My name is Cecilia Lowe Wong. I'm a professor of medicine at University of Colorado and I'm the chair of the diabetes DSN or disease state network for ACE and I'm really happy to be moderating our session today. So we are going to be hearing from Dr. Peter Rieben and he'll be talking about cardiovascular disease and diabetes assessing risks and prioritizing therapies. He's a professor of clinical medicine in the division of endocrinology at the University of Arizona College of Medicine in Phoenix. He's a research professor at Arizona State University and he's also the staff endocrinologist and director of the diabetes research program at the Phoenix VA. He earned his MD at the University of Chicago and did his residency and fellowship in endocrinology at UC San Diego. He has held a number of leadership positions at the ADA and he's also been part of study sections for the NIH, the VA, and the American Heart Association among other organizations. He has a broad clinical and research background in insulin resistance, diabetes, dyslipidemia, and vascular disease and he's been closely involved in the development, conduct, and analysis of large multicenter clinical trials including the VADT and in using these studies and the data from the studies to try to figure out standard and novel risk factors for cardiovascular disease and trying to translate this into more personalized therapy. So I'm really, really happy to welcome him to the podium to give us his presentation. Dr. Weaven. Great. Thank you, Susie. That was for the very kind introduction. Thanks to all of you for sticking around. It's beautiful. You've got sunset coming up. I can't believe that you're here and so I will try to be quick and let you get back outside and enjoy the world out here in San Diego. We're a pretty small group and so I'm gonna suggest, although I've got a lot of slides here because I didn't know what I was going to talk about really more than a month or so ago when they wanted these slides back in, but please stop and interrupt at any time, ask questions, or we can do it at the end, but please feel free to just ask what's got your interest up. So I sort of like the slide that really points out that if you're really going to try to do a comprehensive approach to managing cardiovascular disease, you've really got to do a lot of different things and that's what makes it complicated in both art and science, but I really can't talk about all those today and so I'm really going to talk about this area on the right, glycemic control and some of the medicines and effects they'll have on cardiovascular disease, but obviously it's important to consider all these different factors shown on the slide, lipid management, blood pressure management, antithrombotic care management, lifestyle, and of course this whole issue and effects that really require public health policy to try to address social determinants of health and I think we realize even more now than we did a couple years ago how important those things are and how they can impact our ability to, you know, the disparity in terms of how we handle different events going on. So this is the outline which I just mentioned and so I'm going to just start off with a case just to kind of get us thinking about this. A 58 year old woman with diabetes for a number of years has some hypoglycemia periodically, had suffered an MI around 14 months ago and recovered, has pretty good ejection fraction as a result, is on metformin, insulin as you can see here, and is, you know, on an ACE and is getting a torvastatin, you can see the values or characteristics phenotype on the right, so overweight, in fact obese, blood pressure is good, lipid levels are pretty good, A1C not bad at 7.9, and GFR is good. So that's case one. Case two, same woman but I've just changed a couple of things here, that now after the MI they're recovery wasn't quite as good and there's a lower ejection fraction and if you look on the right the GFR is a little bit lower. So as you know I pointed out, and I apologize I'm starting to lose my voice here, so hopefully this microphone will keep me going. The patient here has pretty good other risk factors, right? You know, blood pressure is good, lipids are good, etc. But this patient does, and A1C is not bad at 7.9, but probably could be a little bit lower, and the question is, you know, how aggressive should you be? Well, turns out that this patient would fall pretty nicely into the category of individuals that would have qualified for the entry to the cohort advanced and the BIDT glycemic lowering studies, and so we can sort of use that information to help guide some therapy. Shown here are the A1C changes that occurred with intensive treatment, the years of follow-up, and the hazard ratios. And so overall no benefit, at least statistical benefit. You can see sort of a modest trend for improvement, but that's what the results showed us despite, you know, these more intensive glucose lowering efforts. In the VADT, we then just try to look looked at this sort of over time. Part of this, we thought, was related to how long you were able to lower glucose for. The VADT study itself lowered A1C about 1.5 A1C percentage points, and that lasted about six years, and then we continued looking at this for 10 years and 15 years. And then, so what you can see is that there was a trend nearly significant for a reduction of primary outcome during the course of the VADT, and this became significant over 10 years, and then started to wane as the A1C levels were no longer different between the groups, because they were