Good morning. Can you hear me? Okay So, my name is Julie Moby and I am the director for the LAO registry and this session LAO imaging the changing landscape of imaging modalities Will have the session and then following the session. There'll be some Q&A, but we ask that you Use your the mobile app to submit your questions and what you just have to select the the session and then at the bottom of the Set once you select it There'll be a little box that says Q&A and that's how you'll submit your questions And then if there are questions, we'll I'll ask the question out loud and then we can do it that way But you have to submit them through the mobile app Or the website well, however, you're Doing the presentation or viewing the slides and everything We are happy to have Dr. Steven Philby with us presenting today and he is an interventional cardiologist and director of the cardiac cath lab at University Hospitals and Cleveland Medical Center he's assistant professor of medicine and Program director for the interventional fellowship training program at Case Western University His research focuses on structural heart disease intervention and percutaneous coronary intervention And he has authored dozens of peer-reviewed publications book chapters abstracts Abstracts and case reports his work has led to several novel novel techniques for left atrial appendage closure So welcome, dr. Steven Philby Thank you Thank you Okay, well it's a pleasure to be here thanks for having me It's always good to get out of Cleveland to Florida whenever possible next time please invite me in February January it'll be great So I'm gonna cover pretty much all aspects of imaging for left atrial appendage closure. I'm gonna try and do it in 45 minutes These are my disclosures Imaging for left atrial appendage closure Just like with any structural intervention is absolutely crucial It's crucial for planning the procedure understanding the anatomy for a device selection and Then in the context of the procedure itself making sure you're implanting in the right location that there isn't Any complications and then follow-up imaging also important has the device moved has it done its job? Is it sealed the appendage are there any late sequela complications? So imaging important all three of those aspects We're gonna start by talking about pre-procedural imaging now, I Would say 80 to 90 percent of US sites employ transesophageal echocardiograms for Pre-procedural planning this obviously involves esophageal intubation oftentimes general anesthesia It's it's a big to-do and it's not a benign process and here's the images that you get in the left panel You see a grayscale image sort of a granular Ambiguous appendage you don't really see that much definition and on the right you see cardiac CTA with multi-planar reconstruction Where you're able to really get a three-dimensional? Reconstruction in the bottom right quadrant there and actually go plane to plane in understanding the exact anatomy of the appendage And it is an incredibly complex structure the left atrial appendage is you know it's classically defined as a few Morphologies, but in practicality, it's as unique as your thumbprint as I described to patients they come in innumerable sizes and shapes and this becomes important as You implant a device which is circular into the orifice of this appendage you want to make sure it can seal properly only about 5% of left atrial appendage ostea or Or origins are circular or round the majority of which are irregular if you are making a size determination based on a diameter such as which you get with a TEE and Physicians implanting physicians will make their device selection based on maximum diameter using TEE You can understand that in ellipse and a circular circle Both have the same diameter But they have very different areas and this lesson was really learned through TAVR you know when we first started doing TAVR we were using TEE for device selection and for those of you that are familiar with Contemporary TAVR programs where you would say how many of those are using TEE now for device selection? none Always using CT, so we just basically carried that same methodology over in Implementing a CT program for our program and then there's additional things that you can use a CT fusion where you map out the left atrial appendage on CT and Then using bony landmarks literally transpose or co-register that image Onto the real-life patient under fluoroscopy subtract the bone And it's like playing a video game So you have the left atrial appendage there And you're implanting the device into that left atrial appendage structure that you've already predetermined based on CT so there's a lot of Advantages to using CT we've looked at Fusion in this small study. We didn't show a difference in fluoro time or radiation dose Versus a conventional strategy not using CT fluorofusion, but nonetheless small study And it's a very attractive downstream technology for added benefit with imaging So there are several advantages that we've already highlighted But in practicality you're able to do everything that you can with TEE with CT And then some you can exclude thrombus you get these really wonderful high dimension 3d Complicated anatomies you're able to plan the procedure you understand the relationship of the left atrial appendage to its surrounding structures You're able to plan your transeptal puncture and for all these reasons we switched from a TEE strategy to cardiac CT It really helped us that we had this this suite where You see on the on the right hand panel there if I can use my pointer So here is our CT. Here's our CT scanner right here, and then this is our hybrid cath lab So patients literally go from the CT suite right to this table from one table to the other you don't need to have such a fancy CT suite But it did help us because it had the juxtaposition of our imaging physicians who office right here Next to that with our interventional cardiologists who office right here, and that created a nice Dynamic that helped to inform our workflows and help us to design some of the workflows that I'm going to describe here and it certainly helped us to adopt the CT program that we have and Here's what we found in our initial experience, so let's go back to 2017 2019 predominantly doing TEE guided left atrial appendage closure procedures in In 2020 with COVID we switched over to a CT planned approach CT with ice And we were pretty good in 2017 2019 mean devices used per case around 1.2 What does that mean that means that we pretty much got it right most of the