Chapters Transcript Video The Evolving Role of Cardiac CT in Clinical Care Dr. Villines presents the cardiac non-invasive functional imaging testing considerations for evaluation of chest pain. Todd Villines, M.D.Professor of MedicineVice-Chair, Clinical ServicesDivision of Cardiovascular MedicineUniversity of Virginia It is such a pleasure to be here. And I wanna thank Doctor Joyner and all of you for the kind invitation to uh get out of the mountains in Central Virginia and drive over to uh uh to see the coast of Virginia. So, thank you so much. And it's really a pleasure to be amongst friends. And, you know, I've known Doctor Joyner for years through, through her leadership in the AC C and as governor uh and just the, the leadership that she's brought in the field of imaging to this area. And so it's really exciting to see the program and actually to see this amazing facility that you all have uh have put together that's really dedicated to cardiovascular care. In fact, I'm envious in that respect um to see the specialties all in this single building working together and to see about, about what you're doing in the field of imaging uh and, and with cardiac CT. And so I'm gonna talk today about the evolving role of cardiac CT and I say evolving because, you know, this is a dynamic field and how we're using CT. I'm gonna show you some of the recent guidelines and evidence and perhaps even some re some studies, maybe that you haven't uh maybe, maybe aren't aware of or haven't read yet. And then talk a little bit about how we operationalize this to um improve access, improve quality of imaging. Um because all of us are being asked to do more. But if you're in the field of imaging, you're asked to do more echocardiography, more ct and this volume is growing uh exponentially. And there's no other modality that, that uh you know, where that is the case or, or no other modality that where that's uh definitely the case in CT. And so these are my disclosures and I'll point out that I, I don't just do cardiac CT. In fact, I spend a lot of my time reading a lot of other imaging modalities like, like most people who do imaging. Um and that we, we, we all do uh some, some background, I've been at the uni uh the University of Virginia since 2019. Uh II, I took a little hiatus um really trying develop software to make ct uh more efficient and effective. And so I, I'm a former Chief Medical Officer, but I realized about uh you know, uh about six months ago that really what I, what I missed was, was doing uh patient care and research and, and uh and clinical work and, and teaching fellows. So, so this is, you know, just an image I think, you know, this is for those of you who don't see coronary CT imaging or cardiac CT imaging, we're gonna focus mainly on imaging the coronary arteries. But we're gonna talk a little bit about other forms of cardiac CT such as ee you know, enabling folks like Dr gentles and others to close appendages or to image people after they've had such devices. Um and how can we approve uh care efficiencies? And so the goal when we talk about coronary CT is a very high accurate, high, highly accurate test to directly image noninvasively the coronary arteries if you think back to 2005. And so, um this wasn't long after doctor Gentles taught me to read EKG S at Walter Reed. So um I, I really, I really thank him for that. But when I was a senior fellow in, in in cardiology, uh this was the front cover of Time magazine and think back to where we were back then, you know, when you think about functional tests, right, this test to be able to noninvasively assess the coronaries for not just stenosis but plaque detection, there was no other modality that could do that. We had calcium scoring, but we know it didn't detect stenosis. Um And the whole goal if you think about it and that at that time, we were very stenosis focussed, it was detecting epicardial coronary disease. That was the focus of functional testing, getting the right people into the Cath lab. And so to be able to do that noninvasively without putting catheters into the body, was felt to be one of the holy grails of imaging. Now, this was the front page of uh the cover of time, how to stop a heart attack. And what was ironic about this is that at that time, we had absolutely no evidence that Coronary CT could do anything about stopping heart attacks or saving lives. In fact, it was criticized for a very high radiation test. It, the images weren't that great. We had uh overall accuracy was modest, but the promise was there. And so this is really what got me interested in this technology, particularly having walked out as a fellow and you know, you need four or 500 catheterizations and you're like, wait a second, I think we need a better test. So many of these tests are normal. All right, we're taking a lot of people to the Cath lab that probably could be diagnosed if we had such a technology. And that was kind of my interest. And so I wanna, I wanna fast forward and I think all of you are probably aware of this document. Uh This is the 2021 multis societal chest pain document or the US chest pain guidelines and where we've come since 2005. And so we'll just highlight where we're at for the treatment. And this is, this document covers as you know, both acute uh and stable uh chest pain. And so this concept of doing initial clinical risk assessment and then using imaging selectively in patients where management will be changed. And so this is where we are today in 2021 guidelines. And fast forward, these are the guidelines that we all currently operate, operate under. Is that for example, in stable chest pain, we see that you have a choice between an an atomic test, coronary CT A or a variety of stress testing. And in fact, going from a test that had literally no evidence back in 2005 for its use, other than you could generate pictures that seem to be reasonably accurate. It's now a class one level of evidence say the highest recommendation of any noninvasive test for the assessment of coronary disease, um uh noninvasively. And so the guidelines give us some uh a little bit of a guardrail. They say, well, where should you use coronary CT? And I think you're already doing this here very effectively. Uh You know, it's certainly in patients where you want to obtain information about uh epicardial stenosis or obstructive disease. But also I think the unique, uh I think the unique strength of CT is its ability to noninvasively assess Blackburn. I'm gonna come