Low carbohydrate does not mean low glycaemic index

[Guest guest]

Hi All,

The below letter is on a paper that compared diets and is probably, I

believe, relevant to other such papers on low carbohydrate diets, and diets

generally. The diets they chose can have a large effect on outcomes.

Cheers, Al.

Br J Nutr. 2002 Aug;88(2):211-2; author reply 213-4.

Low carbohydrate does not mean low glycaemic index!

Wolever TM.

PMID: 12144725 [PubMed - indexed for MEDLINE]

Author Keywords: Subject preparation; Fasting length; Exercise; Pre-meal

effect; Glycemic responses; Glycemic index; Healthy humans

Dumesnil et al. (2001) reported the results of a study from Laval

University, Québec, in which overweight men were fed under metabolic ward

conditions for 6 d on each of three different test diets. They concluded

that: ‘a low-glycaemic index–low-fat–high-protein diet can produce a marked

decrease in ad libitum energy intake without increasing hunger or decreasing

satiety while having rapid and marked effects on metabolic risk variables’.

However, the authors have provided no evidence that the effects they

observed were in any way due to the use of low-glycaemic-index (GI) foods.

The conclusions related to GI are unjustified because the paper fails to

demonstrate that the GI of their low-GI–low-fat–high-protein diet is less

than that of the control (American Heart Association; Krauss et al. 1996)

diet. The GI of neither experimental diet is given. An appendix is provided

giving sample menus from each dietary regimen and showing the GI values of

the foods in each diet. However, the information is imprecise: there are no

weights of foods given, in some cases the specific food is not indicated.

Three foods, wholewheat bread, unsweetened jam and fresh fruits, provide

most of the carbohydrate in the low-GI diet. The GI values of bread and jam

are indicated as being: ‘Undetermined, but probably <51’. The GI of ‘fresh

fruits’ is listed as being 36–53.

We have tested commercial varieties of wholewheat bread available in Canada,

and wholewheat bread baked in our laboratory, on several occasions. The

lowest value ever obtained was 65 (Wolever et al. 1994a), although the value

obtained was usually closer to 71 ( et al. 1983a,b, 1986, 1988;

Wolever et al. 1986, 1994a). Thus wholewheat bread is not a low-GI food. The

published GI values of certain popular tropical fruits, readily available in

Canada, are higher than the maximum value of 53 given by the authors; e.g.

ripe banana (all yellow skin, GI 54), over-ripe banana (few brown spots on

skin, GI 64) (Wolever et al. 1988), pineapple (GI 66) and watermelon (GI 72)

(- & Brand , 1995). Since the fruits used are not

indicated, it is not possible to verify that the GI values cited are

correct.

To the best of my knowledge, there is no published GI value for jam,

although (DJA and C Kendal, personal communication) found

the GI of strawberry jam to be 84. The carbohydrates in regular jam consist

of the sugars in the fruit, plus the sucrose or other carbohydrate added to

sweeten and thicken the jam. Fruits contain varying proportions of sucrose,

fructose and glucose with some, such as pear, peach and apricot, containing

100–160 g sorbitol/kg (Wolever et al. 1993). Dumesnil et al. (2001) do not

indicate what kind of fruit was in ‘Unsweetened jam’. Presumably the jam

contained no added sucrose, but could have contained extra fruit or fruit

juice. It may or may not be correct to assume that it contained no other

sweetener such as sorbitol or aspartame. These different kinds of jam would

contain different amounts of carbohydrate, but the GI may not change.

Removing sugar from jam, or replacing it with extra fruit or fruit juice

would not be expected to change the GI appreciably, if the sugars in the

fruit approximate the composition of sucrose. Sweetening jam with aspartame

would not alter its GI since aspartame is not a carbohydrate. The effect on

the GI of sweetening the jam with sorbitol is a contentious issue depending

on whether sorbitol is included as a ‘glycaemic carbohydrate’.

Dumesnil et al. (2001) indicate that cheese, salad and yellow waxed beans

have a ‘Low GI’ and, in a footnote, further add: ‘ & hellip;no data are

available but GI probably <30’. This represents a fundamental error about

the definition of the GI. Foods which do not raise blood glucose do not

necessarily have a low GI. The GI is an index of the blood glucose raising

potential of the glycaemic carbohydrate in foods (Food and Agriculture

Organization/United Nations University/World Health Organization, 1998).

Thus, the term ‘low-GI food’ is normally used to describe high-carbohydrate

foods. Foods containing no carbohydrate do not have a GI. Foods containing

only a small amount of carbohydrate theoretically could have a GI, but

because they do not contain enough carbohydrate to raise blood glucose, it

is difficult in practice to determine the GI. Since cheese, salad and yellow

waxed beans contain very little carbohydrate, it is not meaningful to

describe them as low GI. The available carbohydrate in cheese, lactose, has

a low GI (Wolever et al. 1985), and so there is at least some justification

in calling it a low-GI food. However, I do not know what glycaemic

carbohydrates are contained in salad and yellow beans, though I would guess

that it was glucose (high GI). Nevertheless, diet GI is not affected

appreciably whether these foods are considered to have high or low GI,

because they contain little glycaemic carbohydrate.

I estimated the GI of the sample diets using previously described methods

(Wolever et al. 1994b). For fresh fruits, I used apple. I took the GI value

of unsweetened jam to be 61. The weights of foods was based on common

portion sizes, adjusted so as to result in total food weight and intakes of

energy, fat, protein, carbohydrate and fibre which were the same as the

average values for the diets given in Table 1 of Dumesnil et al. (2001). The

estimated GI values of the two diets are very similar (Table 1). The

resulting values may be imprecise because of missing information about the

type and amounts of foods in the sample menus, and the need to estimate GI

values for several foods. Nevertheless, there is no evidence from this

calculation that the low-GI diet actually had a low GI.

