Video Why Use Continuous Glucose Monitoring to Improve Diabetes Management in Primary Care? Play Pause Volume Quality 1080P 720P 576P Fullscreen Captions Transcript Chapters Slides Why Use Continuous Glucose Monitoring to Improve Diabetes Management in Primary Care? Overview Hi I'm dr Richard Burke, install executive director of the International diabetes center HealthPartners Institute in Minneapolis Minnesota. And today we're going to talk about an important topic of Y. U. S. Continuous glucose monitoring to optimize and improve diabetes management in primary care. These are my disclosures. I think we should start when we're talking about using C. G. M. To be sure we all agree that we need to improve our diabetes management even further than what we're doing currently. This is a really nice article just out in June looking at diabetes trends in glucose control in the United States from 1999 to 2018. And if you look carefully you can see in an orange A. One C. Under eight and blew a one C under seven. That in 2010 We reached our highest level of achievement of these targets. And then in 2014 we got worse and in 2018 we're still not back to our optimal so we have plenty of room for improvement in improving our dedicated hemoglobin a marker of good optimal management. If I bring this a little bit back home and I look at Minnesota and the clinic where I work at health partners where we have over 50,000 people with diabetes who we track on what we call the D five. These are five measures that you'll be well aware of blood pressure cholesterol. A one C under eight not using tobacco taking aspirin if needed. And you can see just at a glance on this slide that we do well with four of the measures. But look at that A one C. Is an outlier. Only 70% of our patients who qualify for this measurement are reaching an A. One C. Of under 8%. So it's a real outlier and we've been steady at this for the last several years. So I call this a plateau that we're on and we've got to get off this plateau in today's discussion is really around whether see GM can improve that A one C. Can we use this as a tool to optimize our glycemic management. I wrote about this back in 2018 and I really did think and I do think that CGM. Is going to transform the way we manage diabetes. But it's a step by step process. And let's talk about a few of those steps today. And the first step is to understand where we are with finger stick or blood glucose monitoring. We all are aware of this. We want patients to know what their blood sugars are. But look what happens usually in the primary care setting and you know this. Well, People will test in the morning and you'll get a bunch of morning sugar is occasionally has scattered other blood sugar but you really have a hard time from this data. Getting a whole profile and no one wants to poke their fingers six or eight or 10 times a day that it would really take to fill this out. So then along came continuous glucose monitoring in the early 2000s and several companies are now making really small and accurate and easy to use continuous glucose monitors. The two biggest ones are Abbott and decks calm and you see their little sensor patches there and there are certain devices and if I showed you a little blow up of this the transmitter and sits on the top of the skin and there's a little filament that goes under the skin that as the sensor on it measures interstitial glucose translates that into a plasma glucose. And you can have a personal CGM you use every day and you look at your blood sugar is the person with diabetes or professional one where the clinic owns it and gives it to the patient and they use it for 10 to 14 days. And then we have a shared decision making session around looking at the data together and you're going to hear much more about personal and professional see gm in the upcoming sessions on continuous glucose monitoring. But look at the difference in this panel, you'll see three finger stick blood sugars at the top. They look pretty good. They're in range. It's nice that we have three values But you look below and you see the continuous glucose monitoring 288 values in a day and you see that the profile looks very different. There's lows overnight. There's highs in the afternoon so you get a much bigger better picture with C. G. M. As a matter of fact the companies who make the C. G. M. Devices have uh put this these C. G. M. S uh into reports and they turned out initially to be 20 or 30 pages long. And many of us including the I. D. C. Have worked hard to get this down to a one page report. Um And we call this one page report the ambulatory glucose profile. Can we shrink thousands of data points into one page that we can use for clinical management that you can use in your practice. But the question comes up are you sure we really need that extra data. Can't we just rely on the hemoglobin A one C to guide our management decisions. I see it in all the algorithms but look at that A one C. It's not an easy concept of glucose attached to the end of a lien of the beta chain of hemoglobin. Um that patients say well that's okay you give me a number, you tell me you want it under seven and I'm at 67. So I guess I'll go with that even if I don't completely understand exactly what it means. But is that enough? That's 6.7 for you to really guide management safely and effectively. So let's look at this. It's always good to go to examples. And these are three patients I happen to know well having followed them for the last 35 years And these three patients each had an a one c of 6.7. But look to the left and you see a glucose profile and we'll talk more about profiles in a minute. But I think you can tell at a glance this is midnight to midnight and you see that profile number one on the top versus three at the bottom are very different. Yet they have the same A one C. And they're so different that the one on the bottom has nine