back to that a real hot topic in CT. Um And so this idea that we know the majority of patients that we test are not going to have you know, significant epicardial disease. And most of them won't need to go to the Cath lab, but many of them will have atherosclerosis and that's something uniquely detected uh by CT, you have to have the expertise and that's what you all have here. Uh perhaps consider it in younger patients or if you've had a prior functional test that's inconclusive. Or if there's a unique indication such as evaluating anomalous coronaries or assessing uh additional an anatomy in the field of use, such as the aorta uh or particularly known lung pathology. And this is true for both acute and chronic chest pain. We won't spend a lot of time with this but multiple randomized clinical trials in patients who present to emergency departments that you can use coronary CT to effectively evaluate these patients and um risk stratify them uh as compared to standard of care. And so this is kind of where we've come. This is the current guidelines. Now, I think one of the things that I would say that is becoming increasingly, I think recognized is that we do a lot of chest imaging, not just cardiac ct of high proportion of your patients have had prior chest imaging, non gated chest CTS, whether it be contrast or noncontrast. And I tell the fellows that if anyone has had AC T scan that covers the heart, that is a prior cardiac test and it should be looked at. And I would tell you that uh you know, all of you who, who, who do clinical cardiology probably already know this. But having your packs open when someone comes in for chest pain and they've had AC T scan within the last few years that this data can help refine your pre-test probability that their symptoms are due to epicardial coronary disease. And so this is something that you should look at and this would help guide your test selection. You know, for example, if someone has sheets of very, very dense circumferential calcium on a noncontrast CT scan in the past, perhaps you would do functional imaging. But that if someone has had nondiagnostic or equivocal functional testing, don't repeat that same test again and expect a different outcome, right? This is a in fact, in fact, the the use of CT is now in the guidelines which I've shown you for inconclusive functional tests. It's the preferred test. And so this can be used as a gatekeeper. I'm gonna show you some data showing that you can use coronary CT as a very effective gatekeeper to your Cath lab and probably keep your interventionists as busy or busier involved in the care of patients who actually have coronary disease using CT as a gatekeeper. So these guidelines, I'll show you these are now aligned with international guidelines. The Europeans were a little ahead of us in 2019. Their chest pain guidelines made Coronary CT the preferred test in most patients provided that you had local expertise and particularly in patients who have no known coronary disease. Um patients who have had priory vascularization or established coronary disease, particularly when extensive are probably better served with functional tests. Although that paradigm with uh some of our newer technology photo on County CT technology is probably also going to change as well. And this gatekeeper role again in the European guidelines should be considered as an alternative to invasive angiography if another noninvasive test is equivocal or nondiagnostic. And so this is a particularly important role for coronary CT. Um And uh and I think uh we'll get to kind of how we use this in an interventional way to guide interventions. Um And in acute chest pain, again, very consistent. This is a 2020 ESC guidelines. Again, a class one level evidence a in patients with chest pain where there is a uh either a low to intermediate uh uh likelihood of coronary disease when cardiac opponents are equivocal or if they're negative. And, and, and there's an ongoing suspicion for A CS. And I think in the era of high sensitivity troponins, I think we're learning that of course, that can help us do no testing in the very lowest risk patients. But we're still left with these patients in this so called gray zone or in a terminate zone or have elevated but non dynamically changing proponents where there is an ongoing concern about acute coronary syndrome and doing CT in the emergency department can help you evaluate these patients. I think in a much more rapid way now. And then lastly, the UK, they, the nice guidelines came out in 2016. In fact, they were uh the first real major guidelines that said we're going all in for CT in 2016. They said coronary CT should be the diagnostic test of choice for patients with chest pain. They, in fact, they said don't do any more exercise, treadmill, testing without imaging and reserve functional testing to people who either have stenosis on CT or have known coronary disease. And what you've seen is a trim AAA trend. I'm gonna come back to this. How do we, how do we as a as a program grow to accommodate this changing volume if you look at the UK? Now this is only data for 2019. Let me see if I can get my pointer uh to uh to work here. Um But I think what you can see here is if you look at CT and the salmon, there's been an exponential increase, there's been a decline in, in this case, spect imaging. Uh And I think if you were to fast forward to the, the more recent data, you would see this trend continuing. It's now the dominant form of imaging in most European countries is coronary CT. And so why do CT and I think I've already mentioned this, is that what I like about it? Is it, it allows you to phenotypic, identify all stages of coronary disease and very unique compared to other functional tests. What I mean by that is we can identify patients who have no atherosclerosis. If you look at these patients with an IVIS catheter, very rarely do you miss areas of of of of of plaque even with inter coronary imaging. Um although the spatial resolution obviously is below that of oct and and intravascular ultrasound. But we know the prognosis of these patients is outstanding out to 15 years of follow up all the way to severe coronary disease. Um And so this concept that you can really identify the patients, not only you need to go to the cath lab, but who need more preventive therapies and you know, back in 2005 or 10 or even 15 people said, well, I don't really need this plaque data because all I have are aspirin and statins for preventive