Time Stamps

  • What are clinically relevant limitations of an A1c vs. a CGM
  • 10:49 What do the different parts of the ambulatory glucose profile tell you?
  • 16:24 What are common pitfalls, medications, lifestyle patterns that are important to ask your patient about when seeing them for a visit to manage blood sugar?
  • 21:30 How can you titrate insulin based on when you see the hypoglycemia on the ambulatory glucose profile
  • 27:45 When there is too much hyperglycemia, how do you titrate insulin when adding on GLP-1 agonists?

 

Sponsor: The New England Journal of Medicine Fellowship Program. Learn more about the one-year, full-time paid opportunity here and the application process.

 

Show Notes

Pearl #1: CGM is a reliable measure of glucose trends that can give an indication of recent glycemic control. Hemoglobin a1c is better used for population measures and serum blood glucose remains the gold standard.  

How does CGM compare to other measures of blood glucose control including point of care blood sugar and hemoglobin a1c? 

  • Hemoglobin A1c: average glycemic control over the past three months
      • Diagnostic limitations: 
        • Misses variability over shorter periods of time
        • Does not show hypoglycemia as clearly as CGM can
      • Best used as a measure of population measures
    • A1c is limited has its limitations in different contexts: “Hemoglobin A1c and Glucose Measurements” 
      • Increased red blood cell turnover -> Falsely lowers the hemoglobin A1c
        • hemolysis
        • blood loss
        • iron and vitamin B12 supplementation
      • Decrease red blood cell turnover -> Falsely increase A1c (RBCs have more time to become glycated)
        • nutritional deficiencies 
          • Ex. Iron and vitamin B12 deficiency 
        • hematologic malignancies
  • CGM: reliable measure of recent glucose trends
  • How precise are CGMs?
    • Serum blood glucose remains the gold standard to which all measures of blood glucose are compared
        • Continuous glucose monitors (CGMs) has a MARD (mean absolute relative difference) is around 8-12% compared to serum blood glucose
        • Point-of-care glucometers have a MARD of 5-6%. 
      • Inaccuracies in CGM matter the most when the CGM is reading low 
    • CGM measures glucose in the interstitial fluid
      • This lags behind blood glucose readings by 5-15 minutes
      • Educate patients to check the point-of-care glucose monitors if they are having symptoms of hypoglycemia 
    • What may cause inaccuracies in the CGM reading?
      • Warm-up period
        • CGMs have a warm-up period in which accuracy can be lower in the first 24 hours of sensor placement 
      • Pressure on the sensor (compression artifact) 
      • Certain medications 

Pearl #2: CGMs can provide real-time data as well as ambulatory glucose profiles to help manage patient data in real time. 

When you have a patient with a CGM device, how do you access the data? What does each section of the ambulatory glucose profile tell you? 

  • The Ambulatory Glucose Profile (AGP) is downloadable and accessible through accounts most clinics have.
    • Granular day-to-day information
    • Larger trends over the last two weeks
  • The AGP has three sections:
    • Time in Range
    • Ambulatory Glucose Profile
    • Daily Glucose Profile
  • Time in Range (Top Section): Shows the percentage of time a patient is in:
  • Ambulatory Glucose Profile (Middle Section): Contains the modal day view or ambulatory glucose profile
    • Glucose trends from the last 14 days onto a single 24-hour timeline
        • median
        • 95th percentile
        • 5th percentile 
      • Helps identify consistent patterns of hypo- or hyperglycemia at specific times of day.
  • Daily Glucose Profile (Bottom Section): Displays the daily glucose profiles over the last 14 days, which allows for more granular views of glucose data. 
    • This helps with the identification of patterns related to weekends, sickness, or other life events. 

Pearl #3: Before we get to titration, put your internist hat on and ask detective Q’s on why you may see the patterns that you do and what may be adjusted before med changes. 

What are common pitfalls, medications, and lifestyle patterns that are important to ask your patient about when seeing them for a visit to manage blood sugar? 

  • Walk with your patient through their insulin administration steps: 
    • How do you administer your insulin?
    • How do you store their insulin and and when is the expiration date?
    • What insulin is are you taking and when? 
    • Are you rotating insulin injection sites? 
    • Are you injecting your meal time insulin (if applicable) 15 minutes prior to a meal?
    • Look for evidence of lipohypertrophy on physical exam
  • Diet and physical activity can also have major impacts on glycemic control. 
  • Is there any acute illness that could be causing a rise in glucose?
  • Ask about what is happening during times of hypoglycemia or hyperglycemia. 
  • Alcohol intake can significantly modify glycemic control
    • Increases insulin sensitivity
    • Inhibits the liver’s ability to counter-regulate hypoglycemia
      • Placing patients at a higher risk of hypoglycemic events. 
  • Pairing CGM data review with counseling helps patients understand how their body responds to different factors, making the CGM a learning tool.
  • Real-time feedback from CGM can be a powerful motivator for patients to make positive lifestyle changes!

