 Well, good afternoon everyone. Really the pleasure and privilege is mine. It's an honor to be speaking at this esteemed institution. Obviously the innovation and the transformation of cardiac care that has been let out of this institution is obviously almost unparalleled. So it's a real privilege for me I'm gonna focus on cardiac CT and I'm gonna try to make the case based on the learnings and some of our publications and data from other sites as to why we should be considering cardiac CT as a first-line test for patients who are suspected but not yet confirmed coronary disease. We'll also talk a little bit about physiology and then where I think we're going using non-invasive imaging with CT and non-invasive physiology to not only guide the decision as to whether or not to cath, but actually to map and plan and strategize around interventional procedures. I do have disclosures most relate to the core lab that we run that supports the NIH trials like ischemia, but also industry-funded trials like the partner trial, the low-risk work from Metronic and Edwards. I take no compensation for that. I do provide advice to heart flow around science and education. They make FFRCT. I did academic work for seven years without disclosure, but for the last two years I have provided advice. So we'll start with a clinical case. We do 30 coronary CT angiograms a day. We do it on a dedicated cardiac CT. And what we have here is a clinical case of a 64-year-old male and he has chest pain three times when we talked to him doing three things that always remind my fellows you should never do, right? You should never argue with your wife because it gives you chest pain and you lose the argument anyways. It intermittently wakes him from sleep and that's how I tell my fellows don't sleep. You sleep when you die. You should work when you're alive. You should be productive. But apparently if you sleep you get chest pain. And then if you visit Vancouver, what you'll see, I grew up in the center of Canada. That's flat like Texas. But in Vancouver, we have these hills and he's 64 and his buddy said let's go for a bike ride up. Never mind a hill. It's a mountain. And let's do this. And he said I can do this and of course he does this and he gets chest pain. But when you talk to him, I said to him, you know, did you stop biking? Did you have to take a break? And he said, no, I kept going and the pain went away. So it's a typical chest pain. Nonetheless, he's a he. He's carcassian. He's 64. He's mildly obese and he's dyslipidemic and hypertensive. So he presents to our lab for an evaluation and I would suggest that when he presents the clinical cardiologist, he comes with a specific question. And I think that specific question is most elegantly accurately and most richly answered by cardiac CT in a noninvasive way. I don't think he comes in asking what is my coronary flow reserve. I don't think he comes in asking what is my regional wall motion. I think he wants to know if he does or does not have coronary artery disease and how extensive is it and what does that mean for him and his family and how are you going to treat him. And so when we look at the traditional work up and management of patients with suspected coronary artery disease, according to the stable ischemic heart guidelines here in the United States, you can see that the algorithm is if anything circuitous, you know, it starts with a clinical risk assessment, which is sex, gender and ethnically biased and even regionally biased. You know, the image here would be a great one if I was Dr. Bot who's coming next week in Boston, you know, an older white man, stumbling out of a restaurant in a snowstorm. But if that's the only way people can manifest of those are the only symptoms that are associated with coronary disease, then you would never have coronary disease in Houston. We wouldn't have it in Vancouver. And if they have to be an old white man, that means what women who have different symptoms can't have real coronary disease or people who are ethnically diverse can't manifest with coronary artery disease. So this is a limited, limited assessment. And then the physical exam, which is, of course, a challenge, you know, you have a hard time auscultating an LAD brewery, and then historically you've relied on noninvasive stress tests, which I think are very powerful tools to answer many questions. But in my humble opinion, and according to the data, I don't think they answer the question, is there atherosclerosis for certain they don't answer that question. And I don't think they particularly answer the question as to whether or not there's anatomical epicardial coronary disease. And then ultimately, you're left based on the pre-test likelihood and the results of these noninvasive tests to make a decision around cath. I'm going to highlight now the pre-test risk calculator that was updated in the ESC guidance for chronic coronary syndromes. And I think what this highlights is the historical limitation of diamond and forester risk calculation, which informed the field for 30 years, but in 2019 grossly overestimates the disease burden. If those of you that use diamond and forester would say a 64-year-old white male with atypical chest pain would have a pre-test likelihood of about 60% of having anatomical coronary disease. And yet we scan such patients all the time and we find that their burden of coronary disease is very modest. We look at the ESC guidance you can see right here that if we assume that his chest pain is atypical, his pre-test likelihood is 26%. So it's by no means it's too high to simply pass off and say he doesn't have coronary disease, obviously, but it's not high enough to just send him to the cath lab. In fact, even if he had typical angina, it's only 44% pre-test likelihood of obstructive coronary disease. So I think we need to do better in forming downstream decisions. So should we just simply use stress tests? Well, these are data from the ACC registry data published by Manesh Patel and Pam Douglas almost a decade ago showing that the overwhelming majority of patients that go to the cath lab in this great country aren't found of actionable coronary disease. 