times the hypoglycemia at the one at the top, but one of the bottom has twice the glucose variability. If you measure it by coefficient of variation and you can see that variability with the up and down up and down at the bottom And they have very different time and ranges and that means the time in this target range of 70-1 80. So If we look at this together, I hope you'll agree with me that using the A1C alone really does not give us adequate tool to manage the diabetes to avoid the hypoglycemia. The variability to get time and range A one C has a few other limitations. Um It's really not working as you saw from the very first slide, I started with to get a good A one C. Using a one C as our guide has not gotten us to the ultimate goal. It doesn't show hippo and variability that we just talked about and there are several categories of conditions or patients where it may not be reliable from an accuracy point of view, hemoglobin opera, these iron deficiency, renal kidney disease, different lifespans. And yes people have very different lifespans of their red cell. The average lifespan might match up with the lab A one C. But not if you have shorter or longer lifespans which is really turns out to be important. But yes the A. One C. Is incredibly correlated as a population with complications particularly micro vascular disease. And you all know the D. C. C. T. That showed this better than any study 1983-1993 A one C. Correlates with complications. As a matter of fact, if you look at this nice Graphic from the New England journal several years ago by Ed Greg and he showed that in this era from the 90s to the uh to thousands we were making good progress on some of the complications of diabetes and I doctored up his nice graphic a little bit by putting this red arrow in and said here's where the D. C. C. T. Ended in 1993. And ever since then we've been living in this a one sierra. The D. C. C. T. Show the A. One C correlated with complications. We've stuck with it ever since as our main management tool and population and quality metrics. So if I summarized just briefly in the early 2000s see GM was available and then got approved. It wasn't until six years later that we had the first international conference on standardizing the jeep standardizing glucose report and we called it the ambulatory glucose profile. It was then several years later that the wider community said okay these are the core metrics from the C. G. M. That we really think are valuable for us to all know. A few years later 2019 International Consensus added targets to say where do we want those metrics to be as a measure of good guy Scenic management. And then finally In January of 2020 the American Diabetes Association said Yes you've built a good data base and a good bit of evidence. So we'll put these into the standards of care, 10 metrics time and range targets and recommend the egg report as a one page report to help guide therapy. And then just last month the International diabetes center integrated the C. G. M. Data and the IGP into electronic record hearing the drumbeat of we need the data we need the data we need easy access. So progress is being made on the C. G. M. Front but we come back to the ADP report one page report. Thousands of data points metrics and targets at the top 14 days of a A. G. P. Profile in the middle and the daily views at the bottom. You take those 14 daily views put them all together. You get the middle view of the ambulatory glucose profile. So that's the report we want you to get very familiar with looking at. And those 10 core metrics that we agreed on I think are important for you to know the first two are do we have enough data then the mean glucose then an important one. And you see I haven't highlighted or bold ID. And that's a glucose management indicator. That's what we used to call. The estimated A one C. We gave it a new name uh And I like to call it the personalized A. One C. All of your glucose data. What would the A. One C. B. From the C G. M. Data that you just collected. We have a measure of glucose variability, the coefficient of variation. And then we have five time and ranges. Yes we use the word time and range to note The time and target range of 70- 180. But we also have to times above range and two times below range Categories that you need to get familiar with. We call low and very low high and very low. Some call them level one and level two and then the egg report itself. Well once we had the metrics we really needed to agree on where we want those metrics to be to qualify as optimal or good control or or or effective control. And you see that we want the time and range 70-180 to be over 70% And we like the time above range to be 25% of the time. Very high uh 5% of the time below 70. Less than 4% of the time. Below 50 for less than 1%. So we have targets now to shoot for and a gradual steady process working together. So we come back to the report And we have those targets and let me just blow up the top panel of this version of the report to show you why the American Diabetes Association at least in my opinion, thought that this was a good report because the top panel has all those 10 metrics that were just agreed upon also it also has a target of all for all those metrics. So this seemed to be an important piece of the report. And then I give you the correlation that is 70% in range correlation to an A one c. Of seven Time and range of 50. And anyone see of a just some important facts for you to know how this levels out. So it's important to standardize this data. So we all get used to looking at it no matter what see Gm device we're using. We get used to a standard report so we can efficiently talk this through with our colleagues and with our with our patients. So this is what the a. g. p. looked like back in 2013. Here's what it looked like in 2019. Here's what it looks like today. We put it in color. We move