care anyway. And I'm just gonna use those reflexively anyway. And I think that really has changed in 2024 where we have nearly a dozen compounds that have been shown to reduce heart cardiovascular events. And how do you choose amongst them? All right. Well, many of them injectables, long term therapies, G LP one receptors, SGLT twos, et cetera are driven probably um are, are, it's challenging many times to select which of those therapies a patient should use. And I would argue that we know, conceptually that our therapies that I just mentioned are most effective in those at highest risk. And so plaque burden on CT I believe can help you make those choices and counsel patients who are going to benefit most and who is unlikely to benefit. And as far as accuracy goes, what we know the strength of CT compared to other modalities is it's very high per patient sensitivity. And so when you look at it compared to the ca to the Cath lab, it has a very high sensitivity, we don't miss critical disease. Um And it's um if you look at its its accuracy compared to invasive FFR, it has also a very high sensitivity with a modest specificity. Meaning when we see an intermediate stenosis uh on CT, just like in the Cath lab, um additional testing is sometimes often needed in those patients because stenosis alone um is often insufficient to um to, to show whether that is a symptomatic lesion. And so why do the guideline writers make this recommendation? Well, there are studies like this and I think you're all familiar with, with some of these uh the Scott Hart study which was a randomized clinical trial comparing CT to standard of care. We saw a 40% reduction in M I at five years. Meta analysis have shown similar findings and all of this is not driven by improvements in or increases in revascularization. What we see in each of these trials is about a 40% increase in preventive therapies using CT based on the plaque that we see and, and most of these patients have non obstructive disease. And so this is for me was as a preventative and imaging cardios. Very exciting that we can take, not just answer the question. Do you need to go to the Cath lab or not? Would you benefit from revascularization? But I think we can make better informed decisions about prevention and patient risk. And we'll come back to that in a moment. Now, I mentioned this gatekeep and I mentioned my experience when I was uh a senior fellow um and spent a lot of time in the Cath Lab. In fact, as an invasive cardiologist for the first, you know, 15 years of my career, I've spent a full day every, every week in the Cath Lab and I became convinced that we simply had to do better because about half of our patients, if you look historically at any one of these large databases, about half of our patients who have no known coronary disease who historically referred for Cath. Half of them had no significant angiographic coronary disease at all. OK. And this was because um a lot, a variety of reasons. Now, some of these patients had ongoing symptoms, some of these patients had repeat presentations and they went to the Cath Lab based on those things. But many of them were because of the, the uh the, you know, the the performance of many of our functional tests. Uh you know, the challenge with artifacts and other uh imaging challenges using functional testing simply was a large part of this. And this is something that's been shown in the VA registry, the A CNN CD R and in multiple clinical trials. And when you use CT, you're very uh very much less likely to send people to the catholic who don't have coronary disease in effect from the ischemia trial, which was a landmark clinical trial that tested intervention, you know, treating ischemia with revascularization versus medical therapy. Um And we can go into the intricacies of that trial. But one of the things they did is they said we know what after you have moderate to severe ischemia, we wanna make sure they have coronary disease before we randomize them. Let's do AC T. So over 80% of these patients got a coronary CT. And what do you think they found? These are patients remind you with sight interpreted moderate to severe ischemia. It turns out one in five of them had almost no coronary disease and they were not randomized into the clinical trial using CT. And so I think this, this number should remind us of some of the challenges and why perhaps an an atomic strategy is now favored in the guidelines. And so I wanna remind you, this is a relatively recent clinical trial that I think um really solidifies what the guideline writers had in mind in, in, in, in taking stable outpatients with chest pain and, and, and, and doing coronary CT. And so this is a randomized clinical trial, over 2100 patients, they were randomized to what they call a precision strategy, which is really coronary CT plus provisional FFRCT, which ended up being performed in about 30% of patients if you had an intermediate stenosis on coronary CT, usual testing was functional testing and these were relatively low risk outpatients. These were patients who did not have known uh coronary disease and they randomized them and said, let's just see what happens for a year. And uh they had a pretty unique primary outcome and it was, it was a death as you would expect. It was M I but they included in that invasive coronary angiography without obstructive coronary disease. And not surprising. I think you can predict what this would show based on what I've already shown you um in a one year trial in low risk patients, no difference in death, no difference in M I but a significant reduction, an 80% reduction in normal cath rates. And so think about, you know, if you're, if you're like me at U VA, we have a finite number of cath lab spots every single day for outpatients and those are under high demand. How can we get your interventionalist where those slots are filled with people who actually have Coronary epicardial coronary disease who will benefit from either invasive testing with FFR or intervention. And so this was something that I think uh really solidified what the guideline writers had had, had stated in that ac T first strategy may be more efficient care. And so I mentioned utilization. And so if you look just at patients in this study who went to the Cath lab in the usual care, remember this is functional testing. 