Thus, it is not possible to ascribe any of the effects Dumesnil et al.

(2001) observed to differences in GI. I would not be so concerned if the

authors had referred to their experimental diet as a Montignac diet in the

title and conclusions of their paper. However, for such a reputable research

group to attribute their results, even in part, to the GI does damage by

contributing to the wealth of erroneous information and unsubstantiated

claims about the GI which already exist, promulgated by those who are not

experts in the field.

*All glycaemic index values quoted in this article are given on the glucose

standard, i.e. the glycaemic index of glucose=100.

References

Dumesnil JG, Turgeon J, Tremblay A, Poirier P, Gilbert M, Gagnon L,

St-Pierre S, Garneau C, Lemieux I, Pascot A, Bergeron J & Després J-P (2001)

Effect of a low-glycaemic index–low-fat–high protein diet on the atherogenic

metabolic risk profile of abdominally obese men. British Journal of

Nutrition 86, 557–568

Food and Agriculture Organization/United Nations University World Health

Organization (1998) Carbohydrates in Human Nutrition. Rome: FRAO

- K & Brand J (1995) International tables of glycemic

index. American Journal of Clinical Nutrition 62,

871S–893S

DJA, Wesson V, Wolever TMS, Kalmusky J, Giudici S, Csima A, Josse RG

& Wong GS (1988) Wholemeal versus wholegrain breads: proportion of whole or

cracked grain and the glycaemic response. British Medical Journal 297,

958–960

DJA, Wolever TMS, AL, Giordano C, Giudici S, LU,

Kalmusky J, Josse RG & Wong GS (1986)

Low glycemic response to traditionally processed wheat and rye products:

bulgur and pumpernickel bread. American Journal f Clinical Nutrition 43,

516–520

DJA, Wolever TMS, AL, Thorne MJ, Lee R, Kalmusky J, Reichert

R & Wong GS (1983a) The glycaemic index of foods tested in diabetic

patients: a new basis for carbohydrate exchange favouring the use of

legumes. Diabetologia 24, 257–264

DJA, Wolever TMS, AL, Lee R, Wong GS & Josse RG (1983b)

Glycemic response to wheat products: reduced response to pasta but no effect

of fiber. Diabetes Care 6, 155–159

Krauss RM, Deckelbaum RJ, Ernst N, Fisher E, BV, Knopp RH, Kotchen T,

Lichtenstein AH, McGill HC, Pearson TA, Prewitt TE, Stone NJ, Horn LU &

Weinberg R (1996) Dietary guidelines for healthy American adults. A statem

for health professionals from the Nutrition Committee. American Heart

Association Circulation 94, 1795–1800

Wolever TMS, Cohen Z, LU, Thorne MJ, MJA, Prokipchuk EJ &

DJA (1986) Ileal loss of available carbohydrate in man: comparison

of a breath hydrogen method with direct measurement using a human ileostomy

model. American Journal of Gastroenterology 81, 115–122

Wolever TMS, Wong GS, Kenshole A, Josse RG, LU, Lam KY &

DJA (1985) Lactose in the diabetic diet: a comparison with other

carbohydrates. Nutrition Research 5, 1335–1345

Wolever TMS, DJA, AL, Vuksan V, Wong GS & Josse RG (1988)

Effect of ripeness on the glycaemic response to banana. Journal of Clinical

Nutrition and Gastroenterology 3, 85–88

Wolever TMS, Katzman-Relle L, AL, Vuksan V, Josse RG & DJA

(1994a) Glycaemic index of 102 complex carbohydrate foods in patients with

diabetes. Nutrition Research 14, 651–669

Wolever TMS, Nguyen PM, Chiasson J-L, Hunt JA, Josse RG, Palmason C, Rodger

NW, Ross SA, EA & Tan MH (1994b) Determinants of diet glycemic index

calculated retrospectively from diet records of 342 individuals with

non-insulin-dependent diabetes mellitus. American Journal of Clinical

Nutrition 59, 1265–1269

Wolever TMS, Vuksan V, Katzman-Relle L, AL, Josse RG, Wong GS &

DJA (1993) Glycaemic index of some fruits and fruit products in

patients with diabetes. International Journal of Food Sciences and Nutrition

43, 205–212

Nutrition Research Volume 23, Issue 5 , May 2003, Pages 621-629

Controlling subjects' prior diet and activities does not reduce

within-subject variation of postprandial glycemic responses to foods

Janice E. , Glowczewski and M. S. Wolever

Abstract

Glycemic responses to the same food vary from day-to-day within subjects.

To determine whether controlling the subjects' activities and dinner the

night before a standard breakfast reduces the within-subject variability of

glycemic responses.

Thirteen healthy subjects performed 4 controlled and 4 uncontrolled trials

in a randomized block design. The controlled trials entailed no vigorous

exercise for 24 hrs, consumption of a standard dinner, and set fast length

(± 15 minutes). The uncontrolled trials entailed usual activity, no standard

dinner, and a fast between 10–14 hours), after which a two-hour blood

glucose response to a standard breakfast meal was quantified for both trial

types. The within-subject coefficient of variation (CV) of the area under

the glycemic response curves (AUC) were calculated.

The mean CV of controlled trials, 24.3%, was not significantly different

from that of the uncontrolled trials, 20.4%. However, the controlled CV

values were higher in 10 of 13 subjects (p < 0.05; one-tailed).

The method used here of controlling subjects' activities and dinner the

night before a glycemic response test did not reduce within-subject blood

glucose variability.

Alan Pater, Ph.D.; Faculty of Medicine; Memorial University; St. 's, NL

A1B 3V6 Canada; Tel. No.: (709) 777-6488; Fax No.: (709) 777-7010; email:

apater@...