Video Beyond A1c: Using CGM to Provide Precision Diabetes Management Play Pause Volume Quality 1040P 694P 554P Fullscreen Captions Transcript Chapters Slides Beyond A1c: Using CGM to Provide Precision Diabetes Management Overview CONTINUE TO TEST Back to Symposium Hello and welcome to see G. M. Clinician guided type two diabetes management program. My name is Greg Simonson and I'm director of care, transformation and training at International diabetes center. This is the first webinar of a three webinar series. Our first webinar today is really looking beyond A one C. And using C. G. M. Or continuous glucose monitoring to provide precision diabetes management. The 2nd 2nd webinar is really looking at the practical aspects of starting C. G. M. In your practice. Our last webinar will be how you use C. G. M. And the ambulatory glucose profile or a G. P. Report to improve diabetes management. It is my pleasure to welcome our first presenter for Webinar one Dr Richard Berg install is executive director of International Diabetes Center. He is an adult endocrinologist and really a key opinion leader in how we integrate diabetes technology to improve diabetes care and education. I turned over to Dr. Berg install to lead us in Webinar one. Thank you. Dr Simonson. It's a great pleasure to kick off this series and over the next 20 minutes I'd like you to just think about whether you feel C. G. M. Does provide more precision diabetes management than using the A. One C. Which were used to doing over the last few decades. These are my disclosures at the International diabetes center. We're all about ensuring that every individual with diabetes receives the best possible care and with every individual that means really individualizing our approach. So let's take a look at the options we have available to us over the last few decades. We know that looking at the glucose is a critical part of managing your diabetes. And we've had the A. One C. For the last 40 to 50 years we've had various forms of finger stick glucose testing, Killing in logbooks or sending the data over the cloud. But it's often just blood sugars in the morning and not a whole picture. And over the last 15 to 20 years even we've had continuous glucose monitoring that has really evolved and that's what we want to focus on today and how that can really give you a much better look at your blood sugars over time. Here's a typical log book. I know many of you listening have seen a similar thing. Sometimes it's written on a napkin. Sometimes it's actually in a log book but there's a lot of morning sugars but not a whole picture. When you look at those individual blood sugars you might have they might look good. But down below is a profile with continuous monitoring and you see all of a sudden weight there were lows I didn't know about. There were highs I didn't know about just by poking my finger intermittently. And so along comes C. G. M. That I told you about and the companies that make the C. G. M. Put all this data together and there's a lot of data. Remember we're talking about every five minutes or now even every minute having blood sugars. So the initial reports were 20 to 30 pages long. That was a lot of data. And so the international diabetes centers and other places around the world said, well can we organize this data a little better? And at our center we called this the ambulatory glucose profile. Report a one page summary. A way of getting started with the data. Yes. There's other innovative ways to look at the data but this gives you a snapshot of 14 days so you can really start to hone in on appropriate management. But the question comes up again and again and I'll bet you've asked it out there as well, are you sure we need all that data? We like your single report much better than 20 to 30 pages. But I've got an A. One c. Can I just use that to guide my management? Well, the best way to answer that question is just to take an example. Here's an a one c of 6.7. What should I do? Well let's go to the next side and this shows a couple of cases you see on the left hand side three different glucose profiles and you'll be experts very shortly if you aren't already on how to look at a profile. But I think you can tell at a glance the top profile looks pretty good. It's within that 70-1 80 range mostly. And it's pretty flat. And the one here there's some highs at the midday and the one at the bottom, there's lows overnight and there's highs at the midday and high at the end of the day, they look really different. And I think any of you watching would say. I'm gonna treat these three patients very differently. But you might be surprised or maybe not if I told you that all of them had the same A one C. So I think just knowing these two things, I hope reinforces that for management. C. G. M. A glucose profile is much more helpful than an A. One C. A one C. Might be a good marker of long term risk but not so much for management. As a matter of fact, the lower profile has nine times the rate of hypoglycemia than the upper one same. A one C. The lower profile has twice the glucose variability that spread around the median line than the upper profile same. A one C. And very different time in range is 70 to 1 80. So I hope that reinforces the point for management. I'd go with the C. G. M. Data. There's a few other limitations to the A. One C. Besides not telling you the whole picture, we've used A one C for decades and we're still struggling to get our A one CS are glucose is in the target range that we're trying to achieve for good quality