Asia Pacific J Clin Nutr (1993) 2, 107-110

Invited Editorial

The glycaemic index of foods

Janette Brand Miller PhD

Human Nutrition Unit, Department of Biochemistry, University of Sydney, NSW, Australia.

The glycaemic index of foods

The glycaemic index (GI) is a ranking of foods based on their glycaemic impact¹. It has proven to be a logical and useful tool for comparing foods in situations such as diabetes sport and appetite where fluctuations in blood sugar levels are considered important. In Australia the GI concept is already being utilized in some diabetes education centres. The International Diabetes Institute in Melbourne has produced educational material describing the differences between foods in terms of their GI. In fact Australia may be ahead of the rest of the world in its acceptance of the GI approach. New editions of most textbooks of nutrition and dietetics now devote a section to the subject² but most do not give their blessing citing conflicting early studies. However over the last few years many studies have proven that the GI concept is not only reproducible and predictable but clinically useful in the dietary management of insulin-dependent diabetes mellitus (IDDM) non-insulin-dependent diabetes mellitus (NIDDM) and hyperlipidaemia^3,4.

When Jenkins and co-workers introduced the GI concept in 1981¹ it confirmed what others had also found but had not tabulated in the same way ie that equal carbohydrate portions of starchy foods varied widely in their glycaemic properties from the very high responses produced by potatoes to very low responses produced by legumes. Similarly simple sugars produced a range of responses with sucrose being intermediate. GI seemed a more logical approach to the dietary management of diabetes but it went against the prevailing dietary dogma that starches were all slowly digested and absorbed while the opposite was true of sugars. Indeed the publication of the first GI of foods suggested that the system of carbohydrate exchanges for diabetic diets had little scientific validity.

In the last 12 years nearly 300 separate foods have been subjected to GI testing representing about 200 different kinds of foods from all around the world. The team here at the University of Sydney s Human Nutrition Unit has been responsible for about half of the data. The article by Mani et al. in this issue⁵ of the journal is a further addition to the body of knowledge of GI of foods. In particular their results provide more evidence that 'traditional', 'unprocessed' foods have a low GI. Legumes whole cereal grains and millets produce exceptionally low GI values contrasting with the effects of modern foods such as bread potatoes and many breakfast cereals which elicit high plasma glucose and insulin response¹. Factory processing and milling markedly increase glycaemic responses to foods^6,7. Interestingly the staple carbohydrate foods of Nauruans Australian Aborigines and Pima Indians were primarily low GI foods^8,9. In the past century these population groups have adopted a Western lifestyle and modern foods and now develop non-insulin-dependent diabetes mellitus (NIDDM) in alarming numbers. It is not unreasonable to suggest that the high GI of the diet may have some role to play. A preliminary study in rats supports the hypothesis that high GI foods worsen insulin resistance and therefore the risk of NIDDM¹⁰.

In early 1980s the debate on the GI approach to diabetes management became increasingly polarized with the open expression and publication of directly opposing views on its usefulness¹¹. Its reproducibility application to mixed meals and long-term effects were all open to question. There was a widespread belief that GI was useful only in the comparison of single foods but not where mixed meals were concerned^12,13. In 1986 the NIH consensus conference on diet and exercise in NIDDM recommended against the use of GI in the dietary management of diabetes¹⁴. The main criticisms were: no differences were apparent when individual carbohydrate foods were taken as part of a mixed meal and secondly there were no studies showing long-term benefits.

In the intervening six years since the NIH statement these criticisms have been shown to be without foundation. There are now at least 15 studies on mixed meals and 11 medium-to-long term studies using the GI approach in the dietary management of diabetes. Although several early studies failed to show any differences in glycaemic response when foods of different GI were incorporated into mixed meals^12,13,15, there are now three times as many studies which show that GI is very predictable in mixed meals^16-25. Methodological differences can explain some of the conflict but studies addressing the long-term effects help to answer both criticisms.

