Dear Everyone,

I apologise for the delay in answering Dean's questions about the role of
sucrose in the diet.  I managed to lose the first draft of the reply and
had to start again.

1. Yudkin omitted data to
>make a correlation between sucrose and CHD appear to be there.  Has this
>been documented anywhere?

Sucrose was first implicated as a risk factor for CHD by Yudkin and
although the hypothesis gained popular credibility, it was quickly refuted.
In univariate analysis, Liu et al (1) showed that both saturated fat and
refined sugars were highly correlated with CHD in 20 industrialised
nations.  However, in subsequent multivariate analysis sucrose was no
longer significantly related to CHD mortality.  Thus one of the causes of
the confusion is that saturated fat and sucrose correlate with each other
in population studies (but not in individual diets).

Truswell (2) reviewed 10 case-control studies of sucrose and CHD and found
that none supported the hypothesis. The international community thinks so
little of the Yudkin's hypothesis that 'no prevention trial of CHD with
sugar has been completed, started, planned or even contemplated.'
>

2. but would it also
>not be the case that sugars consumed in nature would also be eaten with
>massive amounts of fibre, thus slowing down the absorption of those sugars?

While it may be true that sugars in nature are often eaten with massive
amounts of fibre, this is not to say that they give metabolic responses any
lower that refined products.  We found that there was a wide range of
glycaemic and insulin responses (an indirect measure of rate of absorption)
among 40 naturally-occurring sugars and refined sugar-containing foods (3).
In Australia, commercial honey, bush honey and refined sucrose have similar
GIs.

There are plenty of sugary foods in western diets with a low GI.  This
includes most of the sweetened dairy products and chocolate confectionery.
We have just completed GI studies of 10 Mars confectionary/chocolate items.
Many have a GI in 30-50 range (the same as legumes). One reason for this
low GI is that the high fat content slows down gastric emptying.  The high
saturated fat content is a concern but not the sugar.  On the other hand
there are tropical fruits with GIs in the 70s. I'm using the GI scale where
glucose = 100.

In a study of the diets of 342 people with diabetes, Wolever et al (4)
found that the GI of the diet was *inversely* related to the proportion of
carbohydrate that was sugars (both natural and refined).  Thus, the higher
the sugars content, the lower the overall diet GI.  He also found that the
lower the GI, the lower their glycosylated hemoglobin (a measure of average
blood glucose levels).

Some studies also suggest that insulin sensitivity is better on high sugar
(vs high starch) diets (5).


3.  sucrose itself has been shown to have a
>glycemic index no worse than that of white bread, does this suggest that
>sucrose is harmless--or might it not just suggest that bread is not so
>benign as it is often thought to be?

Yes, I agree with you on this one.  I am always stunned by the very high
sustained glycaemic and insulin response we see to white or wholemeal
bread. But note that on average sugar-containing foods have a lower GI than
most western starchy staples (3).


>
4.  You also stated your that glucose is the obligatory fuel for the brain and
>fetus, and that we can't make enough glucose via gluconeogenesis alone.
>But Cahill (Cahill, G. and Aoki, T.T., Medical Times 98, 1970) found that
>the brain will  utilize ketones preferentially to glucose, once a brief
>period of adjustment of 1-3 weeks is made.

Yes, it's true that brain can utilise ketones when itmust but the state of
ketonemia is generally regarded as undesirable.  It has been shown to be
associated with impaired cognitive function and teratogenic effects on the
fetus (6, 7).  See below for further details.


5.  >It is also well known that the Inuit eat almost no carbohydrate, and must
>therefore use gluconeogenesis and ketones for most of their fuel needs.
>But certainly they manage to have children.

Yes, I am very much aware of this and it forms one of the links in our
'carnivore connection' hypothesis (8).  We hypothesised that a human
females exposed to diets in which there was little carbohydrate would be
advantaged by a state of genetically determined insulin resistance.  This
would spare glucose for the fetus and increase the capacity of the liver to
produce glucose from non-glucose precursors.

Female dogs fed a carbohydrate-free diet with 26% protein became
hypoglycaemic and ketotic towards the end of gestation and over a third of
the puppies were stillborn (9-11).  Adequate synthesis of glucose from
gluconeogenic amino acids may be accomplished if dietary protein is
sufficiently high.  For example, dogs are able to reproduce on a
carbohydrate-free diet when the protein intake is sufficiently high (10,
11).  This fits well with the evolutionary development of the dog as a
hunter, since the body of prey would have supplied only a little available
carbohydrate but large amounts of protein.  The dog therefore falls halfway
between a carnivore and an omnivore (and perhaps we humans do too).

In true carnivorous animals like the cat, gluconeogenesis is also more or
less permanently 'switched on' (40) with maximal gluconeogenesis occurring
in the absorptive phase immediately following a meal.  Carnivorous animals
like the cat who have evolved and reproduce well on a low carbohydrate
intake, appear to be genetically insulin resistant (12).  Moreover, they
appear to develop NIDDM when exposed to a high carbohydrate diet (13).