no longer in the trial, they were just being followed in an observational fashion. And so it does suggest, it's consistent with the idea that glucose lowering over a long period of time is required to start to get some of these benefits. But you can see, not much different in cardiovascular deaths or mortality. So I think in terms of glucose lowering, you know, we can start to come up with some conclusions from earlier studies at UKPDS and DCCT, that I think it really behooves us to start early, if you can, when these patients are first developing diabetes or younger, and then it requires a fairly sustained substantial glucose lowering to really start to have impacts on macrovascular disease. The microvascular disease may be a little bit more responsive, although even in those three major studies of type 2 diabetes, even the microvascular disease did not improve as much as we would have hoped. But this is the message to date. So as we start to look at basically, you know, diabetes, glucose lowering in diabetes medications, there are a lot of options. I'm not going to talk about all of these. Obviously, insulin has been studied for a long time and generally is felt to have a fairly neutral effect on cardiovascular disease. And so I thought I would sort of briefly remind you of some of the older studies looking at some of these other diabetes medications before we move on to some of the newer medications and their relationships with cardiovascular disease, which probably go beyond, as we'll talk about, glucose lowering itself. So just going back a ways, the UKPDS study studied, you know, intensive glucose lowering in sort of new onset diabetes patients. But they had a subset of individuals that were actually randomized to metformin. They were more obese individuals. And as shown in this graph here, that metformin actually seemed to outperform slightly even intensive, the intensive glucose lowering on these new onset diabetes patients compared to conventional therapy and was associated with a nearly 40% reduction in myocardial infarction. And then interestingly, shown on this slide on the right now, is sort of the long term effects. So the study went on for about 10 years, ended in 1997, and then they continued to follow these patients afterwards. And you can see that metformin continued to be associated with reduced cardiovascular risk in these individuals for a long period of time, supporting the sort of the overall idea of glycemic legacy. Oh, thank you. That's perfect. Great. And interestingly, this held for the many different outcomes that were evaluated in this study. So all diabetes related endpoints, mortality, myocardial infarction, stroke, and microvascular disease. So it's a small subset of individuals. It was only a little over 300 individuals, but pretty impressive results. And there's been some other studies of metformin that I think add to the interest in this. There was the SPREAD diabetes study, which was a randomized study of metformin versus glipizide in individuals that had known cardiovascular disease. And in this group, there was about a 50% reduction in cardiovascular events. There was also the REACH registry, which was a very large registry of patients with cardiovascular disease that were then followed observationally. And then they looked to see whether those on metformin, how they did compared to other glycemic agents. And again, there was a 25% reduction in all-cause death in these individuals. And then a very similar outcome was seen in a post hoc analysis of the SAVOR study, where they looked at the use of metformin. So again, these are more observational post hoc studies, but again, somewhat supportive of this idea that maybe metformin could be associated with reduced cardiovascular disease. And then there's been a couple of real world studies where they looked at large databases. This was a combination, I think, of three large insurance claim databases. And again, showed that compared to other diabetes therapy, that there was a reduction in a number of different cardiovascular events as shown here, including heart failure, which was also interesting. So obviously, lots of mechanisms that have been proposed for how metformin might actually have benefit on multiple tissues, whether so the benefit on the vascular outcomes could be both indirect through metabolic improvements or perhaps directly through effects on the heart or vasculature. But again, can we put at rest our, would you bet on any of these? No. We don't even know how metformin really works in terms of lowering glucose. So I don't think we really understand the mechanisms here, but it's certainly intriguing. So I think, as is summarized here, this is probably true. There's a paucity of evidence from long-term trials on this, but there's a lot of interest in metformin still. And in the VA, we're now conducting a very large study, actually looking at metformin in pre-diabetes patients for cardiovascular outcomes in cancer. And there's a heart failure study that's ongoing as well in patients with diabetes or pre-diabetes. So what about sulfonylurea agents? You know, the UGTPS study decades ago raised the concern about glipiza or, I'm sorry, sulfonylureas as increasing risk, but this was not seen in the UKBDS study. And so it was, the SU agents had similar effect as did insulin. So no evidence of harm in that study. There's been meta-analyses conducted on a lot of these trials that looked at various sulfonylurea agents, and although there's some that show some potential harm, overall these have been fairly neutral. Maybe a slight increase