time so the number of devices you use in The case so if I choose a device based on TEE and I and I don't choose the right device But I end up putting in that device, and I'm like I don't really like the way that it looks Maybe there's a leak. Okay. I'm gonna take that device out and put a new device in so Ideally you want a mean device used per case 1.0. That means you're getting it right every single time you've Appropriately selected the the right device. That's important because the left atrial appendage is on average about 0.6 millimeters in thickness, it's it's literally translucent ex vivo you can see through it So the less instrumentation you have to the appendage the less injurious you Potentially are the less comp the fewer the complications so by switching from a TEE guided approach to a CT guided approach We found that number of mean devices used per case went down and our procedural success rate went up And this was our initial sizing algorithm that we published For those of you familiar with TAVR you can see that this is sort of analogous using an area derived approach for the Determination of the of the size device, and this was our initial publication one of one of our initial publications what we found was a very high rate of appropriately selected devices so 97% Accurately determined by CT meaning that we were we had an accuracy rate of device selection of 97% based on CT Getting it right almost every single time But then we found also that patients were going we're starting to go home the same day They were bored and it was just it just sort of naturally happened at the same time with kovat, right? We wanted to reduce bed occupancy Patients were bored in the hospital. We were the procedures were going so smoothly They were so efficient that the duration of the procedure was reduced that we were starting to send those patients home Just after one night. So same-day discharge Became a focus of our program and so we stepped back and codified that and said, okay We're gonna come up with the same-day discharge algorithm as it's not do this willy-nilly, but really have a pattern an algorithm Whereby patients are evaluated Do they have any growing complications are they able to spend the night with someone at home after receiving conscious sedation Was their minimal recaptures again? We're not instrumenting the appendage that much So if they checked all those boxes they were eligible under our workflow under our algorithm in program to go home and Lo and behold what we found was No complications. They did just as well as the patients who spent the night except they spent one last night and And not only that and despite the fact that we were using CT and using ice these Big ticket expensive items if you will in comparison to a conventional approach which used TEE General anesthesia had the patient spend the night. We saved the patient the hospital 15% on average per patient. So it was a win-win. We reduced bed occupancy We saved money the patients were happier because they got to spend the night in their their own bed and eat food at their house and So this is our most recent methodology we've had several iterations of of using CT in terms of planning the device and the selection of the device and currently what we're using is 100% device selection based on cardiac CT We don't we do appendage angiography as you'll see we go through this I'll go through some cases with you But we make the device selection entirely based on CT and we go for a target compression of around 15 to 25 percent But particularly with the Watchman Flex device and there's been this was an accompanying editorial to one of our papers from Claudia Tondo Basically a call to arms highlighting several of the advantages that I've already spoken to about CT over TEE But it hasn't been as well adopted as as I would hope CT I think there are several reasons for that one of which is a lack of Familiarity with the technology for implanting physicians. They they just don't feel comfortable looking at the images Perhaps it's it's a lack of training and then it was pointed out to me Well, we really don't have the follow-up that we do with TEE Most of the studies all of the studies upon which this device was predicated and improved were based on TEE So we have long-term follow-up data for those but what about CT do we have any long-term? So this was just published this week This is from our center showing five-year follow-up with CT plan procedure single center experience but I want to point out in comparison to protect prevail evolution registry data and CDR registry data that device related thrombus in our center using CT 2.5% it's the lowest published rate of DRT Perry device leak 3% also very low and we know that there are anatomic predictors to both of these Obviously, so getting it right getting the device right understanding the anatomy. I think really speaks to the success and these results It has I mean there is more data that's coming out this coming from Swiss apparel It's a sub analysis showing good success rate with CT plan procedure. So it is gaining some traction in the literature still a ways to go I Would argue that it has improved our procedural efficiency our accuracy I think it's possibly safer because again, you're not doing esophageal intubation You're not sticking something down someone's throat in order to plan the procedure and in in so doing improved patient satisfaction So, I don't know if you recall but was it about a year ago Maybe two that we had the worldwide contra shortage. Do you recall that? And so we started talking to our patients and saying You know, we're not we're not able to to plan your procedure using a cardiac CT We're gonna have to use a TEE, which is the conventional approach. This is the tried-and-true method and This is not a symptom driven therapy so patients don't come for a left atrial appendage closure wanting to feel less short of breath or Wanting to obviate or mitigate their chest discomfort like they might with a TAVR for instance They come because they want to come off of anticoagulation and they've got reason to do so They can wait and so these patients were saying I'll wait What is it gonna be three months four months five months six months and so we're busy center like us We're we're approaching like 400 implants this year So we can't we can't backlog three or four months So, what do we do if you have no contrast or it sort of forced us to come up with an MRI? sizing protocol, which is what we did and you can see the images on MRI are You know, they're pretty good. They're less