62% have no stenosis greater than that that that requires revascularization, at least by anatomy. How about how effective is stress testing at actually determining whether or not the patient has prognostic CAD? Well, we look at data from Dan Berman's lab and we look at nuclear stress testing in particular, in this case SPECT, what you can see here is, I'll use this pointer. What you can see nicely is that 44% of patients with left main angiographically adjudicated left main stenosis were found to have either normal or mildly abnormal nuclear stress tests. So when you have a normal or mildly abnormal, abnormal nuclear stress test, the patient either has no coronary disease or has occult left main disease. This is fundamentally a problem. So why do I think we should start with CT? Well, I think we should start with CT if you feel that the question is, does the patient have or not have coronary artery disease? And you could see that while CT has many limitations, even data from 11 years ago, not surprisingly, as a non-invasive anatomical test, it has better correlation with the invasive gold standard. It's not perfect, but it has very, very high negative predictive value. It safely excludes disease and it has fairly good positive predictive value when it comes to anatomy. So the next question that our patients ask us is, you know, is this a safe test? You know, I heard about all of this concerns that we have with cardiac CT, there's radiation. Can we really be, can we really be using this test as a first line strategy? And the reality is, if we look at the Protection 6 data that was published by Jörg Hausleiter in European Heart Journal and presented at ESC in 2018, what you could see is that the dose from cardiac CT is 80% lower than it was a decade ago. And in fact, the estimated effective radiation dose of cardiac CT in our lab is about 30% that of a nuclear stress test in our hospital. That's not pet, but that's nuclear spec. So by no means am I suggesting we should be doing CT in patients without symptoms or patients without a reasonable pretest likelihood. But if you have a patient with possible or probable angina pectoris or symptoms suggestive of CAD related chest symptoms, I think CT is not only a highly accurate study, but it's a safe study. The next question that the patients ask is, you know, it's great that you've given me an accurate diagnosis, you've done it in a safe fashion, but what does this mean? And a decade ago, my clinical colleagues would come to me and say, you know, Jonathan, your test is really accurate. It's really elegant. It shows me all this plaque. But at the end of the day, I don't know what it means for my patients. I don't know how we should titrate medical therapy. I don't know if it's prognostic. I rely on the treadmill. The treadmill and exercise capacity informs my understanding of patient specific risk. Well, that was a decade ago. We could talk for hours about the overwhelming data that is emanated from the field of cardiac CT and prognosis, but we won't. We're going to highlight one slide. This is from the promise trial randomization of 10,000 Americans between CT on the left, nuclear stress testing on the right. This is the prognostic data that comes to us from Udo Hoffman from MGH. On the left, you see the anatomical testing, and it highlights two of the really key elements of prognosis that are informed by CT over and above nuclear stress testing. We look at the two lowest curves. The x-axis is time. The y is events. The lowest curve is those bereft of atherosclerosis by CT. They have no plaque. You could see that the absolute event rate is less than half of those patients with a normal nuclear stress test. And that's even just at three years. When we map those patients out to five and seven years, you see an even more profound difference. So a negative CT has a very, very good warranty length and very good prognosis. But in addition, let's look at the capillary curve immediately above it. What we see here is those patients with atherosclerosis, but no stenosis greater than 50%. What you can appreciate is the relative risk of those patients is threefold. So if you have plaque, but no stenosis greater than 50%, your mace event is threefold higher than those without atherosclerosis. Let's turn our attention to the Kaplan-Meier curve on the right. Modally abnormal and normal nuclear stress test. What can you tell the patient? I don't know which of you is going to have an event. I don't know which of you has atherosclerosis. You're both at a modestly, you both have a not a very high event rate, but significantly higher than a negative CT, but you don't know who to treat and you don't know who not to treat. Because all you know is that they either have a normal or mildly abnormal perfusion study. Similarly, let's look at those with a severely abnormal CT. The relative risk tenfold. Why? Because the absolute risk of a negative CT is very low and the absolute risk of a severely abnormal CT is quite high. Because even if I over read a stenosis and we send the patient to the cath lab, you're not going to find, if I say there's triple vessel obstructive disease, you're not going to find someone without any atherosclerosis. Maybe I over called one vessel, maybe I over called two vessels, but the patient very likely has significant coronary artery disease anatomically. On the other hand, when we look at functional testing, I'll highlight to you some of the data from our core lab work in the ischemia trial. Ischemia trial randomized 6,200 patients between revascularization based therapy versus medicine in the setting of a severely abnormal stress test. We were the CT core lab. Our role was that with other sites with Maboudov and Jim Min. Our role was the NIH said, before you randomize patients with severely abnormal stress tests, we want to make sure they don't have left main disease. So we would do the CTs in a blinded