6.  ..recent research has also shown
>that athletes, particularly endurance athletes, tend to show performance
>increases on very low carbohydrate, high protein, high fat diets, after a
>1-3 week adaptation period.

My understanding of the literature is that high carbohydrate diets (and
carbohydrate loading) enhance prolonged strenuous exercise because the
level of glycogen stores is the limiting factor here (for review see 14).

High carbohydrate diets (vs high fat diets) have been shown to produce
higher glycogen stores.  When exercise is sufficiently strenuous (above 70%
VO2max), only glucose can be utilised as fuel by the muscle because free
fatty acids cannot enter the cell fast enough.  As exercise progresses (and
the same high VO2max is maintained), glycogen stores gradually fall.  The
sensation of 'hitting the wall' corresponds with glycogen stores becoming
fully depleted.  It is likely that other forms of exercise (sprint events)
are not affected by diet.

Perhaps Loren Cordain (if he's still reading!) might like to comment here.

Summing up
For a major recent review of all health aspects of sugars, I refer readers
to a supplement number 1 of the Am J Clin Nutr 1995 (Volume 62).  Apart
from lingering concerns about TG levels and very high carbohdyrate diets,
sucrose is given a clean bill of health.

I am in the process of writing a paper titled 'Sugar - not a villain after
all'.  In it I provide the accumulated scientific evidence (most of it from
the last 5-10 years) that restricting sugar may do more harm than good.
This is because low sugar diets in HUMANS (not rats!) are associated with:

1. an increase in fat, especially saturated fat, intake (the sugar-fat seesaw)
2. an increase in obesity and overweight
3. an increase in the glycaemic index of the diet
4. a decrease in insulin sensitivity
5. diversion of millions of research and consumer dollars to non-sucrose
sweeteners.

I would be happy to post the summary to the review when it's finished.

Best wishes  Jennie


References

1.  Liu et al. Dietary lipids, sugar, fiber and mortaolity from coronary
heart disease.  Arteriosclerosi 1982; 2: 221-7.

2.  Truswell AS. Sugar and health: a review. Food Technol Aust 1987; 39:
134-40.

3. Brand Miller J, Pang E, Broomhead L.  The glycemic index of foods
containing sugars: comparison of foods with naturally occurring versus
added sugars.  Brit J Nutr 1995; 73: 613-623.

4. Wolever TMS, Nguyen P, Chiasson J,  Hunt JA, Josse RG, Palmason C,
Rodger NW, Ross SA, Ryan EA, Tan MH. Determinants of diet glycemic index
calculated retrospectively from diet records of 342 individuals with
non-insulin-dependent diabetes mellitus.  Am J Clin Nutr 1994;59: 1265-9
214.  Wolever

5. Piatti PM et al.  Insulin sensitivity and lipid levels in obese subjects
after slimming diets with different complex and simple carbohdyrate
content.  In J Obesity 1993; 17:375-81.

6.  Veneman T et al.  Effect of hyperketonemia and hperlactatacidemia on
symptoms, cognitive function... during hypoglycaemica in normal humans.
Diabetes 1994; 43: 1311-7.

7.  Reece et al.  Multifactorial basis of the syndrome of diabetes
embyropathy.  Teratology 1996; 54: 171-82.

8.  Brand Miller J, Colagiuri S.  The carnivore connection: dietary
carbohydrate in the evolution of non-insulin dependent diabetes.
Diabetologia 1994; 37: 1280-86.

9.      Rosmos DR, Palmer HJ, Muiruri KL, Bensink MR (1981) Influence of a
low carbohydrate diet on performance of pregnant and lactating dogs. J Nutr
111: 678-689.

10.     Kienzle E, Meyer H, Lohrie H (1985) Influence of carbohydrate-free
rations with various protein/energy relationships on foetal develpoment,
viability of newborn puppies and milk composition. Advances in Animal
Physiology and Animal Nutrition 16: 78-99.

11.     Blaza SE, Booles D, Burger IH (1989) Is carbohydrate essential for
pregnancy and lactation in dogs?  In: Berger IH, Rivers JPW (eds) Nutrition
of the dog and cat. Cambridge University Press, Cambridge, pp 229-243.

12.     MacDonald ML, Rogers QR, Morris JG (1984) Nutrition of the domestic
cat, a mammalian carnivore. Ann Rev Nutr 4: 521-62.

13.     O'Brian TD, Hayden DW, Johnson KH, Stevens JB.  High dose glucose
tolerance test and serum insulin glucagon levels in diabetic and
non-diabetic cats: relationships to insular amyloidosis. Vet Path
22:255-261.

14.  Sherman WM.  Metabolism of sugars and physical performance.  Am J Clin
Nutr 1995; 62: 212S-227S.