care. The A one C. Has some other issues of not really showing hypo or variability which I showed you the last slide and many individuals have different characteristics that make the A. One C. Not very accurate. Iron deficiency liver disease renal disease changes in the red cell lifespan which I'll come back to in a minute. So that made me a few years ago to sit down and write this commentary saying I really think this C. G. M may transform the way we manage diabetes but it's going to be step by step and indeed I think I was right. It's been step by step. It's taken us several years to get to where we are now but we're really entering a phase now where I think C G. M is really starting to be an effective tool and just look where the C. G. M. Has come. We now have different manufacturers that are making C. G. M. That are small that are convenient that are accurate that are linked to your phone or to your watch. There's even one company that has an implantable C. G. M. But the systems are getting so much better and so usable that I think now we're ready to show that they are effective and to put them into our daily workflow. So let's start with that first statement are affected. It's nice to have a good system but what are the studies show if you use the systems, do you actually improve the A. One C or the equivalent the time and range. And this is a meta analysis of randomized trials. And it does show that the A. One C. Gets better time and range improves the few places where the A. One C. Seems to stay about the same. Is because in those cases we were dealing with trying to reduce the hypoglycemia. The individual was already in a decent overall glycemic control or A one C. And I'll reinforce that on the next slide because many of you primary care in particular are really concerned about hypoglycemia. So here's the study a series of studies that show using C. G. M. Hypoglycemia was improved. This isn't millie moles. Here's in milligrams per deciliter whichever you're comfortable with. But this shows that over and over again using C. G. M. Less hypoglycemia. So improved A one C. Time and range and less hypoglycemia makes me feel like, yep this is a good tool let's learn what it's about and how to use it. The remaining question does does it prevent some of the longer term costs and more more morbidity and mortality ease? And here's a study that I was part of looking at a large database and we looked at thousands of people who were on insulin use and and before they used C. G. M. In the blue and then they got C. G. M. And we looked for six months after. They used C. G. M. And we compared what were their acute inpatient outpatient emergency visits before and after using C. G. M. What were their hospitalizations before and after. And you can see a pretty dramatic separation in these two curves. You know as 61% reduction in acute events and a significant 30% reduction in hospitalizations. So now we're putting the picture together to say this is therapy worth getting to know and getting built into your practice. Not to show you one last study because this was really an important one. It's people who are on insulin particularly with type two who struggle the most to get their glycemic control where we want it to be. It's just tough regulating that and adjusting that insulin. And so here is a study. People on insulin not a goal comparing C. G. M. Two B. G. M. And you can see that after eight months the C. G. M. Group had a 80.4% lower A one C. Than the B. G. M. Group. And so this was an important study to start getting C. G. M. Covered. Um more uniformly for people on insulin whether it's basil or basil bolas insulin. So we have good devices. We have good evidence, we're getting better and better insurance coverage for the systems due to randomized trials like this. And this one was a not a randomized trial. But this was one of those real world studies. It's really important. This is out of the Kaiser system which you know is a big respected healthcare organization and this is from the California Kaiser group. And they took now thousands of people with Type one and type two diabetes again before and after they're on C. G. M. For a 12 month period and both Type one and type two is using C. G. M. Reduce their A. One C consistently uh And and and and and statistically significantly and they also reduce their hypoglycemia that would lead to emergency room visits or hospitalizations. So now the database is pretty robust from R. C. T. S. To real world studies to saying this seems like a pretty good tool step by step to improve care. So now let's spend the next part really looking at that at that data that comes out of the C. G. M. Just so everyone is familiar and the best way to look at it is what does the american diabetes association say? And their standards of care are the 10 core metrics of the hundreds of possible metrics of C. G. M. They highlighted 10 that we all as clinicians using C. G. M. Should get to know and I'll just walk through them rather quickly. But just to give you a sense You need adequate data 10-14 days, 70% of the days you are measuring it. And that's not just made up as somebody's idea of adequate amount of data there. Studies behind that to say say when you get to 14 days on the dotted line, you're starting to level off with your time and range and you're mean glucose and your G. M. I. That's a pretty good indicator. Not much different than if you go 30 days or 60 days out. If you're really interested in hypoglycemia, you might collect a little more data 30 days or so. But