An attempt to rationally define a glucose target for non-diabetic CGM users:
Per these linked works (both digs from second but good background in first) glycemic excursion is strongly associated with oxidative stress.
care.diabetesjournals.org/content/31/Sup…
jamanetwork.com/journals/jama/…
Per these linked works (both digs from second but good background in first) glycemic excursion is strongly associated with oxidative stress.
care.diabetesjournals.org/content/31/Sup…
jamanetwork.com/journals/jama/…
The table shows that in ostensibly healthy controls excretion of this oxidative byproduct is of a level that corresponds (in the DM2 group) to a mean amplitude of glycemic excursions of 40 mg/dL.
Note extrapolation in one group from another. These aren’t perfect data and lowish n (21 iirc). Also we don’t know if this is really the optimum—is there a lower bound where the oxidation vs MAGE curve flattens out?
(Or if we reduce oxidation too much will this be deleterious?)
(Or if we reduce oxidation too much will this be deleterious?)
Finally, not having played with my Dexcom G6 yet, I’m not sure if MAGE is easily derivable from the Dexcom (or Freestyle Libre) interface.
In the first reference the authors argue that measures only of s.d. are insufficient.
In the first reference the authors argue that measures only of s.d. are insufficient.
As a side note, note that the oxidative stress was measured via a proxy of an end product of… wait for it… arachidonic acid.
(The n-6 intake <==> AA level connection is tenuous but dietary n-6 ~= oxidative stress has been shown by a few groups. In humans, no less!)
(The n-6 intake <==> AA level connection is tenuous but dietary n-6 ~= oxidative stress has been shown by a few groups. In humans, no less!)
More data:
sciencedirect.com/science/articl…
MDA is malondialdehyde, NGR and IGR normal and impaired glucose regulation.
NGR had lower MDA, which correlated with MAGE of 1.6 mmol/l, which is 28.83 mg/dL. So concordant, another target in the 30-40 mg/dL range for putative normals.
sciencedirect.com/science/articl…
MDA is malondialdehyde, NGR and IGR normal and impaired glucose regulation.
NGR had lower MDA, which correlated with MAGE of 1.6 mmol/l, which is 28.83 mg/dL. So concordant, another target in the 30-40 mg/dL range for putative normals.
link.springer.com/article/10.100…
Another one where the normal glucose regulation group had a 1.60 mmol/l MAGE. Rather noisy data in their log-transformed plot of oxidative byproducts vs. MAGE.
Another one where the normal glucose regulation group had a 1.60 mmol/l MAGE. Rather noisy data in their log-transformed plot of oxidative byproducts vs. MAGE.
A rather tightly glycemically controlled mean value for normal Chinese subjects in this graph.
"Reference values for continuous glucose monitoring in Chinese subjects" is what ref 9 refers to here. They didn't report MAGE, but it would have been pretty low.
"Reference values for continuous glucose monitoring in Chinese subjects" is what ref 9 refers to here. They didn't report MAGE, but it would have been pretty low.
1.54 mmol/l glucose is about 28 mg/dL for their normal control MAGE value. Again in that 30-40 mg/dL range.
Of note in this one is that their HOMA-IR values (presumably HOMA1) were super low, IMO, including for the insulin-resistant-by-definition T2DM!
Of note in this one is that their HOMA-IR values (presumably HOMA1) were super low, IMO, including for the insulin-resistant-by-definition T2DM!
One last one for now:
academic.oup.com/jcem/article/9…
Only looked at T2DM and didn't report as MAGE (since not a CGM study--finger sticks at 0-120 min at 30 min intervals instead). Thus the lowest category for A1C being < 7.0…
academic.oup.com/jcem/article/9…
Only looked at T2DM and didn't report as MAGE (since not a CGM study--finger sticks at 0-120 min at 30 min intervals instead). Thus the lowest category for A1C being < 7.0…
This is an interesting one despite its lack of direct relevance to non-diabetics because it shows there are measurable benefits to both lower mean glucose levels (A1C) and incremental glucose peaks.
IGP: max incremental increase in glucose obtained at any point after a meal.
IGP: max incremental increase in glucose obtained at any point after a meal.
So through this I hope I've shown so far a few things:
a) Normal MAGE is on the order of 30-40 mg/dL.
b) MAGE is positively correlated with oxidative byproducts, and by extension from incremental glucose peak with CIMT.
c) No lower limit clearly established yet.
a) Normal MAGE is on the order of 30-40 mg/dL.
b) MAGE is positively correlated with oxidative byproducts, and by extension from incremental glucose peak with CIMT.
c) No lower limit clearly established yet.
sciencedirect.com/science/articl…
Association with advanced glycation end products when groups split between < 40 and > 40 mg/dL MAGE. I would have much rather liked to see a scatterplot of MAGE versus Glycer-AGE, but this is enough to see that even this relationship is noisy.
Association with advanced glycation end products when groups split between < 40 and > 40 mg/dL MAGE. I would have much rather liked to see a scatterplot of MAGE versus Glycer-AGE, but this is enough to see that even this relationship is noisy.
One big possible confounder is dietary consumption of AGEs:
ncbi.nlm.nih.gov/pmc/articles/P…
AGEs are apparently coveted in food production because they are tasty...
sciencedirect.com/science/articl…
A conundrum for the ages.
ncbi.nlm.nih.gov/pmc/articles/P…
AGEs are apparently coveted in food production because they are tasty...
sciencedirect.com/science/articl…
A conundrum for the ages.
sciencedirect.com/science/articl…
Large n. 22 year followup. Mortality data, and reporting of data in ranges rather than just cutoffs (e.g. 140 mg/dL).
Large n. 22 year followup. Mortality data, and reporting of data in ranges rather than just cutoffs (e.g. 140 mg/dL).
Difficult to summarize, but lowest all cause mortality generally with 1 hour post-prandial glucose values on the range of 100-120 mg/dL, very roughly.
This begs the question of whether it's the glucose at action here or whether these were just reflections of metabolic health.
This begs the question of whether it's the glucose at action here or whether these were just reflections of metabolic health.
And if the latter true then does "cheating" by lowering carbs so as to never broach 120 mg/dL actually do anything?
care.diabetesjournals.org/content/21/3/3…
3 studies summarized. Whitehall Study reference ranges low (50 g glucose load as opposed to 75 or 75/90 g for the other two).
From the Paris and Helsinki studies 2 hour glucose under 122 and 113 mg/dL associated with lowest all cause mortality.
3 studies summarized. Whitehall Study reference ranges low (50 g glucose load as opposed to 75 or 75/90 g for the other two).
From the Paris and Helsinki studies 2 hour glucose under 122 and 113 mg/dL associated with lowest all cause mortality.
The most interesting thing about these data to me is from the Helsinki study. Their 80th percentile range was pretty low, and was _not_ associated with the lowest all-cause mortality. This is weak evidence that there may be a floor––lower at all costs may not be ideal.
diabetes.diabetesjournals.org/content/36/6/6…
Possibly extra relevant to me being half-Japanese. Risks expressed relative to quintile 1. Quintile 2 doesn't seem that bad. 50 g glucose at 1 hour.
Quintile 1 was mean 99.7 (40-114) mg/dL, quintile 2: 123.9 (115-133) mg/dL. Again that 100-120 range.
Possibly extra relevant to me being half-Japanese. Risks expressed relative to quintile 1. Quintile 2 doesn't seem that bad. 50 g glucose at 1 hour.
Quintile 1 was mean 99.7 (40-114) mg/dL, quintile 2: 123.9 (115-133) mg/dL. Again that 100-120 range.
@threadreaderapp unroll
