, and you could use it actually in some cases, especially if you cannot put the coil or cannot put the plug. So we use this one and this outcome in the end, we have no leak here. You see the color flow going through the main vein and in the 3D, you have no more space that we've seen in this area. This is my final slide. So if you see a visible communication between left atrium and left appendage, on the follow-up, 45 days, 60 days, whatever days, and then you diagnose the device leak. You better have to understand the mechanisms and what you need to do. You have to plan for that, you know. Now, the question here, if the left atrium appendage is thrombosed, you have clot, there's nothing on, let's say, cascan. Usually if I have a TE and showed leak, I get a cascan. If the cascan showed no contrast beyond the device and have everything, the device is full with thrombus and closed, I will not do anything. That, I would say, this is complete occlusion. So first, is there contrast beyond the device? No contrast beyond the device. Is there a thrombus and dark color on the CT? Yes, then I'm okay, we're done. You know, that's the 74% of the cases. Now we have to deal with this. If you have it, now we have the issue, we have the leak, now we have to know how big is it and try to find the largest area and the echoes have been contracted with the contrast, with the colored Doppler, or on the CT, then try to understand, is it uncovered lobe, is it a big channel? Then try to figure out what to do with that. If it's small, is that three millimeter, it's proximal, do you wanna just continue to watch that and closer, because really, there's no good data on that. Maybe Dr. Afzal and Dr. Bhutta paper will say, this is gonna regress, just conservative treatment and follow up in the three months. But if it's more than three to five millimeter and you have big lobe, like our case, you have to cover that. And then you have to come to make a decision. We talked about that yesterday. Bleeding risk or no bleeding risk? Well, patient cannot be on anticoagulation for a long time because he had a GI bleed and he bleed all the time and all his endoscopy didn't show anything, maybe AV malformation. And then you can decide on the approach to do it, what you wanna do. If you think patient has no bleed, you can do oral anticoagulation re-image. Hopefully it regress. If it doesn't regress, now we have to work on this. If it's narrow landing zone, low profile, I'm gonna put COIL or RF. If I have a big space and if I have lobe, I might put another device or I put an occluder. I hope I give you a good overview of how we deal with peri-device leak and how it's really specific from system to system. We use a lot of COIL at Ohio State. We don't do ablation. And it's regional to regional, but we don't have really good comprehensive data. And hopefully this will be reported by the NCDR and hopefully we have some information later on about long-term outcome. With this, I will stop here and I'll be happy to take any comment or any questions. Thank you. So we do have a couple of questions and time for those. So two part, what do you consider gold standard for 45-day follow-up imaging? And then the second part of that is, do you foresee a change of practice or recommendations to change TPE from 45 days to 90 days? Thanks. That's a very good question. That's a very good question. And that actually will give a great, again, we call it patient focus practice. If you heard in the morning on the great lecture from the transparency lecture today, really what matter is, what is the best modality? And that the reason we're doing this FIFE Center, multi-center study started soon. And we'll know more about, I believe, started soon and we'll know more about, I believe the CT will be better now. How I choose T or CT, I will tell you it depend on the scenario. If I have a patient has history of stroke, and patient has a history of systemic embolism, I'm gonna go with a CT. I'm gonna be absolutely clear that he's among the 71, 74% on the NCDR. Very OCD approach, and we should, all of us. Yeah, it's sensitive. Why not? We can use it. And it depend on the program. Now, we are blessed at Ohio State University. We have a very stellar, a great CT program, cardiac CT and TE. It depends where you are. If you trust that your team can deliver, sometimes you have to educate them. They have to train them. You have to do that. So it depends. I prefer personally CT. That's me. Now, my other eight electrophysiologist colleague in the electrophysiology might disagree with me. The chief actually all the time like TE, but I like CT. And I don't know. So the change in practice or recommendations, do you? Yeah, so I- From 45 to 90 days? Yeah, I think the CT will be dominating, I think. Now, and imagine the CT also has high resolution and it give you, it's evolving and faster. Now, yesterday we heard about the AKI. You know, for our group from Ohio State University, we have a great program there. And I get call all the time from them. If you have a kidney, abnormal kidney function, you can hydrate this patient. They have special protocols. So that could be implemented. So if you have this protocol about avoiding acute kidney injury and do CT, I think, I believe CT is gonna dominate and take over the TE. And what about the timeframe? So it doesn't matter, 45 days, 90 days, we're seeing some- That's a big question also we ask about. I think I would go with 30, with, sorry, oh, it's three months. So I'm an addict of three months because all the data we took yesterday about healing process of the left epithelial occlusion device has been extrapolated from the dog. And we said the dog healing is faster than humans. So really, I'm in favor of to wait three months. Now, it depends. Some people, they are a bleeder. They don't, you know. So these are the outlier. I am a big advocate to go for three months and CAT scan. Okay, for patients who have a large period device leak requiring re-intervention, what is the preferred device to close the period device leak? That depend on the CT criteria. If I have a big lobe and I have a way I can anchor the end of the plug, I prefer plug. If I have a device, let's say the end of the device is closing and I cannot put the wire through that, that will be plug. So if you have a big leak and uncover lobe, I would go with the plug, a vascular plug. If you have a big lobe, now sometimes you might have two appendage and one of them was forgotten. So you can put another watchman in. So usually I put a device, either plug or smaller left appendage occlusion device. If it's a smaller channel and you have a space to go behind the device, I'll put the coil to avoid embolization. I think we have time for one more. Do you see any difference in leaks between general anesthesia and ICE procedures? That's a great question. I do a lot of ICE, you know, and we did a study actually looking at multiple outcome, multiple variable outcome compared to leak. If you have a TE or you have side effect of the anesthesia, links of stay, complication from the TE during the anesthesia, because if you have a patient, put them under general anesthesia, put the TE in, and we found no differences on that, except the time for ICE dependent closure device take longer. Reason of that, because I'm doing the ICE and I'm deploying. So same person doing the two job. And somehow in academic and other places they are, we have a promise called efficiency of the electrophysiology lab, very common. And when they see the anesthesia, you know, they go faster when they have no anesthesia, take them longer time to come. So it's really depend on the efficiency of the lab, but I would say, you know, and it depend on the operator. If you do it all the time, you know, how good, how comfortable you feel with doing the ICE. So our data was not that big. Again, all the data are small. What we did actually recently added when our data collection define where is the device is being placed. Is it at the landing zone or patient left with a shoulder? You know, I believe that's gonna give us more information regardless if you are TE or ICE. If you put it a little bit higher, there is a concern that it might, you know, immigrated more toward the left atrium and you start having some leak under the fabric. So if you can put it in the landing zone, great. If not, maybe try different devices. Amulet is still available and you can achieve that goal. Thank you. And we're at time. So thank you very much. It's been wonderful.

residual leaks

left atrial appendage occlusion

imaging modalities

transesophageal echocardiography

cardiac CT

stroke prediction

peri-device leaks

occlusion devices

multi-center trial

personalized strategies