Pearl #4: Titration of insulin with hypoglycemia

How do you use CGMs to avoid hypoglycemia? 

  • Hypoglycemia is our main safety parameter and is important to minimize when patients are on insulin or sulfonylureas,
    • Goal to keep hypoglycemia less than 3-4% of the time!
  • CGMs have safety features with alarms that cannot be turned off to alert patients to hypoglycemia.
    • Alarms are set at ~54 mg/dL
  • Patients with frequent hypoglycemia are at risk of hypoglycemic unawareness, resulting in a lack of symptoms when blood glucose is downtrending
  • STOP the sulfonylureas if you see hypoglycemia on the patient’s CGM!
  • If a patient is on basal and/or prandial insulin -> look at the ambulatory glucose profile to identify when the hypoglycemia is occurring 
    • If at night ->consider decreasing the basal insulin as it may be too high 
      • A common pitfall in diabetes management is treating post-prandial highs by increasing the basal insulin dose 
      • Patients who are on doses of  greater than 0.5 units/kg/day are at the higher risk of over-basalization 
    • If during a specific time of the day (such as the afternoon or mid-day) -> adjust by reducing prandial insulin by 10% 
    • If BOTH at night and during the day -> decrease the total daily insulin dose, reducing both basal and mealtime doses
      • Don’t forget to ask the patient about what is going on during these hypoglycemic episodes– keep an eye out for any alcohol use, concern for compression artifact while sleeping or other interference. 
  • There is a lot of value to frequent patient check-ins when titrating insulin and virtual visits/ digital communication can be a big win for making changes. 
  • Hypoglycemia risk factors include hepatic dysfunction, renal dysfunction, longer duration of diabetes, and older age.

Pearl #5: Titrating insulin using CGMs when starting patients on GLP-1 agonists. 

How much should you decrease insulin when starting patients on GLP-1 agonists? What trends can you expect to see when looking at the glucose profile? How do you counsel patients? 

  • In general, the goal is for patients to have a Time in Range (TIR) of around 70%.
    • When patients have a TIR of <70%, which suggests worsened glycemic control further action is indicated
      • On the app this means that a patient has more time in the yellow range and less time in the green range 
  • The first step to treat hyperglycemia is actually to initiate a GLP-1 agonist or dual GIP/ GLP-1 receptor agonist as the first line injectable.  
    • GLP-1 agonists have other benefits
        • treat obesity
        • treat OSA,
        • treat MAFLD
        • cardiovascular protection
    • Low risk of hypoglycemia with GLP-1 agonist
    • Less frequent dose titration with GLP-1 agonist 
  • Initiating a patient on GLP-1 agonist therapy who is already on insulin therapy also comes with risks!
    • Patients become
    • How to start a GLP-1 in patients on insulin?
      • If A1c of either or less and a time in range of 50% or more (good glycemic control)
        • reduce the total insulin dose by 20%
      • If A1c over 8%, and the TIR is less than 50% (poor glycemic control)
        • no need to adjust the insulin dose
        • Just add the GLP-1!
          • The GLP-1 agonist dose will be titrated upward over time while insulin requirements decrease over time. 
          • GLP-1 agonists tend to dampen postprandial glucose spikes, potentially reducing or eliminating the need for mealtime insulin in some patients on basal-bolus regimens. This can simplify the treatment regimen.
  • What is the patient cannot start a GLP-1 due to insurance or side effects? 
    • Consider where they could benefit from additional insulin coverage! 
      • If they are on basal insulin only:
        • consider adding prandial insulin
          • this is often the go-to approach when a patient has too much time below range but also not enough time in range. 
      • If they are on basal–bolus insulin:
        • consider increasing the total daily insulin dose by 10% divided between basal and bolus dosing
      • If they are not on any injectable therapy:
        • consider starting at 0.1 units/kg per day, and increase by 10% every 2 weeks until TIR is >70%
  • After any titration of insulin or addition/up-titration of a GLP-1 agonist, it is important to revisit the AGP after about two weeks to assess the impact and make further adjustments as needed.
  • Remember that CGM provides actionable trends, even if the absolute precision has some variability and can be important to alerting patients of hypoglycemia!

Transcript

Dr. Kristen Flint: Diabetes is a specialty where you’re really walking with your patient. You only see them once every three to six months. They’re out in the real world managing their diabetes 24 7. And so if you don’t include them in the medical decision-making that affects them, you’re going to be at a real loss.

