Geeking Out About Continuous Glucose Monitors
Options
Dragonwolf
Posts: 5,600 Member
So, I hopped over to http://eatingacademy.com to refer some people to Dr. Attia's "about me" section, which has some cool stuff on it, when I noticed that he actually updated his blog (which he admittedly doesn't do often, because he's too busy kicking *kitten* and taking names, basically). Nothing particularly out of the ordinary for the most part. He left NuSI to open his own private practice (I'm so jealous of San Diego and NYC people right now...), he eats Paleo for all intents and purposes, stopped competing in stuff but still works out, yadda yadda yadda.
Then, down at the bottom of the "What I Eat" section is this gem:
This is awesome to see! I've made mention before that a technological revolution is on the horizon for the medical industry. I know this in part because I've been involved with some of the early trailblazers of that revolution, and in part because I've had my ear to the ground about this stuff (because frankly, it's really cool). However, about things that are actually available to the public, I've only heard rather nebulous stuff -- stuff in testing, but largely only available to a select few, elite doctors. Granted, Dr. Attia is arguably one of those "select few, elite doctors," but he's got it as a "patient" and not as yet another piece of medical equipment for a practice.
This also sheds a ton of light on glucose numbers -- like the fact that he has tiny, long-lived red blood cells, which make him appear prediabetic even though he's not. This will also be huge for diabetics (especially those who are insulin dependent) for better and more accurate monitoring.
Now, if only we could talk someone into insulin testing of some sort...
Then, down at the bottom of the "What I Eat" section is this gem:
The biggest “news” on my eating front is that I now wear a 24-hour continuous glucose monitor (CGM) 24/7. This was the result of one of the most fortuitous flights of my life. In the fall of 2015 I was flying to NYC and half way through the flight, needing a short break from work, I went to one of my favorite watch sites. The fellow next to me made a comment—clearly he was part of the cognoscenti—and we got ultra-deep into watch idiotness. After a while I asked him what he did only to find out he was the CEO of Dexcom, the company that makes the best CGM device on the market. Fast forward a week and Kevin has introduced me to his amazing team (Christy Pospisil is awesome!) and I’m hooked. CGM is a game-changer and it does warrant more discussion than I can provide now. The insights have been staggering. I’m pretty obsessed with it (shocker, yes) and I aim to keep my 14-day running glucose around 90 mg/dL with spot-check standard deviation less than 10 mg/dL. By keeping average glucose low and glucose variability low, I can reasonably assume my insulin AUC (area under curve) is low.
Below is a printout of my last 14 days. As you can see my measured average glucose was 92 mg/dL, which imputes an A1C of 4.8%. At some point I may write about the dozen insights gleaned from CGM (and I think I mention a few in the podcasts), but here’s one: measured A1C is probably directionally valuable (you know, the difference between, say, 5% and 9%), but that’s about it. If your RBC (red blood cells) live longer than 90 days—mine live much longer since I have beta thal trait—your A1C will artificially reflect a higher average glucose. Conversely, if your RBC are large, the opposite occurs. (For those wondering, MCV, which is part of a standard CBC, shows you RBC size).
My A1C in standard blood tests routinely measures 5.5% to 6.0% (courtesy of my tiny RBCs), which poses a problem when applying for life insurance (prediabetic is defined as 5.7% to 6.4%). But with CGM, which is calibrated 2-3 times daily, my imputed A1C, which is much more reliable, varies from 4.6 to 4.9%. Big difference, huh? As an aside, I can’t talk about my beta-thal without hearing my med school roommate, Matt McCormack referring to them as “shite for blood” in the best Scottish accent ever. As if it’s not bad enough having an artificially high A1C… you gotta have shite for blood.
And that’s the least amazing part of CGM. I’m not sure I’m at liberty to discuss the next generation of CGM. Admittedly, not too many people want to wear the device I wear, but in two years, well, that’s when it will get amazing.
And that’s just the tip of the iceberg when it comes to why this device is adding insights and actionable data at a geometric rate. In two years this device will evolve into something everyone can wear.
This is awesome to see! I've made mention before that a technological revolution is on the horizon for the medical industry. I know this in part because I've been involved with some of the early trailblazers of that revolution, and in part because I've had my ear to the ground about this stuff (because frankly, it's really cool). However, about things that are actually available to the public, I've only heard rather nebulous stuff -- stuff in testing, but largely only available to a select few, elite doctors. Granted, Dr. Attia is arguably one of those "select few, elite doctors," but he's got it as a "patient" and not as yet another piece of medical equipment for a practice.
This also sheds a ton of light on glucose numbers -- like the fact that he has tiny, long-lived red blood cells, which make him appear prediabetic even though he's not. This will also be huge for diabetics (especially those who are insulin dependent) for better and more accurate monitoring.