fashion, make sure there was no left main disease, but the interventionalists were wise enough to say, you know, we're tired of getting these nuclear stress tests and finding no actionable coronary disease. How am I going to beat medicine if you send me someone who has no disease to treat? So we also had to exclude the absence of nonobstructed or the presence of nonobstructive disease. And this has been published in the American heart journal, 18% of the CTs we were sent in the setting of severely abnormal stress tests core lab adjudicated. So these are the most abnormal stress tests did not have a single stenosis in their coronary arteries greater than 50%. So when you have a severely abnormal stress test, you could tell your patient you have either severely abnormal coronary arteries or you have no stenosis greater than 50%. So that is not the diagnostic certainty I want to practice with. And that's why I think an upfront understanding of anatomy is so powerful. We're not going to talk a lot about plaque, but clearly there's a wealth of data now coming out about the role of CT, not only for plaque quantification, but also plaque characterization without the elegance and the spatial resolution of IVAS, obviously, but the opportunity to noninvasively evaluate the entire coronary tree, not just the proximal segments, allowing us to understand are there adverse plaque features like positive remodeling or low density plaque is there napkin ring sign, how severe is the plaque burden, is there plaque in the left main, which we all know our findings of that help us identify the host at risk of downstream infarction. So it allows us to better understand not only the presence or absence of disease, but the pattern and extent of disease that helps inform prognosis. The final question before we turn our attention to physiology is this one, you know, what should I do doc? It's great you told me I have this disease, but should I just go get my affairs in order? Or, you know, what are you going to do differently because of this disease that you've helped me diagnose? So what we've seen now from multiple meta-analyses and at least one randomized trial is that coronary CT helps inform medical management in a fashion that improves clinical outcomes. This is the five-year Scott Hart data published in the New England Journal in August of last year showing a randomization of 4,000 scots between traditional care and CT, and there was an absolute reduction of 40% my cardiovascular death or MI in those patients that were treated on the basis of CT as compared to traditional care. Now, many people questioned the results and I'm sorry that the reference here is cut off, but actually David knew me publish a paper last week. Phil Adams is the lead author and I helped them with this project, they were kind enough to let me. What they really wanted to do was dive deeply into what is the mechanism of risk reduction in the stratification or in the treatment of patients following CT. And what you can see here is in blue is CT, in red is traditional care, on the left is statin, on the right is platelet, anti-platelet therapy, and you can see amongst those patients that were treated and were randomized to the CT strategy, the use of medical therapy was significantly higher than compared to traditional care. But this belies the real power of CT because I think the real power of CT is not just that more people are on medicine, it's that I think the right patients are on medicine, that instead of treating on the basis of the risk of the disease, you actually titrate medical therapy based on the presence or absence of disease. As I always say, I would rather have every risk factor and no atherosclerosis than no risk factors and a lot of atherosclerosis. So what you can see here on the top is for anti-platelet, the bottom statin, and you have these three curves or graphs and what you can see mapped out is the assigned risk score, which is a population-based risk that also looks at socioeconomics and environmental exposures, it's a Scottish risk score, and the vertical is the y-axis is the percentage on said medical therapy. The gray bars are those without atherosclerosis, the orange are those with stenosis greater than 50%, and purple is presence of atherosclerosis but no stenosis. And what you can see here is that in the absence of atherosclerosis, they down-traited medical therapy even in those that were deemed to be high risk. Now, I wouldn't suggest we do that in a 40-year-old, but if you have a 65-year-old man and he has no atherosclerosis and he has a borderline LDL, I don't know what you're treating. I really don't, you know, certainly 70, 70-year-old woman. On the other hand, if they have non-obstructive atherosclerosis, even if they have a low risk, their population-based risk score is low, it's only 5% 10-year-old. Why do you treat it? Well, you're treating the disease, the disease that causes myocardial infarction. People don't want to be treated on the basis of the risk of the disease. I think they'd rather be treated on the presence of the disease, and I think that results in improvement in downstream clinical outcomes. The other thing we saw in Scott Hart is that more patients went to the cath lab after CT initially, but over five years the rate of cath was the same, but the yield of the cath was better in the CT arm than the traditional cararm. The traditional cararm, you're more likely to undergo a cath when you represent with ongoing chest pain and you go to your cardiologist and say, doc, do I have coronary disease? I'd pain again. And you know what the cardiologist says? I don't know. So they get the answer. They get the answer with the cath lab and they find the patient doesn't have coronary disease. How about revascularization? The revascular rate in the CT arm in blue was higher within one year. Out to five years the revascular rate actually normalized and there were more revascularizations that occurred in the stress testing strategy, but most of those were due to urgent unstable presentation or