the time and range barred over to the left because people read from left to right will just keep updating it. But the core metrics are the same. The profile is there the daily views So that's good. We've come a long way and agreeing on metrics um and in standardizing them and organizing them into a report. Now we need to learn how to analyze it so we can eventually act on it. So let me spend just 12 minutes on analyzing the report again. Ah telling you you're going to hear much more about this for my colleagues and the subsequent cmi programs on continuous glucose monitoring. So here's the latest version so very latest version of the inventory glucose profile. See GM report. And here's how I like to do it in the very quick analysis manner. You start at the very top and look at that panel and say is there a glucose control problem? Is there room for improvement? And by that I mean look at the green and the red the time in range bar. We want more green. We want less red And we'll even tell you the numbers we want the green to be over 70 if at all possible and the red to be under 4% under 70 and under 1% for under 54. And look at this patient, 46% In the green and 10% in the UNDER 70 and 5% of the UNDER 54. So there is a problem here, both on the high and the low end. So then you go to the middle panel and say okay I agree we've got some things to work on. But where is that problem? Where does it show up throughout the day and here in the middle aged profile, going from midnight to midnight With the time and range 71 80 marks right here. You can see this patient is low and the nice coloring, the shading um shows you this at a glance low overnight. I in the evening we need to work on these areas start with the hippos first because that's a short term emergency emergency short term complication. And how do we want this profile to look? We'd like it to be F N. I. R. I know that sounds funny but it's flat narrow and in range. And then we go to the bottom when we look at the patterns and we see a weekend is different than weekdays. Uh Is there any pattern to look at in particular? Does that low show up multiple times as it does. So let me come back just for one more minute to this flat narrow and in range. Just so visually you can see what I mean. Here is a patient that we look at their egg profile and then several months later we looked at it again and I won't go through all the details of the case. I just want to show you that this profile was up and down there, there wasn't as much green as we wanted, there's too much red. And then when we made changes in their therapy they came back and this is what I mean. We got more green, we got less red and we got a profile that's pretty darn flat and narrow and almost all in range. So that's what we're aiming for in a step by step process and we'll go through many more cases in subsequent sessions. So then once we have a way to analyze it, I think you're next question would be at least my next question would be well if I do those things and make that analysis and action plan, what are the studies show? My results will be well, here's a meta analysis. Yes. Although C G M is a relatively new tool. We have enough randomized controlled trials are good observational studies to be able to show that if you use C G M versus sMB G. Or usual care on the left, the A one C goes down where it doesn't go down. That's because those patients had a decent A one C but they got less hypoglycemia are on the right. Those same studies the time and range consistently went up. So the evidence is pretty good in trials that if you use C. G. M. Both type one and type two you can make improvements in the A. One C. And the time and range. I know primary care is often worried about hypoglycemia as they should be as as we all should be. So here's a nice study just addressing hypoglycemia can using C. G. M. Versus sMB G. Reduced hypoglycemia. These are all in millie mole. So let me put an overlay here of the milligrams per deciliter. And you can see look at all these studies that moved this to the left meaning favored the C. G. M. To reducing hypoglycemia. Study after study after study. So the next question will probably be well that's good. But in addition to getting better numbers I'd like to keep people out of the hospital or not have emergency room calls. So this was a study that there's that the I. D. C. Participated in a database study where we looked at thousands of patients with type two diabetes before they got to see Gm. In blue or six months after they got C. G. M. In red and ran this database and said There was and showed there was a 61% reduction in acute events that was inpatient or outpatient emergency room calls or visits to the urgent care or the er and 32% reduction in those who ended up getting hospitalist. So it's not only numbers it's actual events that really are meaningful and scary and costly. And then finally you might ask okay how about the long term A. One C. Is such a good marker of long term complications at least for a population of people. How does C. G. M. Stack up and I won't go through again in detail. I'll just tell you we looked at the D. C. C. T. And I know that was finger sticks but they did four seven point profiles a year for 10 years and we looked at all of that data and put it together and said if we looked at the time in range of those finger stick values the more time and range the less retinopathy The more time and range here's 70% versus 10% the less micro albumin area or renal disease. It correlated with those complications just as well as the a. one c. So now we have a short term marker, a complication risk marker, a long term marker um That holds up ready darn well so now you might say okay so you standardised you organized we've got a quick analysis plan and if we act we're likely to see decent results but my problem or your problem probably is I need the data I need the data to act on. So