Glycemic index and glycemic load: measurement issues and their effect on diet–disease relationships

Abstract

Glycemic index (GI) describes the blood glucose response after consumption of a carbohydrate containing test food relative to a carbohydrate containing reference food, typically glucose or white bread. GI was originally designed for people with diabetes as a guide to food selection, advice being given to select foods with a low GI. The amount of food consumed is a major determinant of postprandial hyperglycemia, and the concept of glycemic load (GL) takes account of the GI of a food and the amount eaten. More recent recommendations regarding the potential of low GI and GL diets to reduce the risk of chronic diseases and to treat conditions other than diabetes, should be interpreted in the light of the individual variation in blood glucose levels and other methodological issues relating to measurement of GI and GL. Several factors explain the large inter- and intra-individual variation in glycemic response to foods. More reliable measurements of GI and GL of individual foods than are currently available can be obtained by studying, under standard conditions, a larger number of subjects than has typically been the case in the past. Meta-analyses suggest that foods with a low GI or GL may confer benefit in terms of glycemic control in diabetes and lipid management. However, low GI and GL foods can be energy dense and contain substantial amounts of sugars or undesirable fats that contribute to a diminished glycemic response. Therefore, functionality in terms of a low glycemic response alone does not necessarily justify a health claim. Most studies, which have demonstrated health benefits of low GI or GL involved naturally occurring and minimally processed carbohydrate containing cereals, vegetables and fruit. These foods have qualities other than their immediate impact on postprandial glycemia as a basis to recommend their consumption. When the GI or GL concepts are used to guide food choice, this should be done in the context of other nutritional indicators and when values have been reliably measured in a large group of individuals.

Introduction

The glycemic index (GI) concept was introduced by Jenkins et al. (1981) in the early 1980s as a ranking system for carbohydrates based on their immediate impact on blood glucose levels. GI was originally designed for people with diabetes as a guide to food selection, advice being given to select foods with a low GI (Jenkins et al., 1983). Lower GI foods were considered to confer benefit as a result of the relatively low glycemic response following ingestion compared with high GI foods. The GI concept has been extended to also take into account the effect of the total amount of carbohydrate consumed. Thus glycemic load (GL), a product of GI and quantity of carbohydrate eaten provides an indication of glucose available for energy or storage following a carbohydrate containing meal. Although GI is usually tested on individual foods, there are methods described whereby the GI and GL of meals and habitual diets can be estimated (Wolever and Jenkins, 1986; Salmeron et al., 1997a). In addition to a role in the treatment of diabetes, low GI and GL diets have more recently been widely recommended for the prevention of chronic diseases including diabetes, obesity, cancer and heart disease and in the treatment of cardiovascular risk factors, especially dyslipidaemia (Jenkins et al., 2002).

The usefulness of GI and GL has been questioned on several counts: failure to consider the insulin response (Coulston et al., 1984), the high intra- and inter-subject variation in glucose response to a food (Pi-Sunyer, 2002), and a loss of discriminating power when foods are combined in a mixed meal (Flint et al., 2004). Furthermore, foods with a high sugar (sucrose) content and those containing both carbohydrate and fat may have a low GI, but may not be regarded as particularly appropriate choices because of their energy density and nature of dietary fat (Freeman, 2005). This review considers the reliability of the measurement and the practical application of GI. Its value in relating dietary attributes to chronic diseases is considered in other papers in this series.

Definition and measurement

GI is defined as the blood glucose response measured as area under the curve (AUC) in response to a test food consumed by an individual under standard conditions expressed as a percentage of the AUC following consumption of a reference food consumed by the same person on a different day (FAO/WHO, 1998). The test food and reference food (usually 50 g glucose) must contain the same amount of available carbohydrate (Figure 1). It is important to standardize GI testing conditions, and the procedure for the measurement of GI is described in detail in the 1998 FAO/WHO report on carbohydrates in human nutrition (FAO, 1998). Hundreds of foods have been tested for GI with the aim of ranking foods within and between food categories. A GI classification system is in common use in which foods are categorized as having low (<55), medium (55–69) or high GI (>70) (Brand-Miller et al., 2003a).