in mortality, and certainly hypoglycemia is at increased risk. And so that's a well-known indicator or risk factor for cardiovascular disease. And so obviously one reason to be more cautious with these agents. What about TZDs? Obviously these affect a lot of different tissues and organs and have been associated with many different metabolic factors and decreased plasma fatty acids, decreased glucose, improvements in lipids, and direct effects on the vasculature. And so I think all of these would lead you to believe that they may be beneficial for cardiovascular disease. You know, there was an initial meta-analysis in 2007 which suggested that rosiglitazone may actually increase cardiovascular risk, and that led to many of the FDA changing our whole approach to drug approval and led to many of the CVOTs, which are now yielding very interesting results. And then it was quite concerning. However, that wasn't seen with pioglitazone, which was actually showing trends for reduction in cardiovascular disease in the smaller studies that were put together in the meta-analyses. And then there was the PROACTIVE study, which actually examined the use of pioglitazone in individuals with type 2 diabetes and with cardiovascular events or very high risk. And unfortunately, they picked an outcome, composite outcome, as shown here that included a lot of softer call and peripheral vascular disease events, which are tougher to actually, I think, improve. And so it led to a trend, but non-significant, a 10% reduction in vascular events. However, when you sort of pulled those out and looked at some of the more typical endpoints for vascular outcomes, there was, in fact, a significant reduction of around 16% in the PROACTIVE study. So a long, you know, study was done a long time ago, but really sort of suggested that these agents might be useful. Then there was the IRIS study, which actually took patients without diabetes but had insulin resistance and had had experienced a prior ischemic stroke or a TIA and randomized them to pioglitazone or placebo. And you can see over the course of five years here that there was a nice reduction around 25% in fatal and non-fatal stroke or fatal and non-fatal MI. So that was certainly a very positive study in this group of individuals. So I think there's, you know, a lot of studies suggesting that there may be some benefits for the TZDs in terms of cardiovascular disease. And the FDA eventually removed even the safety restrictions on rosiglitazone after that initial controversy, but certainly rosiglitazone is not the agent that anybody would recommend to use given its less favorable record of safety than pioglitazone. So, but you must be aware, right, of the risks of using pioglitazone. Those are shown on the right here. There is increased weight gain with these agents. It may be useful in terms of it can actually decrease the fat in the liver, which can be good for NASH protection, but it does lead to some weight gain. And this can be difficult when many of our patients are already overweight. And then if you're adding to that, that can be certainly concerning. But in combination with other drugs that may be associated with weight loss, that combination may be more useful. Obviously, the risk for heart failure, fluid edema is present, and certainly there is some risk for bone fractures. So, the recommendations from a number of organizations really are that you should avoid the TZDs and this, really, pioglitazone in patients with heart failure or at risk for heart failure, but otherwise, you know, certainly a drug to consider as long as you take into account the risk for some of these other adverse effects that I just described. It may be useful in those very high-risk patients that you might have on a SGLT2 inhibitor or a GLP-1 receptor agonist, and that may also counter some of the weight gain that can be associated with TZDs. Lots of, you know, interest in the DPP4 inhibitors, of course. Originally, because they thought, well, we're going to raise the incretins, and we're going to have these sort of GLP-1 independent effects, such as SDF1-alpha, which was thought that might be really an interesting molecule in terms of its ability to improve vascular genesis and bring sort of healthy cells to the vasculature. But as I think many of you know, the meta-analyses and the individual studies of these agents just didn't pan out. And in fact, there was a signal for increased heart failure, and as shown down at the bottom here, which certainly, they're useful for glucose lowering, but certainly do not reduce cardiovascular disease. And as shown here in this meta-analysis, there is still an overall signal for heart failure, although most of this is really related to the data from the SAVER study with saxagliptin that tended to be the one that was most closely associated with increased heart failure. Big issue, though, because heart failure is a big problem in diabetes, and it certainly increases as we age, and certainly much more common in individuals with diabetes. So certainly something that has to be considered when you're using those agents. So not recommended, saxagliptin, certainly not recommended in those with diabetes and high risk of heart failure, but certainly can be used where appropriate for glucose lowering in those other individuals. What about GLP1 receptor agonists? Lots of favorable mechanisms have been proposed that lead to both better metabolic milieu, as well as, perhaps, direct vascular or cardiac effects. And you've heard lots of talks on those already. And there are differences, right? So