Karol Watson, MD joins Georgia Heart Institute as a Cardiovascular Grand Rounds speaker during the CME lectures for 2022. Dr. Watson is a professor of medicine and cardiology at the David Geffen School of Medicine at UCLA. Dr. Watson discusses coronary calcification as it applies to observation in research, unraveling molecular mechanisms in vascular calcification, and using CAC as a research and clinical tool. To receive CME credit for this presentation, please visit the survey link. [https://www.surveymonkey.com/r/2GGNYSY]
Thank you. Yeah. Okay. Yeah. Right. I will. Yeah. How interesting. I'm gonna get to see em. You can come up here. Good morning everybody. Thank you so much for being here. We really do appreciate you. Um Getting here in the morning and being here to welcome our guest speaker. It's great to see you all. I know we haven't had grand rounds in a couple of months, so we're glad to be back. Um So just a quick announcement about CMI before we get started, this program is provided by Georgia Heart Institute with support from our industry partners. The planners have disclosed no relevant financial relationships with commercial interests through grants and research support. The presenter has a relationship with Amarin Amgen and Boehringer Ingelheim. These relationships do not influence today's presentation to claim cmi credits today. Please answer the survey evaluation. If you are viewing online, the link will be posted into the chat. Those attending in person will receive the link or QR code at the end of this session. If you have a question for the presenter, please hold until the Q and A segment. Online viewers may type questions into the chat and then we'll read them at the end. And now I'd like to ask Dr Burkle to introduce our guest speaker, Dr Watson. Thank you, Suzanne. Good morning. It's a great pleasure to introduce my old friend Dr Carol Watson. Uh Dr Watson is professor of medicine and cardiology at the Geffen uh School of Medicine at U C L A. She holds the Jon Matsumura Term chair in medicine and was honored to be named the cardiologist of the year by the California Chapter of the American College of Cardiology. In 2017, Dr Watson received her undergrad degree from Stanford University and graduated magna cum laude from Harvard Medical School. She also holds a phd in physiology from U C L A. She completed her internal medicine residency and her cardio oncology fellowship program at U C L A and was also chief fellow at the same institution. Dr Watson is currently the director of the Barbra Streisand Women's Health women's Heart health program at U C L A, Co director of the preventive cardiology program at U C L A, something that we both hold close to our heart and director of this U C L A fellowship program in cardiovascular diseases as well. So Dr Watson is also a principal investigator of multiple NIH trials and serves in multiple NIH steering committee. So it's an honor for me to present Dr Watson who will be speaking on coronary calcification, basic mechanisms, research applications and clinical utility. Carol Thank you everyone. I thank you doctor work well for this invitation. And um he and I go battle, we go back a ways and we were really to the grateful for your friendship and your advocacy. Let me tell you about the story of the last 20 years of my life. Basically from when I was a first year cardiology fellow like you and um to now because I have been involved in coronary calcification in some way, shape or form for all of that time. So these are my disclosures and nothing that I say today will be affected by this Start talking about how I even got into this and observation. And it led to a question and that question led to the next 20 years of my life. Um first, as a basic scientist unraveling the molecular mechanisms of vascular calcification, then I really decided that I needed to look more broadly of populations to figure out how vascular calcification and coronary calcification can be used as a research tool and as a clinical tool. And then I want to finish up with some future questions so well. First, there was this observation and that observation came from my mentor back then. And still to this day, Dr Linda Diemer, Linda had always said, you know, she was a biomechanical engineer who then became a physiologist and she looked at a lot of pathology specimens because she was interested in the mechanical properties. And she said you know, one of the things that really influences the mechanical properties of the arthroscopic plaque is calcium. You have this big lipid fluffy core and right next to it, you have this big hard rock, it CALC that's a really unstable interface, like that's going to influence a lot of the, the plaque um stability. So she said, you know, I'm really interested like what is this calcium doing here? And she said, you know, looked through the research and for centuries, people had seen that and they said, oh, it's just a, a passive process of aging. No, you know, no one. It's not important, don't worry about it. But she said, you know what? I think there's something there. And when she looked at more and more specimens, she actually noted some real stark similarities between vascular calcification and osteo genic tissues like bone and cartilage. So she said, you know, maybe there's really something more here. It's not just passive, maybe there's something active going on. So these are some of the slides that she showed me initially. You, these are actually specimens from pathology sections of atherosclerotic plaques. You can see what looks exactly like wavy amorphous bone there, things cells that look like cartilage, things that look like fat. She could see all different cell types there and things that look like bone marrow. So she said, you know, it's hard to imagine that you just passively laid on calcium in these tissues form So I think there's something there. And then she said I could even see things that look like multi nuclear hated osteoclasts, which suggests that there's active remodeling of the calcification in atherosclerotic plaques. So it's not just that this is a passive deposition of calcium. I think there's really something active there. So the next step and that's where I came in was trying to unravel the molecular mechanisms. So I was a first year cardiology fellow is part of the star program at U C L A. That's a program where you do your cardiology fellowship and get a phd at the same time. So I asked Linda to be my mentor and decided to carry out my phd under her and I was going to look at the molecular and cellular mechanisms of vascular cast. So Linda Diemer was my primary mentor. I also worked with the postdoc Christina, but at the time, who was just a post, not just like she was a postdoc, but then she subsequently also went on to do her cardiology fellowship. And Yen Tinted, who was a