Figure 1
figure1

Example of an individual's data used to estimate glycemic index (GI). Area under the curve (AUC) refers to the area included between the baseline and incremental blood glucose points when connected by straight lines. The area under each incremental glucose curve is calculated using the trapezoid rule (note: only areas above the baseline are used). GI=AUCFood/mean (AUCReference) × 100.

Glucose, a monosaccharide, induces a large glycemic response and is often used as the reference food and assigned a GI of 100. Some polysaccharides, such as those present in instant potato for example, may also result in large glycemic responses when consumed in an amount containing 50 g available carbohydrate because of rapid and near complete digestion and absorption in the small intestine. From the International Tables, the GI for instant potato, determined as the mean of six studies, was 85 (Foster-Powell et al., 2002). Sucrose, a disaccharide of glucose and fructose, has a somewhat lower GI of 68, resulting from the fructose component which has an exceptionally low GI of 19. Adding protein or fat to a carbohydrate containing food can also lower overall GI (Miller et al., 2006). Resistant starch and dietary fibre are largely undigested and not absorbed in the small intestine and therefore contribute little to postprandial glycemia. However, a lowering of glycemic response has been found when purified extracts of fibre, particularly of the type that forms a viscous gel in water such as guar gum, are added to a test food in sufficient quantity (Jenkins et al., 1976; Doi et al., 1979; Wolever et al., 1991; Tappy et al., 1996). GI cannot be predicted from the fibre content of a carbohydrate containing food or from the terms wholemeal and wholegrain for which there are no universally accepted definitions. For example, from the International Tables, the mean GI of wholemeal bread from 13 studies is 71, while that of white wheat bread (mean of six studies) is 70 (Foster-Powell et al., 2002). Whole grains, when largely intact, have been found to lower GI (Jenkins et al., 1986, 1988; Liljeberg et al., 1992; Granfeldt et al., 1994, 1995), but wholegrain products contain a variable proportion of intact grains.

GI does not take into account the amount of carbohydrate consumed, an important determinant of glycemic response. For example, watermelon has a high GI (Foster-Powell et al., 2002) and may not be considered a good food selection as part of a low GI diet. However, watermelon only contains 5 g of carbohydrate per 100 g, thus it would have a minimal glycemic effect. GL takes into account how much carbohydrate a serving of a food contains and may be determined by indirect and direct methods.

The indirect method involves multiplying the GI of a food by the amount of available carbohydrate in the portion of food consumed. This method implies that GL is directly proportional to the amount of the particular food eaten. This is perhaps counterintuitive, because the blood glucose AUC does not increase in direct proportion to the amount consumed. For example, eating six times the amount of bread results in an approximately threefold increase in AUC (Brand-Miller et al., 2003c). In other words, as the amount of food increased, the rate of increase in AUC declines, an effect shown in Figure 2 (Venn et al., 2006). Therefore, it is implicit in the calculation of GL that the AUC for both the test and the reference foods are attenuated to the same degree with increasing amounts consumed.

Figure 2
figure2

Blood glucose area under the curve (AUC) responses to increasing amounts of glucose and granola bar tested in 20 people.

Glycemic equivalence is a method of directly determining GL. For each subject an AUC for glucose is calculated for a range of doses of the reference food measured on different days. A standard curve is constructed for each subject with increasing amounts of the reference on the x axis with its corresponding AUC for blood glucose on the y axis (Venn et al., 2006). The AUC in response to a food consumed at any portion size, typically a usual serving, is compared to that individual's glucose standard curve as depicted in Figure 3 (Venn et al., 2006). Using this technique, glycemic equivalence is the amount of glucose that would theoretically produce the same blood glucose AUC as that particular portion size of food consumed. Major drawbacks of the direct method are increased time and cost required to determine the GL of a food. The reference must be tested at several doses in each subject and the GL of a food cannot be estimated from currently available GI values. Data from our laboratory support the premise that GL is linearly related to the amount of food consumed that is, GL calculated using GI × available carbohydrate agrees well with GL measured directly, at least when food is consumed over a range of usual intakes (Venn et al., 2006). A GL classification system is used in which foods are categorized as having low (10), medium (>10–<20) or high GL (20).