they're less precise the the resolution is is less crisp perhaps than with CT and With this is one case series that we published based on our early experience with MRI pretty good so remember when I said 97% accuracy for CT based procedures this comes out to be about 80% which is around as accurate as C is T So it's about as good as T MRI it requires a special protocol and obviously you need you need a cardiac MRI unit to do it and we develop this with our imaging physicians this protocol, but it turns out to be a pretty good alternative for those patients who who Can't undergo a CT based on their chronic kidney disease provided they have no device or a cardiac implantable device that's MRI compatible if if all of those are exclusionary then of course we would still do a TE but it ends up being a very minute a very small number of cases that we actually plan this procedure with TE And how is our success rate? Well Really good in comparison to national average and These numbers are actually a little old. We are averaging about 30 minutes a case. That's skin-to-skin We've done it as in as few as 20 minutes per case We'll do five cases myself in a day and that's with trainees. So we have a fellowship training program so I'm teaching people to do the procedure and Really what it's led for our program to have exponential growth I mean patients as you know vote with their feet they seek out programs that they know are excellent and that they've heard others have very good outcomes with and So when I started this program the leadership of this program four years ago We were doing 50 implants per year and this year we're approaching 400 implants So Here's our workflow we get cardiac CT and just about everybody we give antibiotics conscious sedation We get femoral access to vein venous access sites. We of course use heparin for anticoagulation We do a single transeptal puncture with the Bayless system under ice Sheath exchange we do a pigtail angiography and we get three ice fuse. We'll speak a little bit about ice in just a minute and And then of course we we have to satisfy the pass or close criteria for device deployment and We get hemostasis typically with a perclose and a vascate So we do to use a vascular closure device and typically we'll give protamine what's exciting about This space and imaging is that there are now purpose-built CT platforms Specific for left atrial appendage closure. This is just one example called true plan, but Pheops is another one Three Mencio we've used and Terra Recon actually has more specificity for left atrial appendage closure You can get there with each one of these platforms But as we get more evolved and have very specific programs Really? It makes this whole planning procedure so easy So you're able to choose the angle the device you're able to choose the sheath you you gain so much information in planning the procedure This is real though, let me see if I can go back one Okay, so I'm gonna take you through sort of just a planning for left atrial appendage closure I'm finding the left atrial appendage moving my crosshairs to the the osteum or the origin of the appendage and Sort of squaring that up in two of three of these NPR planes and on the third It's gonna give you an on FOSS view and that upper left axial plane And I'm gonna drop my landing zone here and this is all done with a program right there So it automatically determines my landing zone. It gives me a size. I'm able to choose. It typically will select one of two sizes for me, and I'm gonna make my determination based on an area-derived diameter measurement, not a maximum diameter. And then once I do that, it's gonna actually place a device into, you know, sort of a device at that compression into the left atrial appendage. I can then move that device around. So I'm sort of trying to think, how is this device gonna behave in the procedure? All of this information gained with about two minutes on a CT work station. And here's a 3D reconstruction in which I can pick the angle that I want to implant this device. So I know going into the procedure, there's no guesswork. I know exactly what angle is gonna give me the best angle of deployment so that I can deploy this device and have a very good image and representation of how that's going to look so that I get the best result for closure. And so to that example, here you see on the right is an angiogram, so a pigtail in the procedure, injecting X-ray dye into the appendage to visualize it. And on the left panel is the angle that I had predicted based on CT. So you can see exactly representative of what I had selected. Importantly, I can also pick my transeptal. This is one of the areas for misadventure for this procedure, is if you puncture the septum, you're going across the septum, you puncture in the wrong location, or you puncture poorly, you can end up with a pericardial fusion. That patient's gonna end up needing perhaps drainage in some sort, either percutaneous or surgical. You want to avoid that. So having a good, being very meticulous about your transeptal. And also, where do you need to puncture the septum to get it into so that that device fits properly into the orifice? Because as I mentioned, so many different size shapes, but also angles of orientation, and you want to be coaxial with your deployment. Here we see on the left, this is in, so in the procedure itself, I'm injecting dye as the devices fit in, and you can interrogate it using ice here, and look for leak as well. So CT, I think, superior to TE, defining the anatomy, allows for better pre-procedural planning, including the transeptal puncture, improved patient experience, I think improved success, efficiency, and resource utilization. I really think it should be the preferred modality for planning appendage closure. I think one of the barriers, lack of familiarity and comfort with the technology, but as we have more purpose-built software systems, as I've spoken to, we're going to see expansion for CT in this space. Let's move a little bit now and talk about intra-procedural imaging. Now I've already mentioned that we, our approach is to use CT for planning, but then also intracardiac echo in terms of the procedure itself. So a CT ice guided strategy, to me, in my mind, that's coupled. If someone is going to be using ice to guide their implant, my strong bias would be that they do CT before. So it's CT and ice together, because you do lose a little bit of imaging quality with ice. Sometimes the imaging views are more difficult to obtain, especially if the operator is new to ice, and sometimes