phd scientist still in the lab to this day. So my first job was she said, find me a cell based model of vascular calcification. So I was like, okay, not knowing anything about what I was doing. We started by looking at um bovine aortic smooth muscle cells and human aortic smooth muscle cells. We are heart transplant program. We had a lot of access to human aortic smooth muscle cells. And we got some bovine areas, smooth muscle cells. So we just cultured them to see if any of these cells were gonna calcify. And it turns out that many people had seen for many years that whenever you have these cultures over time, they start to form these nodules. And again, a lot of people said you can tell when yourselves are getting old, they start to form these nodules and you just throw them away, but we didn't throw them away. We actually looked at them and we stained these nodules and they were staining brightly for calcification. So like that's interesting. And one of the things you could see was that if you start with a very, you know, stable smooth muscle cell culture undifferentiated over there on the left, you can see when they were about to form nodules. So these are staying with CASA looking at calcium. So all cells have calcium. So you see a very diffuse, you know, staining of calcium, then when the cells start to condense and then they go on to form nodule formation, you can see where the calcification and the bank a sustaining gets more intense and then these calcified nodules form. So one of the first things we did was said, okay, is this anything else turning on at the same time that this calcification is going on and we really, really interested in classic bone proteins. So the first thing we looked at was alkaline phosphate, classic bone protein. It's, you know, it is seen in other tissues as well, but it's always seen in classifying tissues. So you could also see where this calcification and calcified nodules were about to form by the turning on of alkaline phosphates in the pre nodules stain and the nodule stain. And then when you get really mature nodules with a lot of calcium, you see a lot of alkaline phosphate staining, but still we weren't sure. Is this just a morpheus calcium or is this more like bone or bone? Is, you know, the calcium is not just calcium, it's hydroxy appetite, it's a very specific type of bone. So we wanted to see if the calcification in these nodules had any appearance of hydroxy appetite and that's what we found. So again, this is just looking at a section through a nozzle and stained for Van Casa. For calcium, you can see the calcium there, we could take an X ray of the nodules and you see a little skeleton. And then we did electron microprobe analysis to see exactly the mineral constituents of the nodules. And they actually had calcium and phosphorus for us in the exact same molecular ratios as you would see in hydroxy appetite, we were getting more and more convinced. So now we wanted to see if, as we saw these cells differentiate from undifferentiated to pre Nigel to nodule forming cells. If we could see bone specific proteins turn on. So the next thing we did was look for collagen types. Now, if you look at vascular smooth muscle cells, they have collagen type one, that's their predominant collagen type. Um So I'm sorry, collagen uh type four. So what we saw that if you have calcified cells, they have predominantly collagen type one. And that's the type of collagen that you predominantly see in calcified tissues like bone and cartilage. Collagen type four is the type of collagen that you see in basement membranes of all sorts of cell types. So it was more commonly seen in smooth muscle cells. And what we could see is if you had a rapidly calcify ng um smooth muscle cell culture, You would see a lot of collagen type one. If you had slowly calcified cells, you didn't see any or non calcified cells, which were presumably just with muscle cells, you didn't see any. Also, we also looked at fiber connected also pretty specific for bone and calcified tissues like bone and cartilage. You would see a lot of fiber and pectin in the rapidly classifying cells a little bit in the solid classifying failed cells and very little in the non classifying cells. Then if you look at collagen type of for the type that's seen primarily in smooth muscle cells, you really wouldn't see much in the rapidly calcified cells. You would see a little bit in the slowly classifying sales and you would see a lot in the non classifying cells. So we could see this turning on and off of bone specific collagen types as these cells were classifying. Then we started looking at more and more specific bone type proteins, osteopenia, tin, that's a real classic protein that's seen in really only calcified tissues. It's kind of never seen in smooth muscle cells. And so we wanted to see if we could see that in our classifying cells. And we also saw that. So again, if you just look at smooth muscle cells, you just don't see any osteopenia 10. But if you look in different clones of our calcified vascular cells, you would see a lot of osteopenia. So again, getting more and more specific and then the cynic winnin and probably one of our most important papers came when Christina and I found bone morpho genetic protein too in calcified vascular cells. This had never been seen in vascular cells before. So it's a part of the T G F beta super family. It's well known and deserve both bone and cartilage formation expressed by a variety of cell types um including sales of arthroscopic lesions that was new, had not been shown before. Um And we showed that and that's one of the big advances that are lab made. Um So both BMP two and BMP four are inducers of inflammation and mineralization. So we felt that this is getting more and more specific. Now for bone like tissues. So we used in inside to hybridization to see if we could show um signal for BMP two. And we saw, if you had the negative control, you saw nothing. Um The positive control, you see a lot. And then when we looked at our cell types, you could see a lot of BMP two as well. So we were getting pretty convinced now that we have found a group of osteoblasts like cells that live in the artery wall. Um This is something that had not been seen before. Um But we, because of, of Linda's observations that she would see these tissue types that looked at all the world just like bone and cartilage. It made us look down this line. And so we said, okay, there are these cells here but like why and what stimulates them. So the next step, we said, let's look at for stimulatory proteins that can cause these vascular smooth muscle cells to change into osteoblasts like cells. One of the first things we thought, you know, we're looking at an arthroscopic plaque, what's big in plaque, cholesterol. And we looked at regular all cholesterol didn't see a lot. But