Glycemic index and obesity1–4

Janette C Brand-Miller, Susanna HA Holt, Dorota B Pawlak, and Joanna McMillan

ABSTRACT Although weight loss can be achieved by any highly palatable foods; and social and economic influences are

Downloaded from https://academic.oup.com/ajcn/article-abstract/76/1/281S/4689499 by guest on 04 July 2020

means of energy restriction, current dietary guidelines have not all likely contributors to the rising prevalence of overweight
prevented weight regain or population-level increases in obesity and obesity. In this report, we further suggest that current dietary
and overweight. Many high-carbohydrate, low-fat diets may be recommendations to increase the consumption of carbohydrate-
counterproductive to weight control because they markedly dense foods are counterproductive to weight control. Our
increase postprandial hyperglycemia and hyperinsulinemia. hypothesis is that a high-carbohydrate diet based on foods that
Many high-carbohydrate foods common to Western diets pro- promote a high glycemic response [ie, high-glycemic-index (GI)
duce a high glycemic response [high-glycemic-index (GI) foods] alters appetite and energy partitioning in a way that is
foods], promoting postprandial carbohydrate oxidation at the conducive to body fat gain.
expense of fat oxidation, thus altering fuel partitioning in a way Reducing fat intake has been the primary focus of dietary pre-
that may be conducive to body fat gain. In contrast, diets based vention and treatment of overweight and obesity for > 20 y. The
on low-fat foods that produce a low glycemic response (low-GI most concentrated source of energy, fat, is efficiently stored as
foods) may enhance weight control because they promote sati- body fat. Furthermore, high-fat foods are relatively less satiating
ety, minimize postprandial insulin secretion, and maintain than are isoenergetic portions of high-carbohydrate or high-
insulin sensitivity. This hypothesis is supported by several inter- protein foods (4, 5). Intervention studies have shown that ad libi-
vention studies in humans in which energy-restricted diets based tum, high-carbohydrate diets facilitate greater weight loss than
on low-GI foods produced greater weight loss than did equiva- do high-fat diets (6). Many epidemiologic studies have shown
lent diets based on high-GI foods. Long-term studies in animal that relatively high dietary fat intakes correspond with increased
models have also shown that diets based on high-GI starches obesity rates (7).
promote weight gain, visceral adiposity, and higher concentra- Whereas clinical studies have shown that a reduction in fat
tions of lipogenic enzymes than do isoenergetic, macronutrient- intake can produce clinically significant weight loss in overweight
controlled, low-GI-starch diets. In a study of healthy pregnant persons, the results are typically modest (6). A reduction of
women, a high-GI diet was associated with greater weight at 10–20% of energy intake from fat produces, on average, weight
term than was a nutrient-balanced, low-GI diet. In a study of diet losses of 5–7 kg in obese persons, but weight regain often occurs.
and complications of type 1 diabetes, the GI of the overall diet
was an independent predictor of waist circumference in men.
These findings provide the scientific rationale to justify random- RATIONALE FOR HIGH-CARBOHYDRATE, LOW-GI
ized, controlled, multicenter intervention studies comparing the DIETS
effects of conventional and low-GI diets on weight control. Adherence to standard dietary advice to reduce fat intake
Am J Clin Nutr 2002;76(suppl):281S–5S. while increasing carbohydrate intake generally increases the
glycemic effect of the diet. Both the quantity and quality of a
KEY WORDS Glycemic index, obesity, overweight, insulin, carbohydrate influence postprandial glycemia, and the interac-
glucose responses, carbohydrate tion between the 2 may be synergistic. A typical Western, high-
carbohydrate diet based on high-GI foods such as potatoes,
breads, and low-fat cereal products is digested and absorbed
INTRODUCTION
Rates of obesity and overweight continue to increase in most
1
Western countries despite the efforts of governments and health From the Human Nutrition Unit, School of Molecular and Microbial Bio-
care providers to prevent and reduce this trend. In the United sciences, University of Sydney, Australia.
2
Presented at a symposium held at Experimental Biology 2001, Orlando,
States, Europe, and Australia, > 40% of adults are now over-
FL, 1 April 2001.
weight or obese, a rate almost double that of the early 1980s 3
Supported by grants from the National Health and Medical Research
(1–3). The underlying reasons for this global pandemic are com- Council (NHMRC) and the Sydney University Nutrition Research Founda-
plex. Whereas genetic susceptibility plays a part, changes in the tion (SUNRF).
genetic makeup of the population cannot explain the dramatic 4
Address reprint requests to JC Brand-Miller, Human Nutrition Unit,
rise in obesity rates over the past 10–15 y. Reduced physical School of Molecular and Microbial Biosciences, University of Sydney, NSW
activity; abundant, easily available, and affordable energy-dense, 2006, Australia. E-mail: j.brandmiller@biochem.usyd.edu.au.