Dr. Shreya Trivedi: That’s Dr. Kristen Flint, an endocrinologist at MGH. Welcome to the 5 Pearls Podcast, bringing you high-yield, evidence-based pearls. Today we are talking about continuous glucose monitors or CGMs and insulin titration based on CGM. I am Dr. Shreya Trivedi and i’m joined by…

Dr. Becca Easly-Merski: Hi Dr. Becca! And I’m an internal medicine resident at BIDMC.

Dr. Shreya Trivedi: So, lets get into it. Remember to test yourself by pausing after each of the 5 questions. Remember the more you test yourself, the deeper your learning gains. 

Dr. Becca Easly-Merski: Pearl #1 – CGM  vs. A1c

Dr. Shreya Trivedi: What are clinically relevant limitations of an A1c vs. a CGM?

Dr. Becca Easly-Merski: Pearl 2 – Ambulatory glucose profiles

Dr. Shreya Trivedi:What do the different parts of the ambulatory glucose profile tell you? 

Dr. Becca Easly-Merski: Pearl 3 – High-yield questions

Dr. Shreya Trivedi: What are common pitfalls, medications, lifestyle patterns that are important to ask your patient about when seeing them for a visit to manage blood sugar? 

Dr. Becca Easly-Merski: Pearl 4 – Management of hypoglycemia with CGM

Dr. Shreya Trivedi: How can you titrate insulin based on when you see the hypoglycemia on the ambulatory glucose profile?

Dr. Becca Easly-Merski: Pearl  5 – Management of hyperglycemia with CGM

Dr. Shreya Trivedi: When there is too much hyperglycemia, how do you titrate insulin when adding on GLP-1 agonists?

Pearl 1: CGM  vs. A1c

Dr. Becca Easly-Merski: I am imagining sitting in the clinic and facing my patient who is showing me their blood sugar from their CGM app on their smartphone. But I want to take a step back– Do I care more about what number I am being shown on the cellphone or do I put more stock in the hemoglobin a1c? 

Dr. Kristen Flint: So the A1C is reflecting the blood sugar control over the past three months because it’s directly related to the life cycle of the red blood cell. Meanwhile, when you look at A CGM, you’re looking at daily profiles or profiles for every 14 days. If a patient has made recent changes to their medications or made significant changes to their diet, their exercise levels, their lifestyle, you’ll see that reflected more quickly in that 14 day measurement you get a sense of what’s going on with the blood sugar over a shorter, more recent timeframe compared to the A1C, which is going to take three months to turn over.

Dr. Becca Easly-Merski: So every lab test has its own limitations depending on the context, which may be the underlying theme of this episode.

Dr. Shreya Trivedi: Let’s start with the limits of A1c. So a to simplify A1c, it measures the glucose on the RBCs. This is some good spaced repetition from a prior Mind the Gap episode on patient states that increase blood turnover like hemolysis, blood loss , iron and vitamin B12 supplementation can falsely lower hemoglobin a1c… so in high cell turnover states, there is less time for the glucose to hang out the RBC, so you can get lower A1cs.

Dr. Becca Easly-Merski: On the other hand there are times where there is low blood turnover like nutritional deficiencies and other hematologic malignancies — these are conditions  where the bone marrow just isn’t producing as many new cells. So there are more old red blood cells around and they have more time to hang out and meet up with glucose, which can falsely elevate the hemoglobin a1c. 

Dr. Shreya Trivedi: I think patients with renal dysfunction may be a good group to highlight. Their A1c can be higher or lower than actual bc they’re gonna less cell turnover with iron deficiency they might have concomitantly and at the same time have higher cell turnover with epo injections. And there is also all these glucose shifts after dialysis so we might not be captured with A1c.

Dr. Becca Easly-Merski: So for all these reasons in patients who have severe renal dysfunction the CGM can help us be more precise about their blood sugar control.

Dr. Shreya Trivedi: So we sat down with Dr. Marten, an internist at Park Nicollet Clinic and Director of the International Diabetes center, to talk about the discrepancies he has btw the A1c and CGMs.

Dr. Thomas Martens: Can think of many individual cases where maybe the A1C is at goal, you get some CGM data, it’s like, oh, there’s way too much hypoglycemia here. And there are people who just do not meet their glucose targets. If you’re looking at A one Cs, we’re all sort of beholden to A1C as a quality measure. But I can think of a number of individuals who have CGM data that looks essentially excellent. Their time and range looks great, their time below range looks very good, yet their A1C  is over eight. And what do you do with that? You can’t really push insulin based therapies too much harder or you’re going to have hypoglycemia. You just have to live with the fact that that hemoglobin A1C may not be reflective of where this individual’s at from a glucose standpoint and don’t overtreat the glucose to get your A1C below eight if the glucose data looks really really good based on CGM.

