Activity Transcript

Davida F. Kruger, MSN, APN-BC, BC-ADM: Hello, I'm Davida Kruger, a certified nurse practitioner in diabetes at Henry Ford Health System in Detroit, Michigan. Welcome to this program titled, "Why Should We Use iCGM in Patients with Type 2 Diabetes?"

Joining me today are Ashlyn Smith, a certified physician assistant at Endocrinology Associates in Scottsdale, Arizona; and Dr Eden Miller, a family medicine specialist focused on diabetes in Bend, Oregon. Welcome to our discussion, Ashlyn and Eden. We're going to have some fun today.

Let me first start with talking about the background or giving a brief overview of the evolution of blood glucose monitoring to continuous glucose monitoring (CGM). Without dating myself too much, I like to say that I've been in diabetes for the past 39 years. When I first started in diabetes, we didn't even have blood glucose monitoring -- we had urine testing. When we think of blood glucose testing, we have to think about accurate measurement of capillary glucose, and that's an advantage. It's relatively inexpensive, easy to train patients to use, widely used, and people are pretty familiar with it. Some disadvantages are that it's subject to user error and misrecorded data. There is some training and checking required, and we have to make sure the patient's hands are clean so that we're measuring glucose from the patient's blood and not from food the patient has been eating. It provides limited data at a single point in time and sporadic measurements that require multiple tests per day for effective clinical use. At best, we're probably going to get 2 to 4 blood sugars a day, and we're not getting overnight blood sugars. It may be inconvenient, and patients do complain about it being uncomfortable or painful. Then there is the quantity of tests we can get from our healthcare providers and the insurance companies.

Now, when we think of CGM, it provides a comprehensive picture of glucose variability, including times when glucose levels are not normally measured: overnight, after meals, and when blood sugars are low (most patients don't want to stop when blood sugars are low; they just want to treat it). There are no missed readings because you're getting blood sugars around the clock, 24/7. CGM provides metrics to individualize treatment and care for that patient. It's simple to use, and pre-calibration does remove, for the most part, the need for daily fingersticks. Now, the disadvantages are that, in some instances, it can be expensive (it depends on insurance coverage); and it can be complex. It requires some education, but once we provide some education, it's fairly simple for the patient to use. Also, the patient has to use it for it to be beneficial. Some of the devices -- I think only 1 CGM left on the market now -- require fingersticks for calibration. Otherwise, you don't have to calibrate these, and you don't have to do a fingerstick to adjust medication. It is always on the body, which is a good thing (or a bad thing) because the patient doesn't have to go look for a meter. Replacement of sensors is anywhere from 7 to 14 days; with the implantable device, it is 90 days.

So, there are advantages and disadvantages, but pretty much with the market right now, we really want to see people move -- whether they have type 1 diabetes (TD1) or type 2 diabetes (TD2) -- to the benefit of using CGM, which is so much more beneficial for both the patient and the provider.

I'm going to turn now to Ashlyn. Can you talk a little bit about the rationale for using iCGM in patients with T2D? We've thought about it for a long time with TD1, but I think most of us are moving in that direction for T2D.

Ashlyn Smith, MMS, PA-C: Absolutely. Thank you, Davida. Yes, I agree. It's easy to think about using CGM technology in our patients with T1D, but we may be missing a large part of the population that has T2D and is experiencing hypoglycemia. I think there's a misconception or a false sense of security where providers and patients can be lulled into thinking that because it's T2D, we're not dealing with as much hypoglycemia risk. Unfortunately, with the sheer number of people who have T2D, we actually end up seeing that most of the severe hypoglycemic events that occur are in patients with T2D. Hypoglycemia of any variety is distressing. It can cause a delay in advancing therapy, and it can lead to missed medication doses. It can be a big barrier and a source of distress. So, we need to not forget about our patients with T2D and to keep hypoglycemia in our minds for them, as well.

Unfortunately, we see that as the disease progresses and as our patients get older, the risk of hypoglycemia increases. We talk a lot about β-cell decline and don't necessarily talk about α-cell decline, but we can see that hypoglycemia risk does increase as people get older and as their diabetes has been around longer. Of course, there are patient-specific criteria: certainly, anybody who has kidney disease or has a variable lifestyle or is, say, experiencing the pandemic and suddenly has a change in their living situation or their working situation. Or the gyms are closed, etc. In these stressful times, that change in their habits can increase hypoglycemia risk in patients with T2D.