70% of those patients got no revascularization. They got a diagnostic angiogram and, and, and, and, and, and or an invasive test with FFR in very small numbers. Most of these were just a diagnostic angiogram and they went home, only 30% got revascularized. These numbers were reversed in the CT arm. And so your interventionalist are busier in the sense that overall cath rates were really not that different. So CT was not cannibalizing your Cath lab. It was changing the the the type of patients who came to the Cath lab. These were patients who actually had known disease. And now I will say in the CT world, there's an increasing recognition that even in the interventional world, you can use the CT data perhaps to even plan your interventions with regards to legion length, side branch angulation, calcifications. And there's international courses now that I work with folks like Carlos Collette and others um uh around the world that are now kind of using and embracing this data because as an interventionist, you're gonna see more and more people who come to your Cath lab having CT data already uh in the chart. And so, um and so this is something I think that as a healthcare system we can kind of embrace now, I'm gonna talk briefly about CTF far because I understand that is a technology that you all are considering. And I think it's important as you bring this into your uh into your health system that you understand where it has a role. But I think it's also important to understand where its limitations are. And I think that's something that, you know, ha having, having worked in this deep into the technology of this uh this space and developing CTFFR. It's something that I think um I, I find personally interesting and I think all of, you know, that invasive FFR um for patients who come in with stable coronary disease uh where they have inter intermediate lesions is the standard of care or some other type such as Ifr. Uh uh I think all of you are aware of that. In fact, the guidelines go so far. It's such a dichotomous decision tool that if you FFRIFR is normal, it's considered class three harm to intervene on those patients. Whereas if it's abnormal as you know, it's considered appropriate. And so this is a really important decision tool in the Cath lab. And so you can imagine the interest if you could get some directional assessment of physiology from CT uh why that might be useful? And so this is the whole concept of, you know, we've already done AC T can we based on the anatomy, the stenosis, um et cetera. Can we infer some changes in physiology from this data we've already acquired or do we need to, for example, see a functional test? Imagine a patient who's had a coronary CT that has a 60% mid led lesion or a proxy led lesion with unconvincing symptoms. And that's not an infrequent scenario if you look at large registries and our own institution. Um Most people who undergo coronary CT about a third to 40% are normal, another 30 to 40% are not obst clear, clearly, not significant epicardial disease, but the rest have some degree of stenosis. And this may be a decision tool. And so this is a little bit of a background about how this works. I think you're all probably familiar with this but this um uh y you know, takes this 3D data set. It creates a 3D model by extracting out the coronary arteries. And then it uses a supercomputer that models um coronary blood flow. It models what the microvascular resistance would be if under maximum uh hyperemia. For example, if you gave adenosine at high dose, um and it uses some equations of blood flow and resistance um based on the geometry of this coronary tree to generate an output that looks like this. And so this is the the one of the I think the dominant commercially available um provider in the field heart flow, they've been around now for several years. Um And this is kind of the output that you get is this assessment. It's an estimate, right? It's an estimate of invasive FFR. And I would point out that it's a very imperfect estimate. And so I think it's important that as you get these reports back and use them in your clinical decision making, I think it's important to understand where it does well and where perhaps it does not. And I'll go through this. So I use, I view this as a decision tool. We have had FFRCT at U VA since 2018. And in fact, we were, we were actually required to get it. In fact, many of the payers would not cover coronary CT or they would make you do preauthorizations unless you had FFRCT. So the payers were convinced that you would, you would reflexively send too many people to C A. This might help that. Um And so that's really where this is at. So what's the data for its use? And these are older studies and this just uh the most uh recent study was called the NXT. This is the study that led to FDA approval and you can see overall sensitivity and accuracy. It like any other noninvasive test is imperfect. And so these aren't 95 99%. Um but where do they help and what they do and what you probably can't see in the bottom left here is that they helped improve overall specificity. So if I look at a lesion and say that's a moderate lesion, um go to the Cath lab, right? It's a flip of a coin whether they actually have an abnormal invasive FFR, right? This made this a little more tilted in favor of being accurate. Uh if you included FFRCT and this is what led to FDA approval and it has been compared to other tests. So this is a specific study. This is a unique study. Um They took 208 patients and they did literally everything to these patients. I actually still to this day, I don't know how they did this study. They took patients to the Cath lab, they did three, they did three vessel FFR, they did coronary CT AC T, they did spect they did uh pet scanning. Um and they said, OK, what's most accurate compared to invasive FFR? And it turns out when you concluded CT plus FFRCT and you excluded all the bad cases, right? So they cherry in a way they kind of cherry picked, they took out 25% of the cases that they couldn't process. And I can tell you in looking at this pop, I've seen every CT in this study by the way, and there's a lot of bad CTS. So that was a little bit of the fault I think of the, of the site. So there's a lot of challenges in image quality and I'm gonna come back to that what that means in our, you know, our CT labs here. Um But when you have good image quality and you run FFRCT, you can see that it, it actually performed as well if, if not slightly better than what we think is uh a very high quality noninvasive test. And that is pet imaging uh compared to invasive FFR. But I think it's