you're just limited because of this probe moving in the left atrial space. Can't get all the images that you can with a TE, but by having CT up front, you have so much more information. You can, it's a little bit forgivable, if you will. We use conscious sedation, and again, we deploy all of our devices with ice. Surpass Registry looked at this, looked at operators using ice, and initially there was some concern because there was a slight increased risk of pericardial effusions using ice. As it turns out, those operators had very little or relatively little experience using ice. In a more dedicated ice LAA study, and you'll see the central figure on the right-hand panel of this slide, showed that there were no complications in a group of 100 patients with more experienced operators using ice. I still think there's a learning curve with ice, certainly. But once you get there, I think it's worthwhile, and especially if you couple it with a CT ice strategy, you're going to see the results that I've already highlighted from our center. No matter how you image, you have to get multiple views. You want to see as many angles of this device as possible to ascertain whether or not there's a leak. So if you're doing TE, you're going to be getting 0, 45, 90, 135, or thereabouts. We use 2D ice with an 8-fringe catheter, and we get what's called the mitral valve view, which is around the 60 to 90 degree view. We get an aortic view, which is sort of the short axis, about 45 degrees, and we get the pulmonic view, which is the 135 or equivalent. Of course, if you're using 3D or 4D ice, as you were, you can get numerous, innumerable views. I mean, just there's no, really, there's no limit to the number of views that you can get. We use 2D ice at our center for the fact that the 3D system is very costly per catheter, and also we get what we need and have been able to demonstrate really great results. So I don't see, at least in my center, I don't see a value added, at least at this point, to moving from a 2D to a 3D ice. The 3D ice, I would just say, is very attractive, especially for newer implanting physicians, because you basically, you park this catheter in the left atrium, and you don't have to move it around. You just, the imaging tech can get all the views that you need by orienting the crystals from the control time. This is what I see during the procedure. Here you see on the left panel, that's ice imaging. I'm showing you that I'm puncturing the interatrial septum. You see the ballus wire in the left upper pulmonary vein. And then once we dilate the interatrial septum, I take my delivery sheath back, and I literally thread the ice catheter through that hole. It's like threading a needle sometimes. And you put the ice catheter in the left atrium. So we use a single transeptal technique. And then once I put my ice catheter in the left atrium, then I go back in with my delivery sheath, so that all of the equipment is now in the left atrium. And you get very good images with ice. Here you can see the Watchman Flex device being deployed in the left atrial appendage at a good location. So I would challenge you to say, could you get TE images better than this? I think that ice can get very, very good, high-quality images to support your procedure. And then here it shows that we can interrogate the device just as you would with TE, looking for leaks. So I look at color Doppler. And again, I'm going to get multiple views. This showing the mitral view. We call it the mitral view because this is the mitral valve here, shown in this echo picture. You know, after implant, the majority of centers will use TE. It is unquestionably the gold standard for imaging after left atrial appendage closure. But we had such great experience using CT that we shifted to do CT afterward as well. So we're using CT before to plan the procedure and now CT to survey the device to make sure that the appendage is closed. We started out imaging all patients at 45 days. Why? Because we wanted to be consistent with the registry. We thought that the LAO registry was requiring as such. And as it turns out, we had a lot of ambiguity with those images. So when you image with CT early, what happens is that you're going to get a lot of contrast in the appendage. It's going to look like it's going to be a leaky device. When in fact, when you put the device in, what happens is there's an endothelialization that occurs over the face of the device whereby skin tissue essentially from the left atrium grows over the fabric aspect of the device and completes the closure. And that process, as we think, probably takes about three or four months to occur. So if you image with CT before that has occurred, it's going to look like the device is leaking. It's going to look ambiguous. It's going to end up like we were, end up getting a CT at 45 days and then repeating it with a TE thereafter. So at the present time we image at four months unless the patient has a contrast allergy. It is helpful that you make a statement to some degree in the chart because this is sort of a newer technique. And you say something like a surveillance CT is required at four months to assess device position to rule out leak and to assess for any device-related thrombus. Once we did that, the peer-to-peer request sort of went away as we're moving into CT and perhaps insurers are not as well comfortable or familiar with this imaging technique. It is important that you work with the imaging readers if those are cardiology or radiology and come up with binary conclusion statements and not get too complicated. We want to really know about position, thrombus and leak. These are the three things that I want to know in my conclusion statements. When you get radiologists sometimes reading this or perhaps even imaging physicians, they want to write these exposés on how the device is looking. It becomes very flowery and confusing for our end users who often are nurse navigators and other clinical support team. They're trying to make a decision. Okay, is this something I need to worry about? What are the action items here? So we like to keep our conclusion. They can say whatever they want in the body of the report, but my conclusion statements are is the device in good position? Does the device have thrombus on the face of it? And is there a leak that we have to worry about? These are the three things that I try and emphasize with our team. This is our CT protocol post-implant. It's available to you for a slide if you like. And in building a program, again, it's important to be collaborative. I had meetings