Hello, everyone and welcome to ATT D 2024 in Florence Italy. It's really our pleasure to welcome you today and to have you a part of this symposium CGM, expert boards. Moving beyond A one C, this is gonna be an interactive, a fun session. We really look forward to your participation. We just have an excellent uh faculty today. Joining us, it is the CME certified symposium and there is commercial support uh by Abbot Diabetes Care. We really have a excellent program faculty today. I'm very excited, my friends, colleagues. I think you're gonna enjoy this uh interaction we have today. Uh You look at our title here and I'm starting off with the move beyond a one C component of this discussion. Then we're followed by and I'd like to, to welcome Doctor Jennifer Sheer, um who's a professor of pediatrics, uh and pediatric endocrinology at Yale University in New Haven. Good friend of mine, an amazing researcher and clinician. Uh welcome Jennifer and she's gonna be talking about these advanced technology portion uh of this uh uh system and what's new and innovative. And then, and then I'd like to welcome Doctor Meta Munshi who is a director of the Joslyn uh Geriatric Diabetes Program, a Geriatrician at Beth Israel Deaconess Medical Center, Associate Professor at Harvard. Uh and it's just perfect timing to have her fill in the spectrum um of individualized personalized care um in this symposium. So great faculty get ready for some interaction and discussion. Uh And let's launch into it. I'm starting off the talk. I am said as covering the first portion of the beyond A one C and a little bit of a road map of improving outcomes for diabetes care. So the first step on the road map that I'd like to address is my disclosures there they are. Um And the second step is really to, just to say in broad terms, what amazing progress that we've made in the CGM field with uh small and accurate and easily usable sensors. Uh So it's really an exciting time to explore the effectiveness of this technology. I was given the opportunity last fall to put together a road map for CGM since we've been doing this for 25 years now, uh it seemed time to map where we started, where we've come and what we still have to achieve. Um So take a look at this uh sometime if you have a chance innovations, innovations implementation, and the road map from the artist's point of view looks like this starting with CGM approval and we're not going to deal with all these details. But since we're in an interactive interactive symposium today. I thought I'd lay it out this way where it's a little more visual and just say yes, it starts 25 years ago with the approval in the US, at least US centric view of CGM. Where would you put the final step? Number 12. If you made your way around this road map, what's the end point of a road map? Well, for me, it was pretty obvious. Uh but it's a fun exercise to do. I thought we should end with what I'm calling the quintuple aim. And I didn't make up the term. It's a term in health care, but it really applies nicely to diabetes, equity and diabetes care, good quality diabetes care, reducing patient burden, clinician burden and working on getting the cost of diabetes care and those complications down. How can CGM influence this? That's gonna be what our symposium will reveal uh and discuss with you today. Yes, there's a lot of details. No, we won't discuss all of these, but it's a fun exercise to follow this track of developing metrics using metrics innovating on those metrics trying to reach this quadruple aim, quintuple aim. Sorry, I'm gonna talk about a couple of, of these components and the digital quality metrics. As I, as I mentioned, Doctor Munchi will talk about really personalizing diabetes care to a particular segment of the diabetes um uh field and uh and doctor Sheer will be talking about one of my innovations that I listed uh continuous ketone monitoring. So let me start over here with the CGM, uh digital quality metrics. Uh You're all familiar with GM I and time and range and time below range. Um And I'm gonna talk a little bit about time and tight range, but we have to start with just the elephant in the room. Well, isn't it a one c that we're all dealing with today? And this is a slide or a va variation you've seen. But we just have to start here to say, is CGM really the way to manage diabetes? Yes, it's a good metric for your risk of complications. But there are some accuracy problems you all know about uh hypoglycemia, hyperglycemia, uh hype and variability of glucose are not well reflected in an average glucose measure and sometimes not well appreciated. Is this number three, we've been using CGM. Uh We've been using a one C sorry for uh decades now, the guide management and look at the right hand side of the slide. We're still not at what anybody would say is an acceptable level of glycemic management overall. So using A one C has not led us down the path to really effective management is CGM any better. And you've seen some of these, this data before, but just to say CGM achieves a better A one C than using A ONE C to guide management. That's what that upper right hand shows CGM reduces our risk of hypoglycemia and CGM, which the insurers are particularly interested in reduces acute events, emergency room visits and inpatient hospitalizations. So it's filling the bill quite effectively. And if I put my little grid up here to say, how are A one C and CGM doing on measurement? Well, they both are pretty good measurement tools to tell us about our risk. But how about