Jean-Louis:
>>From Ellenberg and Rifkin's "Diabetes Mellitus", 5th Edition, 1997,
>GIP (Gastric Inhibitory Polypeptide, or Glocose dependent Insulinotropic
>Polypeptide), is a gut hormone which stimulates insulin release (some
>hormones stimulate, others inhibit insulin release). The effectiveness of
>GIP to stimulate insulin release is dependent on the plasma glucose level
>(the higher the glucose level, the higher the effectiveness). The exact
>contribution of GIP to the insecretin effect is currently inclear because
>its insulin-stimulating effect may vary with species.

This confirms the connection that is usually made between glucose and
insulin levels and how if blood glucose levels are lowered by adding fat to
carbohydrate, then GIF will trigger the release of only small amounts of
insulin.

>These results suggest the following mechanism:
> -Fat stimulates GIP release
> -GIP stimulate insulin secretion when blood glucose level is elevated
> -therefore, despite a lower blood glucose with butter+potato than with
>potato alone, the consequence of a higher level of GIP is that the insulin
>curve is almost unaffected by the addition of fat.

But as you have pointed out, higher levels of GIP can cause the insulin
response of future meals to be elevated.

>The aim of that article was to test the conjecture that the RATE of rise
>of GIP, rather than the STEADY STATE LEVEL achieved, stimulates insulin
>secretion when fat and carbohydrate are ingested together. To do that, the
>authors administered a 5 g iv bolus dose of glucose 15 minutes after the
>meal, when the RATE of rise of GIP was highest but the LEVEL still LOW,
>and observed the insulin response. It appears that the peak insulin after
>the mixed meal was 60% higher than after the carbohydrate meal. That seems
>to confirm their hypothesis.

Yes, but in the study you mentioned before this, equal amounts of fat and
carbohydrate were administered causing no surge of insulin:

>[Collier et al., The effect of co-ingestion of fat on the glucose, insulin
>and gastric inhibitory polypeptide responses to carbohydrate and protein,
>Am J Clin Nutr 37:941]:
>The effects of co-ingestion of 50 g fat (butter) with 50 g carbohydrate
>(potato) compared with potato alone: markedly decreased glucose, very
>slightly decreased insulin, markedly increased GIP.

Whereas in the experiment you are mentioning now the fat/carbohydrate ratio
is much smaller which is maybe why there is a rise in insulin levels.  This
suggests that only if small amounts of fat are consumed with carbohydrate
will insulin levels rise. See below:

>Similar experiments and results with 75 g carbohydrate (438 g potato or
>121 g lentils) with or without 37.5 g fat (46.5 g butter). Of course, all
>the effects are more important with the rapidly absorbed carbohydrate
>(potato) than with the slowly absorbed carbohydrate (lentils).

Confused? - do not read any further. ;-)

>It should be noted that, in a practical situation, nobody will inject you
>some glucose after lunch, so, that the insulin peak is higher after such a
>practice following a mixed meal shouldn't be of concern. More important
>for us "consumers", the first two articles of Collier show that eating fat
>with carbohydrate indeed prevents skyrocketing of blood glucose levels,
>but that we shouldn't feel too protected by the co-ingestion of fat, since
>it barely affects insulin levels.

If co-ingestion of fat barely effects insulin levels, surely we should
indeed feel protected?  If insulin levels already are low, fat can hardly
be expected to lower them any further.

>Moreover, I would add, from [Collier et al., concurrent ingestion of fat
>and reduction in starch content impairs carbohydrate tolerance to
>subsequent meals, Am J Clin Nutr 45(5):963-9, 1987] that, even if
>ingesting fat with carbohydrate at a meal doesn't harm you, it can at the
>next meal.

This is what I find the most interesting.  I still wonder why GIF seems to
have a delayed effect.

Peter:
>> If the effect is almost zero or even positive it is not clear to me why the
>> carbohydrate tolerance to future meals will be impaired.  Of course there
>> is the GIP factor but I doubt that more than a handful of very educated
>> scholars really understand the mechanism behind it.

Jean-Louis:
>Perhaps the exact mechanism hasn't been found yet. There are many
>experimental data about response to different carbohydrates with various
>quantities of protein and/or fat, and all of those are not consistent with
>each other, nor do they provide any theorical models that I know of. For
>instance, [Westphal et al., Metabolic response to glucose ingested with
>various amounts of protein, Am J Clin Nutr 1990 (62):267-272] found that
>protein (very lean beef) with 50 g glucose barely changes the insulin
>curve, whereas [Spiller et al., Effect of protein dose on serum glucose
>and insulin response to sugars, Am J Clin Nutr 1987 (46):474-80] found
>that 58 grams of carbohydrates (maltodextrin+fructose+lactose) ingested
>with various amounts (16,25,34,50 grams) of protein (1/3 milk proteins and
>2/3 soy proteins) approximately doubles the area under the insulin curve
>compared with the carbohydrates ingested alone.

An increasingly confusing picture. The second example does support the
notion that mixing carbs with proteins can inhibit the production of human
growth hormone (insulin inhibits HGH) and in the case of the lean beef we
do not know if maybe not enough was ingested to balance the 50g of glucose.

Jean-Louis, I appreciate all your work and effort in digging up all this
interesting research.  It is becoming clearer and clearer to me how so many
scholars selectively pick the research they need to confirm their already
preconceived notions.  When Loren Cordain posted his message on Paleodiet a
couple of months ago about how mixing fats and carbohydrates caused insulin
levels to rise, I was already suspicious.  Supplying scientific references
can be a very effective way of shutting up any possible critics, for only
the smallest minority is going to check them anyhow or even check in the
literature to see what else has been published on the subject.

Best, Peter
[log in to unmask]