Figure 3
figure3

Example of an individual's standard glucose curve generated using glucose doses of 12.5, 25, 50 and 75 g. A test food is consumed and the resulting area under the curve (AUC) used to impute the glycemic load. AUC refers to the area included between the baseline and incremental blood glucose points when connected by straight lines. The area under each incremental glucose curve is calculated using the trapezoid rule (Note: only areas above the baseline are used.)

The relationship between GI and GL is not straightforward; for example, a high GI food can have a low GL if eaten in small quantities. Conversely, a low GI food can have a high GL dependent upon the portion size eaten. This effect is demonstrated in Table 1, in which various foods from the International Tables have been selected (Foster-Powell et al., 2002). A ‘serving size’ of watermelon, a high GI food, has the same GL as a serving size of high fat ice cream, a low GI food. Mashed potato and macaroni may be contrasted with the lower GI food (macaroni) having a higher GL per serving. Foods having very different nutrient profiles can have similar GIs and GLs per serving, such as parboiled rice and a chocolate bar. GI and GL can also be positively related to each other, for example comparing porridge and corn flakes, in which the higher GI food (corn flakes) predicts a higher GL per serving. Although a food is assigned a fixed GI value, any food could have a low, medium or high GL because GL is dependent upon the amount eaten.

Table 1 Examples of GL arranged by classification taken from the international tables (Foster-Powell et al., 2002)

The glycemic load of a diet can be calculated by summing the glycemic loads for all foods consumed in the diet. A low GL diet could be achieved by choosing small servings of foods relatively high in carbohydrate having a low GI. Alternatively, a low GL diet could comprise foods having a high fat, high protein, low carbohydrate content. The heterogeneity of foods that could be used to construct a low GL diet indicates that food selection should not be made on GL alone. Knowledge of other qualities of the food, for example fat content, type of fat, energy density, fibre content and appropriate serving size should be taken into consideration.

GI and GL labelling

Voluntary GI labelling of foods by food manufacturers occurs in several countries. Products may need to meet nutritional compositional requirements to be eligible for GI testing and labelling, such as a limit on the type or amount of fat contained in the food. However, compositional requirements are not standardized either within a country, where more than one laboratory may provide a GI-testing service, or among countries around the world. Standardized eligibility criteria would give consumers, health professionals and regulators more confidence in the suitability of a food to display its GI. It could be argued that GL should be labeled because GL more closely reflects the glycaemic impact associated with consuming an amount of the food.

Factors affecting the measurement

Postprandial glucose concentrations are dependent upon several factors. In people with impaired glucose tolerance and diabetes the glycemic response measured as blood glucose AUC is increased compared with healthy individuals. However, GI is the AUC in response to a test food relative to that of a reference food and given that each person acts as his/her own control the GI of a food should not differ in those with and without abnormalities of glucose metabolism. GI testing has been carried out, and values published in international tables, using normoglycaemic individuals as well as those with impaired glucose tolerance (Foster-Powell et al., 2002). Despite broadly comparable results Brouns et al. (2005) have recommended using people with normal glucose tolerance for the determination of GI because variability in glycemic response is greater in people with impaired glucose tolerance or diabetes.

The use of a test food referenced to a standard could be used as an argument that GI is a property of food, rather than a characteristic of the individual consuming the food.

However, postprandial glycemia may be influenced by the extent to which individuals chew food prior to swallowing (Read et al., 1986; Suzuki et al., 2005) as well as the expected biological variation in rates and extent of digestion and absorption. These variables may not apply equally to test and reference foods; a reference food commonly used is a glucose beverage. The observed intra- and inter-individual differences in GI and GL which are apparent even when measured under standardized conditions may be further exaggerated by differing physical and chemical nature of apparently similar food products (Wolever, 1990). For example, gelatinization, the process of rendering starches water soluble; retrogradation, a realignment of starch molecules during cooling and storage; starch type; and dietary fibre, are all factors with potential glycemic-modifying effects. Some of these factors are affected by cooking times and methods, and the temperature of the food consumed, potentially providing a source of variability both in GI measurement and in day-to-day variability of glycemic responses to the same food.