there's the short-acting GLP1 receptor agonists, exenatide and lixenatide, and then, of course, a whole variety of longer-acting agents. And it turns out that the longer-acting GLP1 receptor agonists tend to have slightly better effects on glucose reduction and greater effects on body weight. And semaglutide tends to be the star among those with a slightly better response overall for glucose lowering and body weight reduction. So what do the cardiovascular studies look like for these individuals? For these individual agents, you can see them all shown on the left. If you throw out the ELIXIS study, which was conducted in individuals that had had a recent acute coronary syndrome event, all the others pretty consistently showing reduction in MACE, as you can see. And then on the right is the meta-analysis, which actually looks at some of the overall MACE composite outcome and then individual events. And you can see that for the GLP1 receptor agonists, they're generally supportive of reduction for each of these. And I think what's unique is the effects on stroke, where it's a fairly substantial reduction in risk, much less in a more minor reduction in hospitalization for failure, but pretty consistent across the spectrum of cardiovascular-related events. Shown here is just to give you an idea for looking at these different components, three-component MACE, cardiovascular death, MI, and then stroke. You can see the hazard ratios for the summary of these. And I think that really gives you an idea of the degree of improvement. And again, as I was pointing out earlier, what's interesting is that stroke is, in fact, reduced more than these other outcomes as a meta-analysis. And it's interesting, when you look at the SUSTAIN-6 trial, it was really a substantial reduction in stroke. So benefits, very effective agents for a lot of metabolic aspects, glucose, body weight, pure cardioprotective. We didn't really talk about it. There's been a lot of talks on the renal benefits, mainly proteinuria for the glipinoceptor agonist, and obviously the low risk for hypoglycemia. And now we actually have an oral version of this. And there's an ongoing CVOT to actually look at this to see whether it can actually reduce cardiovascular disease outcomes in a large set of individuals. Obviously, the limitations, I think you're well aware of, nausea, vomiting, diarrhea. The GI ones are the main issues. And I think the pancreatitis risk has not really panned out, although certainly this is an agent you would not want to, I think, use in someone who is a high risk for pancreatitis or has had pancreatitis in the past. And pancreatic cancer has not panned out as a true adverse effect. So based on this type of good data, the recommendations from the guidelines are really starting to come together. And both European and American guidelines are now consistently saying it really should be considered, potentially along with SGLT2 inhibitors, as first line treatment in patients with type 2 diabetes and at high risk or with cardiovascular disease. Moving on to SGLT2 inhibitors, there's obviously been a lot of interest in these agents over the last three to five years. And their potential mechanisms of action for how they work are expanding almost monthly. And you can see some of the benefits that have been proposed here. Body weight changes, reducing hyperglycemia, increasing ketones, which may be particularly useful for the heart. There's also been proposed effects through natriuresis on the blood pressure and arterial stiffness. And so all of these have been proposed as potential ways in which we may benefit the heart. And that's what we found in terms of outcomes. When you look here, the four original studies looking at their benefits on cardiovascular outcomes. I think shown here, I think it's important to keep in mind that these studies do vary in terms of their percentage of individuals at risk for cardiovascular disease. With Empiregg having nearly all individuals having established atherosclerotic disease, whereas the DECLARE study, for example, had fewer people with known atherosclerotic disease at baseline. And that, I think, is influencing, to some degree, the results that you see here. But fairly consistent benefits across these studies. Although certainly not as dramatic, I think, as was seen for the GLP1 receptor agonist here. You're looking at MACE. The most impressive was really Empiregg, but less impressive some of the other agents with crossing one over here. All-cause mortality, again, very impressive with Empiregg's study. And milder effects seen with some of these other studies. The most dramatic effects, really, and consistently are for hospitalization for heart failure shown in this line here, where you can see they're all very consistently showing reductions in that outcome. But less so for the more typical atherosclerotic cardiovascular disease events. And then there was this whole second series of studies that were done in those with heart failure. And there, they really do shine as well. You can see really nice reductions in the combination of cardiovascular mortality and hospitalization for heart failure for each of these studies. The SOLIST study was an agent which was both an SGLT2-1 and 2 inhibitor. So slightly different drugs than the others, but in terms of heart failure, pure and just as successful. And you can really see here all the SGLT2 studies lined up. And you can see a lot of them. And then you look at the outcome of CVD death and heart failure hospitalization. You can see really consistent benefits. I mean, this is very