Ms Kruger: I think you make an excellent point that we don't think of hypoglycemia in our patients who have T2D. I think we think of them as insulin resistant and that they're just not going to have hypoglycemia. But the statistics prove that they absolutely will have hypoglycemia. The other thing that's really relevant to our conversation is that hypoglycemia is not dependent on the HbA1c level. We always think of a lower HbA1c level as greater risk, but, in fact, people who have a high HbA1c level are also at risk for hypoglycemia. Your points are so well taken.

Ms Smith: Absolutely. Again, there's that false sense of security in thinking that a higher HbA1c level means they're probably not having lows when, in fact, that's the population who has the most hypoglycemia. There are a lot of reasons for this. It could be dysglycemia, a kind of rollercoaster pattern of blood sugar, that's keeping the HbA1c level high. It could be missed doses of medication. Perhaps there's medication intolerance or a cost issue or an issue with remembering their medications. Then they suddenly take their medications and end up having a hypoglycemic event. Or we, as providers, unknowingly are escalating their medications, not realizing it's more of a stepped medication issue. Then they end up taking that higher dose and have low blood sugar. That's just one of many reasons why we see hypoglycemia with higher HbA1c levels.

Ms Kruger: The beauty now is that we do have CGM. I mean, it's been around for longer than we talk about, and the devices seem to get more and more accurate every year. Could you take a few minutes and review for us what devices have been cleared by the FDA for use in the United States?

Ms Smith: Absolutely. This is a rapidly evolving area of medicine, and I think it is important to note a couple of the subtle differences because that's going to be important for patient selection and patient preference. For instance, we have the Libre 14-day and the Libre 2 systems on the market, which do not require fingersticks. We also have the G6, which does not require fingersticks. If your patient is specifically looking to avoid daily fingersticks, I would focus on one of those. If they are looking for something that has a little longer duration, I would call your attention to how long each of these are worn. With a Libre 14-day system, it's, of course, a 14-day sensor. The Libre 2 is also a 14-day sensor. The G6 is a 10-day sensor, so that duration is a little shorter, but it does have integration with some of the insulin pumps. If we're thinking pump therapy, we need to keep that particular sensor in mind. Then we have Guardian Sensor 3, which has integration with the CGM. It has a 7-day sensor. The Eversense CGM is a 90-day implantable CGM, so if you really want longer duration of action, that's the way to go. Now, those latter 2 CGMs that I mentioned do require at least 2 fingersticks a day for calibration. The first 3 CGMs -- the Libre 14, Libre 2, and G6 -- should not be calibrated. They're factory calibrated, so it's important to educate our patients not to attempt to calibrate them with fingersticks because it can actually make them less accurate.

Ms Kruger: Excellent. Thank you so much for that overview. Can you explain to us what is meant by integrated CGM, or iCGM?

Ms Smith: Davida, you said it so well when talking about the evolution of where we are and being able to assess glucose levels. This is the level that we're at right now. With an iCGM, we're talking about something that is either a continually sensing CGM or an intermittent-scanning CGM. Either way, it's something that, essentially, kind of plays well with others. They have integration either with a separate receiver or an app so the patient can intervene and make sense of how their glucose is doing, or they integrate with an automated insulin delivery device. These CGM systems can be used with very current technology and can be used to make treatment changes.

Ms Kruger: Thank you so much for reviewing that for us. I'm going to turn to Eden. Eden, we really are excited when we think about using CGM in T2D. That, of course, accounts for the majority of people with diabetes in the United States and across the world. Can you talk to us about some of the data that support using CGM in patients with T2D?