when we looked at oxidized, cholesterol and oxidized possible lipids, we did see a lot of stimulation. So oxidized fossil lipids are as you know, the sort of sina qua non of atherosclerosis. LDL alone doesn't do a lot of damage. But once you get oxidized LDL, that's where all the damage has happens. So, ox P A P C is one of the most prominent possible lipids in oxidized LDL. So we were able to show that this oxidized fossil lipid specifically was the most potent inducer of nodule formation. So as we increased our concentration of ox P A P C, we increased the amount of nodule formation. And you could actually see that graphically that the number of nodules is a nice biomarker for the amount of calcification. We could see that as we increase the amount of oxidized possible lipids, we see more and more nodules in each well, was not seen in negative controls or with non oxidized cholesterol. So oxidized cholesterol did cause these nodules to form. We also saw that oxidized cholesterol induced alkaline phosphate based activity. And we saw that oxidized fossil lipids induced collagen. One synthesis, the type of collagen you see in bone forming tissues and and calcified tissues, but um not so much in college in force. So here's looking at collagen one in our negative controls, you would see in our controls, you would just see a little bit of collagen one. But once you added the oxidized fossil lipid collagen, one activity just took off. We also did some really interesting studies and I'll come back to this later because it's a question I still have, which I never really got back to. We looked at a classic osteo plastic cell type M C three T three E one cells. So cbc's where our classic calcified vascular cell type M C three T three E one cells are classic osteo plastic cell types. And what you see is in control cells not treated with anything osteoblasts, make a ton of college in one that makes sense. This is what you see in bone forming cells. When you expose them to oxidize fossil lipids, they get a lot less collagen one. So this brought us to a question that we actually had always wanted to come back to, but haven't yet. You will always see this paradox between osteoporosis and vascular calcification. A lot of it is probably mediated by age. But the question because really why are tissues laying down calcium in the vasculature where you don't need it while at the same time taking it away from the bone where you do need it. That's the question that we still want to come back to. Well, we found a lot of inducers of calcification. So oxidized fossil lipids were the first we looked at, but we found that a lot of inflammatory mediators could do the same thing like TNF alpha oxidative stress BMP to the classic bone protein M S X, which is a nuclear transcription factor, osteo calcium and phosphates rank L lost upon a lot of different things. But remember those early pictures I showed you where Linda was looking at pathology sections and she saw a bunch of different tissues that look like bone cartilage. And that one of the things we also thought, okay, we are finding these cells that have osteoblasts like properties. But what if there really is more of a just a musical kind of stem cell, these these nascent cells in the artery wall that can really transform into a bunch of different cell times, not just bone may be fat and maybe cartilage. So we started looking for more thinkable stem cells cells and we looked via flow. So telemetry. So bone derived unsinkable still in stem cells are known to be positive for CD 29 and CD 44 And negative for CD 45 and CD 14. So we wanted to see if our calcified vascular cells had similar cell markers. And what we saw if you look, we saw that the M1 cells are positive control as you would expect are positive for CD 29 and CD 44 and negative for CD 45 and CD 14. But the M2 cells are calcified Josko cells. We actually saw a very similar pattern. So we saw positive for CD 29 and CD 44, negative for CD 14 and negative for CD 45. So it really did make us think we were dealing with not just an osteoblasts like cell type in the outer wall, maybe really just amazing Kable cell stem cell that could then transform into a number of different things given whatever stimulus, you expose it to. Okay. So that was one of the things that we really were excited about because if you think about it, if you have this musical stem cell in the artery wall, that means a lot of different things could happen. And then maybe you could actually figure out how to modulate them and maybe get those cells to transform into things you really want them to do. That's something that's further down the line. But it does bring up some really exciting possibilities. And this is when we actually got a lot of press because this was really the first time that anyone had ever thought of vascular calcification as anything but just this amorphous very boring process. Again, that's me as a first year fellow many years ago. But we were really excited because this was something new and different Or was it? So if you look back through some of the really old pathology treatises, this is from Burqa from like 1880 or something like that, he basically wrote in these treaties. Um that vast calcification is quote unquote an ossification of the arteries. So this is when he was using rudimentary stuff or microscopes to look at this. He said this is just like bone. So again, this is one of the things I wanted cardiology fellows to realize past is prologue. A lot of things that we think are new, like inflammation, the hot new thing. A lot of people a lot of many older pathologies and cardiologists had really thought of these things a long time ago. So, one of the things that I then became interested in was using CAC coronary artery calcification as a research tool. So, what a lot of investigators had shown was that the calcium identifies plaque. And in fact, that's really what we think the value of getting a CAC score is, it really does tell you one question is there. Atherosclerosis, chronic plaque there, yes or no. And it's proportional typically to the amount of plaque doesn't tell you exactly where the plaque is or how cyanotic it is or anything like that. But it tells you there's atherosclerotic plaque that's there and the more calcium there is the more plaque there is. So that was at the point where I was heavily ensconced in my basic science lab and I did love it, but I really needed to answer more questions and the most of the questions I needed to answer required populations. So it was at that point that I sort of transitioned from basic science to translational and population science. And to do that, I used the um the power of the mesa study, the multi ethnic study of atherosclerosis. So, Mesa this is a multi center study sponsored by N H L B I A Longitudinal cohort study investigating the