Dr. Shreya Trivedi: So it sounds like there are lots of opportunities to get more granular trends with CGMs that aren’t as influenced by red blood cell turnover. But how does CGM data compare to the gold standard serum glucose? How accurate is the CGM?

Dr. Becca Easly-Merski: Yeah that was something I was curious about as well so I sat down Dr. Zach Taxin, an endocrinologist at BIDMC who gave us a sense of what percentage off the CGM may be from the actual blood glucose… and for those in the biz, this the absolute relative difference between the serum glucose vs. CGM.

Dr. Zach Taxin: But for most of the CGM devices, it’s like eight to 12%, probably 10 to 12%, maybe some of them are getting down to the eight to 10% range. So that means that the relative difference between the CGM value and whatever your gold standard is, a whole blood or usually a serum glucose, you’re looking at an eight to 10% on average difference for each paired value. 

Dr. Shreya Trivedi: So these CGMs are not perfect but can give good ballparks and give trends without having to prick our patients TID.

Dr. Becca Easly-Merski: And I was surprised to learn from Dr Taxin that even with our glucometers that we are so used to using in the hospital and that our patients are using outpatient, there is an 5-6% difference between our point of care glucometer testing vs. the gold standard serum glucose

Dr. Shreya Trivedi: I really appreciate that comparison that none of these machines, even our glucometers are perfect and are an estimate of the truth. And so back to CGMs, I think some of that variability comes from how exactly the CGM measures glucose in our bodies. 

Dr. Kristen Flint: So CGM measures the glucose in the interstitial fluid between the cells, so it’s not actually measuring the glucose in the blood. That glucose has to move    from the bloodstream diffuse through the vessel wall into the fluid between cells, and that’s what gets measured. That takes five to 15 minutes for it to be measured. And so the true CGM reading is going to be delayed by five to 15 minutes compared to the corresponding blood glucose.

Dr. Shreya Trivedi: Oh i can see teaching point about the CGM lagging by 5-15 mins being relevant to tell patients about say they’re feeling woozy, lightheaded, they might be falsely reassured by their CGM read of their glucose that might actually reflect reality from minutes ago vs. what it’s actually dipped down to. 

Dr. Becca Easly-Merski: Absolutely, and there are are also a few other caveats to give patients a heads up about when interpreting their glucose numbers on a CGM.

Dr. Kristen Flint: So first, some of the CGMs will be less accurate over the first 24 hours. And so I tell a lot of my patients, the CGM is going to be great, but the first 24 hours might not look quite right. So if you have any concerns about high blood sugar or low blood sugar or you’re having symptoms, just do a finger stick to confirm. Usually after the first day, things are pretty stable.

Dr. Zach Taxin: I mean a pretty common one is so pressure on the area which compresses the space or reduces the blood flow or alters the dynamics of the interstitial space or actually just injures the device itself. Those can obviously cause pretty significant errors. The other very common one is when you insert the device, if you nick a very superficial small van or even capillary, sometimes there’s a little blood there, it can clot and then you’re not going to get accurate readings.

Dr. Shreya Trivedi: Other fun facts is high doses of vitamin C and tylenol can also affect the CGM’s accuracy. But we are talking about super high doses of vitamin C (greater than 500 mg doses, which is the equivalent to 10 oranges) 

Dr. Becca Easly-Merski: Or tylenol in doses greater than 1g at one time or greater than 4g per day are also associated with inaccurate readings, but we may want to avoid these doses anyway 

Dr. Shreya Trivedi: Let’s review what we just covered. Continuous glucose monitors give us more granular data on trends in blood sugar that we might not uncover with just an A1c. But CGM is a tool and isnt perfect, there is a warm-up period for the 1st 24 hours or so and can have about 10% variation from a patient’s actual serum glucose and may lag by 5-15 minutes. 

Pearl 2: Ambulatory Glucose Profiles

Dr. Shreya Trivedi: Okay, now onto a very quick and practical dive into CGMs 

Dr. Becca Easly-Merski: So say a patient has a CGM that either you prescribed or they were able to purchase over the counter, and they say hey wanna look at it? And you are like yes of course but then you go to  take at look at the app, you are like wow so much data–where do I even start? 

Dr. Shreya Trivedi: There is so much granular info there!  The best thing to look at is something called an Ambulatory Glucose Profile, which is downloadable and most clinics have accounts to access. 

Dr. Becca Easly-Merski: So there are three sections of the ambulatory glucose profile and we will go through the sections one by one and you can follow along in the show graphic. The first section gives the percentage of time a patient is in range. The  green part is the “Time in Range”, yellow is “Time Above Range”, and red is “Time Below Range”. 