Of the 11 medium-to-long-term studies (2-12 weeks) which have specifically used the glycaemic index (GI) approach to determine the clinical gains in diabetes or lipid management^26-36 all but one produced positive findings. Altogether 156 subjects (63 NIDDM 45 IDDM 42 hyperlipidaemic 6 normal) have been studied all on an outpatient basis in a cross-over design in which patients consumed both types of diet for an equal period of time in random order.

An overview analysis of these 11 studies⁴ showed that on average low GI diets reduced glycosylated haemoglobin by 9% fructosamine by 8%, urinary C-peptide by 20% and day-long blood glucose by 16%. Cholesterol was reduced by an average of 6% and triglycerides by 9%. Improvements were found in well-controlled poorly controlled and overweight NIDDM subjects and applied to both adults and children with IDDM. One can criticize these results as modest improvements but so too were the changes to the diet. They were not exceptionally high in fibre or low in fat and the subjects did not have to lose weight. In most studies only half the carbohydrate was exchanged from high to low GI which meant that foods such as bread and potatoes could still be eaten on the low GI diet. Furthermore the findings applied to free-living subjects not to institutionalized or metabolic ward patients whose food intake can be strictly controlled but is not necessarily realistic. In our study²⁸ compliance was high on the low GI diets and patients remarked that they felt better on them.

A recent study from the Hammersmith Hospital in London UK is also cause for confidence because it was a large study in a typical clinical setting. Sixty newly diagnosed NIDDM subjects were randomly assigned to either standard dietary advice or standard plus low GI advice for 12 weeks³⁷. The low GI group not only had a significantly lower GI but also achieved a lower fat intake and high carbohydrate and fibre intake. There was a significantly greater fall in fructosamine and cholesterol in the low GI group.

GI and sports performance

Diabetes is not the only area where GI of foods may be important. The sport physiologists have suddenly discovered the GI concept and have recommended high and low GI foods for different situations³⁸. Thomas et al.⁴⁰ found that consumption of low GI foods before prolonged strenuous exercise may increase endurance time by as much as 20 minutes. The findings are relevant to all forms of prolonged strenuous activity in both humans and animals including sport and even national defence.

GI and satiety

There is another area in which GI is relevant that of satiety and weight maintenance. Flatt³⁹ has hypothesized that meals producing a high respiratory quotient (RQ) result in carbohydrate being burned at the expense of fat and less deposition of glucose as glycogen. The lower glycogen stores are thought to result in hunger developing sooner rather than later and therefore greater food intake. Both situations predispose to weight gain. Our research has shown that high GI foods produce higher RQs than low GI foods both before and during exercise and therefore have the potential to promote greater weight gain⁴⁰. We and others have also found that higher glucose and insulin responses are consistently associated with lower satiety and vice versa^41-43. These findings provide support for Flatt's hypothesis and may help to explain why modern diets (with their high GI foods) are particularly associated with overweight and obesity.

What criticisms of GI still remain?

The GI approach has been criticized because some foods have been rated as good or bad simply on the basis of their GI. It was never intended that the GI be used in isolation. The fat fibre and salt content of a food are particularly relevant to diabetes. Some people have argued that GI makes high-fat foods appear in a falsely favourable light because fat slows gastric emptying and blunts the glycaemic response to the carbohydrate. High-fat foods do tend to have a low GI including ice cream and Mars Bars^TM but fat may not be the main reason. Carbohydrates such as lactose and sucrose produce moderate glycaemic responses by themselves. Jenkins and colleagues have argued that the GI approach should be applied only to low-fat starchy foods⁴⁴. There is some dissention about recommending high-carbohydrate diets for all individuals with diabetes because of their tendency to lower HDL cholesterol^45-47. However this criticism of high carbohydrate diets is overcome by use of low-GI high-carbohydrate diets.