Reliability of GI values for individual foods

The 1998 Joint FAO/WHO Expert Consultation on carbohydrates suggested that for the determination of the GI of a food, six subjects would be required; although the basis for this number was not given (FAO, 1998). More recently, it has been recommended that a sample of 10 should be used, on the grounds that it allows for a ‘reasonable degree of power and precision for most purposes’, although it was acknowledged that more people would be necessary if greater precision was required (Brouns et al., 2005). However, there are strong indications that using 10 people is insufficient to obtain reliable estimates of GI, particularly if GI levels are high, because variance increases with the mean. Large variation in glycemic response between- and within-people makes it difficult to show differences among foods. For example, Henry et al. (2005) tested eight varieties of potato in groups of 10 people and reported mean±s.e.m. GIs ranging from 56±3 to 94±16. Despite a wide range of GIs, it was not possible to demonstrate statistically significant differences among the potato varieties. Because of the large variation there is the potential to miss-classify foods into categories of low, medium, or high GI.

In an inter-laboratory study, seven laboratories tested the GIs of centrally provided foods, each using 8–12 participants (Wolever et al., 2003). A range in mean GI values among laboratories was obtained for each of the test foods; potato 65.2±44.6–98.5±20.6; bread (locally sourced) 64.2±15.4–78.9±26.1; rice 54.8±24.1–85.0±28.6; spaghetti 36.4±35.8–69.9±18.8; and barley 23.2±24.6–47.1±49.7. Rice would have been classified as low GI (54.8±24.1) by one laboratory, medium GI (62.6±25.0, 63.3±8.1, 68.4±48.0) by three laboratories, and high GI (76.9±12.9, 85.0±28.6, 87.0±75.9) by the other three laboratories. It appeared that results were more consistent when GI was calculated using capillary blood obtained by finger prick rather than venous blood. A much better estimate was obtained when data from the laboratories using capillary blood were combined. Using a pooled sample of 47 participants, mean GI for rice was 69 with narrower confidence intervals (95% CI: 63, 76). Thus, in addition to indicating the most appropriate method for blood sampling, the data demonstrate the enhanced reliability of a measurement when studying large numbers of individuals. Most of the variation in GI differences between laboratories was attributed to random within-person variability (Wolever et al., 2003). There is no ready explanation for this random day-to-day variability in glycemic response that occurs even to repeat challenges of the same food under standardized conditions.

Within-person variability can be reduced to some extent by increasing the number of replicates for each subject. The current recommendation is that the reference food should be tested two or three times in each subject (Brouns et al., 2005). However, within-person variability is also present for the test food. Using data obtained from our laboratory in which a test food and a reference food were tested three times and four times, respectively, in 20 people, we have calculated sample sizes necessary to be confident of a difference of 10 GI units between foods. The sample size is dependent on the level of GI. For a difference of 10 units, between 30 and 40 for instance, it was estimated that 25 people would be required if the food was tested once and the reference food three times, or 19 people if the test and reference foods were both tested twice. These estimates are comparable with the sample size estimates shown by Brouns et al. (2005). The same likelihood of detecting differences of 10 GI units between foods having GIs toward the upper end of the scale (70 and 80), would require a sample size of 114 people testing the food once and the reference food three times, or 86 people if the test and reference foods are both tested twice. Increasing the sample and/or repeating the test food would appreciably increase the cost of testing, perhaps a necessary expense, if more precision in GI measurement is to be achieved.