prevalence correlates and progression of subclinical cardiovascular disease. And one of the things. So I actually joined Mesa um after its inception. But one of the things that the initial investigators wanted to do was sort of update our Framing Him knowledge base. So we know we use the Framingham database for so many things to predict a theocratic events and things like that. We learned so much from Framingham, incredibly important study, but it really is from a small town in Massachusetts, which didn't have a lot of ethnic diversity. And so they said, let's kind of update the database that we know and get a study that is about a quarter, black quarter, quite a quarter, Asian and a quarter Hispanic. So our job as the six U S field centers was to recruit healthy men and women between the ages of 45 80 for about 50% men and 50% women. And we wanted again, 25 25 25 25%. We didn't exactly make those percentages, but we got 34% white, 28% African American, 22% Hispanic and 12% Chinese. And these are the field sinners. So U C L A is the only one on the West Coast. Everything else is eastern, but we've all been working together to try to figure out some basic uh population data on atherosclerosis. Now, the idea was to look at subclinical atherosclerosis and figure out how that related to atherosclerotic progression and events. And so the main subclinical atherosclerosis measures that were initially thought of in MESA were coronary artery calcium score CAC or crowd interval media thickness. And when MESA was begun, nobody had any idea which one would be best. Both are not invasive, both are fairly easily measured, both could be used widely. So the idea was to figure out which one was most predictive of future cardiovascular events and Kaplan by kind of a lot. So this is data from our very first seminal paper out of mesa looking at the hazard ratios for cardiovascular disease, coronary heart disease and for each standard deviation increase in either coronary calcium or corroded I M T. So when looking at total cardiovascular disease that involves heart attack, stroke, heart failure events, things like everything. Um you could see that for each increment increase in coronary calcification, each standard deviation increase, there was about a two fold increase in events that look pretty good. Um For corroded I M T, it was only 1.2 and it wasn't um as statistically significant. So we weren't as excited about that. Now, when looking at coronary heart disease, specifically, coronary calcium really performs a lot better than credit I M T. Um So you can see for each standard deviation increase in coronary calcium score, there's about a 2.3 fold increase in coronary heart disease events. Interestingly, when you look at stroke credit I M T actually uh predicted that better. So you could see that the for each standard deviation increase in credit. I M T, there's about a 1.4 increase in stroke events for coronary calcium is only about 1.1 and it wasn't statistically significant, but because we were looking mostly at coronary heart disease events that clearly one. Um So we definitely came across Corner Castle is a really great predictor of cardiovascular event, especially coronary heart disease events. Um Again, I M t wasn't as good. Um It was really Corner Qassem is really good for myocardial events and interestingly crowded. I M T actually prevent predicts stroke events a little bit better, but we were interested in coronary heart disease. So that's where we um put most of our energy. So then we really decide to use as our research tool. So we can, you know, we can always wait for cardiovascular events to see who lives and dies and has cardiovascular events for an outcome, just takes a long time. But if we use corner calcium, we can get data a little earlier and probably as reliably. So one of the studies I want to show you is the study we did using Cock as a research tool to figure out if air pollution is harmful for the cardiovascular system. So this is the study we did association between air pollution and coronary artery calcification within six metropolitan areas in the U S six U S field centers for mesa. Um A longitudinal cohort study published in the lancet. Now, the reason we care about air pollution is because of what we call PM. 2.5 particulate matter, 2.5. Those are those super, super, super tiny micro particles. It gets sucked deep down into your lungs. Um Their most common sources, our roadways and, and where I live in California, it supports the big ships and things, but it can be found in all sorts of things like power plants, forest fires, which California is burning right now. Um And even outdoor barbecues, I can tell you since doing this research, I just can't, I really enjoy the outdoor barbecues like I used to because I realized that every time I'm there, what I'm doing is sucking these little tiny micro particles deep down into my lungs. So all the major bigger particles get filtered out by all of your defense mechanisms. So that's good. But those tiny micro particles never get filtered out and they end up taking up residence and deep down into your lungs which causes this inflammatory cascade. So, again, larger particles deposit in the upper airway, nose and throat, they're just cleared out smaller particles that can penetrate deep into the lungs. They retain their or and they may even enter the bloodstream. And these inflammatory stimuli have the capacity to elicit widespread inflammation far beyond the lungs, including in the cardiovascular system. So we there had been some data before we started our study that long term exposure to this fine particulate matter PM, 2.5 and traffic traffic related air pollutants which mostly are PM 2.5 were associated with cardiovascular risk. But most of these studies were of short duration, they were just a single point in time. Um And there's, there's really interesting city variability. So we wanted to know Um if longitudinal e you could see a relationship between the amount of PM 2.5 and other pollutant exposures that people experience and their cardiovascular um risk. So again, wanted to test a priori the association of air pollutant concentrations with progression of coronary calcium. We were still looking at crowded I M T at this point as well. So we looked at that as well. Spoiler alert. It didn't show anything but we wanted to look at air pollutants PM 2.5, the most important when we were interested in. But we also looked at nitrogen oxides, nitrogen dioxide and black carbon. And to do this, we use utilize the Mexico. So we repeatedly measure CAC in almost 7000 participants. Um 2000 to 2005, 2010 common credit I M T was measured twice in everybody between 2002 and 2010. And so we could look longitudinal e we then had to figure out how to do the monitoring and that was really pretty tricky tricky. We just had to do home, outdoor