myocardial infarction. So David would posit that if you are going to revascularize someone who has new onset or worsening symptoms, the time to do it is early. And there's evidence to suggest that may be of greater benefit rather than letting the patient sort of sort itself out and then represent at a later time. The last slide from this important paper is here and that's looking at non-anginal and possible anginal chest pain. Their two-year data suggested that the greatest benefit of the CT-guided approach was in those with possible or probable angina pectoris. At five years you could see that the risk reduction here in blue is similar whether the patients had possible or non-anginal chest pain. So what we're learning over time is not that we should test everyone by no means am I suggesting that, but our historical reliance on angina typicality to really understand risk is I think limited. I think it's limited again by gender presentation, sex, and by ethnicity. So I think if you see CT-based atherosclerosis the patient's at a heightened risk. So we've seen a proposal that maybe it's time for a change in the guidelines that CT is more accurate it provides prognostic information and it enables better titration of medical therapy in a fashion that improves outcomes and as a result the European guidelines at least came out two months ago as you know in Paris and suggested that CT should be considered as a class one test for patients with suspected coronary disease but not yet confirmed CAD. Regardless of baseline almost regardless of baseline risk with stress testing really reserved for those that have the highest pretest likelihood but when you look at their pretest likelihood calculator that they propose us using that's just a very small subset of patients. So whether the US guidelines will shift in that direction at ACC maybe in the new year I don't know but it certainly makes sense to me. So the last thing I'll turn my attention to is this last question and that is whether or not the patient should be revascularized and we know that in the absence of prognostic CAD even if I'm so accurate on my CT interpretation and I interpret a 71 per cent stenosis and I send it to my cath colleagues in Vancouver they need to confirm or document physiological significance either invasively or with some sort of non-invasive test and we know that CT historically hasn't been able to provide that because the same reasons and geography can't we know that the totality of randomized data to date doesn't support stenting strictly on the base of anatomy we know that from courage we know that from Barry 2D. What we also know is that at least in Europe and I think increasingly in the US that the class 1 there's class 1 guidance for the use of invasive physiology to adjudicate lesion specific ischemia or lesion specific really physiology to better determine which lesion is likely to derive benefit from revascularization these are the five year fame to data published in the New England Journal highlighting that if you that you could safely defer physiologically non-significant disease but once you document physiologically significant disease that there's a benefit in revascularization over and above medical therapy when it comes to urgent unplanned revas and even when it comes to potential downstream MI. So the question that we're faced with is can we actually provide both anatomy and physiology from arresting CT so about a decade ago some science guys from Stanford came to me and Jim Min at the SCCT and said they wanted us to do some research on this concept of integrating computational fluid dynamics onto CT to extract physiology pressure flow shear stress from arresting coronary CT and I thought this is crazy how can we possibly do it without administering adenosine how can we do it without shifting our protocol but it turns out that there's a wealth of science behind this that goes back to Newton's days and fluid dynamics and the whole concept of using computational fluid dynamics to calculate pressure and resistance and flow is obviously applied you know what is a tranquility base what's the base in Houston where the shuttles go off from but you know they use that stuff too NASA uses this stuff and nonetheless we never had an anatomical model for coronary arteries to actually solve these equations but with the advent of 64 slice CT we now have patient specific anatomy branch points aortic root aortic root geometry that helps inform the flow conditions not only in the aortic root but of course in the coronary arteries and we also know that humans like all mammalian species once you have the branching of the coronary tree down to 0.5 millimeters you can actually even extract the cascade of the microcirculation we know that athletes for example like Deshaun Watson who has I've never seen his heart but one would imagine he has thicker heart muscle than I do so he's gonna have more resting flow I would imagine he has larger coronary arteries than I do so he'll have lower micro circulatory resistance he'll have a better nitrate response because he exercises and we know that you know according to Murray's Laws that the coronary arteries adapt to accommodate homeostatic shear stress so there's rich physiological information in the CT and then adenosine can be modeled based on the expected response to adenosine from Bob Wilson's lab at the University of Minnesota in 1990 now it's important adenosine is computed so it's a conservative calculation because the pressure is going sorry hyperemia is is always achievable when it's computed it's not always achievable invasively so the computational FFRCT may trend lower than the invasive physiology because there's some patients with diabetes or the like where you may not achieve hyperemic response so what data do we have to suggest that FFRCT is accurate for the discrimination of lesion specific physiology well we have data that was published by the group from Amsterdam from the pacific trial where they used oxygen labeled water pet SPECT and cta on 208 patients and did