hba1c comment

[Wuzzy]

My previous comment is innaccurate, HBA1C appears to be caused
by recalled carbohydrate intake more than it is by body weight
in healthy humans. r=0.15 for carb and r=0.10 for bmi.

Fasting glucose on the other hand is probably not a marker of
carb intake.

Given that this is the case, can anyone advise me why
HBA1C is not recommended as a diagnostic? but only for
management of bgs..

AT any rate I'm pretty convinced it acts as a marker of
glucose intake, this suggests that blood glucose does spike
after a meal enough to make it into protein structures.

oddly fiber appears to protect against this, this might just
be due to math: eg., the non-digestible part of carbohydrate
protects against spikes..

[Pbeyer]

wuzzy wrote:

> My previous comment is innaccurate, HBA1C appears to be
> caused by recalled carbohydrate intake more than it is by
> body weight in healthy humans. r=0.15 for carb and r=0.10
> for bmi.
>
> Fasting glucose on the other hand is probably not a marker
> of carb intake.
>
> Given that this is the case, can anyone advise me why HBA1C
> is not recommended as a diagnostic? but only for management
> of bgs..
>
> AT any rate I'm pretty convinced it acts as a marker of
> glucose intake, this suggests that blood glucose does spike
> after a meal enough to make it into protein structures.
>
> oddly fiber appears to protect against this, this might just
> be due to math: eg., the non-digestible part of carbohydrate
> protects against spikes..

Hi Wuzzy Are you speaking of glycosylate Hb in normal, obese
or diabetic populations-- behavior and use might vary in each
group. Its normally used in type 1 and 2 for evaluation of
chronic control of blood sugar, but it has its pitfalls. For
example, one can signifacantly improve one's HbA1c in a short
period by fasting, extreme exercise, etc. Meds may also
interfere/alter its response. Pete

[Eric Bohlm]

mypcos@hotmail.com (wuzzy) wrote in
news:d996c21a.0301162038.734f2afd@posting.google.com:

> My previous comment is innaccurate, HBA1C appears to be
> caused by recalled carbohydrate intake more than it is by
> body weight in healthy humans. r=0.15 for carb and r=0.10
> for bmi.

Those are *awfully* weak correlations: if variations in
recalled carb intake account for 2.25% of the variation in A1C
and variations in BMI account for 1%, then you've got 96.75%
of the variation left unaccounted- for.

If "healthy" excludes pre-diabetic (IFG or IGT) subjects as
well as diabetics, then I'd have to assume that most of the
variation in A1C is going to be due to variation in
susceptibility to glycation rather than variability in glucose
availability for glycation, simply because BG level is going
to be under very tight homeostatic control.

> Fasting glucose on the other hand is probably not a marker
> of carb intake.

It's chiefly a measure of hepatic glucose production, cerebral
glucose uptake, and adipose glucose uptake (OK, in very active
individuals there's probably extra muscular glucose uptake to
replace depleted glycogen stores).

> Given that this is the case, can anyone advise me why HBA1C
> is not recommended as a diagnostic? but only for management
> of bgs..

Probably mostly for practical reasons: although there's an
effort (the NGSP) to standardize A1C measurement
methodologies, right now there's no full standardization and
A1C values from one lab can't be directly compared to those
from another; the normal ranges go all over the place. And the
fact that an A1C test is more expensive than a fasting BG test
(the latter is included in standard blood panels).

> AT any rate I'm pretty convinced it acts as a marker of
> glucose intake, this suggests that blood glucose does spike
> after a meal enough to make it into protein structures.

IIRC, attempts to correlate A1C with post-prandial glucose
readings have been disappointing; apparently short-term spikes
don't create much glycation (which is a two-step reaction with
the first step being readily reversible). Yet there is some
evidence associating PP spikes with risk for macrovascular
disease, *and* A1C doesn't predict macrovascular complications
as well as it does microvascular complications.

> oddly fiber appears to protect against this, this might just
> be due to math: eg., the non-digestible part of carbohydrate
> protects against spikes..

Or slower absorption, or delay of absorption until
second-phase insulin secretion is under way.

[Wuzzy]

need to run more tests, feel like I don't know much, but I'll
try to comment:

Eric Bohlman <ebohlman@earthlink.net> wrote in message
news:<Xns93065D4D4D3C5ebohlmanomsdevcom@130.133.1.4>...
> mypcos@hotmail.com (wuzzy) wrote in
> news:d996c21a.0301162038.734f2afd@posting.google.com:
>
> > My previous comment is innaccurate, HBA1C appears to be
> > caused by recalled carbohydrate intake more than it is by
> > body weight in healthy humans. r=0.15 for carb and r=0.10
> > for bmi.
>
> Those are *awfully* weak correlations: if variations in
> recalled carb intake account for 2.25% of the variation in
> A1C and variations in BMI account for 1%, then you've got
> 96.75% of the variation left unaccounted- for.