The insulin response to a food is also important and does not always correlate with the glycaemic response⁴⁸. Some rice varieties for example may produce a high GI but a substantially lower insulin index compared to white bread⁴⁹. The clinical significance of this is not clear but it may mean that we should be producing an insulin index of foods as well as GI. The insulin index may be more relevant to individuals with impaired glucose tolerance hyperinsulinaemia and high blood pressure ( Syndrome X ). Another criticism of the GI is that the usual serving size is often not a 50g carbohydrate portion. Some foods may contain so little carbohydrate that even a high GI is of little practical significance such as the case with carrots (GI=92) but there are few other foods in this category. Theoretically it is possible to show that the expected glycaemic response to the usual serving size correlates well with the GI of a 50g carbohydrate portion. Another factor which has been hindering the practical application of the GI has been the lack of a comprehensive list of GIs including common supermarket brand names and ethnic foods. We have been addressing this issue and plan to publish an extensive list comprising nearly 300 separate food items representing about 200 different types of food in the near future.

Australians are renowned for giving things a fair go and the GI appears to be no exception. The ground work has been well prepared and the only remaining problem is how best to explain it all to the patients. Various teaching strategies are being explored at present. The GI concept is arguably one of the most logical ways to teach individuals how to manage their blood glucose levels. I think it is time that we acknowledge that the GI of foods is a clinically proven method of determining which carbohydrate foods are best.