important to understand the limitations. I always tell people it's easy to be right um when it's normal. OK. And what I mean by that is um if you look in the clinical NXT trial, this is a distribution of invasive FFRS. Um And you can see that the majority of the cases, 80% of the cases in that study were normal. OK. And so only about 20% of the cases were abnormal. There was much more dispersion or le lower levels of agreement as your FFR got below 0.8. And so that's something that's important to understand is that um And in fact, when you think about the algorithm, I'm gonna come back to that the algorithm of FFRCT that they, that they use at heart flow, it's really designed uh in, in many ways to not miss things. And I wanna come back and show you kinda why that is, in fact, if you look at it, this is a AAA meta analysis that looked at the overall um diagnostic accuracy. And you can see it was pretty accurate when it was really low. It was pretty accurate when it was really high. But in this kind of middle zone, this gray zone, you can see again, it was kind of a flip of a coin. And I think we can get some insights in fact about this algorithm by looking at its own registry data. And this is something that's important to note that it generally, if you look at heart flow FFRCT, it generally overestimates the severity of the invasive FFR by a few points. And people have gone through and said, well in the L AD, it's, it's, it's about 0.04 right? So imagine if your FFR is 0.75 or 0.76 it's very common to go to the Cath lab and that be 0.81 or 0.82. OK. And that's something that systematically, if you look at all these studies, you've seen a slight overestimation of the physiologic severity and that's been our experience as well. OK. So interpret these numbers with that in mind. And so these values that you receive back from heart flow are not meant to be used as I in in a vacuum, right? You have to look, where is this lesion? What does this lesion look like from a black standpoint, what are the symptomatology? And yeah, if the FFR is normal, you're probably unlikely to find an abnormal invasive FFR. So in our clinical practice, it's very helpful when our heart flow comes back normal, we send about 8% of cases to heart flow. So the majority of Coronary CTS we do not send and we can talk about the finances of, of what we, you have to spend to, to, to, to send these for processing. But it's very common that if we get ffrcts back in the 0.7 to 0.8 range, that when they go to the Cath Lab, they are in fact normal and so be prepared for that and understand that it is only the accuracy is not 95%. It is helpful to improve your specificity. If it's very abnormal, unless you have really bad image quality, it's usually abnormal when you go to the cap LA. So again, understanding the limitations of the technology and the guidelines. Now, for the first time, incorporated, if you look in the center box here, they incorporated FFRCT as a two way indication for patients between 40 90%. Now, personally, I don't think I need FFRCT and a 90% for L ad lesion, I know how to manage that. Um And those people are generally going to go to the Cath Lab, right? But suffice it to say that's what is in the guidelines. It's a two, a recommendation to use in patients where there's, there is equipoise about how to manage the patients. Now, if you read the text of the guidelines, they do say this is for proximal to mid coronary arteries where you would consider Reva reservation. So I always tell the readers, you know, if this is uh somebody where they're never going to go to the Cath Lab for whatever reason. Um Well, I don't know what, maybe, maybe they shouldn't have got the corner CT, but that's you probably don't need to do FFRCT. If this is a small diagonal that you would never Reva right? You don't need to run FFRCT and spend $1000 on that to get that answer, you can probably manage these patients medically. So use medical decision making when, when to send this for FFRCT, but in proximal mid corna of sufficient diameter where revascularization would be considered. I think it is a helpful technology. Now, the question I always had was that, well, can't you, you know this, this technology, what I didn't show you in the algorithm was plaque, right? They're not considering how bulky the plaque is. Um And we know that plaque adverse plaque characteristics, if you have longer lesions that have high risk plaque where there's positive remodeling, they are more likely to be ischemic in the Cath lab. And we've seen study after study show that can we develop an FFRCT that incorporates that? And there's some interesting work. Now going on. Uh in fact, I think you're gonna see this uh uh coming down the road where um you can develop a technology that not just includes the an atomic coronary tree, but also includes the disease, the atherosclerosis. And this is just a one machine learning study that was published in 2021 where they said, OK, what predicts an abnormal FFR if we look at all of the features um using a software called Medis, which is a research plaque analysis software. Um And what they showed is that in fact, minimal Luminal area percent at aroma volume, how much of that diseased artery is made up of plaque, the volume is made up of plaque, how much of it is a low attenuation or higher risk plaque? Um And so there are some plaque features that were very predictive of having an abnormal FFR above and beyond stenosis. And in fact, stenosis was one of the, one of the lower predictors of an abnormal FFR and heart flow does not incorporate these plaque features. And I think that's something you as a reader have to look at uh uh and consider. Um and, and there have been, you know, this is a study that was, that was published. I was a steering committee member of this study called the Creedence Study. They said, ok, let's test, can we construct a plaque based model that will predict an abnormal invasive FFR? How does it compare to stress testing. And how does it compare to heart flow FFRCT? And so these were over 600 patients, they did three vessel distal FFRS. And that's an important, important point. They did distal invasive FFRS. Um They did not do lesions specific ischemia. Um But what they showed is that coronary CT when you include not just stenosis but actually plaque features was more predictive of ischemia by invasive FFR than was spec now, this is not a big surprise. Um uh But what what was interesting is that if you look at uh hard flow on