with my radiology imaging colleagues in which we went over cases, showed them the device, and then periodically every couple of months if there was a question whether or not there was a leak in one case or the next, we would sort of highlight those. Our agreed-upon strategy has really been if we had any, especially in the first year or two that we were doing imaging with CT, we had a low threshold to send the patient for TE if there was any question. If there was a question about a leak, TE. If there was a question about a DRT device-related thrombus, TE. TE obviously remains the gold standard until we have more data with CT. When I'm looking at a cardiac CT after implant for surveillance, I'm looking at four things, position, thrombus, whether or not there's a pericardial fusion, and whether or not there's a leak. I want to make sure that that device is not tilted. I want to make sure that that device appears to be as it was when it was implanted, so it hasn't moved. I'm looking at the compression, whether or not there's apposition or contact of the device with tissue all the way around. And specifically, are there gaps? Are there areas of leak? Device-related thrombus is relatively uncommon but still very problematic. If you develop a thrombus on the surface of the device, of course that is an itis for an area for stroke if that thrombus embolizes. So it's a very precarious finding. It's something that we don't like to see, but it does happen. It's been reported anywhere between 1 and 16 percent depending on studies. Again, in our center, we're less than 1 percent, and I think largely based on our planning and our approach to patients, but that's a single center experience. On CT, it appears to be this low attenuation area here. You can see it sort of looks black, like this black blob, this low attenuation. So this device is, what happens, of course, when we put the device in, if we seal the appendage, there's blood pool in the appendage, right? So if you seal that, what happens to that? It thromboses. It becomes clot. So what we want to see on a CT is the appendage is thrombosed. But I don't want to see thrombus outside of that. I want to see thrombus behind the device. And here what we see is this globular area there, low attenuation consistent with thrombus. And the confusing thing is that what happens, and this is a slide from Boston Scientific, the confusing thing is that when you put the device in, this is the Watchman device, there's actually a cove or an indentation, a virtual space around the threaded insertion, and then a cap of skin tissue grows over that. So that's a potential space. So the endothelial cap goes over that, so there's blood inside there. You can see here, and that can kind of give the appearance of device-related thrombus. And this confuses a lot of centers that are using CT, especially if they're just getting started with CT, because they can over-call DRT. And so what you have to understand is that, and these are all animal models where they've looked at CT for canine models and said, you know, is this DRT, explanted it, and no, these are all normal healing, normal endothelialized devices. But what happens is that endothelial cap extends over the face of the device, and underneath that skin tissue, underneath that endothelial cap is blood, which then thrombosis and gives the appearance of low attenuation. But you sort of have to look at where that low attenuation is occurring in relation to the device. And there's been some effort to sort of characterize that. Now these are, this is new data, relatively new, but this is an important study that attempts to do that in which they tried to characterize the morphology of this low attenuation. Is it sessile? Is it pedunculated? They even came up with a schema for measuring, is it three millimeters or more beyond the shoulders of the device? And practically speaking, we don't do that. I don't measure millimeters. I go with morphology, and I try to sort of understand, okay, where is the endothelial cap going to occur? You know, it's not going to go beyond the thread and insertion very much. If it extends beyond that, it really is concerning to me. Both of these would be very concerning and obviously consistent with DRT, whereas this is more consistent with healing, you see. So we make these morphological assessments, and if we have an idea if it's DRT, of course, and we would send the patient for a TE if there's any question. Moving on, we're going to talk about peridevice leak, and it really starts with whether or not there's contrast beyond the device. And here you can see that this entire appendage is filled with contrast. Here's the device itself. It's not black, so there's no thrombus in here. There's contrast coming in, and this is exactly where we measure the leak, this most narrow part. Understanding where that contrast is coming from. Is it coming through the face of the device? Is it coming around the device? When was the study done? Is the study done one week, two weeks, 45 days after the implant? Well, it could be coming through the face of the device we talked about. It takes time for that endothelialization process to occur. Really following the contrast, understanding where that contrast, and trying to go through the images, scrolling through the NPR images on your CT to try and understand the pathway of the contrast. This is how we read the CT scans. And if the device is tilted like this, and this is just a model that I created, that you understand that this device, this is the Washman Flex device, doesn't have fabric covering the entirety of the device. So if the device is tilted and there's a so-called mitral shoulder, so a little bit of a shoulder or part of the device hanging out, it could expose the seam where the frame and the fabric meet. And that is an area where blood can get through and create a leak. And so what you see sometimes in that situation is a layer of thrombus with contrast behind it. We call that in our center, at least, the macchiato sign. So if we see the macchiato sign, and we know that we're looking pretty carefully for leak if we see macchiato. If we do see contrast, if we're concerned about a leak, when do we send the patient for TEE? Again, the gold standard. So we're imaging everybody with CT, if their kidneys can afford it. If we see a leak that's greater than three millimeters or anything that's ambiguous. And in the early days we sent almost everybody for TEE, but we're doing it less and less often now as we feel more comfortable interpreting these studies. Of course leaks on a TEE