Am J Clin Nutr 2002;76(suppl):281S–5S. Printed in USA. © 2002 American Society for Clinical Nutrition 281S
282S BRAND-MILLER ET AL

rapidly, resulting in a high glycemic load and increased demand expression of the rate-limiting enzymes and alters the potential
for insulin secretion (8, 9). In insulin-resistant persons who for fat oxidation. Reduced capacity to oxidize fatty acids is pres-
consume high-GI foods, postprandial hyperglycemia and insu- ent in some obese human subjects (18) and obesity-prone rats (19).
linemia are magnified, possibly contributing to  cell exhaus- Reduced rates of fat oxidation were linked with greater weight
tion and the development of type 2 diabetes (10, 11). On the gain in several prospective studies (20, 21). Whether high-GI
other hand, low-GI, high-carbohydrate foods may maintain diets, which induce chronic hyperglycemia and hyperinsuline-
insulin sensitivity and increase the weight-loss potential of ad mia, can reduce the body’s capacity to oxidize fat and signifi-
libitum, low-fat diets (12). cantly increase body fat storage is still questionable, although
Low-GI foods may benefit weight control in 2 ways: 1) by some evidence supports the hypothesis.
promoting satiety and 2) by promoting fat oxidation at the
expense of carbohydrate oxidation. These 2 qualities of low-GI
foods stem from the slower rates at which they are digested and DIETS BASED ON HIGH-GI CARBOHYDRATES
absorbed and the corresponding effects on postprandial glycemia ENHANCE FAT STORAGE AND WEIGHT GAIN
and hyperinsulinemia. Even when appearance and nutrient con- Acute short- and long-term studies in humans and animals

Downloaded from https://academic.oup.com/ajcn/article-abstract/76/1/281S/4689499 by guest on 04 July 2020