the right versus a plaque model that when you incorporated plaque burden and adverse plaque characteristics, that these things actually were just as predictive of that normal FFR. And so there's um certainly research like doin day group that are looking at machine learning, imagine getting a coronary CT and analyzing it for things uh not just stenosis. And in this study, stenosis was very predictive but low density noncalcified plaque volume was the second most predictive um of ischemia, this low attenuation plaque and this is something that's not incorporated in current FFRCT and maybe in the future uh future iterations uh will. So I want to transition into, you know, we, I've shown you the guidelines not just for coronary CT, but for FFRCT, how can we optimize patient access? Because I can tell you our CT lab is being bombarded. I mean, we are we, we are growing at 30 to 50% year over year growth as the guidelines have shifted payer approvals have uh softened. And this is a test that is being increasingly or uh utilized well. And this is a study actually that, that I was involved in that did a na international survey. So don't just take my word for it. We went out and surveyed over 600 sites and this was uh a study sponsored by the International Atomic Energy Agency who has a health division that is interested in equitable access to imaging around the world. And we looked at PRE versus post COVID. We went out and got one month of imaging data from 600 sites. And what was interesting if you compare uh and this is kind of a baseline uh with COVID in blue. This is in 2020 everything went down, right? We weren't doing much of anything but we looked at recovery and so we looked at sites in 2021 mid, mid 2021 and said what imaging modalities have recovered. And not surprisingly, you can see that stress ecg stress spect and traditional testing modalities like stress. Uh you know that I just mentioned stress echo, none of those had actually recovered their volumes had per we, we had not even made it back to where they were in 2019. And that tests like coronary CT A stress MRI and pet have exploded and are increasingly being used. And so this shouldn't surprise you. This is pro this may have happened. I irregardless of COVID because of evolution of the technology and changes in guidelines. But it shows us that these uh technologies are gonna be in increasing demand. So one of the things that we try to struggle with at U VA was how do we get these patients through the scanner and get them home so we can increase our throughput, you know, is it going to take us 45 minutes or an hour to scan somebody? Can we do it faster? And so we asked the question with a dual source scanner. So for those of you do imaging, this is a scanner that has two radiation sources. And so it acquires images in half the time as a traditional scanner, it's much faster. And I believe you have that here. Um And so one of the things we realized is that, you know, the traditional guidelines recommend beta blockade with a target heart rate of as close to 60 or lower that you can get it. And this sometimes is time consuming. And how do you do this? Do you write a prescription before the patient sees you and have them take it the morning prior to seeing you about an hour prior to their appointment. Do you have them show up and then give them an oral beta blocker? And anecdotally, we've all seen that oral beta blockers seem to be more effective than IV. Do you have them show up, start an IV and then give them IV beta blockers. We know it's common for people to get on the scanner, NPO with a scanner starting and their heart rate goes up a little bit right now. Or you give them nitroglycerin and their heart rate goes up a little bit. So having beta blockade is standard of care in coronary CT. But we've known for years that you can image people at a relatively motion free part of the cardiac cycle. And this is at in cysto or during what's called the isovolumic relaxation phase of the heart. And so the the least coronary motion is in mid diastole when there's already been filling of the ventricles and you're waiting on a P wave. And this is shown here. This is where we traditionally image people, but there's also a small window at insist and that tends to be at any heart rate, a more predictable area to image. So you can image we image people uh last week at a woman at 100 and 20 beats per minute. And we got beautiful images because we imaged. It insists there's no di there's very little diastole left in that patient who's 100 and 20 beats a minute. It's very challenging. There's no motion free period. And so we started doing this. Uh And in fact, a good friend of mine at Mass General said that's what they've been doing for five years. He said we don't give anybody beta blockers at Mass General because we just can't, we don't have time. We can't stop and do this. And so they image everybody at insists. And so we started doing this, uh at insist and I'll show you kinda how we do this uh in the next slide. And so, um we don't get beta blockers for routine Coronary CT. We get a lot of, uh referrals in Charlottesville from kind of outside providers from all over central Virginia. If the heart rate is less than 70 regular, you can either do in systolic imaging, which is a lot what a lot of our radiologists recommend because they don't even want to mess with beta blockers. I still favor diastolic imaging of these people because I think you still get a little better image quality. But if you're over 70 or equal to 70 you would target insist. And generally speaking, we use a millisecond target between 250 to 400 milliseconds after the R wave regardless of heart rate. And so we studied this and in fact, what we saw is that our appointment times went this appointment time is our times that patients are in the hospital from start to leave. Ok. We actually, we actually turns out we actually track that and so they went from 94 to 69 minutes. Uh We saw our scan times in the scanner drop by five minutes per patient I I in the scanner. And this is preliminary data and our patient throughput in a day for outpatient coronary CTS doubled. Now our volume is going up anyway. But what this allowed us to do is actually schedule more outpatients and increase access to care. I will tell you that what I went through and we did, we did blinded um image quality scores. The image quality is a little lower within systolic imaging. I will tell you, I, I if you've got a stable heart rate of 