monitoring and home indoor monitoring. We also tried to do personal monitoring with people wearing these little backpacks that had air monitoring turned out that nobody like the backpack and nobody really wore them. So we had to abandon that part. But what we decided to do was to sample 100 homes in each city. You have to monitor them in different seasons. So for instance, in California, you have to monitor during forest fire season. In the east, you have to monitor during winter season where you have all the furnaces going. Um So we did that and then that was 100 homes in each city. Then for indoor monitoring, we chose about 50 homes in each city. And again, in two different seasons, we asked them to choose whatever room most people conjure, congregated in the living room, whatever it is, just put your monitor there. So we could see what the home, indoor pollutant levels were like we um collaborated with the air pollution monitoring group at USC to do this because monitoring is really pretty tricky. Um So they were experts at doing this and when you do, well, you can get these really, really rich geospatial maps of where your pollutants are. You can actually see to the, you know, the street, the house exactly where the air pollutants are. One of the other really interesting things we did at U C L A was we wanted to um just look at air pollution. Um And so we enrolled socio demographically matched participants, one from Dennis that's a really good air quality area right near the ocean, but still demographically similar to people in Riverside, which is right near um the most polluted area in California. But the socio demographics were about the same. So we wanted to figure out all things being equal. Just the difference being the air pollution. Can we see a difference? Um One of the nice, so where we live in southern California, the ocean breezes do a great job of clearing out air pollution if you happen to live near the ocean. So that's where Venice's riverside. So when the ocean breeze blows the air pollution in, it blows it to the San Gabriel Mountains and then it gets trapped, riverside, sits right at the San Gabriel Mountains. So all that air pollution gets blown in from the ocean gets stuck by the mountains. And so it just sits there, server side has some of the worst air quality in the country. Um So we could see again these really, really rich maps of PM 2.5 and nitrogen oxide concentrations in mesa. And we can see by each state. So we could say Baltimore than Winston Salem. And the number of black dots are correlated with the amount of pollution. You can see Los Angeles down there to the left bottom. A lot of black dots, You can see New York City in the middle and the right. A lot of black dots. So that's for PM 2.5 for nitrogen oxide, you can see similar maps Then looking at it graphically, you can see again, the greatest population of PM 2.5 was seen in California. The greatest population of nitrogen oxide was actually seen in New York, but California was very close nitrogen dioxide. Also New York, California was close and black carbon was about the same between California and um New York. But what we were able to see is that for each increment and average PM 2.5 concentration that a participant was exposed to. There was an increase in their coronary calcium score that was almost linear as you see the blue line above. Um we saw an increase also for nitrogen oxide that was significant but nitrogen dioxide and black carbon were not significant at all. Both PM 2.5 and nitrogen oxide, the more air pollution, those pollutants you were exposed to the more corner and calcium you had indicating higher cardiovascular risk. So in mesa CAC increased on average by 20 for Pakistan use units per year crowded I M T increased by about 12 microns per year. But as I said, crowded, I'm t was not specific, not significant. So we didn't really look much further than that. So for PM 2.5 for each five micrograms per cubic millimeter increase in PM 2.5 CAC increased 4.1 activist in units per year. And that was in that linear relationship as I showed you um when nothing was really related to crowded I M T. So we didn't look at that further. The other thing we were able to see in Mesa was looking at each one of these lines is an individual participant in Mesa. And if you look over time of this follow up, you can see there's a general trend for a reduction in average pm 2.5 levels. So it really does suggest that some of the policy initiatives are having some impact. Okay. And another study done that was really interesting using CAC was our study on HDL. So, you know, epidemiologically HDL levels are inversely associated with coronary heart disease. So the higher your HDL, the lower your coronary risk, the lower your HDL, the higher your risk. But we've seen in many populations that deals are heterogeneous. So for instance, if you have HDL, April A one Milano, they have really low HDL levels, but they never got heart disease. And we had seen in some of our cohort studies at U C L A. People with really high HDL levels over 100 and they get a lot of vascular disease. And we've also seen that trials of raising HDL with drugs like niacin and vibrates have not been successful at all. So we wanted to know our hypothesis was that HDL is not protective against atherosclerosis in everyone and the individuals that we thought it would be least protective in for individuals with the metabolic syndrome. So what we did, we categorized the mesa cohort according to the presence or the absence of the metabolic syndrome. And then each group was further stratified by HDL level. We considered low HDL HDL is importing intermediate HDL between 40 and 59 high HDL anything above 60. We then looked at them for their cardiovascular event rates. So not just CAC but heart event rates. Um and then we dissected a little bit further. So this is what we saw. Look first at the far left, the entire cohort. If you looked at the entire court, it looked kind of like what you expect to see when you think about HDL. If you had really low HDL HDL is on 40, you had the highest event rates. If you had HDL, that was intermediate, your event rate was lower and What we were actually a little surprised about in the entire cohort. If your HDL was above 60, it didn't look like it was better. It just looked like it was the same as people who had sort of an average HDL. Now, look all the way to the right people who had no metabolic syndrome at all. And that was, you know, the four classic metabolic syndrome parameters, dislike body mia um insulin resistance, obesity and high blood pressure. These people look like exactly like what you would see HDL. Less than 40 highest event rate. HDL 40-59 Intermediate event rates and lower event rates if your HL was greater than 60. So that looked like what we thought of as the classic HDL