three vessel non-biased FFR what they showed us in their primary analysis is that pet was better than SPECT which was better than cta but what was also interesting in their primary paper published in jama cardiology was that our historical approach of using ct and then SPECT in Canada didn't improve accuracy you just traded the sensitivity of ct for the specificity of SPECT but that that hybrid approach didn't augment accuracy in this sub-study what they showed is that FFRCT was superior to all other tests and importantly those other tests were performed in core labs the ct's were read by myself and jim min in new york the pet was read by johannie canute in finland and the specs were read by richard underwood in london so when you have an analyzable data set of which 88% were analyzable in the trial the FFRCT was superior to these other non-embasable tests now image quality matters it matters i think broadly for the use of ct and it certainly matters for the use of FFRCT here you could see an image from our lab excellent image quality but also an image from our lab historical thankfully terrible image quality our accuracy for coronary ct compared to invasive angiography is over 90% when we have images like this when we have images like this i hope i'm traveling so one of my colleagues has to read it because i have no there's nothing you can do with that sort of image it's just it's not diagnostic and the same thing holds true for FFRCT when we looked at the rejection in over 13 000 cases one of my former fellows looked at that and you could see a very nice relationship between heart rate and rejection heart rate and image quality heart rate and diagnostic performance so as the heart rate goes up the likelihood of rejection goes up your accuracy anatomically goes down so there's no cutting corners in my humble opinion to me i strongly believe in adhering to best practice for ct acquisition and that is a heart rate below 60 and appropriate utilization of nitroglycerin we use two sprays to 0.8 milligrams of nitrates we wait five minutes and we try to ensure maximal coronary vasodilation it makes our job easier when we read the ct's it makes our job easier when we're looking at the overall vascular health of the patient and you know i go to some sites and they tell me we don't need to do all of these things when we do ct but they're practicing in this historical approach where they're doing a very modest number of ct and they're doing it on you know 35 year old women women with very atypical symptoms well of course you don't need optimal image quality if they don't have atherosclerosis but if we're going to do a ct on a 71 year old diabetic male you really want to have perfect image quality so we looked at i don't know why this is cut off but here you could see we did an in-person comparison at 0.8 and 0.4 milligrams and you could see here that at 0.8 milligrams of nitro significant improvement in the coronary vasodilation the coronary volume as we extracted it was significantly higher at 0.8 than 0.4 so assuming the patient's blood pressure can tolerate it we strongly believe in using two sprays of nitroglycerin the last slide on image quality that i would point out is this holds true not only for physiology people say well this is heart flow's problem or in our lab where we're doing you know advanced analytics around atherosclerosis say well that's your problem you need better image quality well the reality is even if you want to simply do ct to determine whether or not they have anatomical coronary disease the image quality is essential that the diagnostic performance is better consistently when the heart rate is below 62 so when i go to some sites and they say we image at 75 it's no problem i asked them for their data they inevitably don't have any data to show me that and i'm confident that they're simply under serving their population by by doing what they think is best by getting them through the scanner quicker but with suboptimal image quality so how do we interpret the data so when we started doing ffrct clinically i realized this is much more complicated than when i was involved in the trials in the trials it was easy there was the ffr wire there we had to co-localize the physiology on the ffrct and it was right or it was wrong now we don't have the ffr we have a non-invasive test that's now providing us three vessel physiology computed and it allows us to answer a number of questions the first question is should this patient go to the cath lab with the hopes of revascularization in a fashion that will improve angina over and above medicine or potentially reduce mi that decision should be made by the pressure distal to the focal lesion and whether or not that value is below 0.75 in single vessel disease or 0.8 maybe in double vessel disease and also the translational gradient so is there significant pressure loss across the lesion in this case it drops from 0.9 to 0.83 the translational gradient is modest and it's above 0.83 i would say the patient has an anatomical stenosis but it's not causing it's not physiologically significant so then people will say to me well but jonathan does that mean that's a false positive i'll say no that's not a false positive this is important information that 0.7 is shouldn't be used to help guide the decision around cath but it is telling us something very much about the patient's vascular health that patient's vascular health is poor they have diffuse athero they have a long wrap around lad and they're dissipating pressure in their lad this patient doesn't need to go to the cath lab this patient needs medical therapy here's the case from our site big dominant right high flow conditions a lot of flow more resistance to flow as in this disease vessel and particularly as you get distally it drops below 0.8 but no focal treatable lesion this patient's treated medically is now out over two years in fact post ffrct this is a nice case of opportunities to learn when you come to big centers like this here's an example of a case where they called