Freestyle libre 3 installed

[mccoy]

At last, I took the time to order and install the new Freestyle Libre 3 sensor, its main advantage being its continuous data acquisition feature with one-minute sampling intervals.

I'm a little baffled since I'm seeing no significant signals, perhaps just noise. This morning I checked my FBG with a traditional strip glucometer and it was 84 mg/dL. The Libre 3 indicated 76, but then it indicated 91. There is some variability. Does it capture the peaks? I don't know yet, I only see noise on the graph so far, maybe because I'm eating no sugars, cereals or other readily absorbed carbs. At lunch I had a green banana at the end of a meal consisting of raw salad, walnuts and seeds, and nonfat mozzarella. No peak at all. AT breakfast I had 750 gr of kefir. Nothing.

Maybe later on I'm going to indulge some cacao with honey and see if the graph budges at all. 

[IgorF]

There is a delay, or I would say rather a hysteresys loop with interstitional glucose comparing to serum. A bit on it https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2903977/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3837059/

I have a bad feeling that CGMs are becoming more and more "algorithmically compensating" the same way as wrist bands for the things they measure. I hope they both are better than weather forecast like - "70% chance tommorow will be the same as today".

Also, another thing - a triggering level, some small amounts of sugar will not trigger the fast insulin response, thus will not be tackled and will raise the glucose. I have a feeling that some fundamental mechanism is in place, the same way as pulse+tone for dopamin, gh, ghrelin-leptin and other hormonal stuff in many areas.

In any case, since libre 3 is also doing it now continously mayby I will also pick it as my next one, if dexcom 7 will be more costly.

 

EDITED to ADD:

I forgot about another thing to be taken into account - ISF and blood could act as 2 compartments with bidirectional flow, this should also be taken into account when observing some strange curves or 50%/50% results distribution from similar experiments, at least I used this explanation in the past to lower my own confusion

https://www.liebertpub.com/doi/10.1089/dia.2016.0112

Edited July 8, 2023 by IgorF

[mccoy]

The GCM has sensed the glucose trough due to today's zone 2 exercise (30 minutes). At last it gave some signs of life! Coffee with one abundant teaspoon of honey was not seen by the GCM at all. 

Edited July 8, 2023 by mccoy

[Ron Put]

3 hours ago, mccoy said:

The GCM has sensed the glucose trough due to today's zone 2 exercise (30 minutes). At last it gave some signs of life! Coffee with one abundant teaspoon of honey was not seen by the GCM at all. 

My experience with the Freestyle 1 was that it was rather random for the first few days at the beginning and the last week or so before the end of the recommended use period. My second one seemed to consistently lower by about 10 point than the first one, so I am not entirely confident about the granular accuracy of these things. I used it with the January AI app, which was next to useless, IMO, even though it's probably one of the better ones in the marketplace.

Based on my use of two sensors, I decided that overall, my spikes and returns were normal. I also did the 75ml glucose challenge that January AI provided, and it sent me up to 170+, then crashed me in the 50s about 3+ hours later. I had fasted for maybe 18 hours, and it was not a pleasant experience. But I am really not sure that the challenge has much value without measuring insulin, which also goes for the regular measurements.

I think the companies that sell these to non-diabetics oversell the validity of a flatline  without the perspective of corresponding insulin secretion, and some use it to push keto marketing since that's all it lends itself to.

Keto, based on what I know and understand, is detrimental to overall health and longevity in the long-term, and GCMs for non-diabetics are essentially a business model to push it and maintain subscriptions, as well as to sell aggregated data for additional monetization.

McCoy, can you buy the Freestyle 3 without prescription in Italy? Did you just walk into a pharmacy, or did you have to order it from Freestyle?

[mccoy]

46 minutes ago, Ron Put said:

McCoy, can you buy the Freestyle 3 without prescription in Italy? Did you just walk into a pharmacy, or did you have to order it from Freestyle?

In Italy you can buy directly from Abbot, I never saw it in the pharmacy but I should ask, also there is no prescription needed and those who are T1 diabetics have it for free. The cost is about 65 Euros per sensor (2 weeks duration).

46 minutes ago, Ron Put said:

My experience with the Freestyle 1 was that it was rather random for the first few days at the beginning and the last week or so before the end of the recommended use period.