are assessed by the vena contracta. So you can see the color jet here coming through. And there's such a thing called gutter flow where you see, you can see color coming, but it doesn't quite come all the way down around the device. It sort of stops right here. And it's not uncommon to see gutter. You can have gutter both with TEE and you have gutter flow with CT. Let's go through just a few case examples in my remaining minutes here just to give you a flavor of how we approach these cases. So again, real-time looking at a left atrial appendage closure with a Watchman device with CT. And I'm going to approach it very similar to what I did for the pre-procedural planning. I'm squaring up across two of the three axes, getting an on-fos few here. Right away you can see that this appendage is black, right. So we know that that's blood. So the appendage is thrombosed. Now I don't see any contrast distal to it. This is all black or dark as well. All this is low attenuation. This is a normal study. This is what you would expect. This is a thrombosed appendage. This is a closed appendage. This patient is good to go. Case number two. Again, I'm highlighting a little bit of a mitral shoulder here. So I'm going to keep my eye on that. But still the appendage appears to be thrombosed. And I'm getting a little glimpse of the distal appendage here. The device itself has thrombus in it. And distally it appears to have thrombus as well. And I'm going around. You can go around and you can scroll around the edges to see is there apposition. In other words, is the device touching the appendage on all angles? I mean, the resolution with this sort of imaging methodology is very, very clear, very accurate. It's actually more sensitive than TEE for the detection of leaks. So right now what we have to figure out is, what does a 3-millimeter leak on CT mean versus a 3-millimeter leak on TEE? When all of the data that we have, most of it is really predicated upon TEE. So we sort of know what to do with those leaks. What does a leak mean on CT? So our challenge right now is to compare the two, and so we're going to embark on a study, a multi-center trial in the next couple of months. Our center and a few other centers looking at comparing CT and TEE in about 400 patients in sort of apples to oranges comparisons, saying what's this versus this. This is another closed appendage, so a good result for that case. Let's go with an abnormal now. So again, same workflow. I do things the same way every single time. So the approach is squaring up across the ostium. And right away, I'm looking at this little glob of low attenuation, and that is very concerning to me for DRT. And as a fact, as we scroll around, there's a couple of areas that are very suspicious for device-related thrombus. This is the way that DRT, one of the ways in which DRT can appear on a CT. And so what we did in this center, and you can measure the Hounsfield units, so low attenuation, so less than 100 consistent with thrombus. And in this situation we got a TEE. And here you can see, I don't know if that shows you, but it's a nice representative example. I showed you the ice images before. If you look in comparison, this looks much more granular, grainy, doesn't have quite as much resolution even as ice. But this is a TEE. And here you can see a mobile echo density on the face of the device. This patient had a DRT. In our center, we've not had many DRT. I've already mentioned that. The majority of these will go away if you put them back on anticoagulation for a few months. About 70% will go away if you put them back on anticoagulation. So our practice has been to put the patient on anticoagulation if they're not on anticoagulation and then reimage in a few months. I've included an appendage angiogram in this case just to show you how complex some of these appendages are and how difficult they can be to close. You can see how large this appendage is and all of the trabeculation and pectinate tissue. All of these little nooks and crannies are areas where thrombus. The more complicated the appendage is, the more area for thrombus to settle into, the higher the risk of stroke. So we have a device. It's sort of tilted. We tried five, six, seven times, recapturing, putting another device in, trying to get the best result possible. Our puncture site was pretty high. So we see we're coming into the appendage at sort of a high angle. It would have been better to come from below. But I think this is the best area the operator could get on the septum. Coming in at this angle, this is the best that we could do. And did a really good tug test. We do a tug test to make sure that we anchor the device. We don't want the device to embolize. Here's what the CT looked like afterward. Now a contradistinction to the images that I've showed you before, there is contrast all in this device. So that this device is in its tilted. You can see that this is exposed here. So there's a leak coming here. There's contrast behind the appendage. There's leak here. So there are a couple areas of leak. The device does not fit that well into the appendage. It's not fit in a coaxial manner. And it's tilted so that there's a two millimeter leak here and a two millimeter leak here. And we always engage with the referring physician after discussion with the patient and with the referring physician decision was made to continue this patient on oral anticoagulation indefinitely. Okay, next case. Macchiato, right. So here you see thrombus on the face of the device. You see contrast behind concerning. Now there's a little bit of a mitral shoulder. I'm concerned about this area here. So I can move my crosshairs over there. I can scroll in with CT and target. By the way, these are all real-time workups, literally each one taking two minutes to accurately determine if there's a leak, where it is, how big it is, and then of course the decision making to follow. But the following important information, you get accurate imaging to support your decision making. There's really no ambiguity. So there's a little bit of a leak here. And so what did we do? We sent this patient for a TEE. And this kind of speaks to, you know, how big is a leak versus, in CT versus TEE, it's a trivial leak. I mean, this is a very, very small leak. We're going to let that one go. That's a one millimeter leak or more, maybe even that difficult, maybe difficult even to quantify on TEE. But certainly this is not a concerning leak. That patient could move to DAPT. Case number six. Right off the bat, you see there's contrast in the left atrial appendage