this is sound data to say if you've got two weeks you can make a good clinical decision uh in most situations you know what the mean glucose is. And then here's another term I want to spend just a little extra time on called the glucose management indicator that maybe a new term to some of you. But I think you'll recognize the older term that used to be called the estimated A one C. We changed that name in consultation with the FDA for several reasons that I'll get to in just a minute. But I like to now call it the personalized A one C. I'm going to take a minute to explain what I mean by that because I think this is an emerging concept in the use of C. G. M. And in diabetes management personalizing that care. So we did a paper describing this new term, the glucose management indicator and we put the formula out. You can get it on the websites and you can you can see how it's calculated and convert a mean glucose from C. G. M. This is all C. G. M. Data. Now there used to be correlations done in the past but it was a combination of finger sticks and C. G. M. S. And now it's all C G. M. Based the G. M. I. Does it agree with the A. One C. Well this is very important to understand this principle. It does agree pretty precisely about 20% of the time and it varies a little more by 0.3 about half the time. And on occasion it'll vary 0.6 or even 0.7% difference from your G. M. I. Versus your A. One C. And some people call that a mismatch. And yes they are different but it doesn't mean that one of them is wrong and one of them is right. It just means they're measuring slightly different things. And I just want to tell you, I think the G. M. I. Is really your best indicator of your actual glucose levels your tissue exposure to glucose because you're measuring that average glucose converting it into an estimated A one C or G. M. I. The A one C. As we talked about just previously varies with the red cell lifespan. How much you'd like eight the red cells, some genetic or biologic factors as well. Let me show you one more slide that talks about this lifespan because that's the main reason you get a difference from your G. M. I. And your lab a one C. Is this red cell lifespan. So just look at this slide for a minute. Here we have red blood cells. And over on the left side if your red cell has a short lifespan it's getting hemolytic anemia. Or you have a valve in place that's chewing up your red cells so they don't live very long. Well you've got glucose in your bloodstream but the red cell doesn't have very long to get duplicated for the glucose to attach. So the the A. One C. Is going to underestimate the true exposure to glucose because you're getting exposed. But you don't have time to go like it. The red cell before it turns over and a new one comes out over on this side you've got red cells that last longer than usual. They just hang around so the same level of glucose just has longer time to attach onto them. And you've got to adjust that A one C. Down a little bit because it's not that you have a higher glucose level in the blood. It's just you have a longer living red cells. If you have an average lifespan of your of your red blood cell the two are gonna match very nicely. So let me show you that in one more form here here's a little table. Here's the table about how often they match up to each other. The lab and the G. M. I. Let's say you have a lab a one c. Of 75 and your G. M. I. Is 81. That's a 810.6 difference that happens about, you know, um 19% of the time they're that much different. This indicates that there's a .6 difference. But if you treat that A one C. You would say well I'm doing pretty well, I'm pretty satisfied around 75. So I'm gonna hold and just let it be maybe make a little change. But really your exposure to glucose is at level of 8.1. So you're underestimating that you're putting this person at risk for complications on the other side. If your lab A one C. Is 81 and your G. M. I. Is 75. And you say I gotta get that glucose down because I'm treating the A. One C. You're gonna put this person at risk for hypoglycemia because the real glucose exposure in the blood is more along the level of 75. So that's I think an important concept of why I think the G. M. I. Is a better guiding principle. As long as you're in a steady state, you haven't just changed therapy last week or just had an acute illness with hyperglycemia for those two weeks that you're measuring the G. M. I. Yes, you can go to a table. There's a nice table from Dr. Hirsch and this consensus report on Type one that talks about other reasons that the there may be a mismatch between a one C. And G. M. I. And you can look and try to see if there's something you could correct to get them closer. And I've done that on many occasions but I would say 80 90% of the time. It's not one of these factors. It's just the difference in the lifespan of the red cell and the glucose uptake and like like location rate within the hemoglobin within the red cell. So I'll end with this on this this point. This is where the field is going. We're not there yet. But there are now efforts being made to take a 14 day tracing, taking a one c. Look at those two and come up with a personal correction factor for that individual personalized factor and say we got to apply this because it always seems to be about a 20.6 difference. So your real A. One C level that you should work off us is closer to 74, not eight point oh. And there's a paper about that you can refer back to later. It takes this glucose uptake across the red cell and it takes