tent are matched, low-GI foods typically induce higher satiety provide evidence that a high-GI diet affects appetite and nutrient
than do their high-GI counterparts and are followed by less partitioning in a way that promotes body fat storage.
energy intake at subsequent meals (12). Corresponding with the
progressive refinement of carbohydrate-rich foods, such as apples Acute studies
or wheat grains, is a step-by-step increase in the food’s GI rating Obese teenage boys who consumed fixed high-GI breakfasts
and, with it, a reciprocal decrease in satiety (13). Similarly, and lunches scored higher on hunger tests and overate a standard
mixed meals with low GIs were found to induce greater chole- ad libitum diet during the remainder of the day compared with
cystokinin secretion and greater satiety over a 180-min period (14). those who consumed lower-GI meals containing similar amounts
These differences are likely to be clinically important. A 50% of energy and nutrients (22). Voluntary energy intake after the
increase in a meal’s GI (eg, from 50 to 75) resulted in a 50% high-GI meals was 53% greater than after the lower-GI meals
decrease in satiety. Overall, 16 of 17 studies confirmed that low- and was associated with reduced concentrations of fatty acids and
GI meals increase fullness to a greater extent than do compara- glucagon and higher concentrations of plasma epinephrine and
ble high-GI meals (12). growth hormone. Elevations in counterregulatory hormone con-
To many, it may appear paradoxical that a food producing a centrations suggest that the reduction in major metabolic fuels is
low blood glucose response should be more satiating than an biologically significant.
isoenergetic portion of food producing a high blood glucose High-GI meals produced lower fatty acid concentrations
response. However, because the former is characterized by throughout the day (23) and lower rates of fat oxidation than did
slower rates of digestion and absorption in the small intestine, low-GI meals of similar composition (24, 25). In young rats,
nutrient receptors in the gastrointestinal tract are stimulated for high-GI starches produced not only higher respiratory quotients
a longer period of time, resulting in prolonged feedback (through for several hours than did low-GI starches, but respiratory quo-
signals such as cholecystokinin and glucagon-like peptide-1) to tients reached > 1.0, indicating that maximal rates of carbohy-
the satiety center in the brain (15). Additional mechanisms may drate oxidation and glycogen synthesis were reached and de
also account for the differences in the satiating effects of high- novo lipogenesis was taking place (26). In healthy subjects,
and low-GI foods. After the consumption of a high-GI meal, when a high-GI meal was ingested 1–2 h before exercise, there
insulin concentrations rise dramatically, leading directly to a was a greater shift in substrate utilization from fat to carbohy-
rapid reduction in both glucose and fatty acid concentrations, drate compared with when a low-GI meal was ingested (25).
often below fasting concentrations. Hence, between 3 and 5 h Thus, at least for several hours postprandially, both at rest and
postconsumption, concentrations of 2 major metabolic fuels cir- during exercise, fat utilization is depressed after high-GI meals
culating in the blood are low simultaneously, a situation that compared with low-GI foods.
could be interpreted by the central nervous system as “low fuel
status” (12). Spontaneous requests for meals by time-blinded Short-term studies
human subjects were found to correspond with transient declines Differences in the glycemic load (GI  carbohydrate content)
in blood glucose concentrations during the postabsorptive state (16). of a diet were shown to affect energy expenditure when Agus et
Meal requests were also associated with so-called dynamic al (27) compared the effects of energy-restricted diets with high
declines in blood glucose concentrations immediately after the and low glycemic loads on hormonal and physiologic adapta-
peak blood glucose concentration induced by carbohydrate con- tions in a group of overweight young men for 9 d. They found
sumption. Thus, the marked hyperglycemic and hypoglycemic that during the high-glycemic-load diet, energy expenditure and
effects of high-GI foods could partly explain the lower satiety serum leptin concentrations declined more, nitrogen balance
observed in the postprandial period. tended to be more negative, and voluntary food intake was
Differences in GI also dictate differences in fuel partitioning greater. Limited availability of the major metabolic fuels after a
and oxidation. Postprandial rises in glucose and insulin con- high-GI meal elicits marked increases in the counterregulatory
centrations increase carbohydrate oxidation acutely through the hormones, some of which have proteolytic actions. This in turn
rapid activation of key rate-limiting enzymes. For example, could favor the catabolism of lean body tissue. Howe et al (28)
malonyl-CoA, an intermediate of glucose oxidation, strongly found that consumption of a low-GI diet based on high-amylose
inhibits fatty acid transport into mitochondria, resulting in starch for 14 wk appeared to increase protein retention to a
decreased fatty acid oxidation (17). Longer exposure to chronic greater degree than did a diet based on high-GI starch in both
hyperglycemia and hyperinsulinemia results in decreased normal and hyperinsulinemic men.
 GLYCEMIC INDEX AND OBESITY 283S

with 48% more muscle glycogen (P = 0.02) after 24 h of recov-

ery from the exercise (32). Finally, 12 healthy pregnant women
who were randomly assigned to consume ad libitum high- and
low-GI diets at 8 wk gestation gained significantly more weight
by full term with the high-GI diet than with the low-GI diet
(19.7 compared with 11.8 kg, respectively; P < 0.05; 33).

Long-term studies in animal models

Studies in animal models have the advantage of providing
information on both the mechanisms and effects of longer life-
time feeding of high- and low-GI diets. Feeding rats high-GI
starch over 5 wk resulted in higher epididymal fat weights (an
index of visceral adiposity in rats) and larger adipocyte volume
than did feeding low-GI starch ad libitum (34). Expression and