management and by management, I mean, guiding motivating, facilitating effective management. I don't really think A one C has that component where CGM does. OK. Back to the grid for one minute of the, of the roadmap GM I time and range time below range. I can do a lot using those three metrics for management. But people say, well, aren't there any new metrics? What else is hot today in the field? Well, this time and tight range, uh people are really saying you should add that to your road map. Well, it is on my road map but it's uh it's one we need to discuss for a minute. And I want to try to convince you that GM I time below range, time in range um time and tight range may fit in. We need to add average glucose and CV into our discussion. So let me talk about time and tight range. Did you know in the past few months there are six articles on time and tight range. Two of them published in the end of 2023 4 of them published just within the last few weeks. So let me just take you through a couple very quickly. Uh 13,000 people using an automated insulin delivery device published in diabetes care um at the end of 2023 and on and, and, and press in 24 look what they showed from baseline to um overall and with optimal settings in this automated insulin delivery system by Metronic, they went from time and range from 62 to 80 time and tight range, which they were really interested in from 37% up to 57%. And they established that they thought the time and tight range target should be around 50 which is equivalent in their study to uh time and range of about 70 75%. Then there was a nice study by Roy Beck. I worked with him on this paper. Uh looking at how do those two correlate because people want to know how correlated is time and range the time and tight range. So they have a really nice figure 94% for the R or 0.94 for the R uh measure of correlation. Um And you can see that they correlate, they vary a little bit depending on the coefficient of variation. And I'll come back to that. Here's one of their tables. I know this is way too much. So I'm just gonna show you this part to say time and range from 20 to 80%. What is the time in tight range you're likely to get? And here it is below, you can see it varies from as little as 11 to almost 30% difference, but somewhere in 20 to 25% difference, but that difference is depending on the coefficient of variation or the time below range. Now, if I just show you that one example of 70% you'll see that time and time range may be 45% or it could be as low as in the thirties or as high as near 50 depending on how much variability in the glucose. So keep that in mind, that's an important uh factor of correlating these. Then there's two more studies that I think are really important. These were uh sponsored by um uh avid diabetes care who who pulled together look at the number 20,000, 20,000 users, type one and type two. And because they had so many, they actually could develop a model of, of how these um data points from CGM interact and correlate with one another. This is the biggest uh correlation. I learned a ton working with them on this, of how this correlation of these metrics work. And I'm gonna weave it into my next this talk and the next one. Um but they had a table two. And again, I'll just emphasize 22,000 patients. We had type ones, we had type twos, um on pumps or uh and we had type twos on basal insulin and people not on insulin. So a large spectrum of people with type two diabetes, which really hasn't been looked at before and correlated. And you see the mean glucose down below. It's interesting that the mean glucose can be very similar and the time and tight range can be different again, depending on the coefficient of variation. How much those date those numbers spread out. OK, let me take this figure and I know we said we're gonna do a lot of cases. This is a sort of an odd case way to do it, but I'm gonna show you this one. And this figure um you may say, what is this? But it's time and range, plotted against average glucose for these 20,000 people put into a model and, and it comes out just remarkably uh interesting to me. So here's my case. Uh in this, in this particular situation, here's a person with a uh average glucose of about 210 that would be an estimated A one C or approximate A one C uh on the upper, upper eights. And they, and you say, and I say, well, gosh, we got to improve this person. What should we do to optimize this person's management? Um Well, I'm gonna introduce CV and average glucose here. You might say, well, gosh, let's just improve their, their glucose variability because right now, they're on this line of glucose variability of 40% they have 40% variability in their glucose. So let's just improve that. That'll get them better. Do you know what happens when you improve the CV? If your average glucose is 210, you actually reduce your time and range. That was an amazing learning for me. I thought I knew something about these metrics. But if you improve the CV, when you're running 210 you just tighten it up and you actually have less time and range. So this person needs to improve their average glucose. Uh If they have that level of variability uh to get a better time and range, to get up from 40 to 50 to 60% then they can increase their variability, they can reduce their glucose variability and go from 40 to 30% variability. Um And they could move right up there. Now, I know many of you are saying, oh, what do you mean reduce the variability? That's impossible. That's hard. That's a theoretical. No, I