Dr. Thomas Martens: And the top part, the yellow part is the time above range. And so just in 10 seconds, what I do is I take a look at this top portion.  The metrics is the time in range at goal is a time below range too high. And so what’s that goal? Time and range? That’s time. With a glucose between 70 and 180 milligrams per deciliter, you want that to be over 70% of your day. And why 70% of the day? Because very loosely, that correlates with an A1C around 7%. And so 70% time and range, you want to get above that if you can. 

Dr. Shreya Trivedi: So ideally we want to see that green part of the ambulatory glucose profile say 70% in range (or above that we are happy that too) btw glucose 70-180 mg/dl.

Dr. Becca Easly-Merski: And caveat to the 70% threshold that our reviewers pointed out is in older patients.  Older patients just have higher risk of hypoglycemia so we are more relaxed about goals. And so we are okay if old patient’s time in range is 50-70% so they  have more leeway.

Dr. Shreya Trivedi: And then moving on the time below range (the red part) be less than 4%.

Dr. Thomas Martens: Obviously you might say, well, don’t we want the time below range to be zero? And we sort of do in real terms. If you look at a glucose profile for somebody with no diabetes, a normal glycemic response, you see time below 70, so that’s maybe 1.1% of the time. If you look at a population-based study, and so people can drop below 70, it’s not dangerous. We worry about it in diabetes because if you’re treating people with agents that can cause hypoglycemia, which is insulin cellia therapy, you want to minimize it because we worry that you’re going to drop further than 70, right? So 4% is the threshold you want to be under 4%. 

Dr. Shreya Trivedi: Okay so The reason the target time below range is not 0% is in the real world, people drift below 70 all the time so some dips accepted and i guess expected.

Dr. Becca Easly-Merski: And if you pop over to the right of the ambulatory glucose, you’ll find a line called the Glucose Management Index aka GMI, which gives an estimates the A1c based on the last 14 days. And this can help with patient counseling, particularly if they really have been so used to using A1c for feedback and giving them an idea of what an A1c would look like based on the last 14 days

Dr. Shreya Trivedi: So the GMI is just a number, basically an estimate of the A1c but then you pop over to the So that second and middle of section ambulatory glucose profiles. This is gonna be glucose trends of  the last 14 days merged together on a 24 hour timeline. 

Dr. Thomas Martens: And so we’re looking at the modal day view, the ambulatory glucose profile view. We’re looking at patterns. There’s a median line, there’s margins 95 and 5% line. And we’re seeing if there’s a pattern to hypoglycemia, if there’s a pattern to hyperglycemia, because those are targets for intervention. 

Dr. Shreya Trivedi: So this middle parts lets us ask what does a typical day look like for this patient? Do they morning hypoglycemia or postprandial hyperglycemia?  Do they have lot of lows at night time over the last 14 days?

Dr. Becca Easly-Merski: The third and bottom section is the daily glucose profile which is a snapshot of each of those 14 days.  So you get a better idea of–are they having more hyperglycemia on the weekends? or were there a few days of significant hyperglycemia?  and was that was possibly around a sickness or some other life event. So now that we reached the bottom of the ambulatory glucose profile, lets summarize, I think this section was new learning for me. I learned to not be intimidated when someone gives you their phone and empowered me to find the ambulatory glucose profile. I look at the time in range to target 70% for the green time in range section, which correlates to an A1c of 7% and then most importantly look to see that the time below range is less than 4%. 

Dr. Shreya Trivedi: We also gain information about certain times of day that patients are higher and lower that can guide questions that we may ask in history. We can also look at the 14 days in the week to help find days where patients are off goal. 

Pearl 3: High-Yield Questions

Dr. Shreya Trivedi: Now before we go ahead and make any changes to a patient’s insulin based on the time in range being too low or more hypoglycemic events, we gotta investigate what other factors may be going on.  

Dr. Becca Easly-Merski: I love this part. And we kind of got into this in the last pearl, when we see certain trends we can ask whats going on at night time when you’re sugars go low? and when is  you last meal? 

Dr. Shreya Trivedi: Or ask if their sugars are higher on the weekends uncover life stressors on the saturday or sundays.

Dr. Becca Easly-Merski: And this is  the beauty of medicine and CGMs is we get learn about each patients patterns. 