we have this little gap here and we call it. Can somebody please be sure we integrate that data so it's readily available to me as soon as I wanted to talk with the patient in the clinic or virtually. And that's what we worked hard on over this last year trying to fill out this model virtual care that you all know so well after we've all lived through this pandemic era of taking sensor data, going to the phone, getting it into a cloud and having an aggregator to get it into electronic records so that the healthcare professional and a person with diabetes can talk about it. So it's nice to have this data we want to get it readily available And so we tackled this by starting with C. G. M. And we started with the Abbott Lee brae Cloud and we use the Redox to integrate the data. We got discrete data elements in the electronic record. We got an A. G. P. Report that shows up automatically at the push of a button in the patient's chart. So we're able to do this with a big team um at the International diabetes centres and HealthPartners Institute working with Abbott and Redox and all of our partners without our throughout our institution. And this was presented at the A. D. A. In june saying yes it's possible there is high hopes for all of us to figure out a way to get this into electronic record to get it into a flow sheep. So you can track data over time, you can highlight it if they're high or low You can see how they compare. These are those 10 core metrics they're all available along with an egg report. So you can have a good management discussion. So in closing I'm going to just try to convince you that we're moving from an A. One C. Management era to a C. G. M. Management era. I hope the data I presented gives you confidence in that move doesn't mean we'll never use a one C again as a marker of long term complications. But for day to day management of that patient in front of you see GM time and range offers so much more information, a more effective tool. So I'm pretty convinced that we're still on the right track for using CGM to transform diabetes care. And I've said it's going to be step by step. We've gotten several steps in place and let me just ask this last question in the last two slides today, what is the american diabetes Association and the Association of Clinical Endocrinologists recommend for you in primary care, diabetes educators, um even specialist seeing people with diabetes who's a good candidate for C. G. M. I'll just take the last minute to show you their their standards of care documents. So the ADA has the standards of care that comes out every january. You probably know and this is their chapter on technology and their section on continuous glucose monitoring. They agree that a one C. And time and range and time. The low range are really important metrics. And they spend the first several recommendations about sMB G. And how maybe it can help us get a better A one C. But then they come to recommendations 7.8 and they say using C. G. M. Focus on training and supporting the patient. 1st use the G. P. Listen to the patient's story. Uh Then they come to the next recommendation they say for people on MD. Or pumps. Type one or Type two. This is really good evidence level A. Or B. They should be used all type ones and type twos on MD. C G. M should be considered as a key tool. Then they come to their next expanded to part of 7.9 and 7 10. Their 10th recommendations says other insolence may be appropriate as well. Which could include basil insulin. But that's A level. See only because we need more data and you're going to hear about some more of this data and the upcoming talks that maybe that basil insulin will move to a higher level of confidence and be clearly covered as well pregnancy. Good day to building we can get it to the next level by just doing more studies and showing how valuable the C. G. M. Can be. Professional C G. M. I just mentioned once more. It's very broad. You can use professional see Gm on any patient with Diabetes you want to to reduce their a. one c. or hyper or hypoglycemia. But its level C. Meaning we need more data to show if we use professional C. G. M. Like this the outcomes improved like we've seen in these other trials and finally probably S. U. Using patients uh patients who want to use CGM to guide diet choices, nutrition choices. Uh That probably will work. But we need more data to get this into standards of care. And even now in the hospital there's an amazing data building and then finally I won't dwell on this. But the american association of Clinical endocrinologists Justin may put out their guidance and instead of showing you the text, I'll just make a little table and say they agreed pumper MD C. G. M. Recommended they have a category called problematic hypoglycemia. For no good reason a person is getting hypoglycemic or reason that they haven't been able to adjust. You see GM pregnancy on insulin used C. G. M. Pregnancy not on insulin. They thought we needed a little more data um to to build that database and less intensive insulin baby recommended we need more data. But that data is really coming. We have amazing study just done to show that baselines and seem very effective and you'll hear more about that. So the standards are building. I think it's time that we really give serious thought to. See Gm as a tool to optimize and personalize diabetes management as we standardize, organized, integrate, analyze and act on the data. So I hope this was helpful to give you a start. I thank you uh, for your attention and I hope you'll tune into the additional c c g m c m E um, sessions that will come. We will delve more into case studies, virtual care professional and Personal CGM. So thank you very much for your attention. Published September 9, 2021 Created by Related Presenters Richard Bergenstal, MD Executive Director International Diabetes Center- Park Nicollet Minneapolis, MN, USA