impressive. No doubt, this appears to be a class effect for the SGLT2 inhibitors. And then relatively recently was the Emperor Preserve study, which was looking at those individuals that had ejection fractions above 40%. So sort of moderate, suppressed, or preserved ejection fraction. And you can see the inclusion criteria over here. And so they had to have some degree of heart failure and lower, I'm sorry, at a moderate to preserved ejection fraction. And then they had to have some evidence of heart failure based on these biomarkers. And you can see really nice difference between that and the placebo group. And again, this was mainly associated with the improved reduction in hospitalization for heart failure with less impressive effects on cardiovascular death. And then just shown here is the results from that study. And I think what's impressive is just how fast these curves separate. If you were looking at the GLP1 receptor agonist curves, they would be separating at 12 to 18 months depending on the study. But here, for this particular outcome, you're seeing really fast improvement. And this is really consistent across most of these agents. This was a meta-analysis done a couple of years ago. It's a little bit out of date because we have really the Rewind study with the GLP1 receptor agonist, which had a slightly different outcome. But this was interesting because it broke down the studies based on whether they had a baseline presence of cardiovascular disease or just with multiple risk factors. And what this meta-analysis was showing is that really the effects seem to be greatest for both GLP1 receptor agonists and the SGLT2 inhibitors and those that had known atherosclerotic disease compared to those that were at high risk. Although I think the Rewind study has shown that overall there was some benefit even in a population that was more of a mixed risk at baseline. So the recommendations for the SGLT2 inhibitors are really to be considered as first-line treatment in patients with type 2 diabetes and high risk or established cardiovascular disease. And I think particularly good in those with established cardiovascular disease. And there's been sort of a gaining consensus that these agents really should be used for part of the treatment for heart failure, both in individuals with diabetes as well as individuals without diabetes. And so what about the combination of these agents? And I think that really is something that one would think based on these very different mechanisms of action that these GLP1 receptor agonists and SGLT2 inhibitors have, that there would be great potential for combinations of these that might have greater effects. So we don't really have the trials to look at that. But there are some real-world studies where people have gone and looked at the big databases to try to see if this is the case. This is one that was recently published that actually looked at both a cardiovascular composite endpoint up on top. And it looked like adding the SGLT2 inhibitors on top of the GLP1 receptor agonists appeared to be associated with significant improvement. And when you looked at incidence of heart failure and hospitalization, of course, that was even more impressive and suggesting that this combination really may have benefits. And of course, this will, I think, continue to be examined very closely as people go back and try to look at these observational data sets to try to clarify if that may, in fact, be the case and how much better the outcomes may be. So I think the guidelines, as I mentioned, have really started to coalesce. Whether it's the ADA guidelines or ACE guidelines, really suggesting that in addition to all the lifestyle behavior efforts that really, for those individuals that have evidence of other comorbid conditions, like heart failure, cardiovascular disease, kidney disease, that you really should be considering SGLT2 inhibitors or GLP1 receptor agonists, independent of glycemic control. Because these agents just have effects that go beyond the benefits that can be explained by glycemic control. As I sort of alluded to at the beginning when I was showing you, there's more modest effects that occur with the glucose lowering studies from ADVANCE, ACORD, and the VADT. And as I said, the same thing really with ACE guidelines. So coming back to those cases, so we have our diabetic patient who is obese, had an MI, has good ejection fraction, GFR is pretty good, is already on a number of medications here that other is addressing her other risk factors pretty well. What might be a useful direction to go? And I think the answer or answers are pretty obvious. I think that a GLP1 receptor agonist would be a good agent to add on for this patient. This would probably more directly affect her high risk for subsequent atherosclerotic cardiovascular events. It would also help with her obesity in terms of increasing her weight loss. And you would probably have to, though, be careful about cutting back or plan on cutting back on her insulin because there were those episodes of hypoglycemia. We don't really know when they were occurring. So we'd have to get a better history to try to determine how to best adjust those. But often, you'll decrease your insulin by 10% to 20% or so depending on what the basal A1C is doing and how at risk they are for hypoglycemia. You would not go wrong, however, with adding a SGLT2 inhibitor if the GLP1 receptor agonist was not possible for one reason or another. And it would not be unreasonable in some situation to consider using pioglitazone, although in this particular case, I