Eden Miller, DO: Yes, thanks so much. It was my distinct privilege to present a couple of different trials of some retrospective data at the American Diabetes Association (ADA) meeting as posters in 2020. I and some colleagues had the opportunity to look at 2 different kinds of cohorts. The first study was persons with T2D who were using the FreeStyle Libre system and were taking long-acting insulin or noninsulin therapy, so your oral antidiabetic therapies or glucagon-like peptide-1 (GLP-1) receptor agonists, not just pills. We looked at those on basal insulin and those on noninsulin agents. We also looked at the association between the FreeStyle Libre system and acute diabetes-related events -- all the things that come along with having diabetes, such as hypoglycemia or hyperglycemia -- and all-cause hospitalizations in persons with T2D who weren't on bolus insulin. It really gave us the idea to look at this from a scientific standpoint, not just from a utility standpoint. I was equally as impressed by reviewing those data. For inclusion criteria, we wanted people to have had an HbA1c test prior to starting CGM. Then we looked at the HbA1c level at 180 days and 360 days after initiation. We used various data sources to retrospectively look at it, like Quest Diagnostics and all the places where you're going to get robust data. That was what was so interesting about it. I was equally surprised, even though all 3 of us use this in a clinical setting so much and are finding its utility just by observation. Looking at the left-hand side of the slide, we see that as an overall group – lumping together those not using insulin and the basal insulin cohort – there was a 0.8% reduction in HbA1c levels in the first 6 months. Then continuing on, we had maintenance of a 0.6% HbA1c reduction. If we subsetted the data out (on the right-hand side), the T2D noninsulin group had the greatest reduction in HbA1c level. This really kind of goes to the point both of you made, Davida and Ashlyn, about how we used to always see CGM as being for those who use insulin. We found that there was a secondary benefit even for noninsulin users with a 0.9% decrease at 6 months and sustained at 0.7% at 12 months.

Now, if we shift to look at acute diabetes-related events or hospitalizations and all those things that come with complications, they did a similar data analysis using the same claims I mentioned earlier. They took an interval timeframe and looked at those 18 years and older with T2D. They made sure they weren't using either rapid or short-acting insulin. Those are the treatments that really kind of put our persons at risk for hypoglycemic events. They looked at enrollment out to at least 6 months' pre-purchase of CGM and analyzed overall acute diabetes-related events and all-cause hospitalizations. After implementing this particular CGM, the FreeStyle Libre system, those persons with T2D not on any bolus insulin had a 30% lower rate of all acute diabetes-related events and a 13% reduction in all-cause hospitalizations. Now, we could conjecture what this could be from, and we might see all different possibilities, whether it was avoiding hypoglycemia or making phone calls to their healthcare provider. It really shows that this kind of illumination can really give scientifically based evidence that there might be nice utility for these people.

Ms Kruger: Thank you. That's wonderful. Can you talk a little bit about the DIAMOND study and how that was developed?

Ms Smith: First of all, Eden, I love what you're saying about the HbA1c. I think what's so remarkable about that HbA1c reduction just with using the FreeStyle Libre is that that's not adding a medication.

Dr Miller: No.

Ms Smith: We are so often thinking about escalating medication -- maybe adding insulin therapy or whatever we can -- to bring that HbA1c down. This simply gives more data so patients can have a more informed diabetes course. They're seeing what their glucose readings are doing, and we aren't having to add medication. We're just giving them a tool that makes their diabetes a little bit better for them. What's also great is that we saw the most reduction in HbA1c levels in those who weren't on insulin therapy, suggesting you do not have to reserve CGM just for those who are on insulin therapy. We all know that that diabetes progression does occur, so even when we're dealing with T2D, we may see individuals who end up on insulin therapy and need multiple daily injections (MDI).

In reference to the DIAMOND study you mentioned, Davida, this was a randomized controlled trial of patients on CGM. When we looked at those with T2D, we saw that there was a reduction in their HbA1c levels across the board, even in those individuals with T2D who were taking MDI of insulin. This is again a reminder that even though they have T2D, they are at risk for hypoglycemia, and CGM is a safer way to help improve their HbA1c. We saw that in this randomized control trial. This kind of study really helps us have confidence in utilizing CGM in this population, as well.

Ms Kruger: Thank you so much. Do you want to talk a little more, Ashlyn, about older individuals and the use of CGM? I think we tend not to think about that group for technology. I don't think 65 happens to be old, but...

