On Fri, 15 Sep 2000, Denise LePage wrote:

> One theory is that high levels of ketosis trigger an exaggerated insulin
> response.  This is the feedback mechanism that prevents ketoacidosis.  When
> ketone levels are high in the blood stream, insulin is the mechanism to
> remove it.  Therefore, someone on a highly ketotic diet will actually be less
> able to lose fat than someone in fairly low ketosis.  This is all theory from
> Lyle McDonald, from anectodal evidence, but seems to ring true.

Interestingly, this has been my experience as well.  On the
Anchell diet I seem to be in trace-mild ketosis most of the time.
My previous experiments with ketogenic diets tended to be of the
deep ketosis sort, and I didn't have great success with weight
loss.

I think *part* of the deal is that, for deep ketosis to be
measured with ketostix, the diet must already be pretty
hypercaloric.  If the ketones in the blood are being burned, not
many will actually be dumped into the urine (although some will,
since ketones are so volatile).  If the ketostix are turning deep
purple it means that *lots* of ketones are entering the urine.
But why should this happen?  It means that many more ketones are
being made than can be used.  If calories are high enough, then
they won't even get dumped out into the urine fast enough, and
that's when the serum level will rise and insulin will be
stimulated to start storing fats in adipose tissue.

So you can indeed gain fat while in ketosis, and the "deeper"
your ketosis, in terms of ketostix readings, the more likely you
are to do so.

> To clarify, as I understand it, the body can convert approx. 50% of protein
> to glucose if there is an excess.  That figure may be wrong, as I don't have
> my books with me.  It does not convert it directly to fat.  If the body lacks
> glucose, it will convert protein.

>From what I have read, it converts 58% of protein to glucose no
matter what.  This is from a physio textbook.  I've always found
this difficult to believe, however.

> The brain and some other tissues need
> glucose.

Parts of the brain need glucose; other parts can adapt to
ketones.  The cells of the retina, red blood cells, and
fast-twitch muscle cells need glucose, and accept no substitutes.

> If adequate glucose is provided, it will not usually be necessary
> for the body to convert it via neoglucogenesis.

Yes, this is what one would expect.  It seems odd that 58% of
protein would be converted even in the presence of adequate
glucose.  If the 58% conversion rate is accurate, regardless of
glucose intake, then this would perhaps be evidence that for a
very long time glucose intake was seldom adequate, so that it
made sense to convert protein at this constant rate.

Todd Moody
[log in to unmask]