Mixed meals

The ranking of meals by GI has been found to reflect the ranking of the major carbohydrate component in the meal. For example, baked potato was found to have a higher GI than rice (Jenkins et al., 1984). When these foods were incorporated into meals, there was a tendency for the postprandial glycemic ranking of the meals to be maintained according to the GI ranking of the food that provided the major carbohydrate source (Wolever and Jenkins, 1986). However, there is debate as to whether summing the individual GIs of foods in a meal can be used to reliably calculate the GI of the meal. Flint et al. (2004) used GI values taken from the International Table of GI (Foster-Powell et al., 2002) to predict the GI of 13 simple breakfast meals, each providing 50 g available carbohydrate. There was no association between the GI calculated from the Tables and the measured GI. It was pointed out that the energy content of the meals had not been standardized (Brand-Miller and Wolever, 2005); however, it might be argued that GI testing is standardized to an available amount of carbohydrate, not to energy content. In another study, a closer relationship was reported between calculated GI and glycemic responses to various breakfast meals, but the agreement was not entirely consistent (Wolever et al., 2006). Two of the meals, a bagel with cream cheese and orange juice; and a meal of rye bread, margarine, cereal, milk, sugar and orange juice each contained approximately 69 g available carbohydrate. The mean±s.e. glycemic responses to the meals, measured as AUC, were similar (148±14 and 143±13 mmol/l min, respectively). However, using published values the calculated GIs of the meals were predicted to be 67 and 51, respectively. Sugiyama et al. (2003) found that the ingestion of milk with rice resulted in a significantly lower GI than when rice was eaten alone. When cheese was added to potato, a dramatic lowering of the potato's mean±s.e.m. GI from 93±8 to 39±5 was found (Henry et al., 2006). Thus while it is clear that combining foods does influence GI and that the addition of protein and fat to a carbohydrate containing meal can appreciably reduce the glycemic response (Collier and O'Dea, 1983; Nuttall et al., 1984) there is insufficient information to accurately predict the effect of different combinations of foods. Aggregating the GIs of individual components of a meal does not reliably predict the observed GI of the meal as a whole.

Published glycemic index and glycemic load values

Care must be taken when using published GI and GL values. Variability of GI and GL among apparently similar foods has led to recommendations that some foods, for example rice, should be tested in the geographical region in which they are consumed. This may be necessary to account for differences in variety and cooking conditions (Foster-Powell et al., 2002). Testing in specific populations may also be important if GI is not solely a property of food. Mettler et al. (2007) found that the training state of athletes affected GI of the same food. It was suggested that Flint et al. (2004) who reported that combining the GI of single foods did not predict the GI of mixed meals, chose incorrect GI values from published tables (Wolever et al., 2006). This problem is likely to occur when multiple entries for the same food are presented. For example, in the International Tables baked Russet Burbank potatoes eaten without added fat (that is butter) are listed as having GI values of 56, 78, 94 and 111 (Foster-Powell et al., 2002). Multiple entries for the same foods also complicate research activities and food selection for individuals trying to follow a low GI diet. For example, boiled carrots have mean GI values listed in the International Tables of 92, 49 and 32 (Foster-Powell et al., 2002). If GI were a major criterion in food selection, a value of 92 might have discouraged people from eating boiled carrots. On the other hand, low GI values of 32 and 49 would have suggested boiled carrots as a highly suitable choice. It has been suggested that reliability of the estimates might have contributed to the differences in GI (Foster-Powell et al., 2002), an argument in favour of studying large numbers of individuals under the standardized conditions described above.

Dietary instruments

Questions have been raised regarding the appropriateness of the dietary instruments used when examining the relationship between GI and GL and various diseases in observational studies (Pi-Sunyer, 2002). Food frequency questionnaires used in several studies were not designed specifically to obtain information on GI and GL. Individual foods were not assigned GI or GL values, rather they were collapsed into categories. Most of the published studies have not described how foods were grouped, but an Australian study gave some insight into the process (Hodge et al., 2004). The food frequency questionnaire used in the Australian study had a category of ‘cereal foods, cakes and biscuits’. Within that category were 17 items, two of which were ‘muesli’ and ‘other breakfast cereals’. A GI value of 46% was assigned to muesli and 62% to ‘other breakfast cereals.’ The use of a single value to describe the GI of breakfast cereals seemed inappropriate given that the GIs of Australian cereals range from 30 (All-Bran) to 85 (Rice Bubbles) (Foster-Powell et al., 2002). Pi-Sunyer has drawn attention to the even broader groupings that were used in the Health Professionals' Follow-up Study and the Nurses' Health Study. Categories used were—all whole grains; all refined grains; all cold breakfast cereals; all fruit; and all fruit juices (Pi-Sunyer, 2002). Such observations do raise some concerns about the degree of confidence that can be placed on the findings of these cohort studies with respect to dietary GI or GL and disease outcome. On the other hand one might argue that misclassification leads to an underestimate of the true association between GI and GL and disease (see paper by Mann in this series). The use of dietary questionnaires specifically designed to gather information on GI and GL, such as those currently being developed, may be expected to generate data which can be interpreted with greater confidence (Flood et al., 2006; Neuhouser et al., 2006).