would probably not want to start off with that because of the degree of obesity. And you might consider that depending on whether there was continued concern about risk for cardiovascular disease down the road. So that's case one. Case two, the patient had a lower ejection fraction and also had a lower GFR, indicating some type of renal disease. And I think in this type of case, we probably would want to reverse the order of those possibilities. We wouldn't want to use pioglitazone given the low ejection fraction. And we'd probably want to start with the SGLT2 inhibitor. And epigliflozin would be a perfectly good drug to use in this setting. And potentially, you might want to add a GLP1 receptor agonist down the road. But this would probably be a very reasonable approach for a patient like this. Obviously, when you're considering SGLT2 inhibitors, there's obviously things you need to consider. Obviously, we don't know as much about these agents in pregnancy or nursing. Certainly, type 1 diabetes patients would be a greater risk for ketoacidosis. And I think that there are some people that are trying these in type 1 diabetes patients. And certainly, we're all seeing our type 1 diabetes patients now gaining a lot of weight and starting to look more like our type 2 diabetes patients. And so I think there's certainly continued interest in whether some of these other agents that can be associated with weight loss and improved glucose control may be used safely in type 1 diabetes, but they're certainly not approved for that. Obviously, the history of frequent genital mycotic infections or urinary tract infections would be concerns for the SGLT2 inhibitors. And of course, the risk for Fournier gangrene, which is a rare but pretty serious side effect. OK. In the last couple of minutes, I want to talk about just some new concepts in CVD and diabetes. I think we've pointed out that even lowering A1C by itself doesn't appear to be enough to reduce the risk for all of the complications that we'd like to see improved in our diabetes patients. And there's other reasons to think that glycemic variability may be relevant. Even our type 1 diabetes patients that are at A1C goals seem to be at increased risk for cardiovascular disease. There's pre-diabetes patients with good A1C levels are developing complications. And we have lots of theoretical reasons. And some lab-based studies demonstrating a glucose fluctuation may, in fact, lead to mechanisms that will cause vascular injury. And finally, glucose variability is closely linked with hypoglycemia, which we know is a risk factor for both micro and macrovascular events. And so what do we mean by glucose variability? There's a lot of different ways to look at this. One of these is sort of long-term glucose variability, looking at the visit-to-visit change in your A1C, your change in glucose. These are just different ways that you can calculate these. You can calculate standard deviations, coefficient of variation. You can do ARV, which actually determines the fluctuation from visit-to-visit as opposed to the fluctuation over long periods of time. And this gives you an idea of what this might look like. So here's a patient that, over time, is actually improving their glucose level. But there's a lot of fluctuation that occurs. And if you then, even just measuring periodically those visits, if you capture that level and then you calculate the variation, you can then get an estimate of the glucose variability over time. And in this particular case, it shows you an example of the A1C is getting slightly better, but the variation is not improving over time. So we looked at this in the VA diabetes trial. So in this case, we're looking at coefficient of variation of variability, as I just showed you in that prior slide. And you can see that the higher your glucose variability, the greater your risk for developing cardiovascular disease events. And shown in the right here, you can see that the risk in these hazard ratios shown here, this is for the cumulative mean glucose. This is whether you look at the maximum glucose that these people achieve. So not significant and only slightly increased risk. But when you start looking at the variability of these measures over time, you start to find that they're associated with increased risk. And this is adjusting for the cumulative or mean values. So this is above and beyond the average A1C or above and beyond the mean glucose over time. And as I pointed out, glucose variability is associated with increased risk for hypoglycemia. So in the VADT or wherever we look at this, we find that the higher your variability, the greater your risk for hypoglycemia, which itself, as you know, is an important determinant of cardiovascular risk. And so, but interesting, even when we adjusted for hypoglycemia events occurring during the VADT that the risk for cardiovascular events was still increased with glucose variability estimates. And so it's not explained completely by hypoglycemia events. And we looked at mortality, same thing. The higher glucose variability shown in quintiles here, the higher your risk for death. And we found that actually glucose levels themselves were better than the fluctuation in A1C. And that's been sort of our pattern, although some investigators have found that variation in A1C still gives you some increased estimate of risk. So will treatment of glucose variability improve outcomes? It's obviously a big and important question. We don't really know the answer to this yet, but I think we're getting closer. I was hoping to actually have some