Dr. Kristen Flint: I have another patient who came in to see me actually last week and she said, I’m having really high sugars after meals. So we looked at her CGM, and sure enough, after she ate, she would have a blood sugar of 300 and it would last for a couple of hours, but then it would come down. And so I asked her what she was doing with her insulin and she said, well, I was eating my entire meal and then taking the insulin, what they told me to do in the hospital. In the hospital, they never knew how much she was going to eat. 
And so they had to see what she ate before they covered it. Sometimes we do the best that we can in the hospital, but it’s not quite right in the outpatient setting, because she was taking the insulin late, she was having the carbs peak and the insulin wasn’t even active yet. So we were seeing the sugars get up to 300 and then it took some time for the insulin to work, but when it did, it brought down her blood sugar. So I said, okay, why don’t you try to move the insulin administration up to 5, 10, 15 minutes before you eat and we’ll see if that helps your blood sugars instead of just going up on the insulin dose. And sure enough, in looking at her cgm, I can see that now her blood sugars are much better controlled with the meals, all because we move the timing of the insulin. And so even though you can get daytime hyperglycemia, nighttime hypoglycemia, unless you talk to the patient about what they’re doing, you’re not going to know what to do constantly. 

Dr. Shreya Trivedi: Love the point about diabetes management is really walking with your patient, and asking “walk me through how you’re taking insulin” and not just jumping to treat numbers.

Dr. Thomas Martens:You have to think through your troubleshoot and put on your internal medicine hat. Is that expired insulin? Insulin does go bad. It’s sort of a fragile protein. So if it freezes, if heats up in the summer, sometimes it loses activity. If it’s expired, it loses activity. And so perhaps it’s expired insulin, maybe there’s something else going on. Acute illness will often cause a rise in glucose values. Lipo hypertrophy. If you inject insulin in the same site over and over, insulin is an anabolic hormone basically. So what you’ll get is you get fat buildup, and that is called lipo hypertrophy. The problem with lipo hypertrophy, aside from having a lump there cosmetically, is that if you’re injecting insulin into that, it’s not real vascular. The insulin doesn’t get absorbed, right, and you start to get variability in absorption or not absorption. And so rotating sites is important.

Dr. Becca Easly-Merski: Yes! The differential we should be thinking of is how is the insulin being stored? Is the insulin expired? Was there an illness?  Are you rotating the insulin injection site? 

Dr. Shreya Trivedi: The other factor that surprised me and i started asking patients more about when i see patients who have hypoglycemia is to ask about their alcohol use 

Dr. Becca Easly-Merski: Yeah fun fact: alcohol can actually increase insulin sensitivity and on top of that, alcohol curbs the liver’s ability to counter-regulate hypoglycemia. 

Dr. Shreya Trivedi: Yes no fun! And then last but not least, i think the most important Q to ask/investigate is to ask about any lifestyle changes in their diet or exercise. Have they made any changes? I actually gave 10 day CGMs to family members as an xmas gift. And in reviewing their sugar it was such a great ah-ha indian food that was always passed down as “highly proteinaceous” spikes their sugar so fast and I love giving them a chance if they are open to it to make food changes before jumping to change meds around.

Dr. Becca Easly-Merski:  Yes, CGMs are much more powerful information than the family whatsapp group or instagram reels they might be getting info from. let’s refresh what we just covered. To summarize, the next time you see a patient with significant abnormalities on their AGP do pause and ask the patient how they are taking their insulin- could there be a mix up with their mealtime or basal insulin? Are they refrigerating their insulin? Is it expired? Could alcohol be playing a factor into hypoglycemia? And most importantly ask them about any changes in their diet or activities that could explain any changes that you are seeing in the last 14 days. 

Pearl 4: Management of hypoglycemia with CGM (Part 1)

Dr. Shreya Trivedi: Now on to the actual management from all the info we get from the CGMs. And first and foremost, we need to make sure our patients our safe, and that means taking action if they are having more than 4% hypoglycemia

Dr. Thomas Martens: I think most of these CGMs have alarms set at 54 that you can’t turn off appropriately. You want people to know that and act on that if they’re drifting down. One of the problems with hypoglycemia is that if people are having frequent hypoglycemia, they start to lose their perception of hypoglycemia. They start to develop hypoglycemic unawareness, and the big risk for hypoglycemic unawareness is hypoglycemia. Your glucose meter in your brain resets itself and you just don’t feel or perceive the symptoms as you start to drift down.

Dr. Becca Easly-Merski: And so that brings us to the first branch point in the management of hypoglycemia. If you see hypoglycemia and you have to ask if they are on a sulfonylurea. If they are, the sulfonylurea has to go.

Dr. Shreya Trivedi: Yes a hard do not pass go with a patient still taking that glipizide or glimepiride. Ok So say they are not on a sulfonylurea and are on insulin but having hypoglycemia > 4% of the time, what do you change? The basal insulin? The mealtime insulin?

Dr. Becca Easly-Merski: This is where peaking over at the bottom of the ambulatory glucose profile is so helpful. We need to see when in the day the low sugars are happening. Is it happening at night? Or a specific time of the day or throughout day and night?