spent a long time just putting this out to figure out to say, how do you reduce variability? Well, just ask these three questions, the type, the timing and the amount or the intensity of their medications, their food, their exercise, their stress and their sleep. And if you ask those questions, you can really start to reduce that variability. Um I highlight just a couple like are you using a drug that attacks post meals and minimizes hypoglycemia. Are you giving a bolus if you're on insulin before the meals and not missing those bolas? Are you counting your carbs properly? Could you take a walk after a high glucose to minimize that excursion? All of those, reduce your variability. Now, I'm not the only one to think of a table for this. Uh you know, Adam Brown had 42 factors that affect glucose and most of these affect glucose by affecting the consistency or the variability. So there's our, there's our case patient who improve their average glucose then worked on their variability. And let's say they wanted to just improve their average glucose some more and move along this curve. They start to run into the fact that just improving the average glucose without reducing the variability, you're gonna start falling off this curve and having hypoglycemia and you might wanna actually go to time and tight range curve. So here's the same curve, time and tight range versus average glucose. And here's our same patient when you get to an average glucose of under 150 around 140. The time and tight range curve is a lot steeper. You'll get a lot more increase in numbers than using time and range. So it's a good time to switch over to think about time and tight range. When your average glucose is down under 140 then you can go up from 40 to 50 to 60% as opposed to edging up uh and very slowly in the time and range. And you can increase, uh you can improve the glucose by reducing the variability further. So that's one indication of when to use time in tight range. And I'll come back to it in, in my cases, in a moment. Ok. Is that, does that solve the issue about time and tight range? Well, let me just put my grid up again now and say time and range versus time and tight range, which is a better measurement. Well, they both are very likely to measure your cardio, your, your complication risk. We need a little more data in time and tight range because it's brand new. So we haven't analyzed all of our data for that factor, but the data is starting to show it is going to be just as good and some think it should obviously be a better marker of complications because it's really in that normal range for management, which one is better to guide management. Well, I showed you lots of graphs and tables and charts that I know were hard to read. And I don't think that's the real answer. The real answer is how about looking at an A GP for those six studies that we talked about, there have not been a single GP shown and I think it's time to show some GPS to decide which is the most effective. So here's a case with a patient, uh whose time and uh and range targets are 70 to 180. Here's the same patient, 70 to 140. And all of a sudden you can see where you might have been happy here. 71% time and range pretty good. A one c when you plot it in time and tight range, it's so obvious that they're above the target and there's room to have a discussion about where to intervene, where it's a little fuzzier here. So I like that when you get down near 7% plotting it this way to guide or motivate one to improve. Here's another one, same thing. And this one, this patient was eating just to keep the postprandials under 180. I call it eating to target. But when I said no, no, that's ok. You could, you don't have to eat up to that target. You can, if we plot it this way, maybe a few less carbs at breakfast at dinner would, would actually be better. So I like this plotting when you get down near the normal range, maybe we should just move the time and tight range for everyone. Well, look at this. If your A one C is nine, it doesn't matter which one you use, I would just stay with time and range if your A one C is 8.5, it doesn't matter if it's, if it's eight, it really doesn't matter if it's 7.9. It doesn't matter only when you start to get close to seven. Do you start to see that the time and tight range? A GP yields more um guiding or facilitating information? Here's another one where you're down close to seven and the time and tight range reveals where your action might be focused. And the daily views a 170 to 140 show the same thing they highlight where action may be necessary. So we'll close by doing the grid again. And just saying, I think both of these are good for measurement. I think for management, they both are good too. But if, if you're managing to, to a time and range of 70 an average glucose of 150 this is fine. But if you wanna get a time and tight range of 85 you wanna get an average glucose of 130 using time and tight range can really be helpful. I think eventually we're gonna have a GPS for each of these and I'll close by just saying time and range or A one C for management. I think it's time and range. Both of these are good guides and it really depends on what your goal is to which one can help you uh achieve your targets. And so I think there's a place for time and tight range when your targets are tight. Um It can help facilitate that management. So I thank you for your attention and, uh, we'll delve into questions when we get into our cases to follow. Thank you very much.
Related Presenters