me a friend called me from california said what do you think of this case no right coronary artery essentially a dominant circ and an lad without any meaningful diagonal system a long lad diffusely diseased a mild stenosis here a mild stenosis there and a wrap around lad and he said well jonathan we'd send the patient of the cath lab and they didn't find a focal treatable lesion and i would send to my friend i'm not surprised there is no focal high-grade anatomical lesion there's no focal delta ffrct across any lesion but did they wire the distal vessel and he said no and i said well that i think and i'm not saying they should have but i don't think this is a false positive you have a big lad with a lot of flow you have no diagonals to off load that flow in the distal lad there's still a lot of flow and increasing resistance as the vessel tapers so it's not surprising in a diffusely atherosclerotic wrap around lad without diagonals the physiology may be abnormal in the distal territory in the absence of focal lesion so again this is how we read it if you're deciding whether or not to go to the cath lab you look at the the value distal to the stenosis you look at the translational gradient if you're looking at overall vascular health we integrate it in the extent and severity of atherosclerosis to better inform how sick the patient is so what is the data around clinical integration we have data from the advanced registry is which is about 5 000 patients that were enrolled across 38 centers and what we saw is that ffrct modified decision-making in a large number of patients now that modification often was uh they needed an additional test and they no longer needed an additional test so that's a soft modification but there were many cases that were modified from needing a cath to going for medicine from going to medicine to needing a cath so it re stratifies your decision-making in a significant way when it comes to prognosis i think the real message from non-invasive physiology is the same as invasive physiology what have we really learned from fame i think we've really learned that it's safe to defer anatomically significant disease when it's not physiologically significant that those patients get good treatment benefit from medicine alone when it's positive i think we need to learn a lot more around the integration of plaque and the depth of positivity around who needs to be revascularized but these kaplan meyer curves i think highlight that here you can see even in the setting of a stenosis greater than 50 percent in the absence of physiological significance the likelihood of needing to be revascularized at one year was incredibly low seven percent there were three m is in the 1600 subjects with negative physiology one had a prior stent and the culprit was in the stented vessel one had aneurysmal coronary disease with thrombus so i don't know whether i would value you know uh a physiology in such a case and the third had an ffrct of 0.82 in the lad so very low heart events in the absence of physiologically signatory and disease now people say to me do you send just 50 to 70% lesions do you send 70 to 90 and what i say usually is that i practice medicine i don't send all 50% lesions i don't send all 80% lesions i send some of both and the reasons i send some it really depends on anatomical location dominance of the vessel length of the lesion pattern of atherosclerosis but i would highlight just some data that manesh patel just put together looking at these 70 to 90 percent lesions and what was interesting in advance one quarter of the stenosis sent into the advanced registry where what we would call anatomically severe by ct and interestingly one out of four were physiologically negative suggesting that they don't need to go to the cath lab but even more importantly i think is that a large number 60% of the total lesions and 75% of the remaining were strongly positive physiologically they had ffrcts that were not in the gray zone but were below 0.71 suggesting that we can no longer send a patient now just saying hey there's a stenosis go figure it out but maybe we can actually defer cath even in those with anatomical disease and then send the ones that are going to the cath lab the large majority with clear anatomy and physiology to provide potentially a roadmap the way we do taver to help guide revascularization so just building on that data from down the road from neils johnson we know that invasive physiology is in binary in clinical practice that an ffr of 0.6 is a lot worse than 0.79 which is not much worse than 0.81 well we've seen data from manesh patel again in jack imaging that as the ffrct goes down the event rate goes up and similarly data from manesh from abdul ibnid and brian co looking at the nxt trial out to five years the lower the lower the ffrct the higher the event rate without a death or am i in any of the subjects with the negative ffrct through five years so the warranty is growing the data is growing the prognostic value of ffrct again i think is really in a safety of deferral and increasing risk with lower and lower ffrct values now you may rightfully be thinking everything i've shown you right now is observational we don't have randomized data i think the randomized data is essential there's a trial called forecast that's done enrolling in the uk that'll be presented probably next year 1400 patients and in the us and globally there's a trial called precise which is a randomization of standard of care versus a precision based arm using ct ffrct but i think what's really elegant in this trial is they'll be using jim udelson's promise risk score to identify those patients at lowest risk and those patients will be randomized well they'll just be randomly assigned and the lowest risk patients in the precision arm will have no test and in the standard of care they will have whatever tests they would normally do so i think this will be the first trial that will allow us to answer not only whether or not ct ffrct and those 20 to 100 patients is better than