50 do diastolic imaging, but it allows you to scan people, you probably would otherwise say you can't scan. Hey, the heart rate's 90 do something about that. And we've actually used it with very good success and you can scan almost anybody as long as the heart rate's regular. Now, another area that has changed our practice and this is going outside of the coronary realm is looking at the uh you know, how do we assess the left atrial appendage and people who come in and need cardioversion. And I, you know, and I came to UV in 2019, I said guys, we have this scanner that can do this in less than a miller of radiation. If it's me, I would take AC T over a tee from the patient. That's just me. Now we've known this for years and this is just an example of how we see sometimes slow filling this patient's laying on their back. Remember this is an anterior structure generally. And if you image them at first pass, you sometimes do see in patients with a FB some decreased filling. But if you image them between 30 to 60 seconds later, the majority of these will fill in. And if this were thrombus, you would see a persistent filling defect and you can do hands fill gradients and things. Um And we've known from multiple studies that it's very, very accurate, right? If you do CT with delayed imaging, you do not miss thrombus. And so, in fact, I would even argue as somebody who does a lot of tee, I did three yesterday uh for various reasons. And these were part of a clinical research protocol where you, they mandated a tee to clear the appendage. And I can tell you I always feel confident, more confident looking at AC T uh than I think I do nowadays uh with a tee. But regardless what we said is when, when COVID hit, it was, I've been screaming, we need to do this, we need to do this. And people say, oh, you know, the fellows need tee experience and you know, we're not ready to do this. We were already doing them pre a by the way, um when COVID hit, everyone said, let's just do CT, right? And so we can go down and do a really low dose flash CT scan and it's now we never went back. So it is now the default test in our hospital. No one almost ever gets a tee for clearing the appendage unless they've got a valvular condition by which you would need to do a tee anyway. And we now do it 24 hours a day. So our house staff have gotten really smart. You know, someone comes in on a Sunday and it's a nuanced set atrial arrhythmia and they're still in it on Monday morning. Uh and they're not gonna go for ablation. I show up for rounds and they've already had their CT scan. And our anesthesiologist love it because they don't have to do it, sit there for 20 minutes doing a te they can just, we can just sedate the patient, cardiovert them and they go home. We also can identify coronary disease during the scan. It turns out um and uh get people out of the hospital faster. And so this has now become our default test for chlorine dependence and we're now studying how it affects our cost. So in the last few minutes, I'm just gonna talk briefly about plaque. Um I talked about high-risk plaque features and this goes back to kind of seminal papers that have shown that in 1000 patients who did not have obstructive coronary disease if you had positive or expansile remodeling and a dark plaque low attenuation below 30. If you had both of those features, not only about 5% of people had it, but you had a 20 fold increased risk of future A CS. And this has kind of led us to understand that these high risk plaque features, things that I just mentioned plus body calcification do improve or increase someone's risk. And this goes back, if you look at studies on oct patients with high risk plaque features are much more likely to have a thin cap fibber atheroma. Um If you have multiple high-risk plaque features, particularly if you have positive remodeling, low attenuation, you're a 10 to 16 fold, likely to have a thin cap fibro aroma defined on OCT. Then if you don't have those features. And so while we can't look at CT and say that's a thin CAPP fibro aroma, we can't image down to that resolution. Probability suggests that these are indeed higher risk plaques. And so in our coronary reports, you see things like high risk or what used to be called vulnerable plaque, these positive remodeling expansile low ct attenuation napkin ring sign where you have both low attenuation in the center and a peripheral enhancement and spotty calcium less than three millimeters. And so these are now in our reports if you're reporting using what's called C AD reds 2.0. Um we now report not just stenosis severity but plaque severity with a comment on whether the patient has high risk plaque features if they have atherosclerosis. So this is something we're already doing. The question is, can we actually look at this plaque and quantify it? And there's now software that uh are out there will do an intense area of research. And I'm not gonna have time to go through all of this. But we learned, for example, in the Scott Hart clinical trial that if you quantified the amount of that low attenuation plaque, if you looked at the proportion of plaque that was uh made up or the proportion of the volley of the vessel wall that was made up of this plaque. Uh If you had more of it, you had a significant increased risk of having a myard infarction. This was superior to stenosis, it was superior to calcium scoring. And so a lot of people have said, well, gosh, we're not quantifying the plaque that we see, perhaps we should start doing that. And there are a variety of software that are now being tested. Um And this is a kind of an emerging field in CT is whether when we do a Coronary CT instead of just saying there's a mild amount of plaque here, a little bit of plaque here. This we kind of we actually, if you look at our CT reports, they read a lot like Cath reports currently, should we be more quantitative, particularly if patients may come back for a follow up CT. Um we where we can truly assess whether their disease has progressed. And so, um you know, this is a very uh evolving area and something that I've, I've, I've been involved in and it's based a lot of the interest in this is based on studies where they do repeat CTS. And you say, well, gosh, what predicts plaque progression. And it's a lot of the volumetric measures that we currently do not assess. And so I think um this is uh uh an exciting area guidelines currently not recommended. Payers don't pay for it. There is a code for this, but