References

1. Jenkins DJA, Wolever TMS, Taylor RH, et al. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981: 34:362-6.
2. Zeman FJ. Clinical Nutrition and Dietetics. New York: Macmillan, 1991.
3. Wolever TMS, Jenkins DJA, Jenkins AL, Josse RG. The glycemic index: methodology and clinical implications. Am J Clin Nutr 1991 54:846-54.
4. Brand Miller J. The importance of glycemic index in diabetes. Am J Clin Nutr, 1994: in press.
5. Mani UV, Prabhu BM, Darnle SS and Mani I. Glycaemic index of some commonly consumed foods in western India. Asia Pacific J Clin Nutr 1993 2(3):111-114.
6. Brand JC, Nicholson PL, Thorburn AW, Truswell AS. Food processing and the glycaemic index. Am J Clin Nutr 1985: 42: 1192-1196.
7. Heaton KW, Marcus SN, Emmett PM, Bolton CH. Particle size of what, maize, and oat test meals: effects on plasma glucose and insulin responses and on the rate of starch digestion in vitro. Am J Clin Nutr 1988 47:675-82.
8. Thorburn AW, Brand JC, Truswell AS. Slowly digested and absorbed carbohydrate in traditional bushfoods: a protective factor against diabetes. Am J Clin Nutr 1987: 45:98-106.
9. Brand JC, Snow, J, Nabhan GP, Truswell AS. Plasma glucose and insulin responses to traditional Pima Indian meals. Am J Clin Nutr 1990 51:416-20.
10. Byrnes S, Denyer G, Brand Miller JC. Development of insulin resistance in rats after low amylose vs high amylose diets (abstract). Proceedings of a satellite symposium of the XVth International Congress of Nutrition 'Obesity, diabetes and the thrifty gene' Lorne, October 1993.
11. Coulston AM, Hollenbeck CB, Reaven GM. Utility of studies measuring glucose and insulin responses to various carbohydrate-containing foods. Am J Clin Nutr 1984 ; 39: 163-5.
12. Hollenbeck CB, Coulston AM, Reaven GM. Comparison of plasma glucose and insulin responses to mixed meals of high-, intermediate- and low-glycemic potential. Diabetes. Care 1988: 11:323-9.
13. Laine DC, Thomas W, Levitt MD, Bantle JP. Comparison of predictive capabilities of diabetic exchange lists and glycemic index of foods. Diabetes Care 1987 10:387-94.
14. NIH Consensus Development Conference Statement on Diet and Exercise. Bethesda, MD: US Dept of Health and Human Sciences NIH, 1986.
15. Coulston AM, Hollenbeck CB, Swiswlocki ALM, Reaven GM. Effect of source of dietary carbohydrate on plasma glucose and insulin responses to mixed meals in subjects with NIDDM. Diabetes Care 1987 10:395-400.
16. Chew I, Brand JC, Thorburn AW, Truswell AS. Plasma glucose and insulin responses to mixed meals. Proc Nutr Soc Aust 1985 10: 194-7.
17. Bornet FRJ, Costagliola D, Rizkalla SW et al. Insulinemic and glycemic indexes of six starch-rich foods taken alone 36 and in a mixed meal by type 2 diabetics. Am J Clin Nutr 1987 45:58F95.
18. Wolever TMS, Jenkins DJA. The use of glycemic index in predicting the blood glucose response to mixed meals. Am J Clin Nutr 1986: 43:167-72.
19. Rasmussen O, Winther E, Arnfred J, Hermansen K. Comparison of blood glucose and insulin responses in 38 non-insulin dependent diabetic patients. Eur J Clin Nutr 1988 42:953-61.
20. Weyman-Daum M, Fort P, Recker B, Lanes R, Lifshitz F. Glycemic response in children with insulin-dependent diabetes mellitus after high- and low-glycemic-index breakfast. Am J Clin Nutr 1987: 46:798-803.
21. Collier GR, Wolever TMS, Wong GS, Josse RG. Prediction of glycemic response to mixed meals in noninsulin-dependent diabetic subjects. Am J Clin Nutr 1986 44:349-352.
22. Colagiuri S, Miller JJ, Holliday JL, Phelan E. Comparison of plasma glucose, serum insulin, and C-peptide responses to three isocaloric breakfasts in non-insulin-dependent diabetic subjects. Diabetes Care 1986 9:250-4.
23. Chantelau E, Spraul K, Kunze K, Sonneberg, Berger M. Effects of the glycemic index of dietary carbohydrates on prandial glycemia and insulin therapy in type 1 diabetes mellitus. Diab Res Clin Prac 1986 2:35-41.
24. Parillo M, Giacco R, Rivellese A, Giacco A, Iovine C, Riccardi G. Acute effects on pancreatic hormones and blood lipids of bread and spaghetti consumed within a meal. Diab Nutr Metab 1988 1:133-7.
25. Hermansen K, Rasmussen O, Arnfred J, Winther E, Schmitz O. Glycemic effects of spaghetti and potato consumed as part of mixed meal on IDDM patients. Diabetes Care 1987: 10:401-6.