Now, if only we could talk someone into insulin testing of some sort...
1
Replies
-
I used to love the idea of continuously monitoring blood chemistry. And why not add urine and fecal analysis too? But I think it'd just make me (more) obsessive.
When the study came out showing that people varied widely in their glycemic response to foods and that you could have a diet personalized for your particular response, Peter over at hyperlipid responded: eat fat.
So there you go. No need for a CGM. Just eat fat.
0 -
I used to love the idea of continuously monitoring blood chemistry. And why not add urine and fecal analysis too? But I think it'd just make me (more) obsessive.
When the study came out showing that people varied widely in their glycemic response to foods and that you could have a diet personalized for your particular response, Peter over at hyperlipid responded: eat fat.
So there you go. No need for a CGM. Just eat fat.
lol, while I agree, I think CGM would be a huge step in getting people in general to do that. Teach them to "eat to the meter" and they should invariably start leaning in that direction.
Seriously, I think about my in-laws with this. Apparently, going from Metformin to insulin, or seeing her own mother undergo quadruple bypass, though "scary" to my MIL, wasn't enough to get her to actually listen to me, or at least read the books I loaned her.
0 -
My daughters insulin pump came with the dexcom CGM. She hates wearing it.
I'm having a very difficult time getting her supplies ordered at the moment, but since we have reached our out of pocket maximum with insurance already this year, the supplies will be covered at 100%.
I will be able to get enough CGM's to wear one every day for the next year if I wanted to, but I can't use them!
The reason is because they communicate with her insulin pump. I don't know what I would need to buy to be able to gather the information it collects for myself. But I actually would love, love, love to do it!
0 -
Is this (CGM) the same as what type 1 diabetics wear? I had one on me for a few days a couple years ago. I remember that the person that put it on me, must not of done it right. It was embedded into my skin, around my hip. The thing was itchy and my skin broke out for a couple weeks afterwards. It was a hassle to take a shower. Also, I couldn't review the data, it was in the meter, when my trial was over, the doctor downloaded the data. I had to keep a hand written log with food and time that I ate it. Then when the doc gave me the data, we linked it up to what and when I ate.
Now, what would be cool, is if it was non-evasive. I know there is a company working on that, maybe this is one of them? IDK. Also, I suspect that the newer ones have bluetooth, and you could probably get real time data. I don't see that he mentioned a brand name or type?
Dan
0 -
Here's his post (I realized I forgot to link directly to it) -- http://eatingacademy.com/personal/2016-update
He doesn't say which one, though I suspect it's the G5 mobile (https://www.dexcom.com/g5-mobile-cgm), which reports to your mobile phone app and would allow you to see the reports directly. There's also a chance he got a next-gen one, since it sounds like he was privy to that kind of information.
From what I've seen, while CGMs seem more common among Type 1 Diabetics, it's still rare among Type 2s, and I think low cost/accessible ones would really go a long way to helping that demographic, and probably do again for that segment what glucose meters did for it a few decades ago.0 -
Awesome. I always like what he has to say. Thanks!
Man... That would be so interesting to wear a CGM for a week. He gets all the best toys!
0 -
I just started a short trial on a Dexcom G4. It's been interesting so far...
The first two days saw a fair amount of disparity between my CGM and glucometer, but it has narrowed over time and more calibrations (4x daily input of BG of data points from the glucometer). The first night I got alarms for low BG, but in fact my meter showed my actual number to be 30 points higher.
Fortunately, this hasn't happened since.
I spent the second evening at a tavern, and the CGM insisted my BG remained steady between 85 and 100, despite the 115-125 readings on my glucometer.
Notes -- The initial period when using a new CGM sensor can be woefully reliable. Not a replacement for a glucometer!
- The CGM is useful for tracking trends, such as BG leading up to early-morning high fasting BG (Dawn Phenomenon), as well for monitoring reactions to calculated amounts of food, exercise, and medications.
- Clearly, a short trial is not long enough to do in-depth n=1 experiments - I'd want it for at least 4 weeks.
- The CGM measures glucose in insterstitial fluid, not blood, so the CGM readings may tend to lag glucometer readings if your BG is on the move.
2 -
I just started a short trial on a Dexcom G4. It's been interesting so far...
The first two days saw a fair amount of disparity between my CGM and glucometer, but it has narrowed over time and more calibrations (4x daily input of BG of data points from the glucometer). The first night I got alarms for low BG, but in fact my meter showed my actual number to be 30 points higher. Fortunately, this hasn't happened since.
I spent the second evening at a tavern, and the CGM insisted my BG remained steady between 85 and 100, despite the 115-125 readings on my glucometer.