payers don't reimburse you right now for doing advanced plaque analysis. Um And there's some early studies, this is one with what's called a, a company called Clearly. And I have no uh no affiliation with any of these companies. Um But where they actually have a staging system based on plaque volumes and they put people in a stage 012 or three kind of. And the thought being that this would become more like cancer treatment where you'd stage people's coronary disease and they showed it was better than calcium scoring stenosis or manual assessments. But what they were able to do is show a, a estimated 10 year A S CV D risk. And so they're proposing that you get this scan, you run the analysis and you get a post scan, 10 year revised A S CV D risk based on your staging system. And that may, you may use this to better treat patients and I won't go for the sake of time through all of the details. But this is uh one such te technology that's out there and they also provide you a stenosis assessment. And so if you're thinking about, you know, this evolution of, and they've shown that their stenosis was as good as three experts. And so you're, you know, this increasing ization of coronary ct, not just acquisition but interpretation. And perhaps this software can help us better uh improve our reproducibility. Now, I'll tell you a lot of, there's a lot of caveats to this and I can tell you the limitations of this software, but this is something that you're seeing now being marketed is um looking at different plaque software and perhaps how we report CT should be looking something like this where you have quantitation behind this plaque that we see. And, and this is just one such example and there are others, this is Nominee Day group that have shown that you can do this and it's very accurate when you compare it to IVIS pla burden using it took about six seconds. This is a software that she's developed. It's not commercial but compared uh pla burden uh and stenosis to both IVIS and QC A respectively. Um And so this is kind of an exciting area in the field. But the caveat I will tell you is that this software is highly influenced by how you scan what you measure in the wall of the coronary artery is highly influenced by the brightness in the lumen. And so how and this is, I'm not, I won't go through all the details just because we need to wrap up. But this is a study which proved this where they rescan patients and how much low attenuation plaque you measure it was markedly lower when you scan at low kvs, you get higher brightness in the coronary lumen. And so you, there's, there are still a significant number despite the excitement about plaque and I'm very excited about this. Um I don't know if I would pay $1000 for those numbers, but that's what the companies might have, you believe. But you need to realize the limitations. If you don't segment that coronary wall accurately from epicardial fat, you will get bad numbers. So we need to always be critical of these technologies and, and plaque is one of them. And the last thing I'm just gonna mention as a teaser and where we're going, I'm gonna, and this will be my uh essentially last slide is that there is growing interest in the use of coronary CT for screening purposes. Um And this was one such study called Scapes. They went out in Sweden and took healthy people, middle age, no symptoms, no malignancy. And they did Coronary CTS and they found that about 42% of them had plaque, uh 5% had stenosis greater than 50%. If you look at calcium scoring alone there was missed disease contrast CT and this was true in both men and women. And so it's, you know, this is just a cross sectional study. These people are coming back for follow up, but there's increasing interest and you can particularly imagine if you're a plaque company, there's increasing interest and should we be doing looking for coronary disease? And I think the argument for this is, it's the number one killer in uh in the United States, it's the number one killer worldwide in all industrialized nations, developed countries and women are more likely to die of heart disease and all cancers combined. We don't screen for it. We don't look for it. Even calcium scoring is generally out of pocket pay. And so there's interest if you, if you're better than calcium scoring even slightly, perhaps given the lower radiation and contrast doses that we shouldn't be doing this when people turn 45 or 50 or whatever threshold you might set. And this is in a huge area of debate. And I say that because if you go to the upper East side of Manhattan or many, if you go to Los Angeles, I can tell you there are a lot of people who pay $1500 out of pocket and go get a coroner ct and there's no guidelines to support that. There's not a lot of evidence to support that. But um it is going on. Yeah, right now. And so we see this going on. And this is obviously when you think about volume and access, there are large clinical trials just starting that are testing this hypothesis. A very large clinical trial will probably answer this question in the next 3 to 4 years. Um And they're gonna take patients and do screening coronary CTS and stratify intensity of preventive therapy based on how much disease you have. Think about that staging system, I showed you that's exactly what's being used. And they're randomizing people to advanced preventive therapies, not just statins and aspirin. So these are P CS K nine inhibitors G LP one receptor agonist, SGLT twos, et cetera. And I mentioned that is because a lot of people say, look at what we go through to screen for colon cancer, which is way down the list on. I mean, it's obviously a very important disease, but it's one form of malignancy where you have. I mean, people can die after a colonoscopy rarely, right? You have an anesthesiologist, it's an invasive procedure. Um and with radiation doses and contrast doses being as low as they are, you can imagine the impetus to do this. And we've had this debate. This is Kim Naira. He's a big calcium scoring proponent. And you know, the the advantages of coronary CT and I'm a big fan of calcium scoring, it's underutilized. But you can imagine all of the additional information that I've mentioned already in this talk that you might get from Coronary CT. And so this is an uh an area that's controversial, but it is one debate we will be having. So with that, I'd like to stop and thank you and I encourage your questions. Published June 4, 2024 Created by
BROADCASTMED