atherosclerosis relationship and look at the people in the middle of the metabolic syndrome. The first thing you see is they all have higher event rates than the others. And that makes sense. That's a high risk cohort. What you see is HDL, less than 40 very, very harmful, really high event rates Much lower for that intermediate HDL group, 40-59. But then when you go to the HDL but 60 equally as high event rates. So that was really kind of the first time that we saw HDL that's high is not always helpful and it may be harmful there since have been several communications showing the same thing, but this is really one of the first. So now this is looking at coronary heart disease events. When we looked at total cardiovascular events, we saw a very similar pattern. Again, if you look at the entire cohort, you see this inverse relationship. If you look at those without metabolic syndrome or really strong inverse relationship. But in the middle group, the metabolic syndrome group, either low HDL, very high HDL are associated with worse outcomes. And when we look at the Kaplan Meier curves, you see sort of the same thing. So those who have, um HDL, less than 40 are shown in blue HDL, 40 to 59 shown as green HDL above 60 shown as red. So looking at the entire cohort, the worst survival is in the blue line. HDL less than 40. That makes sense. And the survival of those with HDL in the intermediate range or in the high range about the same. Look to the right in the middle panel, people without metabolic syndrome, you see similar pattern. The blue line has the worst outcomes. HLS M 40 and similar outcomes in those with HDL in the mid range and HDL above 60. But look over at the metabolic syndrome group, very different pattern. The worst outcomes are shown in the blue line as expected. But also the red line, the very high HDL patients and the people in the intermediate HDL group did the best. It was really kind of is very, very, very um interesting and confusing to us. But as you all know, as I said, the metabolic syndrome is a component of four different things, Disl epidemiology to insulin resistance, obesity and high blood pressure. So we wanted to see what the relationship between the interaction between HDL and each one of those four components. Um So we assess them individually controlling for age sex and race to see which one had the strongest effect on this H G L effect. So the only significant interaction we found was between waist circumference and high HDL having the worst outcomes. And again, this was done by my statisticians and my computer programmers. So I cannot tell you anything more about this. Um But what this is what it looks like. So if you look at everybody, so remember the people in blue are the low HDL people, people in green are the intermediate HDL people and the people in red are the high HDL people and compare it to by waist circumference. So waist circumference, that's the smallest is shown to the left, the highest shown to the right. So you can see across all ranges of waist circumference. The relationship between the low HDL people and the intermediate HL people looks to be about the same doesn't really vary by waist circumference. But in the high HDL people, it really does. So if you start off with high HDL and your waist circumference is low, you have lower event rates than everybody else. Kind of the classic inverse relationship we're used to seeing. But as your waist circumference increases the rates not only equal but then exceed those of the low HDL group. That is a modeling exercise. Their based on that mathematical equation I showed you. So it really showed us that HDL is complex and you still has a really big age still research lab run by Alan Fogelman. So we ran all of this data by him and he said, of course, this makes perfect sense. So according to Allen, he said it still is just a big chameleon molecule. It really is just a big ball of fat that will adopt whatever properties of the environment you put it into. So if you have this big ball of fat and you put it into a healthy fit non metabolic syndrome environment, it's gonna perform pretty well and it's gonna basically take the hit for LDL. So you don't get oxidized LDL. You don't get as much atherosclerosis, you don't get as many events that makes sense. But if you take this ball of fat and you put it in a very unhealthy pro inflammatory environment, it then cannot only not take the hit for elk, it then can become a pro inflammatory molecule itself. So to him, it made perfect sense to us, it was unusual. But since as I said, there have been a number of reports showing the same thing, individuals with very high HDL actually have higher event rates. And what we show is that it's predominantly related to obesity. Then when you go back and think about everything we learned about HDL. We learned from the Framingham Heart study Done in the 50s, that's at a time where very little obesity, very little diabetes, everybody was a little more active. And so if you take HDL and you put it in that healthy population, it's gonna look really good. But again, we have a lot more inflammation, a lot more obesity, a lot more diabetes. Now, so putting a bunch of HDL and that environment may not look so good Okay. Let's look now at CAC as a clinical tool as I told you earlier, cake really does identify this crisis and that really is the one answer. It gives you a thorough cirrhosis. Yes or no. If you have, you have atherosclerosis, the more cake you have the more atherosclerosis you have. So we looked at that and this as well. And as you can see, um this was again from our first seminal paper, if you have CAC 1 200 that's the low cat group, you have lower coronary heart disease event rates, intermediate group through 12 300 and so 31 oh 12 300 they have higher cardiovascular event rates. And then the highest event rates are those who have cracked above 300 where the rate was almost 10 has a ratio for a coronary event. Um So, CAC over 300 in our study was really associated with a very high feature CHD rent. But even in that one oh 12 300 group, the hazard ratio of 7.73. So that's really high too. And so what we, you know, cat is not good, you're just not supposed to have calcium in your coronary arteries. And if you do that means you have atherosclerosis. And as Linda taught me many years ago when you have a big fluffy livid plaque next to a really hard calcium um block, that interface is really unstable. So that's where plaque rupture can happen and cause clinical events. Um Oh, sorry, the age adjusted problems of corny costume is different for different race ethnicity. And that was one of the beauties of, of Macy's. You can see that you can see that in general the amount of CAC levels are higher than whites. They are the lowest in African Americans. So for a given amount of atherosclerosis, African Americans have