the lifespan of the red cell. Does some fancy mathematics in a modeling way and comes up with a apparent location ratio or this personal factor. So just keep your eye out for more and more efforts to tie this data together to personalize your A one C. Your G. M. I. So that's the G. M. I. An important indicator uh to use in practice glycemic variability, we use the coefficient of variation. It's one of multiple ways to measure glucose variability and I won't spend a long time on these. But we've just shown that C. V. Is the most reliable. Clinically major is the up and down and mod is the difference between days that are given given time or on average between days. But the coefficient of variation doesn't vary with the mean glucose. Uh It's it's um it's a steady measure of variability no matter if you're mean glucose is lower or higher. But the C. V. Does correlate nicely with hypoglycemia. And you could picture that if you have a lot of variability, you're likely to be drifting down into hypoglycemia on occasion. So we like the C. V. As a measure of variability. We include that in the ambulatory glucose profile. And then finally the last five of the 10 core metrics are the time and ranges. I know that sounds a little funny to call it time and ranges. But time and range T. IR is really A name for the time. In the target range 70-180 or 3.9 to 10 million moles per liter. Then you have two times above. Range, two times below range. You can call them high and very high low and very low. Or you can call them level one and level two. But it distinguishes this whole time and range bar that I'll show you in a minute. So once you know the 10 metrics you want to know well what is that time and range? How does it correlate to the A. One C. Because yes, I'm very familiar with the A. One C. And I just want to get comfortable knowing what time and range means relative to the A. One C that I've used to. And so I'll just give you these key numbers a time and range of 70%. That's 70 to 1 80 70% is about an A. One C of seven. This is just on average, your patient will vary a little bit from that one way or the other and then 50% timing range is an A one C. Of eight. So you can see that 20% change in timing range 1% change in A one C. So just good to get familiar with those metrics. And then finally, not only do you need the metrics, you need to organize them so you can now look at them in a report and that's what I referred to earlier as the one page ambulatory glucose profile report. This is what it looks like. You're gonna hear a lot more about this in the sessions coming up. I'll just lay the groundwork for it here. There's three panels, These metrics on the top and the targets. You want to aim for a profile of two weeks of data. All put as if it were one day from midnight to midnight with that media line and then the variability around it and then finally the daily views each of the days that you collect data that went together to form the middle panel and to form the statistics on the top. So here's a blow up of that top panel. Just so you you get the notion that these are the 10 metrics that are in the uh A. D. A. Standards of the core metrics. They're all right there at the A. G. P. The five time and ranges and the other metrics that we talked about. So they're all available but it's nice to have the metrics. But you also want to know the targets. Well if I'm 49% I'm in range of my target range. Is that good or bad? So you look over at the target say, well we'd like it to be 70. And here's just another way of showing you that we would like The time and range the green to be greater than 70%. We'd like to time below range under 70 which is a combination of level one and two to be less than 4%, less than 1% under that 54, which is really clinically significant hypoglycemia. So now you have the targets, You have the metrics so you're really ready to analyze it. You just have to realize that for pregnancy it's a little bit different for high risk older individuals. If you look on the right side of this panel we might accept a little less time and range. Like can you just be over 50% but avoid hypoglycemia as as aggressively as you can avoid Hippo. So the time and range targets will vary a little bit for for most type ones, people with type one and type two diabetes without special circumstances. This is the time and range bar targets that we are shooting for. So I'm going to wrap up and just tell you that I think we're really approaching that era now where we got good devices. We've got good outcomes using them and randomized and real world trials. We've got a way to organize the data into an A. G. P. We've even updated the A. G. P. That you'll hear about put some color to it, match the the profile to the time and range bar. So it's just that much easier with the targets right next to the time and range numbers and the A. D. A. Has thought well that is a good way to do it and has put it into their standards of care. So I think we're set now to use this data and you're going to learn more about that in these upcoming sessions how to work it into your practice, how to work through cases using the A. G. P. To improve care. So thanks very much for your attention Published January 4, 2023 Created by Related Presenters Gregg Simonson, PhD Director, Care Transformation and TrainingPark Nicollet International Diabetes CenterSt. Louis Park, MN Richard Bergenstal, MD Executive DirectorPark Nicollet International Diabetes CenterAdjunct Professor, Department of MedicineUniversity of MinnesotaSt. Louis Park, MN