True, my numbers are from epidemiology, not experiment. So
this is very common, similarly long-term dietary B12 only
predicts <3-5% of the variance in serum B12 even though all of
it comes from diet and the "true" correlation is very high.
r=0.15 is considered moderate correlation in nutritional epi,
not low. The reason is that nutrients are awfully poorly
measured. Thus it is still possible that HBA1C is modestly
affected by postprandial (3hrs every meal) glucose.

----LONG WINDED OFF TOPIC on FFQ measurement
error------------------- "Nutritional Epidemiology" has a
chapter on error+attenuation of measured intake. In it an
example: (relying on recall) Aside:the equation of regression
attenuation from Beaton GH is r2=r1/(1+w/b) where
r2=attenuated-pearson's r1=true w=within variance, b=between.

Cholesterol has a w/b of 5. (the within to between variance is
5*) Thus correlation coefficient of saturated fat on
cholesterol as per keys+hegsted (the True r) is 0.15. This
will be diluted all the way to r=~0.02 in epidemiologic survey
as per the Beaton equation and it will never show up
significant!!. ---------------------LONG WINDED OFF
TOPIC-------------------

Couple of additional points:
1)fasting glucose predicts 50% of the variance in A1C but only
if you keep diabetics in there. In healthy persons fasting
glucose predicts very little A1C. Around 2% better
prediction. Weight another 1% and diet another 1%.

If you keep diabetics in the equation, though diet no longer
becomes a predictor, since effect of weight, age and insulin
senstivity becomes to complex but you can now predict 50% of
the variance with these latter.

I like just measuring normal people and excluding diabetics
since you get more "common sense" results and tend to have
larger sample size.

Nevertheless it shows that food does cause glycation and that
it is measurable if done right..

I suspect in diabetics, most of the problem is not from high
carb intake - most of it is from having poor tolerance to
glucose loads mainly from body weight and sedentariness.

>
> If "healthy" excludes pre-diabetic (IFG or IGT) subjects as
> well as diabetics, then I'd have to assume that most of the
> variation in A1C is going to be due to variation in
> susceptibility to glycation rather than variability in
> glucose availability for glycation, simply because BG level
> is going to be under very tight homeostatic control.

mostl likely there is variation in susceptibility glycation..

>
> > Fasting glucose on the other hand is probably not a marker
> > of carb intake.
>
> It's chiefly a measure of hepatic glucose production,
> cerebral glucose uptake, and adipose glucose uptake (OK, in
> very active individuals there's probably extra muscular
> glucose uptake to replace depleted glycogen stores).

I would say also body weight and beta-cell failure.. Usually
I prefer looking at insulin and lipids since glucose is
usually normal in subclinical diabetics. (the majority).
HBA1C something I don't have experience with, currently
testing out ideas.

>
> > Given that this is the case, can anyone advise me why
> > HBA1C is not recommended as a diagnostic? but only for
> > management of bgs..
>
> Probably mostly for practical reasons: although there's an
> effort (the NGSP) to standardize A1C measurement
> methodologies, right now there's no full standardization and
> A1C values from one lab can't be directly compared to those
> from another; the normal ranges go all over the place. And
> the fact that an A1C test is more expensive than a fasting
> BG test (the latter is included in standard blood panels).

Appears you are right, may be mostly a matter of
standardization..

>
> > AT any rate I'm pretty convinced it acts as a marker of
> > glucose intake, this suggests that blood glucose does
> > spike after a meal enough to make it into protein
> > structures.
>
> IIRC, attempts to correlate A1C with post-prandial glucose
> readings have been disappointing; apparently short-term
> spikes don't create much glycation (which is a two-step
> reaction with the first step being readily reversible).

interested, I'll look that up..

>Yet there is some evidence associating PP spikes with risk
>for macrovascular disease, *and* A1C doesn't predict
>macrovascular complications as well as it does microvascular
>complications.

again interesting, suggests to avoid having big carbohydrate
meals and more smaller, spaced out, I think.. Probably a limit
to how much carb person can tolerate at one meal, though liver
does a pretty good job at handling large doses.. It stops
producing about 20g and can store another ~20g before much
hits the blood..(those are ballpark figures, i might be able
to come up with a better upper limit)

>
> > oddly fiber appears to protect against this, this might
> > just be due to math: eg., the non-digestible part of
> > carbohydrate protects against spikes..
>
> Or slower absorption, or delay of absorption until
> second-phase insulin secretion is under way.

this would explain why some epi studies find Glycemic Load
more important than total carb in terms of HbA1C..