It may be that it needs some time to adjust, even though theoretically it should be ready from the beginning

 

46 minutes ago, Ron Put said:

I think the companies that sell these to non-diabetics oversell the validity of a flatline

Presently my signal really looks like a flatline with some noise and my main criticism of the system, more precisely the app, is that I cannot adjust the Y-axis max values, I really don't need values up to 350 mg/dL, they rather are detrimental because I lose track of minor peaks and troughs that are of interest to me and to all other users who are not T2 or T1 diabetics.

Edited July 8, 2023 by mccoy

[mccoy]

Today at lunch I ate the usual vegetables, walnuts and seeds with a little parmesan, and at the end one green banana and two unripe fresh figs. The detected concentration reached a little over 115 mg/dL (red circle below), the highest I saw since I installed it. Maybe they are realistic concentrations and my glucose tolerance after 2 years of restriction has improved. Hypothesis to be verified.

Edited July 9, 2023 by mccoy

[mccoy]

OK, tonight I verified the sensor is all right, it is rather my dietary regimen and much-improved glucose tolerance that are avoiding any significant spikes.

I drank over one pint of pineapple juice, with a declared content of 10% carbs, on an empty stomach. After about 30 minutes the peak reached a little above 180 mg/dL, to decrease soon after.

 

[Ron Put]

On 7/9/2023 at 1:50 PM, mccoy said:

drank over one pint of pineapple juice, with a declared content of 10% carbs, on an empty stomach. After about 30 minutes the peak reached a little above 180 mg/dL, to decrease soon after.

When I first read this, I thought it was pretty good, but something I seemed to remember made me curious and I checked pineapple juice -- as long as it is unsweetened, it in fact has a relatively low glycemic index.

Which may indicate that your high spike may be due to some insulin resistance.

For (my own) reference, I searched for my post about a glucose challenge I did after fasting for about 18 hours while wearing the Libre 1, and you can see the result here:
 

Without insulin measurements we can only guess, but IMO, this may still point to insulin resistance as a result of a high fat diet.

[mccoy]

23 hours ago, Ron Put said:

Without insulin measurements we can only guess, but IMO, this may still point to insulin resistance as a result of a high fat diet.

Ron, I doubt that hypothesis, for several reasons. First, my fasting insulin tends to be very low. Second, my visceral fat is very low, presently I can see a sixpack on my abdomen. Third, I tend to be in a state of catabolism rather than anabolism, meaning If my insulin levels were high I should be more anabolic I suppose, whereas I fatigue not to lose weight and presently I'm just losing weight eating ad libitum. The glucometer peak was pretty sharp and not extended with time. Insulin resistance tends to keep glucose higher for longer times after meals, the opposite resulted in my case.

According to the Mayo Clinic insulin resistance may be diagnosed by the signs or symptoms listed below, which I do not exhibit luckily.

While I understand that your body is very sensitive to all fats, mine seems to thrive with healthy fats like nuts, seeds and EVOO. The hypothesis of intramyocellular lipids causing insulin resistance may be true but not all individuals may absorb a significant amount of lipids into muscle tissues, especially if a normo-caloric diet is followed. 

I didn't post more into this thread because I have to recover the records of all my 2 weeks of wearing the glucometer and attach them here, they are probably representative of a low-carb,  high fats plant-based diet but I included some fruit.

I also wonder how would look the glucose signals of people like yourself who are more accustomed to and thrive with a high-carb, very low-fat diet. The comparison would be interesting and maybe unpublished so far.

 

Quote

 

What are the symptoms?

Very often people with insulin resistance don't have any symptoms at all. It is usually picked up by their doctor during an annual health exam or routine blood work. There are some signs of insulin resistance that your doctor may look for. These includes a waistline over 40 inches in men, and a waistline over 35 inches in women. Skin tags or patches of dark velvety skin called acanthosis nigricans. A blood pressure reading of 130 over 80 or higher. A fasting glucose level equal or above 100 milligrams per deciliter. Or a blood sugar level equal or above 140 milligrams per deciliter two hours after a glucose load test. An A1C between 5.7% and 6.3%. A fasting triglycerides level over 150 milligram per deciliter. And an HDL cholesterol level under 40 milligrams per deciliter in men, and an HDL cholesterol level under 50 milligrams per deciliter in women.