data for you today, because we've been looking at use of CGM in the national VA health care records, because we think, obviously, that may be the best way to really look at glucose fluctuation with continuous glucose monitoring. We identified nearly 25,000 new CGM users across the country. These include both type 1 diabetes patients and type 2 diabetes patients, of course. But in type 2 diabetes patients, we've started looking at outcomes now, comparing them to those individuals that did not get started on a continuous glucose monitoring device. And what you can see is that there were pretty nice decreases in hemoglobin A1C six months after starting your CGM at 12 months, or even if you just look at the latest A1C that occurred over that 12-month period. So these are drops of around 0.2 to 3 in terms of A1C values. Hypoglycemia appears to be reduced as well as in these individuals. And we're now looking at hospitalization, and all hospitalization appears to be decreased in these individuals starting CGM. I was hoping to have the glucose variability data, but we're a little bit behind in that analysis. But remarkably, it looks like death is decreased in these individuals. And I keep trying to tear this data apart, and I can't find any reason that this data is not valid that we're actually reducing death when we prescribe CGM in these individuals. It's really quite remarkable. And this is in type 2 diabetes, where we have much less information. In type 1 individuals, these effects are even more dramatic. Well, glucose isn't the only thing that does vary, and so blood pressure varies quite a bit. And so we've been looking at blood pressure variability as a predictor of cardiovascular disease and diabetes outcomes. And so some other individuals, and you can see a whole series of studies looking over time at variability in systolic blood pressure, in this case, as associated with increased risk for cardiovascular events. And again, in most of these cases, they're taking into account the mean levels of blood pressure. So this is on top of the effects of overall high or low blood pressure. And this is now some of our efforts looking at heart failure in relationship to blood pressure variability. So again, in this case, we're looking at, on the top set of graphs, coefficient of variation of blood pressure variability. On the bottom set, it's looking at visit to visit, the variability between your different visits is another measure of variability. And for both systolic blood pressure and diastolic blood pressure, you can see by quartiles, the risk is increasing for heart failure as your blood pressure variability is higher. Again, when we put this in the models and we look on the model on the far right, we're adjusting for everything we can think of and the mean blood pressure for these individuals over the course of the follow-up period. And we're still seeing this significant increased risk for heart failure in relationship to higher blood pressure variability. Just recently, we've been looking at this in the multi-ethnic study of atherosclerosis. This is a community-based study, and they have all sorts of wonderful measures that we can examine. And so we've been looking at blood pressure variability in relationship to biomarkers, in this case BNP, really good indicator of heart failure. And so whether we're looking at those that are ending up achieving a high over time, over 10 years, elevated BNP at 125, or an elevated BNP of 450, you can see on the models on the far right, that fully adjusted, including for mean blood pressure in these individuals over time, that blood pressure variability is predicting even the subclinical measures of heart failure in the individuals that are relatively healthy when they enter this community-based study. Similar pattern actually was seen with troponin T, which is a marker of cardiovascular injury, that blood pressure variability, again, predicting the rise in troponin as an indicator of heart injury. So to conclude, obviously, comprehensive management of all risk factors is important in our diabetes patients, and that's what makes it such a tough job for us. We have to affect and make an impact on all of these risk factors. As I pointed out, there's modest effects of intensive glucose lowering in type 2 diabetes who have advanced diabetes. And I think that some of the evidence really continues to suggest that you're really going to have better effects with glucose lowering if you start early and you aim for the long-term reduction in A1C, not waiting till they get to their glucose levels and A1Cs that are elevated for years and then starting to make an impact on that. You need to avoid hypoglycemia. That's an important target. And I think you really have to personalize your approach in terms of glucose lowering for individuals. And I would suggest that glucose lowering really needs to be considered three legs of the stool, that you need to reduce A1C, you need to reduce hypoglycemia, and you need to reduce glucose variability. I think all three of those things are probably contributing to development of vascular complications. And then, of course, there's a growing list of diabetes medications that have been proven benefit on cardiovascular outcomes, on heart failure, and, as you know, on renal outcomes. And we want to incorporate those as early as possible in our regimens. And I think we still hold out hope that maybe metformin is actually doing more than just lowering glucose and helping, to some extent, with body weight and may actually have some additional benefit on top of its glucose-lowering