Downloaded from https://academic.oup.com/ajcn/article-abstract/76/1/281S/4689499 by guest on 04 July 2020

activity of fatty acid synthase complex in white adipose tissue,
FIGURE 1. Weight loss during a parallel and crossover study of over-
weight women randomly assigned to consume high-glycemic-index () and glucose uptake into adipocytes, were significantly higher in
or low-glycemic-index () diets for 12 wk each (29). rats fed the high-GI starch diet (P < 0.05; 35, 36). For 32 wk, we
fed 2 groups of young adult rats diets similar in energy distribu-
tion to modern Western diets, comprising 45% carbohydrate, 20%
Medium-term studies in humans protein, and 35% fat as a proportion of energy (37). One group
Several medium-term studies have compared the weight-loss was fed a high-GI-starch diet (100% amylopectin, the most com-
potential of high- and low-GI diets. Slabber et al (29) studied mon form of starch in Western diets); the other was fed low-GI
obese females who consumed 2 energy-restricted diets, one starch (60% amylose starch, the form of starch more common in
high-GI and one low-GI, in a 12-wk parallel arm study (n = 30) traditional diets in nonindustrialized countries). The rats were fed
that was followed by a 12-wk crossover study (n = 16). Both 2 large isoenergetic meals per day to simulate meal feeding in
diets produced weight loss during the first 12 wk (x–: 9.4 com- humans and to maximize differences in postprandial hypergly-
pared with 7.4 kg), but there was no significant difference in the cemia between the diets. A significantly different pattern of
amount lost. During the follow-up crossover study, the low-GI weight gain was observed during the 32 wk of dietary interven-
diet produced greater weight loss (7.4 compared with 4.5 kg; tion (Figure 2). Whereas the low-GI group remained weight-
P = 0.04) than did the high-GI diet (Figure 1). In both arms, the stable, the high-GI group gradually gained weight and were 16%
reduction in fasting insulin concentrations was greater with the heavier at the end of the study. Total fat mass, assessed by meas-
low-GI diet than with the isoenergetic, macronutrient-balanced, urements of total body water, was significantly (40%) higher in
high-GI diet. In slightly overweight men, Bouché et al (C Bouché, the high-GI group (P < 0.05). Each of the fat pads, but not the
SW Rizkalla, J Luo, A Veronese, and G Slama, unpublished heart or liver, was heavier in the high-GI group. More impor-
observations, 2000) showed that consumption of a low-GI diet tantly, the average weight of visceral fat in the high-GI group was
for 5 wk, compared with a high-GI diet of equal energy and twice that of the low-GI group and remained significantly higher
macronutrient content, decreased total fat mass by 500 g (P < 0.05, when expressed as a proportion of total body fat (P < 0.05). In
as measured by dual-energy X-ray absorptiometry), despite no contrast, the subcutaneous fat pads, adjusted for total adiposity,
difference in body weight. The decrease in fat mass was mostly did not differ between groups. We measured substrate utilization
abdominal and was associated with a decrease in ob gene expres- over 6 h postprandially and found different patterns both at the begin-
sion in subcutaneous fat tissue. In a retrospective, nonrandom- ning and at the end of the study, with a marked decrease in whole-
ized study comparing the effects of a low-GI diet (n = 64) with body fat oxidation in the high-GI group. Furthermore, the high-GI
those of a conventional reduced-fat diet (n = 43) in the manage- group had higher rates of hepatic lipogenesis and higher liver and
ment of pediatric obesity, body mass index (BMI; in kg/m2) and red oxidative muscle glycogen stores. The high-GI group showed
body weight decreased more in the low-GI group than in the con-
ventional diet group, even after adjustment for age, sex, ethnic-
ity, baseline BMI, and baseline weight (30). Significantly more
patients in the low-GI group achieved a decrease in BMI of at
least 3 units (17% compared with 2%; P < 0.001). In contrast
with the results of these studies, Wolever et al (31) reported sim-
ilar weight loss in obese patients with type 2 diabetes who were
randomly assigned to receive high- or low-GI hypoenergetic
diets for 6 wk (2.5 compared with 1.8 kg; NS).
Apart from differences in weight loss, other intervention stud-
ies suggest that a diet’s glycemic effect influences fuel storage
within the body. In healthy, lean males, chronic consumption of
high-GI diets, compared with nutrient-balanced low-GI diets,
was associated with higher muscle glycogen (14%) and muscle
triacylglycerol (22%) concentrations (both P < 0.05; 23). Com- FIGURE 2. Weight changes in adult rats fed isoenergetic, nutrient-
pared with an equivalent low-GI meal, a high-GI meal consumed balanced diets based on high-glycemic-index () or low-glycemic-index ()
by well-trained athletes after prolonged exercise was associated starch for 32 wk (37).
284S BRAND-MILLER ET AL