Dr. Shreya Trivedi: Such good questions. Let’s tackle each of those scenarios. Let’s say the patient has sugar dips overnight around 11pm or or low fasting AM glucoses.

Dr. Thomas Martens: Keep an eye out for too much basal insulin over basal ization. Again, you can see that pattern with big exaggerated drop overnight typically arise during the day because people aren’t being adequately treated for their mealtime needs. So watch for over basil or too much basal insulin.

Dr. Shreya Trivedi: So over-basalization. So lets go over what that is. This is( when after meals, we keep seeing hyperglycemia, people will often respond  And i’ve fallen for this trap before) with increasing their basal insulin because of all the postprandial hyperglycemia. However, sometimes the basal insulin is increased too much but too much basal insulin often leaves the patient low at night. 

Dr. Becca Easly-Merski: And as a rule of thumb over-basalization usually happens when the basal insulin exceeds 0.5units/kg/day

Dr. Shreya Trivedi: So take away if a patient has glucose lows at night and you suspect there is too much basal insulin on board, you want to go down on basal insulin And then tackle the postprandial highs separately which we will get into in pearl 5. 

Dr. Becca Easly-Merski: But the diagnostic pause we all need to do, as a throwback to pearl 3 is to check in with the patient about what’s going on do they have an alarm going off  often at night because they’re sleeping on it? 

Dr. Thomas Martens: You do get artifacts sometimes if you’re sleeping on a sensor, sleeping on your arm, that pressure can sometimes be enough to cause the glucose usually to drift down because you’re forcing all the interstitial fluid out from around the sensor so it’s not reading accurately.

Dr. Shreya Trivedi: Let’s get into the other two scenarios. You are reviewing the middle and bottom parts of the ambulatory glucose profile and see low sugars at a specific part of the day, like in the afternoon.

Dr. Becca Easly-Merski: So if t he low glucose happens at a specific time of day, we can reduce the  mealtime bolus dose that is happening right before that time. Consider reducing the mealtime bolus dose by 10%, which is the rule of thumb that Dr. Martens uses and teaches with. 

Dr. Shreya Trivedi: And of course, there are many ways to approach insulin adjustment and what may be appropriate for the patient in front you but the 10% is just a rule of thumb that many do.

Dr. Thomas Martens: Frequent small adjustments. If you’re working with somebody with insulin, think about maybe 10% adjustment to one of the insulins. Think about revisiting that profile in two weeks or so. 

Dr. Becca Easly-Merski: I really appreciate virtual visits or electronic messages is great for those for frequent touch points and doing small adjustments — both safe and patient friendly.

Dr. Shreya Trivedi: Bless virtual touchpoints! Last scenario for hypoglycemia, an unforunate case of if lows occur throughout the day and night, what are we doing then?

Dr. Becca Easly-Merski: If we see lows throughout, then we decrease the total daily insulin dose so basically decreasing a bit of the basal as well as the mealtime doses.

Dr. Shreya Trivedi: Awesome, let me recap what i’ve learned so far. Next time I see a patient’s ABP  has Time Below Range more than 4%, I will first check if they’re on a sulfonylurea. If they are, I’ll stop that medication before touching their insulin.Then, I’ll pull up their ambulatory glucose profile and pinpoint exactly when those lows are happening: I’ve learned to avoid the trap of over-basalization. When I see those postprandial spikes with overnight or fasting lows, I reduce the basal insulin and tackle the post-prandial highs separately as we will talk about in pearl 5.

Dr. Becca Easly-Merski: If I notice lows at a specific time of day, I’ll reduce the mealtime bolus that comes before it. And If I’m seeing lows scattered throughout day and night, I’ll decrease their total daily insulin dose.

Pearl 5: Management of hyperglycemia with CGM (Part 2)

Dr. Shreya Trivedi: Okay now that we talked all about the lows, let’s talk about where we find most of our patients with too much yellow aka time above range on the ambulatory glucose profile and not enough green which is the time in range and as a reminder this is someone who have having time in range less than 70% of the time

Dr. Becca Easly-Merski: What I was surprised to learn is that the first step isnt insulin adjustments, its to add on a GLP-1 agonist like semaglutide or the dual GIP/GLP-1 receptor agonists like tripeptide.

Dr. Shreya Trived: Yes! GLP-1 agonists (which we will refer to since its a mouthful to say both GLP1 and GIP/GLP1 RA) are so effective at both fasting hyperglycemia and postprandial hyperglycemia and there are so many other CV benefits that GLP1s are are recommended even before adding on insulin. 

Dr. Becca Easly-Merski: There is certainly less risk of hypoglycemia and less titration with GLP-1s that makes it much more patient friendly for treatment. 