closure device here. There's a little bit of thrombus on this lateral aspect. But there's contrast getting through somehow. So I'm going to see now, does the device have good apposition? Is the device touching the appendage all the way around? How is contrast getting through there? And it's important to know that this patient was imaged at 45 days. So this was one of our earlier patients that we sent to the CT scanner at 45 days. And it creates this ambiguity. We're not sure. Is there a leak going around the device? Or is there flow going through the face of the device? The flow going through the face of the device could come because it hasn't endothelialized at 45 days. So you wouldn't necessarily expect that device to be completely endothelialized. So in this situation, we repeated the CT scan, and what you see is a different device. Same device, but a different appearance. So now that we have endothelialization occur, the device is thrombosed and the appendage is completely thrombosed. So this is a closed appendage, and you can, of course, test for Hounsfield units. You can test if there's any contrast behind the device. I think TE certainly remains the gold standard for post-implant surveillance. I would have hoped to have impressed upon you that CT is really changing the landscape for imaging, certainly for pre-procedural planning as our center's data has been able to support. But I think CT also emerging as an important imaging modality for post-implant imaging as well. Again, you know, once you move to CT, you're obviating the need for a TEE-supported procedure. You're obviating general anesthesia, it's a one-hour CT versus this whole to-do with your imaging colleagues, getting this all set up for TEE. And patients don't like it. Patients don't want to have tubes stuck down their throat. And by the way, TEE is not a benign procedure. And there was a small study published two years ago in JAK. I think it was just about, you know, 50 patients or so, in which they looked at TEE guided structural procedures. So these were MitraClips. These were left atrial appendage closure procedures. And they did an EGD before and after the procedure. Okay, so baseline EGD and then they did an EGD after they did the TEE procedure. So 80% of patients had some degree of esophageal injury from TEE, 80%. 40% of those had significant injury including esophageal laceration or hematoma. These are old, frail patients, most often in whom you are performing left atrial appendage or MitraClip procedures. So these are patients in whom it may not be the best strategy. If you can get them through a CT, if their renal function is not prohibitively risky, I think it's favorable for the patient, certainly. And also better for the workflow for your center. I think that there is emerging experience to suggest that a systematic approach as I demonstrated going through these CT images in sort of a systematic way allows you to process this information, evaluate them and to come up with a strategy for treating. Some of the key issues to resolve are the timing of the images. When's the best time to get a CT after the device is implanted for left atrial appendage closure? Is it three months? Is it four months? Should we repeat an image at one year, you know, if we do it at six months or four months? And then also leak, we know that leak assessment is more accurate, is more sensitive for CT. So what's a quantification of leak in TE mean for CT? Are they equal? You know, so I think these sorts of things have to be addressed with upcoming studies. And I mentioned our center is going to be part of a multi-center trial upcoming. Hopefully we can inform the literature with that. Physician training and incorporation of AI in the CT software packages, this purpose-built, the easier that we make the CT platforms, the better, the easier it is for physicians to work through this, for implanting operators to work through this, and the more information that we can get from the package, from the software, I think it's going to enable increased adoption of this technology, which has been arguably pretty slow to be adopted, frankly, but I think that's moving and changing. So with that, I want to thank everybody again for having me. It was a pleasure. Certainly enjoyed Orlando. And I appreciate the opportunity to speak about this. And I'm happy to take any questions that you might have. Thanks. »» Okay. Is this on? »» Yes. »» Okay. Yeah, there are a few questions. »» Okay. Here we are. »» We've answered a couple of them, but I'll go ahead and go through some. We still have a little bit of time left. Do you still require a separate imager during the procedure after switching to CT guidance with ICE? »» No. You know, our implanting team is lean and mean. So it's me, a fellow, a circulating nurse, and then someone in the control room. That's it. Keep in mind we're using ICE and we're using conscious sedation as well. So it's a bundled deal. But it is very lean, mean and efficient. And again, one of the reasons why we switched, it was, you know, a necessity for us. Not only was there COVID, but frankly, we couldn't get the anesthesia support. I couldn't get anesthesia to come and do five cases with me on a Tuesday. There's no way. Now we're doing cases three days a week. We can hardly get them for the mitral clip procedure. So we just didn't have the anesthesiology support. And the same thing with the TAVR. We moved from general anesthesia to moderate anesthesia and finally to conscious sedation for our TAVR as well for the same reason. It was just, it came from necessity. But what then we found was, it was a win-win. It was a win-win because it made our teams more efficient, we were able to do more procedures. The patients underwent less anesthesia, so the recovery was easier. And they had fewer complications. »» Do you think the CT would be a good idea for patients that are getting ambulant implants as opposed to the TE? »» Yeah. I mean, although I focused and highlighted a lot on the Watchman Flex device, it just happens to be the device that I predominantly use because it is a Gen 2 device and there's a whole different, you know, that's a whole different lecture. We use both. And I think that there is room for both devices in the landscape of left atrial appendage closure or more frankly. But yes, CT useful for all types of devices, useful for, and as a matter of fact the FEOPS, so we showed the TruePlan and I worked through all that software that you saw, a lot of it was TruePlan. But FEOPS is another imaging modality that has a lot of specificity in