reductions in carnitine palmitoyltransferase 1 (CPT-1) messenger sibly at the expense of lean tissue) when the diet is based on
RNA (mRNA) in the liver, a key regulatory site of long-chain high-GI foods.
fatty acid flux through  oxidation. A concomitant increase in
hepatic acetyl-CoA carboxylase (ACC) mRNA (ACC catalyzes
the formation of malonyl-CoA) was observed. In the liver, mal- CONCLUSIONS
onyl-CoA is both an intermediate in de novo lipogenesis and a The postprandial effects of carbohydrate-dense, high-GI foods
potent inhibitor of CPT-1. Thus, in addition to causing an acute described above may help to explain why low-fat diets have not
effect on fuel oxidation, chronic high-GI feeding characteristi- lived up to their potential to inhibit weight gain when consumed
cally leads to changes in enzyme expression, thereby decreasing ad libitum. Postprandial hyperglycemia and hyperinsulinemia
the potential for hepatic fat oxidation. are consequences of typical Western diets that are not seen in
These findings in animal models have important implications rural societies where high-carbohydrate, low-GI diets are typi-
because they challenge the assumption that “a calorie is a calo- cally consumed. This is especially true of persons who are
rie.” High-GI starch caused increased fat accumulation in vis- sedentary, overweight, and genetically prone, in whom insulin
ceral stores and reduced lipolytic capacity despite the fact that resistance is common. Faster digestion and absorption and

Downloaded from https://academic.oup.com/ajcn/article-abstract/76/1/281S/4689499 by guest on 04 July 2020