lower CAC. It's been shown time and time again. I'm gonna go back to the osteoporosis thing. African Americans didn't have much higher bone density, much lower vascular calcification density. There's this inverse relationship we always see there. So for men tend to have higher levels than women as you see there. And again, it varies by race ethnicity, but whether or not you have higher or lower cac levels, it didn't matter. It's still a really great predictor in every race ethnic group. So look at for blacks, even with um you know, there are lower cac levels in black patients. But as your cat global goes up, your coronary event rates go way up. So it's still a great predictor regardless. Um That's true for Chinese Hispanics, whites. It's true for different age groups. It's true for both men and women. And so even if the numbers, the absolute numbers are different, it's still is a great discriminator and predictor. And so we can see the 10 year risk. It's a really nice Kaplan Meier curves, you can see the rates go up by cac level in every race, ethnic group. So then the question becomes like, should everybody get a cat scan? You can see a lot of people advocating that. Um So we know that cat will tell you if I have osteoporosis. Yes or no. Does everyone need that? Well, maybe, maybe not. So you will see a lot of people advertising saying that yeah, everyone to get this because you just don't know. Um one of the problems though is that a lot of these cat scanners popped up and in various places, so people needed to use them. And so they were advertising them to be used. But I think that was actually before the data suggested whether or not they should be used. So unfortunately, the technology lead and then the data followed, it should go the other way around, but that's how it happened here. Now, one of the things again, it tells you after courses, yes or no. But you can also kind of get an idea about that by just doing a risk, right? Like the Framingham risk or the full cohort risk or so, really, only if you can get improvement upon much simpler ways of determining who has atherosclerosis or not. Would this make sense? Fortunately, in mesa, we were able to show that you get additional value from adding CAC two standard risk scores. So this is data looking at the improvement of a CBD risk, his husband and intermediate risk individuals by adding CAC. So you can see um sorry, very busy uh legend, but the dark gray line is Framingham risk score alone. And then by adding CAC to that get improved discrimination and um prediction. So adding cacti to any risk or were able to show you get improved risk prediction, both for incident coronary heart disease and total cardiovascular disease. So how can we use it? Then? Obviously, we can tell you who needs therapy, more aggressive preventive therapy. Um One of the things we can say is like who might need more aggressive BP. Um blood pressure um reduction. This is modeling from mesa. Um the number needed to treat to prevent one atherosclerotic cardiovascular disease or heart failure event. I'm using target BP of 1 20 or lower as compared to 1 40 is shown here. If you're a CVD, risk is less than 15. Um but your blood and your blood pressure is less than 1 40. The number needed to treat by getting that more aggressive blood pressure lowering goal is about 55. But as your risk goes up and as you had capped that you act, you actually can reduce the number needs to treat. So for a systolic blood pressure of less than 1 60 a low arthroscopic cardiovascular disease risk. If your cat is equal to zero, you really can't, it's hard to prove that getting a really aggressive blood pressure goal reduces events. Your cat is over 100. It's really easy to prove that getting that aggressive blood pressure goal of less than one in 20 will reduce events. So you can tailor your therapy or decide how aggressive to be with therapy. Perhaps by using cac King cock inform aspirin decision. Well, we've shown that if you have a cat equals zero and a coronary heart disease risk of less than 10, the number needed to treat to prevent one coronary heart disease event is over 2000. The number needed to harm to cause one bleeding event Is only about 50. So it doesn't make sense in that case, maybe not. But if you have over there to the right, a cat that's over 100 doesn't matter what your estimated coronary heart disease risk is it's always gonna favor treating because these are such high risk people. Um There's also a question about whether or not it can inform staten decision making. This is from a data uh a paper that I wrote with Bill Greenland and some others published in Jack. We said it probably could be useful for deciding who needs statin therapy or not. This was before the most recent cholesterol guidelines came out and then the most recent cholesterol guidelines, they put exactly this into the guidelines. You can use a CAC score to help you decide who needs statin therapy and who doesn't. For instance, if you have your estimated 10 year risk of less than 5% the column, all the way to the left statins not recommended and CAC score of zero statins not recommended. So really, it's hard to imagine that Stanton would be useful in that population. But if you have a patient population with an estimated risk, that's intermediate, but the cat is positive, that's patient population, you want to use the most aggressive preventive therapy. And there's also this is data from Mike Blaha, a lot of impetus to think about a cat equal zero as a way of de risking individuals. So maybe your for predicted tenure risk is really high, but you have no corner calcium. It's probably at least in the short term safe to say they don't need the most aggressive statin therapy or blood pressure reduction therapy. This is data that Mike showed. So um if you have a cat equals to zero, that is that lowest line, you have much lower cardiovascular risk rate no matter what your predecessor probability is. So no matter what your risk score says, what I want to point out from this is that the risk is not zero equals to zero, but it's significantly lower than those who have the presence of CAC. And again, just showing you from the, the chart we put in our first study, even with equals to zero, there's a measurable event rate in everyone regardless of race, age and sex. So it's not zero risk. It just means a significa lower risk. And if you get a cat equals to zero, we don't exactly know what the warranty on that is. Some people say two years, some people say five years. But if your cat has zero on time zero doesn't mean it's gonna always be zero. So you do have to measure it again. All right, future directions. You know, I mentioned this a couple of times. The calcium part paradox. Why is the body depositing calcium in the arteries where it's not needed, instead of depositing it in the bones where