Ms Smith: Absolutely. As we already talked about, the older an individual gets and the longer the duration of diabetes, the more risk of hypoglycemia we see. Certainly, when we talk about ways to try to mitigate the risk of hypoglycemia, a tactic we use is liberating the HbA1c goal. We de-intensify therapy and target a much less aggressive HbA1c level as individuals get older. There was a prospective study that looked at older patients. The average age was about 76 years old, and HbA1c levels were across the spectrum. We saw HbA1c levels under 7% and HbA1c levels up to 11%, so really a range. Unfortunately, they did not see any difference in the hypoglycemia rates across the HbA1c ranges in these individuals who were on insulin therapy. Unfortunately, that hypoglycemia risk persisted even when their treatment was de-intensified from MDIs to basal insulin therapy. While these things we're doing can, in theory, help with the hypoglycemia risk, we are still unfortunately seeing that hypoglycemia risk persists even with those efforts and with de-intensification of therapy and HbA1c targets.

Ms Kruger: So, what I've heard both of you say -- and I'll just summarize that a little because I know Eden has a great case for us -- is that we need to start thinking out of the box. We need to start thinking about using CGM in older individuals. We need to start thinking about using it more in T2D. We need to start thinking about using it in people taking oral agents as well as in people taking basal insulin and not just those using MDIs and insulin pumps, which is what we've done for years and years. The spectrum is there, and we really need to start thinking out of the box because people with all types of diabetes will benefit from CGM. CGM gives us so much more data and is so much more friendly and more effective in people with diabetes. Our scope is certainly widening. Eden, you have a case study for us.

Dr Miller: Yes. This is a unique case. It's a real case of an individual that I saw in my practice. She was brought in by her daughter and had a long-standing history of T2D. She took 12 units of glargine in the PM and 3 units in the AM. I know we do not frequently split them, but that was done. She was prescribed rapid-acting insulin with her mealtime, but she hadn't been taking it at all. So even though we had this rapid-acting insulin on board, this patient was really trying to do a lower carb diet and was doing different regimens with that. She rarely took her insulin. She presented with mental status change. She came in with an HbA1c level of 7.4%. She had a little bit of renal impairment, but she was on an SGLT-2 inhibitor. I told her, "I'm really curious about you and how your glycemia is doing." She never really reported any significant hypoglycemia -- maybe once or twice a week. I thought, "You know what? I'd really liked to see what her glycemic... all the different places she visited on a different day and see what occurred." And we had had these previously reported changes in memory. I ended up placing a personal CGM on her, and she came back in for the very first time to go over this. It was so interesting because the daughter was reporting that certain times in the morning, she was having this confusion.

Looking at this report, I was stunned! I'd like to highlight a little bit. First of all, you see what we call that ski jump of that glycemia just tanking in the middle of the night, well over 200 to 300 points. Her rate of hypoglycemia was less than 70 mg/dL 22% of the time and less than 54 mg/dL 12% of the time. As you can see, this person was not taking rapid-acting insulin: her glycemia was stair-stepping up as time went on. And she was titrating her basal insulin to fasting plasma glucose at 9:00 AM. Her 9:00 AM level is 70 mg/dL. That's because she had had severe hypoglycemia in the early morning hours. Her own glucagon was putting out that glucose. What was so paradoxical about it is that the reported sugar the patient saw when she would look at her glycemia level was above her hypoglycemic range, but she was visiting those areas so significantly. As a result of this, we made a therapeutic change. We immediately de-intensified her basal insulin to a much looser target: Instead of less than 100 mg/dL, we made it less than 140 mg/dL. We added a GLP-1. We initially removed all rapid-acting insulin, but we then talked about possibly taking 1 to 3 units with dinner.

This was a follow-up CGM a few months later. By any means, we had not fixed all the different levels, but we now have an individual who comes in with a really similar HbA1c (the previous HbA1c level was 7.4%, and now it's 7.3%) but we have significantly affected the rate of hypoglycemia. Now, with this elderly person, we want time spent at less than 54 mg/dL to be zero. We don't want her to achieve that; we have a different target range for her. But this was that preliminary secondary follow-up. We achieved significant risk reduction with her hypoglycemia, and we made medication changes as a result.