traditional care but also can we safely defer testing and patients deem lowest risk using the promise risk score so i think this is a really important trial what are the other areas for continued investigation i'll just close on one area of real interest of mine is microvascular dysfunction we know that there are many women who are deemed to have microvascular disease we've seen over time that these patients may often not have focal anatomical stenosis but have athro what we did was we matched women with pain and without pain for the extent and severity of atherosclerosis and risk factors and what we found is that this very concept of coronary volume to mass so when we extract their coronary volume and we normalize it to mass what we saw is that women with chest pain had significantly lower coronary volume normalized to their myocardial mass this concept that their resting flow is so tenuous that even modest atherosclerosis will allow them to tip towards physiologically significant disease at a lower anatomical threshold and when we talk with the people like noel berry mers from cedars around this this concept that the microvascular in these women may be abnormal but i think we've ignored often epicardial coronary disease in women because of the historical reliance on on on focal diameter reduction rather than the totality of the epicardial coronary system also out of the ed this is great work from kavitha chinayan from bow mount showing how they've started to integrate ffr ct even beyond their ct pathway to help better identify a strategize treatment decision making in the moderate to severe coronary artery disease and the last population i'll focus on before we talk about treatment planning is the peripheral vascular work this is work from latvia that i think is really fascinating i always tease in my hospital if you twinge with chest pain you're going to get a test i i don't even i don't want to even move my hand towards my chest of course because chest pain gets testing but we have patients that come in with peripheral vascular disease that have almost rest ischemia and they have a nuclear stress test before their surgery and they get through the surgery and we think we've done a great job and they die of a heart attack at a year and they're not even properly medically treated so i'm not saying we know exactly how to change our therapy but this great work from from latvia really highlights that ct in in advance of the peripheral vascular interventions may help us identify not just atherosclerosis in the coronary arteries but prognostic cad physiologic cad and better understand patient specific risk to help inform not only medical management but perhaps you know downstream treatment and surgical intervention so i'm going to close with this this is something i'm particularly excited about for those of you that know the kind of work that we've done i had the privilege of joining st paul's about 11 years ago and at that time john web had been doing taver for four years at Vancouver the first trans arterial taver and when i started there i said to him john you know i think we can do ct and we can help inform we can help inform your uh your procedures and he said well jonathan we're doing this echo it's way better when the early early days he just put a small valve in a woman and a big valve in a man so he was thrilled he had any imaging but he was relying on 2d echo and i said it's a non non-circular structure you need three-dimensional imaging and now of course he would never do a taver without a ct he would never i mean almost never and yet he goes to the cath lab all the time based on this traditional approach right you know is he is the patient an old white man and he may have coronary disease but if it's a woman of a different ethnicity and doesn't have a focal stenosis then she doesn't have coronary disease it's very limited relying on the stress test and then relying on us to send the patient there and then sometimes he finds absolutely no coronary disease and he mutters under his breath how he just wasted his time and radiated himself and then sometimes he does the cath and he finds the most sinister three-vessel atherosclerosis and in a rationed capitated healthcare system in Canada he's got to make an ad hoc decision based on the lumina gram about how to revascularize this patient which in my opinion is way harder certainly you know then for him than doing a taver so i think the days of going blindly to the cath lab are they should be over or i think that the time is to start thinking about different ways of proceeding why well we know already 60 percent of patients going to the cath lab don't have actionable coronary disease so we need to do better for those patients but how about the 40 that go that do in the syntax trial 30 to 50 percent of patients were not completely revascularized so of the 40 remaining patients 30 to 50 percent are not completely treated now some of those patients may not you may not be able to treat them but you'd like to have a plan and the reason i think you need to completely revascularize we see consistently this is data from professor akasaka and kwon koo from japan and korea and what we could see here is that if you leave disease untreated they're much more likely to have either urgent revascularization but also at least in this meta analysis 36 percent more likely to have downstream mi if they're completely if they're not completely treated so we need to do better for our intervention list we need to provide the more information so that our interventionist can make more thoughtful decisions i think in advance in you know engaging in the heart team and having these discussions so what data do i have not a lot but i think interesting data this is from syroes and carlos caledos a young interventionist so thoughtful what they did was they looked at whether or not non-invasive ct and non-invasive physiology could help plan by providing a non-invasive syntax score and how it would correlate with the invasive syntax score anatomically or physiologically