Overall comment on reliability

Biological variation, differing chemical and physical structure of apparently similar foods and method of food preparation and consumption may all contribute to the marked inter- and intra-individual variation observation in the glycemic response to foods. GI and GL testing on larger numbers of individuals than previously undertaken or increasing the number of replicates carried out on an individual will improve the reliability and precision of GI estimates. Specifying origin and other details of product (for example, variety of fruit or vegetable) will further enhance confidence in the measurement. However, the usefulness of the index will always be limited to some extent by the variation between and within individuals. It is common practice to place foods into broad categories of GI. Classification works well when there is a large separation in GI values between foods, but there are still uncertainties into which category of GI many foods belong because of the variability around group mean GI values. More certainty in the relative ranking of foods by GI would be attained if larger group sizes were used to estimate GI. The degree to which limitations of currently available data influence present use of the concepts of GI and GL in understanding cause or influencing management of disease is considered further in the section on recommendations.

Glycemic index and glycemic load and human health

The GI/GL concept has been widely advocated as a means of identifying foods that might protect against chronic diseases or be useful in disease management. Potential protection against diabetes and cardiovascular disease is considered in the paper by Mann (2007). Several other health issues are discussed here.

Long-term glycemic control in diabetes

The GI concept was used in the management of diabetes before being used in other clinical situations. Glycated haemoglobin (HbA1C) is measured in people with diabetes to assess overall glycemic control over a period of approximately 2–3 months prior to the measurement being made (Goldstein et al., 2004). Fructosamine, another glycated protein, is also occasionally used as a measure of glucose control over the preceding 2–3 weeks. Several studies have been carried out in people with diabetes to examine the effect on HbA1C or fructosamine of diets differing principally with respect to GI. Data from these studies form the basis of two meta-analyses. There was a modest reduction in HbA1C in people consuming low GI diets, estimated to be 0.33% units (95% CI: 0.07, 0.59) in one meta-analysis (Brand-Miller et al., 2003b), and 0.27% units (95% CI: 0.03, 0.5) in the other (Opperman et al., 2004). Fructosamine concentrations were also lower in favour of low GI dietary interventions. In one meta-analysis, the estimated difference between low and high GI dietary periods was 0.19 mmol/l (95% CI: 0.06, 0.32) (Brand-Miller et al., 2003b), and in the other 0.1 mmol/l (95% CI: 0.00, 0.20) (Opperman et al., 2004). Although these reductions in HbA1C or fructosamine are small it is important to note that these effects are in addition to other dietary changes or pharmacological treatments used in diabetes management. Whether changes in glycated proteins of this magnitude affect long-term health outcomes is untested. Trials using drugs such as acarbose, which lower postprandial hyperglycaemia, suggest that acarbose may be effective in reducing cardiovascular complications in people with type 2 diabetes mellitus (Hanefeld et al., 2004). However, reductions in HbA1C of 0.6–0.8% were achieved in these trials (Hanefeld et al., 2004; van de Laar et al., 2005). Nevertheless, any dietary strategy that resulted in improved glycaemic control would be welcome and given the difference in the acute effect that low and high GI foods have on postprandial hyperglycaemia, the proposition that changing foods in the diet from high to low GI might improve markers of glycemic control is entirely plausible. However, some caveats may be appropriate. In many of the studies included in the meta-analyses described above, the low GI foods tended to have low energy density and a high fibre content, such as whole fruit, oats, whole grain, pulses and pasta (Frost et al., 1998; Heilbronn et al., 2002). Thus, modest changes in glycaemic control were achieved under study conditions that required people to be compliant with relatively major changes in dietary habits. The findings may not necessarily apply to the many low GI functional and convenience foods currently available, which may be relatively high in sugars and energy dense.