 

 

[mccoy]

OK, I was able to save the whole report which is pretty extensive. The Libra3 is steps ahead of its predecessors since now data acquisition is automatic and most of the nights are recorded. The plot below is an impressive synthesis of all the data collected, superimposed on a 24-hours day, so that the variability of the whole 2 weeks period is visible.

Low glucose events, especially the nocturnal ones, are probably due to pressure on the sensor (bodyweight pushing the sensor against the mattress).

The average concentrations tended to rise after lunch and dinner, whereas they practically did not rise after breakfast, which was usually 0.75 liters of unsweetened kefir.

The highest visible peak is my test with pineapple juice, other peaks being pretty moderate and probably in concomitance with eating bananas or other fruit (edit: I'm remembering that I was eating fresh figs as well from my tree, fruit with hi-sugars content even if picked not too ripe).

As the literature suggests, by night the body has less tolerance to glucose, which in my case returned to baseline at 3 AM (could take longer with a carbs-rich dinner, depending on many other factors, individual or not).

Troughs are probably due to exercise or limited sensor sensitivity during the first couple of days.

Edited July 31, 2023 by mccoy

[mccoy]

This plot, which I expanded in height, shows that in 90% of data (5th to 95th percentiles, grey swath), the numbers never went above 125 or below 70 mg/dL (except nocturnal lows), showing what I would consider significant stability of the glucose readings. The homeostatic mechanism which governs blood glucose concentrations seems to be efficient, also it has not been challenged with large glycemic loads.

I have seen perhaps flatter curves, in the case of Peter Attia, but he takes SGLT2 inhibitors, which inhibit the reuptake of glucose at the kidneys level.

 

Edited July 30, 2023 by mccoy

[mccoy]

By the way, I may soon modify my present regime and introduce some more carbs, since I've entered a catabolic phase that is not reverting. In other words, I've lost too much weight to my tastes (mostly fat but muscle mass as well) and seem not able to reverse this trend. I'm back when I was at the end of the latest Longo's FMD. The reason I hypothesize is the low glucose/insulin signal.

As can be observed in the following snapshot, glucose has a double effect, directly stimulatory through the LamTorc-Rag subpath, but also elevating insulin and IGF1, which are growth factors stimulating the PIK3-AKT subpath.

A few years ago I had the same problem, low carb diet, and loss of bodyweight even if training. This time around I tried with more protein, but the effect apparently is the same.

I don't know if there are any ways around it, excluding the use of testosterone and GH. Maybe elevating calories with abundant fats, but it would be detrimental to digestion, heart rate, HRV, BP, sleep, and so on. 

This is an optimization problem that is not easy to solve, juggling with energy/calories, glucose, leucine, and other EAAs, while keeping the MGFs (mechanogrowth factors) active....

 

[Ron Put]

McCoy, I may be totally wrong, but based on what I read, a prolonged high-fat, low-carbs diet will likely result in insulin resistance. This looks like quite a spike from pineapple juice, which if not sweetened is generally fairly low on the glycemic scale.

Low fasting insulin can be misleading on a high-fat diet... To be safe, I'd really invest in a glucose challenge if I were you, with corresponding insulin measurements, just to be safe.

HoMA does not identify insulin resistance in cohorts with a high prevalence of pre-diabetes: a study of Black africans african americans