Notes -- The initial period when using a new CGM sensor can be woefully reliable. Not a replacement for a glucometer!
- The CGM is useful for tracking trends, such as BG leading up to early-morning high fasting BG (Dawn Phenomenon), as well for monitoring reactions to calculated amounts of food, exercise, and medications.
- Clearly, a short trial is not long enough to do in-depth n=1 experiments - I'd want it for at least 4 weeks.
- The CGM measures glucose in insterstitial fluid, not blood, so the CGM readings may tend to lag glucometer readings if your BG is on the move.
Looking forward to more updates on this. I would love to get a CGM.
I have done a test of checking roughly ever 30 minutes of my waking day outside of a longer gap while I was at the gym and another because my commute home from work ran a bit long. In the end, I ended up with nearly 30 readings averaging 103 (range of 91 to 126 with 80% of them in the 100-105 range). All my PP along with my FBG matched within a point or two of what I normally get, so I have to assume the others were fairly typical as well. That would translate to an A1c of 5.2. However, my A1c was yet again 6.1 which would average to a 128. My meter (checking right before and right after lab blood draw) matches the lab, so I am trying to figure out how my A1c could continue to be that high given my I rarely get readings as high as 128, let alone having that be an average.
I have to assume I am plagued with the same issue as the esteemed Dr. Attia in terms of my RBCs. However, it would be nice to know for sure by use of a CGM.1 -
cstehansen wrote: »I just started a short trial on a Dexcom G4. It's been interesting so far...
The first two days saw a fair amount of disparity between my CGM and glucometer, but it has narrowed over time and more calibrations (4x daily input of BG of data points from the glucometer). The first night I got alarms for low BG, but in fact my meter showed my actual number to be 30 points higher. Fortunately, this hasn't happened since.
I spent the second evening at a tavern, and the CGM insisted my BG remained steady between 85 and 100, despite the 115-125 readings on my glucometer.
Notes -- The initial period when using a new CGM sensor can be woefully reliable. Not a replacement for a glucometer!
- The CGM is useful for tracking trends, such as BG leading up to early-morning high fasting BG (Dawn Phenomenon), as well for monitoring reactions to calculated amounts of food, exercise, and medications.
- Clearly, a short trial is not long enough to do in-depth n=1 experiments - I'd want it for at least 4 weeks.
- The CGM measures glucose in insterstitial fluid, not blood, so the CGM readings may tend to lag glucometer readings if your BG is on the move.
Looking forward to more updates on this. I would love to get a CGM.
I have done a test of checking roughly ever 30 minutes of my waking day outside of a longer gap while I was at the gym and another because my commute home from work ran a bit long. In the end, I ended up with nearly 30 readings averaging 103 (range of 91 to 126 with 80% of them in the 100-105 range). All my PP along with my FBG matched within a point or two of what I normally get, so I have to assume the others were fairly typical as well. That would translate to an A1c of 5.2. However, my A1c was yet again 6.1 which would average to a 128. My meter (checking right before and right after lab blood draw) matches the lab, so I am trying to figure out how my A1c could continue to be that high given my I rarely get readings as high as 128, let alone having that be an average.
I have to assume I am plagued with the same issue as the esteemed Dr. Attia in terms of my RBCs. However, it would be nice to know for sure by use of a CGM.
@cstehansen -
Even into Day 5, there is still significant variation between my meter and the CGM.
The meters I use are Contour Next EZs. Though they always read fairly close to each other and to the results of the lab I use, they are still sometimes 20-25 points at odds with what the CGM shows (the meters almost always being closer to the mean/median). Also, regular users report differing results from sensor to sensor and depending on where on your body the sensor is attached. So I'm not ready to put a lot faith in the absolute readings of the CGM.
For people with wide swings, the speed and magnitude of the changes in BG could be very important, but for those with relatively controlled blood glucose that strays out of the ideal range sometimes, I'm not sure a CGM adds much value to a good meter. But I'm endeavoring to keep an open mind for the rest of the trial period....
As for your pesky A1c.... as you know, there are many possible confounders of A1c results. Also, A1c is an average not only of values that are important but also of some that are not. For example, if your sleeping BG tends to run between 90 and 100 rather than 65 and 75, you'll get a higher A1c result, despite the fact that the values responsible for the difference are not of particular concern.
You might request that fructosamine tests be added to your regular labs. The fructosamine test aims to measure glycation of proteins rather than hemoglobin and is supposed to show an average BG over a shorter, more-recent period. Though potential confounding factors remain, they're a different herd than those of the A1c test.
Seems that frequent, regular spot BG testing with a reliable meter is still the best option.0