Glycemic index and glycemic load and blood lipids

Relationships between dietary GI and blood lipid fractions have been assessed in several prospective observational studies. A reasonably consistent finding has been an inverse association between fasting HDL cholesterol concentrations and dietary GI (Liu et al., 2001; Amano et al., 2004; Slyper et al., 2005), although one study found no association (Murakami et al., 2006). Ma et al. (2006) found inverse associations between dietary GI and GL in a cross-sectional analysis, but the associations were lost during follow-up. An inverse association between GI and HDL-cholesterol concentration has also been found in a nationally representative sample of US adults (Ford and Liu, 2001).

Findings from intervention trials differed from those of observational studies. Kelly et al. (2004) conducted a meta-analysis of intervention trials that had examined the effect of low GI diets on coronary heart disease risk factors. Results from that analysis showed limited and weak evidence of an inverse relationship between GI and total cholesterol, with no effect of dietary GI on LDL and HDL cholesterol, triglycerides, fasting glucose and fasting insulin. Opperman et al. (2004) conducted a meta-analysis of 14 randomized controlled trials relating to the effects on blood lipids of altering the GI of test diets. There was a difference in total and LDL-cholesterol concentrations of 0.33 (95% CI: 0.18, 0.47) mmol/l and 0.15 (95% CI: 0.00, 0.31) mmol/l, favoring the low GI diets, but no difference in HDL cholesterol concentrations between people consuming low and high GI diets.

Thus, the reasonably consistent finding in observational studies of an inverse association between dietary GI and HDL cholesterol concentrations is not confirmed by meta-analyses of randomized controlled trials in which people consumed diets that had been designed specifically to achieve differences in GI (Kelly et al., 2004; Opperman et al., 2004). On the other hand, the meta-analyses show differences in total and LDL cholesterol not found in the observational data. There is no obvious explanation for this inconsistency.

Glycemic index and glycemic load and insulin response

The GI of a food is affected not only by the rate of absorption of carbohydrate, but also by the rate of glucose removal from the plasma. When comparing two breakfast cereals with different GI values (131±33 and 54.5±7.2), the rate of glucose removal was a major determinant of postprandial hyperglycaemia (Schenk et al., 2003). It was found that the lower GI breakfast cereal had induced hyperinsulinaemia earlier than the higher GI cereal, resulting in an earlier increase in more rapid removal of glucose from circulation. It has been known for some time that insulin response cannot be predicted based solely on the glycemic response to a food. Collier and O'Dea (1983) found marked differences in the glycemic response to potato with or without added butter, but a very similar insulin response. The effect of GI on insulin response may also depend upon insulin sensitivity. Dietary GI has not been shown to have a marked effect on insulin sensitivity whereas dietary fibre has (McAuley and Mann, 2006).