Insulin resistance can be accurately determined from the insulin sensitivity index (SI) with data from frequently sampled intravenous glucose tolerance tests (FSIGT). However, FSIGT are time consuming and expensive. As an alternative, homeostasis model assessment (HOMA) is an inexpensive way to measure insulin resistance. HOMA is calculated using the product of fasting glucose and fasting insulin. HOMA is founded on the principle that fasting insulin levels are increased in the presence of insulin resistance. However, ß-cell failure can lead to low fasting insulin levels. Therefore, in pre-diabetes, which is characterized by both insulin resistance and ß-cell failure, HOMA could be inappropriately low. Using SI as the reference, our goal was to determine if HOMA accurately measures insulin resistance in a cohort with a high prevalence of pre-diabetes. Sixteen Black Africans (BA) [75% male, age 38±8y (mean±SD), BMI 27.2±3.8 kg/m2] were matched by sex, age and BMI to 16 African Americans (AA). All pairs had oral glucose tolerance tests (OGTT) and FSIGT. Insulin resistance was measured by HOMA and SI ß-cell secretion was determined by the acute change in insulin response to glucose (AIRg) during the FSIGT. BA had a higher rate of pre-diabetes than AA. SI did not differ between BA and AA, but BA had lower HOMA and lower AIRg (Results in Table). In conclusion, SI was similar for BA and AA, but HOMA was significantly lower in BA. Therefore, HOMA did not reflect insulin resistance in BA. The cause for the inappropriately low HOMA in BA appears to be ß-cell failure. Evidence of ß-cell failure in BA was their high rate of pre-diabetes and low AIRg. Our data suggest HOMA should be used with caution in cohorts with a high prevalence of pre-diabetes and ß-cell failure.

Edited August 2, 2023 by Ron Put

[mccoy]

Ron, it is pretty sure that ketogenic diets (and to a lesser extent low-carb diets) cause temporary glucose intolerance, maybe because of some insulin resistance, although I'm not totally convinced. In my case, most probably I did not plunge into the ketogenic state and the diet is not very low carb, although my body may read it as very low, to tell the truth.

I don't know about pineapple juice, I don't remember if it was sweetened but I'm not very sure that filtered fruit juice may have those properties, it is absorbed almost immediately, the sharp peak is typical of sugary drinks, fruit juice, even watery fruits eaten without other foods. 

Now, what the article proposes is some degree of ß-cell failure to explain a low HOMA and that may be a concern, it should be coupled to pre-diabetes though and to a generally lower tolerance to glucose.

I'll try and go down this rabbit hole, for now, I'll just try to resume eating some fruit since it may be that the lengthy low-carb regime triggered some catabolic signal with stubborn weight loss or lowering of the bodyweight setpoint. 

Bottom line: life wasn't meant to be easy!

[Ron Put]

1 hour ago, mccoy said:

don't know about pineapple juice, I don't remember if it was sweetened but I'm not very sure that filtered fruit juice may have those properties, it is absorbed almost immediately, the sharp peak is typical of sugary drinks, fruit juice, even watery fruits eaten without other foods.

Hi, I just did a quick search on pineapple juice and it seems to confirm my impression of it having a relatively low GL::

Pineapple Juice (unsweetened): Glycemic Index (GI), glycemic load (GL) and calories per 100g (glycemic-index.net)

It's why I thought that 180 is a pretty high spike for pineapple juice, but if it had tons of sugar, it would make sense (I just assumed that you would be an unsweetened juice type 🙂

I have no reference for your response, but was comparing it to my spike to 174 after drinking 75ml of pure glucose, which was my highest spike while wearing the sensor, if I remember. For (my) reference, during 2022 my diet consisted of 58% carbs, 15 protein and 26 fat. Most of my carbs are whole and the fiber was 70g, and starch 71g on daily average of less than 1800 calories, according to Chronometer.

Edited August 2, 2023 by Ron Put

[mccoy]

Ron, we may discuss more in depth about pineapple, but the GI concept may be misleading, since even if fruit has fructose in it, which tends to lower the GI compared to some cereals, usually it's consumed in relatively large amounts, so the glycemic load becomes significant. Right now I consulted cronometer and pineapple juice seems to have 10 grams of pure sugars, of which abut 50% is glucose, just drinking 500 gr, which is not very much in a hot day, we ingest 25 g of pure glucose in a water solution, plus 7.5 grams sucrose, which tends to spike blood sugar as well. It's not like a glucose challenge with 75 grams of it, but it is not even a very mild glucose solution. My understanding also is that insulin resistance should delay the spike, that is, the BG values do not return to baseline quickly (in a 75 g glucose challenge, after 2 hours BG should be < 140 mg/dL, then there is no (pre)diabetes, no insulin resistance is deemed to occur.

Another aspect that I found meaningful, is that the lactose from dairy products doesn't seem to affect much blood glucose. Not at all in my kefir-based breakfasts, always = or > 500 gr. This has been a known fact for a while, however, indiscriminate consumption of such products is not advised to diabetics since the whey fraction tends to stimulate insulin secretion, independent of glucose.