what they showed us is that my colleagues in vancouver were right that ct alone wasn't enough to help them really plan their pc i because ct would enrich the population in the cath lab they're more likely to find disease but when you had complex disease ct generally on a per-segment basis overestimates the anatomy compared to cath for sure and the anatomical syntax was over was higher than the invasive but when we look at the functional syntax you can see that the functional non-invasive syntax with ct and ffrct correlated very well with the invasive gold standard of three-vessel physiology and invasive angiography the advantage of course being is that you don't have to do the three-vessel instrumentation and you could have that data before in the in in your waiting area planning the cath or even the night before really mapping out what do you think you're going to see tomorrow how physiologically significant what do i want to stent what do i want to interrogate what do i want to leave alone and so here this is just the breakdown you can see a real shift invasively from anatomy to physiology we know that obviously from fame and from eric van belz work in france and all others but here looking at non-invasively the much tighter correlation between the functional invasive and functional non-invasive syntax so i think we can use you know just just to show you as an example here's how we are using more and more ct never mind the physiology how about the fluoroscopic angles i said to john webb i give you fluoroscopic angles for taver every day but yet you just go deal with the coronaries on your own but look at this case where you can use the ct data set to really elongate the lesion and really provide you a fluoroscopic assessment to go in with the plan in this case i think it's quite telling this is a case from carlos collett who lent me here's an example angiographically and i was naive to this as an imager i said my interventional colleagues know how to profile the lesion but look at this lesion you would think it was 36 millimeters long but using the the fluoroscopic angulation proposed by the ct you can see that that was grossly foreshortened so integrating the fluoroscopy integrating the atherosclerosis to better understand the length of the lesion alone i think can really help our interventional colleagues so that she can plan the procedure in advance and here's another one severe stenosis lad severely physiologically significant we said there was a moderate to severe right physiologically insignificant so instead of saying two vessel disease go figure it out good luck here's two vessel disease stem the lad and leave the right alone in a fashion that's more time efficient and i think patient and improving the safety profile by avoiding the need to invasively physiologically interrogate this right which is convincingly negative and here's the physiology so the last thing is this concept of virtual stenting this concept that one can idealize the lumen and then allow our interventional colleagues to actually map their intervention to better understand whether or not that intervention can be effective so you can imagine and this is work that was done by Eric van bell in a study called bowie that will be coming out this concept that if you take the physiological model you can actually idealize the lumen now it doesn't integrate plaque shift or dissection but you could idealize the lumen and recalculate the physiology to understand whether or not that stent is going to improve the physiology or not this is a case from monash as i showed to one of our colleagues last night two stenosis in this case physiologically significant ffr of 0.77 non-invasive ffr of 0.78 here they decided to stent the proximal lesion to unmask the sorry stent the mid lesion here the mid lesion has been virtually stented and you can see that by virtually stenting the mid lesion you don't move the physiology almost at all here the stent the proximal in the mid and you resolve the physiological significance of the pressure loss here you virtually stent the proximal in the mid and you correct the physiology so you say what where's the value well imagine if i were to give my interventional colleague the fluoroscopic angle the description of the plaque the stenosis and then provide her with the physiology and allow her to map out on the physiological model her intended strategy to better define what her plan is to physiologically treat the patient and in that way this is one one could imagine our our interventional colleagues doing that sort of mapping out the physiology mapping out the stent length to look at a stent length in this case of 38 millimeters and then open up the lumen and determine whether or not that stent would actually resolve the physiology and to that end if we look back at this case imagine if we gave this data in advance of the procedure and allowed our interventional colleague to simply virtually stent the proximal lesion leave the mid lesion and realize that she would achieve the same degree of pressure recovery as she would if she stented the proximal in mid so this is hypothesis generating clearly there's some accuracy study going on we need more data we need to show that this hypothesis actually improves outcomes but you know while i think that in greg stone is going to do that trial while i think randomized trials are needed i think like taver when you have a non-invasive tool it doesn't need to be superior to uh to other historical approaches to be valuable you can show value by way of reducing radiation dose reducing contrast reducing wire usage and improving physiological significance and i think those are the challenges that we all face so i'll close there but i i i look forward to your questions i think there's so many opportunities for sites like this and having conversations with some of your colleagues last night around plaque and around physiology and around ct as a first line test and help and to help guide revascularization strategies so thank you again it was a real privilege