It is such a dramatic case. Instead of needing a referral to neurology for mental status change, we needed to decrease her rate of hypoglycemia and make medication adjustments. You can see by looking at this interpretation that there's still some work to be done to minimize hypoglycemia and educate her relating to the variability of her highs and lows. So, we're not done yet. It's not like you do this once and solve all the problems. It's an ongoing learning opportunity for the patient, as well as for me as the provider.

Ms Kruger: Oh, that was a great case. Thank you. Ladies, as we close now, let's talk about some of the practicalities for CGM, because I know that our viewers are saying, "Oh, what do we do about reimbursement? How do we deal with Medicare, costs, billing, and those things?" We've got some good things to share. Clinical research was done to get us reimbursement for Medicare, and we had to prove that we didn't need fingersticks anymore. That's how we got Medicare reimbursement. Ashlyn, would you talk a little bit about what needs to be documented in the medical record to get reimbursement?

Ms Smith: Absolutely, Davida. For Medicare, that's going to be pretty ubiquitous for all of us, so that's a great place to start. With Medicare, there is a requirement that the patient already has to be checking their blood sugar by fingersticks at least 4 times a day, and they have to be using at least 3 injections of insulin per day. That may be just 1 unit of short-acting insulin, but if they are using 3 injections a day or more, they do meet the criteria. Go ahead and document right in your notes the increased concern for blood sugar variability. It could be that they've had hypoglycemia or hypoglycemic unawareness in the past; or that they have dysglycemia, where every day is an adventure with their blood sugar kind of experience; or perhaps they have kidney disease or there's some other reason why we are concerned about their blood sugar. Dysglycemia is both high and low blood sugar concerns. We do need documentation. If I have a patient who is looking or kind of flirting with the idea of CGM and we talk about that, I go ahead and let them know to prepare to start checking their blood sugar at least 4 times a day because we need either a blood glucose log or a meter download to support this, and we need to have all of that ready for the approval process.

Ms Kruger: Thank you so much. It's pretty cut and dry for Medicare. As long as the patient is taking 3 injections a day and doing 4 blood sugars, we pretty much can get it. I will say that most Medicare (there's some that don't) will require the patient to go to a durable medical equipment store rather than a pharmacy. So, keep that in mind. I don't want you to send a prescription to a pharmacy and think the patient doesn't have reimbursement.

Eden, can you talk a little bit about, on the professional side, how we get reimbursed for the time we spend interpreting this data?

Dr Miller: Yes. I'm just going to throw out the 3 current procedural terminology (CPT®) codes that you need to know or be familiar with. There's essentially 3 of them; both 95249 and 95250 are related to the application. One is for patient-provided equipment, and one is for physician-provided equipment that you provide from your office. As you mentioned, you should be familiar with who you're going to for reimbursement. Whether it's Medicare or Medicaid, there's different minimum billing and time, but these are CPT® codes you can talk with your billers about. Then there's ongoing interpretation. Of course, we would only apply the device once, and those 2 CPT® codes I gave you were just for application. The code that's probably going to be used more frequently is 95251, which is where we're taking that interpretation – the ambulatory glucose profile we showed you a few minutes prior with the case -- and getting this data. That is done either by a physician, nurse practitioner, physician assistant, or clinical nurse specialist. It's that ongoing monitoring, with at least 72 hours of CGM data that we're going to be interpreting and documenting in the chart. This can be a modifier that can be added on. It doesn't have to be face-to-face by the way, you can actually do it outside of the appointment. You could also do it during the appointment, or even as ... now that we're doing new billing codes for total time spent during the encounter. This could be incorporated into that. How you approach it really kind of varies.

Ms Kruger: Thank you so much. Eden, can you talk a about some success factors for the use of CGM, such as team-based education of patients and those kinds of things?