macronutrient and energy intake were precisely controlled. higher insulin responses after high-GI meals dictate differences
in satiety and energy partitioning that, over the long term, favor
Epidemiologic evidence expansion of the fat stores. Carefully designed multicenter stud-
Cohort studies provide further evidence that the glycemic ies to assess the efficacy of high-carbohydrate, low-GI diets in
effect of the diet might influence weight control. In the EURO- the treatment and prevention of obesity are clearly justifiable on
DIAB Complications Study of nearly 3000 adults with type 1 dia- scientific grounds.
betes, consumption of a lower-GI diet was found to predict lower
waist-to-hip ratio and waist circumference independently of car-
bohydrate, fat, and fiber intakes (38). In the CARDIA study of REFERENCES
young adults, low fiber consumption (GI was not assessed) pre- 1. Australian Bureau of Statistics. National Nutrition Survey: selected
dicted higher 10-y weight gain, waist-to-hip ratio, and 2-h post- highlights, Australia, 1995. Canberra: Australian Bureau of Statistics,
glucose insulin concentrations (a measure of insulin resistance) to 1997. (Catalog no. 4802.0.)
a greater extent than did total or saturated fat consumption (39). 2. Flegal KM. The obesity epidemic in children and adults: current evi-
dence and research issues. Med Sci Sports Exerc 1999;31(suppl):
Although fiber and GI are not precisely related, viscous dietary
S509–14.
fibers and foods in which the natural cell wall architecture 3. Seidell JC. Obesity, insulin resistance and diabetes—a worldwide epi-
remains intact (eg, legumes) generally have lower GIs (40). demic. Br J Nutr 2000;83(suppl):S5–8.
4. Blundell JE, Cotton JR, Delargy H, et al. The fat paradox: fat-induced
satiety signals versus high fat overconsumption. Int J Obes 1995;19:
THE HYPOTHETICAL BIOCHEMICAL SCENARIO 832–5.
The following scenario for high-GI compared with low-GI 5. Holt SH, Brand-Miller JC, Petocz P, Farmakalidis E. A satiety index
diets in weight regulation is hypothesized on the basis of factual of common foods. Eur J Clin Nutr 1995;49:675–90.
but incomplete evidence from studies in animals and humans. 6. Astrup A, Grunwald GK, Melanson EL, Saris WH, Hill JO. The role
of low-fat diets in body weight control: a meta-analysis of ad libitum
Consumption of a high-carbohydrate, high-GI diet results in
dietary intervention studies. Int J Obes Relat Metab Disord 2000;24:
recurrent postprandial hyperglycemia and hyperinsulinemia that
1545–52.
is accentuated in sedentary persons who are overweight, insulin- 7. Seidell JC. Dietary fat and obesity: an epidemiologic perspective.
resistant, or both. As a consequence, carbohydrate oxidation is Am J Clin Nutr 1998;67(suppl):546S–50S.
higher and fat oxidation lower throughout the postprandial 8. Foster-Powell K, Brand-Miller J. International tables of glycemic
period, whether the person is at rest or exercising. The expres- index. Am J Clin Nutr 1995;62(suppl):871S–93S.
sion of enzymes involved in lipid synthesis, such as ACC mRNA, 9. Holt SH, Brand-Miller JC, Petocz P. An insulin index of common
is up-regulated, whereas the expression of those involved in lipid foods: the insulin demand generated by 1000-kJ portions of common
oxidation, such as CPT1 mRNA, are down-regulated. Glycogen foods. Am J Clin Nutr 1997;66:1264–76.
stores in liver and muscle are maintained at higher concentra- 10. Salmeron J, Ascherio A, Rimm EB, et al. Dietary fiber, glycemic load,
and risk of NIDDM in men. Diabetes Care 1997;20:545–50.
tions with high-GI diets, but glycogen use and gluconeogenesis
11. Salmeron J, Manson JE, Stampfer MJ, et al. Dietary fiber, glycemic
may also be higher. Counterregulatory hormonal responses (eg, load, and risk of non-insulin-dependent diabetes mellitus in women.
of cortisol and noradrenaline) are also higher because of the JAMA 1997;277:472–7.
hyperglycemic-hypoglycemic rebound after consumption. This 12. Ludwig DS. Dietary glycemic index and obesity. J Nutr 2000;
stimulates gluconeogenesis from gluconeogenic amino acids as 130(suppl):280S–3S.
well as meal initiation in free-feeding individuals. The 0–6-h 13. Holt SHA, Brand-Miller J. Particle size, satiety and the glycemic
period following consumption of a high-GI diet is therefore response. Eur J Clin Nutr 1994;48:496–502.
characterized by a greater dependence on carbohydrate and pro- 14. Holt S, Brand J, Soveny C, Hansky J. Relationship of satiety to post-
tein as sources of fuel and less dependence on fat. Because car- prandial glycemic, insulin and cholecystokinin responses. Appetite
1992;18:129–41.
bohydrate and protein stores are limited, their higher rate of
15. Lavin JH, Wittert GA, Andrews J, et al. Interaction of insulin, glucagon-
usage may stimulate appetite and encourage overconsumption. If like peptide 1, gastric inhibitory polypeptide, and appetite in response to
energy intake and energy expenditure are matched over the long intraduodenal carbohydrate. Am J Clin Nutr 1998;68:591–8.
term, then body weight remains stable. However, even the small 16. Melanson KJ, Westerterp-Plantenga MS, Saris WH, Smith FJ,
energy imbalances that are characteristic of modern lifestyles are Campfield LA. Blood glucose patterns and appetite in time-blinded
more likely to promote gradual expansion of the fat stores (pos- humans: carbohydrate versus fat. Am J Physiol 1999;277:R337–45.
 GLYCEMIC INDEX AND OBESITY 285S

17. Wolfe RR. Metabolic interactions between glucose and fatty acids in 29. Slabber M, Barnard HC, Kuyl JM, Dannhauser A, Schall R. Effects of
humans. Am J Clin Nutr 1998;67(suppl):519S–26S. a low-insulin-response, energy-restricted diet on weight loss and
18. Simoneau JA, Veerkamp JH, Turcotte LP, Kelley DE. Markers of capac- plasma insulin concentrations in hyperinsulinemic obese females.
ity to utilize fatty acids in human skeletal muscle: relation to insulin Am J Clin Nutr 1994;60:48–53.
resistance and effects of weight loss. FASEB J 1999;13:2051–60. 30. Spieth LE, Harnish JD, Lenders CM, et al. A low-glycemic index diet
19. Commerford SR, Pagliassotti MJ, Melby CL, Wei Y, Gayles EC, in the treatment of pediatric obesity. Arch Pediatr Adolesc Med 2000;
Hill JO. Fat oxidation, lipolysis, and free fatty acid cycling in obesity- 154:947–51.
prone and obesity-resistant rats. Am J Physiol Endocrinol Metab 2000; 31. Wolever TMS, Jenkins DJA, Vuksan V, Jenkins AL, Wong GS, Joss RG.
279:E875–85. Beneficial effect of low-glycemic index diet in overweight NIDDM
20. Zurlo F, Lillioja S, Esposito-Del PA, et al. Low ratio of fat to carbo- subjects. Diabetes Care 1992;15:562–4.
hydrate oxidation as predictor of weight gain: study of 24-h RQ. Am J 32. Burke LM, Collier GR, Hargreaves M. Muscle glycogen storage after
Physiol 1990;259:E650–7. prolonged exercise: effect of the glycemic index of carbohydrate feed-
21. Weyer C, Pratley R, Salbe A, Bogardus C, Ravussin E, Tataranni P. ings. J Appl Physiol 1993;75:1019–23.
Energy expenditure, fat oxidation, and body weight regulation: a study 33. Clapp JR. Diet, exercise and feto-placental growth. Arch Gynecol
of metabolic adaptation to long-term weight change. J Clin Endocrinol Obstet 1997;261:101–7.