Dr. Shreya Trivedi: And maybe more clincian friendly too! Let’s walk through some scenarios. So what if the patient is already on insulin therapy and now needs to be started on GLP-1 agonist therapy. 

Dr. Becca Easly-Merski: Yeah this a good Q. I’d be worried about GLP-1’s effect on enhanced insulin secretion, less glucagon production when they already have insulin on board and changes in my patient’s appetite and diet.

Dr. Tom Martens: If they have an A1C of eight or less or a time and range of 50% or more, I typically will reduce the insulin dose by 20% as I add the GLP one. If they’re less well, less tightly controlled, if their A1C is over eight time and range is under 50%, I typically just add the GLP one. You’re typically adding the GLP one at its lowest dose because you got to start low and titrate up. And so we’re reducing if they’re close to goal because we don’t want to precipitate hypoglycemia even with adding the lowest dose of a GLP one. And then as we do our titration steps with the GLP one therapy, we’re going to want to be revisiting where we’re at with insulin therapy because typically we’re backing off on insulin often pretty dramatically.

Dr. Shreya Trivedi: So as a recap, if the patient’s glucose is not well controlled at all, in that their time in range is <50% then you can just feel comfortable add GLP1 to their existing insulin regimen and watch closely. But if their time in range is between 50-70%, then you add the GLP1 BUT back off on total daily insulin amount by 20% and then keep a close eye on the impact of the GLP1 on the sugars.

Dr. Thomas Marten: They tend to really damp down the postprandial spikes. And so sometimes if you’re on basal and bolus insulin, you got to keep an eye on that mealtime insulin. And so if we can use basal insulin plus a GLP one in place of mealtime insulin, that is typically a win. 

Dr. Shreya Trivedi: Oh man that would be such a win if patients could just simplify their regimen to a basal insulin and once weekly injections of one of the GLP1 agonist. Makes me smile to think about. Let’s switch gears. Now, what about patients who aren’t candidates for GLP-1 RAs or dual agonists? Maybe their insurance doesn’t cover it or they couldn’t tolerate the GI SE. How do we approach insulin titration for hyperglycemia in these cases?

Dr. Becca Easly-Merski: So for patients who cannot take a GLP1 and have significant hyperglycemia, the approach that Dr. Martens’ takes really depends on their current Time in Range. If they’re not well controlled so say TIR is less than 50% and they’re only on basal insulin, we can consider adding mealtime insulin.

Dr. Shreya Trivedi: Okay what if they are on basal-bolus insulin regimen and not as well controlled and their TIR is < 50%?

Dr. Becca Easly-Merski: So Dr. Martens recommended if they’re already on basal-bolus therapy with TIR less than 50%, we’d increase their total daily insulin dose by 10% (again slow and steady) and that 10% increase in insulin gets redistributed to maintain a 50:50 balance between basal and bolus.

Dr. Thomas Martens: You’re going to want to advise people. If you’re seeing a lot of lows, I want to know about that, give me a call. We’ll troubleshoot it. We’ll adjust your insulin more. And so as you’re titrating, you want to be revisiting where you’re at with your glucoses and CGM makes it a little easier because you can pull up your website and see where they’re at and do a lot of that even remotely. 

Dr. Becca Easly-Merski: And as for spaced repetition, in our older patients who are more risk of hypoglycemia we are more relaxed about goals we are okay with a time in range of 50-70% so they  have more leeway room

Dr. Shreya Trivedi: So to summarize one approach to hyperglycemia (again each patient is so different): First, consider adding or optimizing a GLP-1 RA or dual GIP/GLP-1 RA before making major insulin adjustments. If adding these agents when TIR is 50-70%, we would also reduce basal insulin or total daily dose by 20%. If TIR is below 50%, keep insulin the same when adding one the GLP1s.

Dr. Becca Easly-Merski: For patients who can’t use weekly GLP injections and are not as well controlled, say TIR less than 50%, we can either add mealtime insulin or increase total daily dose by 10%. And most important is to maintain close follow-up, preferably after two weeks, to reassess and make adjustments as needed.

Shreya: Again, the key is to keep frequent visits to see how our individual patients are responding. That is all for today, thank you so much for listening. If you have ever gotten any value from the podcast, please please please share this with at least one other colleague who might also get something out of the episode

Dr. Becca Easly-Merski: Thank you to our reviewers Dr. Jonathan Li and Dr. Michael Weintraub. Thank you to Dr. Jesse Powell for the accompanying graphics.  This episode was made as a part of the Digital Education Track at BIDMC.

Dr. Shreya Trivedi: Opinions expressed are our own and do not represent the opinions of any affiliated institutions.

References