benefit. So I'm going to stop there. And thank you for your attention. Thank you. Thank you. Thank you. Thank you so much. We have time for a few questions. Peter Fabulous. Great work. Is there any connection between the variability in blood pressure and the variability in glucose? Are these two things that are kind of, is it the same person having the same episodes? Or is it a similar mechanism? Is there some commonality there? Yeah, it's really a good question. One of the things that we think about is, is this just medication adherence problems, right? Is a patient that has high glucose variability, are they less consistent with their medication? Is that carrying over to other risk factors, like blood pressure variability, for example? And so the same person is developing all of these variability issues. We don't know for sure, but when we try to adjust and look for the medication compliance, that doesn't appear to explain it. But I want to take your point and go a step further, which is that we think these things may be additive and that when we now are starting to look at combining the variability that we see with blood pressure, that we see with glucose, that we see with lipids, it actually looks like there may be some additive risk. And so they appear to be somewhat independent of each other in terms of that larger contribution. Could this be related to catechols or cortisol? Your body is sensing something's wrong, and so? Yeah, that's an interesting thought. That may be worth looking at. Another question, is this a good reason not to use a lot of short-acting insulin? I cringe when I see, especially an obese type 2 patient, and they're hammering away with short-acting insulin. I go, why are we bothering? I mean, really. Right. I do think that that type of variability, for lots of reasons, has problems. And I agree with you that that may be another reason that some of these newer agents may be beneficial, because we can lower glucose without getting this type of dramatic up and down, reducing hypoglycemia, but also maybe these long-term effects of variability changes. Because we tend to see less variable glucose, for example, when we're using agents like SGLT to inhibitors or glucose receptor agonists. In contrast, obviously, to insulins or sulfonylureas, exactly. Wonderful work. Keep it up. Yeah. Hello. My name is Henry. I'm from the University of Panama at Birmingham. I have a question. I have read some case reports about they are questioning the safety of GLP-1 receptor agonists in patients with advanced heart failure without CAD, knowing that perhaps GLP-1 receptor agonists may not be safe on these patients. And I don't know if there is data about this, or what's your opinion? When you say advanced cardiovascular disease, what do you mean? Advanced heart failure without CAD. Advanced heart failure without CAD. Gosh, I don't really know. Certainly, GLP-1 receptor agonists can cause heart rate increases. And so I think there's been some concern that, in some individuals, that may be a problem. And maybe that is being captured in this group that you're describing. I mean, the heart rates increases are usually relatively modest. But in some individuals, it can be 10, 15 beats per minute. And so that may be putting some individuals, particularly if they have heart failure, at risk. And so that could be a possible connection that you're alluding to. Thank you. Hi. I just have a quick question about the blood pressure variation. Oh, blood pressure variation? Yeah. Because sometimes we see these patients who come in, and their blood pressure is elevated. When they come to the doctor, they have, quote, white coat syndrome. And then they go home, and their blood pressure is much better there. I was just wondering whether anyone had looked at whether that white coat syndrome really is benign, like sometimes we say, or whether that is part of this blood pressure variation that could cause issues. Yeah, people have looked at different times of the rises in blood pressure that people have. And there certainly have been links with increased risk, even in those that have white coat hypertension. And I do think it may reflect some broader sensitivity to stress and to catecholamie release and other things like that. And so I do try to pay attention to that. I don't think we have super data on that yet, but yeah. Thank you. It does turn out that the blood pressure variability complications or the links with outcomes do seem to be affected by what your baseline blood pressure is. And so if you have low blood pressure, and you have a lot of variability, you're even at greater risk. And so I do think that this could be, in some cases, you could be dropping your diastolic blood pressure during these NADERs low enough that you may be potentially creating some coronary ischemia, because you're dropping below the sort of perfusion pressure required to keep the blood flow through the coronaries. And so I think these people that have a lot of fluctuation, particularly if they have some damage already, and sometimes we treat these people overly aggressive and get their blood pressure down, and then we have some fluctuation, I think that may be part of what we're seeing in these blood pressure variability issues. If there are no other questions, we'll end the session. And thank you so much, Peter, for an awesome session. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you.

cardiovascular disease

diabetes

assessing risks

prioritizing therapies

medications

metformin

thiazolidinediones

GLP1 receptor agonists

SGLT2 inhibitors