Downloaded from https://academic.oup.com/ajcn/article-abstract/76/1/281S/4689499 by guest on 04 July 2020

Metab 2000;85:1087–94. 34. Lerer-Metzger M, Rizkalla SW, Luo J, et al. Effects of long-term low-
22. Ludwig DS, Majzoub JA, Al-Zahrani A, Dallal GE, Blanco I, glycemic index starchy food on plasma glucose and lipid concentra-
Roberts SB. High glycemic index foods, overeating, and obesity. Pedi- tions and adipose tissue cellularity in normal and diabetic rats. Br J
atrics 1999;103:E261–6. Nutr 1996;75:723–32.
23. Kiens B, Richter EA. Types of carbohydrate in an ordinary diet affect 35. Kabir M, Rizkalla SW, Champ M, et al. Dietary amylose-amylopectin
insulin action and muscle substrates in humans. Am J Clin Nutr starch content affects glucose and lipid metabolism in adipocytes of
1996;63:47–53. normal and diabetic rats. J Nutr 1998;128:35–43.
24. Thomas DE, Brotherhood JR, Brand-Miller JC. Carbohydrate feeding 36. Kabir M, Rizkalla SW, Quignard-Boulange A, et al. A high glycemic
before exercise: effect of glycemic index. Int J Sports Med 1991;12: index starch diet affects lipid storage-related enzymes in normal and to
180–6. a lesser extent in diabetic rats. J Nutr 1998;128:1878–83.
25. Febbraio MA, Keenan J, Angus DJ, Campbell SE, Garnham AP. Pre- 37. Pawlak D, Denyer GS, Brand-Miller JC. Long term feeding with high
exercise carbohydrate ingestion, glucose kinetics, and muscle glycogen glycemic index starch leads to obesity in mature rats. Proc Nutr Soc
use: effect of the glycemic index. J Appl Physiol 2000;89:1845–51. Aust 2000;24:215 (abstr).
26. Denyer GS, Pawlak D, Higgins J, et al. Dietary carbohydrate and 38. Toeller M, Buyken AE, Heitkamp G, et al. Nutrient intakes as predic-
insulin resistance: lessons from humans and animals. Proc Nutr Soc tors of body weight in European people with type 1 diabetes. Int J
Aust 1998;22:158–67. Obes Relat Metab Disord 2001;25:1815–22.
27. Agus MSD, Swain JF, Larson CL, Eckert EA, Ludwig DS. Dietary 39. Ludwig DS, Pereira MA, Kroenke CH, et al. Dietary fiber, weight
composition and physiologic adaptations to energy restriction. Am J gain, and cardiovascular disease risk factors in young adults. JAMA
Clin Nutr 2000;71:901–7. 1999;282:1539–46.
28. Howe JC, Rumpler WV, Behall KM. Dietary starch composition 40. Jenkins DJA, Axelsen M, Kendall CWC, Augustin LSA, Vuksan V,
and level of energy intake alter nutrient oxidation in ‘carbohydrate- Smith U. Dietary fiber, lente carbohydrates and the insulin-resistant
sensitive’ men. J Nutr 1996;126:2120–9. diseases. Br J Nutr 2000;83(suppl):S157–63.