Dr Miller: Yes. You really do want to remember the whole team, because we really are trying to shoulder the burden and share the burden. One of the things that CGM can offer is more patient engagement. Allow them to shoulder some of the burden of their own disease. It does allow them to be very informed about their own management. We call that engagement. It also gives them that unique patient-centered approach. We really are still too generalized about our approach to persons with diabetes, but this allows you that very specific opportunity to see how food affects them, as well as stress, medicines, adherence, all that stuff. In addition, there are multiple touch points of people, including certified diabetes educational specialists, clinical nurse coordinators, nurse practitioners, and physician assistants who are all assisting the healthcare team and can look at this data. With COVID as well, we're able to do this remotely in some cases, if devices can be shared, or the patient can see it and communicate that information. I really feel like it's opened up a new way for monitoring in this new tech-based world and with the distanced monitoring we're doing with COVID right now. I think it's here to stay for sure.

Ms Kruger: Oh, I think it's absolutely here to stay. In the first 3 months of COVID, I thought we were going to go crazy because we didn't have as many people connected to the cloud as we needed, so we spent all kinds of time doing that and teaching it. I now think between telehealth and CGM, we definitely can monitor our patients anywhere as long as the patients have the apps and are connected and can upload their data. It really is worth the time to get patients connected. Thank you so much.

Well, it's that time when I'm going to ask you guys to give me concluding remarks. This has been amazing. I'm going to start with Ashlyn. How about a few concluding remarks?

Ms Smith: Well, I would probably say the first is just that reminder to look for hypoglycemia in our patients with T2D. It is happening, and we have a very large number of people with T2D. They're getting older, they may be adding insulin, and they may be having life stressors going on, as we all understand right now. So, we need to be looking about hypoglycemia in our patients with T2D, as well. This goes back to exactly what Eden said, to think about that team-based approach. Diabetes is a multidisciplinary disease, so if there is a limitation in our resources at the clinic level -- we all know time is a limitation -- use your diabetes educators and other resources. There are great videos online to help educate our patients. There are resources that can be used. It doesn't all have to fall on our shoulders as providers. We can use our team, for sure. Then make sure to educate your patients. Make sure that you're going through the process of the different CGM technologies to pick the best option for them because that's who is going to be using it the most. They're going to be using it on a day-to-day basis to see how their blood sugar is changing with different factors and really learning their body much better. That's the ultimate individualization of patient care.

Ms Kruger: Thank you so much. Eden, talk a little bit as part of your closing remarks about who some of the ideal candidates are for CGM.

Dr Miller: Yes. It is kind of hard to narrow it down to groups. I guess I would start at the high level. As you can see, we're very passionate about it. I think all persons with diabetes will get some benefit from iCGM. Of course, the benefits are varied, depending on the different therapies they're using. But who wouldn't benefit from illumination into their own disease in a real-time fashion and taking diabetes out of the past and into the present? That benefit could be intermittent use. It could be professional illumination. It could be ongoing management. You can see why we wanted this to be an introduction. I really do think this will shift the paradigm of how we interact with diabetes on so many levels, not to mention the safety and ease associated with CGM, but also the opportunity in the future to learn how medicines have different signatures or fingerprints and how each person can benefit. For me, it's actually a very freeing thing because it brings better understanding both on the provider level and on the patient level to a disease that can be so challenging to monitor. For those of you who are daunted by the steep learning curve, you'll find this great oasis on the other side of being able to engage and manage your patients more effectively. I know and believe you will find each person is their own unique candidate for it. So, getting started and utilizing it however you feel comfortable doing so as a healthcare provider and then expanding from there.

Ms Kruger: Awesome. I agree 100%. It's like, why wouldn't you offer this to all your patients with diabetes? When I think of how it benefits us as healthcare providers on that side and then how the patient can own their own diabetes on their side, it truly is a win-win. I agree with you that the learning curve for how to interpret the data may appear daunting, but it really isn't. After you've done half a dozen cases or even fewer, you're an expert. The patient, of course, also has to be an expert. We teach both the healthcare provider and the patient to own this data and to really use it in clinical practice. Ashlyn, Eden, I want to thank you for this great discussion. It was filled with so much practical information, and our viewers, I know, will walk away with some great information and hopefully get the excitement we have about iCGM and the desire to incorporate it at some level into their own clinical practice. And I want to thank you for participating in this activity, and please continue on to answer the questions that follow and complete the evaluation. Have a good time. Thank you so much.

This transcript has been edited for style and clarity.