26. Jenkins DJA, Wolever TMS, Collier GR, et al. Metabolic effects of a low glycemic index diet. Am J Clin Nutr 1987: 46:968-75.
27. Jenkins DJA, Wolever TMS, Buckley G et al. Low glycemic-index starchy foods in the diabetic diet. Am J Clin Nutr 1988 48:248-54.
28. Brand JC, Colagiuri S, Crossman S, Allen A, Roberts DCK, Truswell AS. Low glycemic index foods improve long-term glycemic control in NIDDM. Diabetes Care 1991 14:95-101.
29. Wolever TMS, Jenkins DJA, Vuksan V, Jenkins AL, Buckley GC, Wong GS, Josse, RG. Beneficial effect of a low-glycaemic index diet in type 2 diabetes. Diabetic Medicine 1992 9:451-8.
30. Wolever TMS, Jenkins DJA, Vuksan V, Jenkins AL, Wong GS, Josse, RG. Beneficial effect of a low-glycemic index diet in overweight NIDDM subjects. Diabetes Care 1992 15:562-4.
31. Fontvieille AM, Rizkalla SW, Penfornis A, Acosta M, Bornet FRJ, Slama G. The use of low glycemic index foods improves metabolic control of diabetic patients in a 10 week study. Diabetic Medicine 1992 9:444-50.
32. Calle-Pascual AL, Gomez V, Leon E, Bordiu E. Foods with a low glycemic index do not improve glycemic control of both type 1 and type 2 diabetic patients after one months of therapy. Diab Metab 1988 14:629-33.
33. Collier GR, Giudici S, Kalmusky J, Wolever TMS, Helman G, Wesson V, Ehrlich RM, Jenkins DJA. Low glycemic index starchy foods improve glucose control and lower serum cholesterol in diabetic children. Diab Nutr Metab 1988 1:11-19.
34. Fontvieille AM, Acosta M, Rizkalla SW, Bornet F, David P, Letanoux G, Tchobroutsky G, Slama G. A moderate switch from high to low glycemic-index foods for 3 weeks improves the metabolic control of Type 1 (IDDM) diabetic subjects. Diab Nutr Metab 1988 1:139-43.
35. Jenkins DJA, Wolever TMS, Kalmusky J et al. Low glycemic index carbohydrate foods in the management of hyperlipidemia. Am J Clin Nutr 1985 42:604-17.
36. Jenkins DJA, Wolever TMS, Kalmusky J et al. Low glycemic index diet in hyperlipidemia: use of traditional starchy foods. Am J Clin Nutr 1987 46:66-71.
37. Frost G, Wilding JPH. Specific advice to use a low glycemic index diet improves metabolic control in newly diagnosed NIDDM. Diabetic Medicine 1993 10 (suppl 1):s29.
38. Coyle EF. Timing and method of increased carbohydrate intake to cope with heavy training, competition and recovery. J Sport Science 1991 9:29-52.
39. Flatt JP. Importance of nutrient balance in body weight regulation. Diabetes Metab Rev 1988 6:571-81.
40. Thomas DE, Brotherhood JR, Brand JC. Carbohydrate feeding before exercise: effect of glycemic index. Internat J Sports Med 1991 12:180-6.
41. Hot S, Brand JC, Soveny C, Hansky J. Relationship of satiety to postprandial glycemic, insulin and cholecystokinin responses. Appetite 1992 18:129-41.
42. Krishnamachar S, Mickelsen O. The influence of different carbohydrate sources on blood glucose levels and satiety: effect of physical activity on blood glucose response. Hum Nutr: Food Sci Nutr 1987 41F:29-39.
43. Leatherwood P. Pollet P. Effects of slow release carbohydrates in the form of bean flakes on the evolution of hunger and satiety in man. Appetite 1988 10:1-11.
44. Jenkins D, Wolever T, Jenkins A. Starchy foods and glycemic index. Diabetes Care 1988 11:149-59.
45. Reaven GM. Dietary therapy in non-insulin-dependent diabetes mellitus. N Engl J Med 1988 319:863-4.
46. Garg A, Bonanone A, Grundy SM, Zu-Jun Zhang, Unger RH. Comparison of a high carbohydrate diet with a high monounsaturated fat diet in patients with non-insulindependent diabetes mellitus. N Engl J Med 1988 319:829-834.
47. Garg A, Grundy SM, Unger RH. Comparison of effects of high and low carbohydrate diets on plasma lipoproteins and insulin sensitivity in patients with mild NIDDM. Diabetes 1992 41: 1278-85.
48. Coulston AM, Hollenbeck CB, Lin GC. Effect of source of dietary carbohydrate on plasma glucose, insulin and gastric inhibitory polypeptide responses to test meals in subjects with NIDDM. Am J Clin Nutr 1984 40:965-70.
49. Brand Miller J, Pang E, Bramall L. Rice: a high or low glycemic index food? Am J Clin Nutr 1992 56:1034-6.

Editor's note: Of related interest is the Concise review in this issue entitled Clinical nutrition of diabetes (Mark L. Wahlqvist and Richard O'Brien, pp. 149-50).

Copyright © 1993 [Asia Pacific Journal of Clinical Nutrition]. All rights reserved.
Please note: this article has been scanned and reformatted.
to the top