particular for the ambulant device. Again, CT guided. So yeah, CT good for any left atrial appendage closure device. »» Is it acceptable just to do TEE just prior to the procedure itself and no CT done beforehand? »» That's a good question. So there is a school of thought with some implanting centers to do no pre-procedural planning prior to device implant. So patients come in, get on the table and they get a TEE. Then from that point they go for implant. So what it does is it obviates that TEE that was done a month before or two weeks before. It saves the patient an esophageal intubation, a trip to the hospital, general anesthesia, excuse me all that stuff, or conscious sedation as it were. My response to the, and there are some people that believe in this. So this is not, you know, completely clear. But my opinion is, is that you still don't have the understanding of the left atrial appendage anatomy. If you do ad hoc TEE, because you're still doing it based on TEE, and I don't know how much data has been published, but my understanding from speaking to implanting physicians and the clinical specialists who support those sort of ad hoc TEE procedures is that there's a higher abort rate. So what that means is, you get into the procedure, you do a transeptal, you're exposing the patient to a procedure, right, and the device doesn't, you can't put the device in to a satisfactory degree. It's got a leak. It doesn't fit well. And it's because you don't understand the anatomy. So whereas we have screen fails on our CT, I didn't show you a screen fail, but even though we have our CT scanner here and our operating room here, they're next to each other, after that CT is done we huddle, we look and say okay, is this patient amenable to a good result? And we do a good result on this patient, and if not, we say listen, it's a screen fail. I'm sorry, I don't feel I can close your appendage. We have, you know, two commercially available devices, but neither one is perfect. Sometimes you just can't get a good result. And by doing that sort of same day TEE approach, which you have is more aborted procedures, slightly more aborted procedures, so you're exposing the patient to a procedure that's not successful. We have, I think, one of the lowest abort rates in the country if I'm not mistaken. »» I think we can do a couple more. This is related to same-day discharge. So how safe is it for same-day discharge? And how much data or how long did you pull the patient info for same-day discharge? »» That first experience that I showed you was done a couple of years ago when we started our same-day discharge program. At this point we actually did have a long-term follow-up paper that we did with, I believe, three years of follow-up. It was the largest published experience of same-day discharge. There's been another couple of experiences published as well. The short answer is, is it safe to send patients home same-day discharge? It absolutely is safe. We now send home probably 98% of our patients same day. We have a little bit of, we still adhere to that algorithm that if they have chest pain, if there are any groin issues, if they don't have someone with whom to spend the night that night, and I don't want them falling over their bedside table at home, we keep them. And if they express that they want to spend the night, that's most commonly the reason why patients spend the night. But yes, it's very safe. This is, particularly in a conscious sedation procedure, this is similar to the sedation strategy that we use for cardiac catheterization or PCI. It's fentanyl and Versed. Now in patients in whom they're receiving general anesthesia, I think I would feel comfortable sending patients home same day if they recovered from their anesthetic relatively early. So if they have an early case under GA and they're able to be extubated within a short period of time after the procedure and they're sitting up eating, I still would feel comfortable sending those patients home same day under the same circumstances. They have someone to drive them, someone with whom to spend the night, et cetera. But there's a lot of data out there on same day. »» And I'll ask one more question and then we'll have to wrap it up. What is the best cutoff time to check for leak after procedure, 45 days, 4 months? »» I think it depends on what imaging modality you're using to test for leak. I think if you're using TEE, that 45 days is perfectly suitable and it's fine. If you're using CT, you can image it 45 days, but you're going to have a lot of appendages with contrast in it because the device isn't endothelialized. TEE doesn't detect for that flow through the device. TEE will show you peridevice leak, but it won't show you, it's not sensitive enough to show you flow through the device. So if you're doing CT, which is more sensitive, you're going to see contrast beyond the device at 45 days. So to give you an example, when we started doing CT imaging at 45 days, just to be in line with what we thought was the expectation, 30% of our left atrial appendage closures had contrast in the appendage beyond the device, 30%. Now I just told you our leak rate is 3%. So what happened was, all of those patients then went back to me, even if they were read by a radiologist, my nurse navigator then calls me and I have to reread it. And then we're sending patients for TEE to sort of figure out the difference and whether or not this patient truly has a leak or whether or not it's coming through the face of the device. So I would be careful about getting a CT at 45 days. I think that some centers still do that because they think that this is a registry requirement at 45 days. My understanding is that imaging is required after implant, as long as you image after implant. And then that timing is unclear. The study that we're doing, this multicenter trial that we're doing, is going to image at three months. We made the decision based on the CHAMPION trial because we were part of the CHAMPION had imaging at four months. So we just, to be consistent with that, we just crosswalked over with the four-month timeframe for our center. But for this trial it's going to be three months. It will also be interesting to see if there's any difference between three and four months. Practically speaking I don't think so. So it depends on what imaging modality you use. »» We are out of time. Thank you very much. »» My pleasure.

interventional cardiologist

CT

TEE

left atrial appendage closure

visualization

device selection

procedural efficiency

post-procedural imaging

AI-assisted CT platforms