it is needed. And I showed you from our studies, a lot of the inducers of vascular calcification actually inhibit bone calcification. So we don't know why. And if we could figure that out, we could do exactly the opposite. Get people to lay down more bone in their bones and less in the arteries, but we can make a lot of money. Um The other thing is vascular calcification and aortic calcification share a lot of the same mechanisms. So when you look at uh calcified valves that they take out, you see the same kind of nodules that are calcified, same kind of nodules we see in our vascular smooth muscle cells. And what can we do to modify that? That would be huge as well. But what we know now there's this sub population of RT well sells, it has the capacity to become osteoblasts like cells producing that ossification of the arteries that Birkoff talked about hundreds of years ago. These cells can be induced to become pasta genic by inflammatory stimuli, pro afra genic stimuli and especially things that we see in our throats crisis like oxidized lipids. Cac is a really powerful research tool because it's a marker of atherosclerosis as a clinical tool. It can be very powerful too. It's really should be used as a decision made just for you just to decide how aggressive to be with your preventive therapies. Everyone always asks, well, how do I get rid of this vascular calcification? And the answer is once you've made that transition from the vascular smooth muscle phenotype to an osteo lawsuit phenotype. We've never seen them go back. So we don't know if you can. Um But it's certainly something that I think future studies should look at. I just want to acknowledge my co investigators at U C L A with MESA and the MESA participants and a lot of our partners like the air pollution study, huge amount of resources went into that. This is a collaboration between N H L B I and the E P A to get a lot of this work done. But thank you again for your attention and I'd be happy to take any questions. Does anybody have a question? You do have to text? Thank you. Care for a magnificent, magnificent presentation. I have a quick question. I think that your data on HDL and metabolic Center is fascinating showing how counter intuitive is to think that patients with metabolic syndrome, a high HDL will actually have a higher event rate. Has it been looked at to see if there's anything to do with HDL sub fractions and to see if there's any difference between the functionality of these. That's a great question. But no people have done a lot of um subclass distribution to see if there's something you can predict that would or would not be helpful. And you could probably find every subclass showing positive or negative. The data is all over the place. That's what I'm trying to say. There's no obvious. Um Right now, we haven't figured out a way to do that. Yeah, we do have a question on one of our from one of our virtual viewers. Okay. Endurance athletes without traditional risk factors have been found to have higher levels of coronary calcification. Yet higher CAC does not pretend a bad prognosis in those athletes. What are your thoughts about the mechanism of coronary calcification in this subset? That's a fantastic question. My co director of preventive cardiology at Greg Monroe, he's a marathon runner and I keep telling him how crazy that is and he keeps arguing with me. So here's my thinking, the big three oxidative stress, inflammation, cellular damage, nothing good happens from those three. And that's exactly what you're doing when you're doing endurance running. And those are exactly the things that we see cause deduction of vascular calcification, oxidative stress, inflammation, cellular damage. So I think you're doing exactly those things that can lead to laying down vascular calcification. But I think you're also getting such great, um, cardiovascular fitness with that, which we all know is beneficial that it does seem to out um, way it. But you're exactly right. When I, a lot of patients who have been healthy their whole life, they come to see me because my God, someone got a random corner custom score and it's 2000. I like lived this perfect life. I do all this great stuff. I eat clean. I do all this exercise. And then I asked him, have you ever been a distance runner? Like, yeah, I'm like, that's why you have this. But the good news is that this is the one subgroup of patients in which corner calcium does not correlate with increased events. That's a great question. Excellent. We do have one more. Um Dr Sandi said fantastic presentation given that CAC increases after statin statin therapy. How do you use CAC after the therapy started? That's a great question. You guys have the best question. That's a great question. The answer is you don't catch, gives you one answer. Coronary answers, questions, yes or no to help you decide on intensity of therapy. And once you get that answer, don't test it again. So what happens is as you guys know, the CAC score is just looking at density of calcium in the box als and pixels in your C T scan. And what that is looking at is you're looking at the calcified portion, the non castle portion portion, everything. So what happens when you put someone on a statin, you suck out the cholesterol from that area. All that looks like is you've made the calcium a lot more dense because you haven't changed that with your stand. But you have sucked out the non calcified person, the lipid portion. So now your calcium looks more dense. Yurcak score will go up. But that is again when talking back to Linda Demers, she was a biomechanical engineer. She's like what you've done is you've made a much more stable interface because that hard calcium next to that big fluffy plaque is just so unstable. You can suck the cholesterol out of that. Your tax score looks like it goes up. But now you have a much more stable interface. Thank you. One more question from Dr Samity. What dietary and therapy advice do you have for patients with osteoporosis and a high cac score? Another great question. So all these patients are on calcium and vitamin D. Um We don't recommend that and I think the US preventive services task force no longer does for most people because um when we looked in vitro, when you add calcium and vitamin D to vascular smooth muscle cells to calcify much more quickly. So again, you just don't know what you're getting when you do that. And I think studies have shown there have been a number of studies of calcium supplementation and postmenopausal women at increased risk of heart attack. So again, you just don't know what you're getting. Thank you. There's no other questions online. Thanks so much for the invitation and thank you to my friend for inviting me. I appreciate it. I'm gonna invite you to L A next time. Thank you so much. So awesome, so much.
Related Presenters