Glycemic index and satiety

An important justification for the claim of an overall health benefit of low GI foods is that low GI foods may aid weight control because they promote satiety (Brand-Miller et al., 2002). Ideally, weight loss studies comparing low and high GI diets would need to assess differences between diets based on ad libitum intake to show that the apparently greater satiating effect of low GI foods led to a reduced energy intake. Holt and colleagues have carried out the most comprehensive study investigating the relationship between GI and satiety, reporting the same work in several articles (Holt et al., 1995; Holt et al., 1996; Holt et al., 1997). Iso-energetic (1000 kJ) servings of 38 foods were tested for satiety rating and glucose and insulin response. The food with the highest satiety score was boiled potato. When comparing a high GI food (potato) with a low GI food (white pasta) on an iso-energetic, equi-carbohydrate (49 g) basis, the high GI food had the highest satiety rating. The opposite was true when comparing oranges (lower GI) and white bread (higher GI), where the lower GI food had the higher satiety rating. Porridge and natural muesli had similar glycemic and insulinemic scores, but porridge had a greater satiety index than muesli (P<0.001). These results suggest that there is little or no relationship between GI and satiety, at least when comparing food portions of equal energy content. Rather, energy density appeared to be inversely related to satiety, presumably because of the high bulk required to obtain a serving containing 1000 kJ when low energy-dense foods were tested. When iso-energetic, iso-volumetric carbohydrate-containing beverages were tested, high GI beverages resulted in lower energy intakes during a subsequent meal, while low GI beverages were found not to suppress appetite and food intake in the short-term (Anderson et al., 2002). A review of the effect of glycemic carbohydrates on short-term satiety has been published (Anderson and Woodend, 2003). One conclusion was that high GI carbohydrates suppress short-term (1 h) food intake more effectively than low GI carbohydrates, whereas low GI carbohydrates appeared to be more effective over longer periods (6 h).

How dietary GI and GL affects satiety and food intake over a number of years is not entirely clear. The effectiveness of dietary GI and GL on weight loss or maintenance is covered by van Damm in this series. The results of several observational studies have shown little difference in body mass index (BMI) across categories of GI and GL (Salmeron et al., 1997a1997b; Hodge et al., 2004; Schulze et al., 2004). Murakami et al. (2006) found a positive association between GI and body mass index in Japanese female farmers, but no association between GL and body mass index. On the other hand, Ford and Liu reported inverse associations between GI and GL and body mass index in a nationally representative sample of US adults (Ford and Liu, 2001). These contradictory findings might suggest that dietary GI and GL is not a major determinant of dietary energy intake over the long-term. A plausible reason is that GI appears not to be related to energy density. Potatoes and lentils for example represent foods with widely differing GIs but comparable energy densities of around 3–4 kJ/g. On the other hand, cakes, cookies and fresh oranges have similar GIs in the low to medium range, but energy densities some 10-fold different (Holt et al., 1996).

Recommendations

The FAO/WHO Report on Carbohydrates in Human Nutrition suggests that the concept of GI provides a useful means of selecting the most appropriate carbohydrate containing foods for the maintenance of health and the treatment of several disease states (FAO, 1998). Since the publication of that report some of the limitations of the GI and GL concepts have become increasingly apparent. With regard to measurement there is clearly a need to study a larger number of subjects under standard conditions to obtain more precise estimates of the GI and GL of individual foods. The introduction of instruments for assessing dietary intake in epidemiological studies that have been designed to include more direct measures of GI and GL will enhance the confidence in findings from such studies. Despite these reservations it does appear that distinguishing between foods with appreciable differences in the indices may produce some benefit in terms of glycemic control in diabetes and lipid management. However, caution should be exercised in food choice based solely on GI or GL because low GI and GL foods may be energy dense and contain substantial amounts of sugars or undesirable fatty acids that contribute to the diminished glycemic response but not necessarily to good health outcomes. This may apply especially to some of the manufactured products that have been GI and GL tested and are available in many countries. Given that most of the studies which have demonstrated a health benefit of low GI and GL involved the use of naturally occurring and minimally processed foods it would seem to be appropriate for such products to be further tested for their health benefits directly, rather than on the basis of their functionality (that is, a low glycemic response). Although some data suggest that the low GI effect is not explained by the dietary fibre content of the foods it remains conceivable that food structure or composition explain some of the health benefits. GI may be a useful indicator to guide food choice if for example bread with a high GI is replaced on a slice-for-slice basis with a lower GI bread, thereby achieving a lower GL. However, the complexity of the relationship between GI and GL is probably not well understood whereby GI and the amount of a food eaten are both important determinants of the postprandial glycemic response. For the present it would seem appropriate that when GI or GL are used to guide food choice, it should only be done in the context of other nutritional indicators and when values have been measured in a large group of individuals.

References

  1. Amano Y, Kawakubo K, Lee JS, Tang AC, Sugiyama M, Mori K (2004). Correlation between dietary glycemic index and cardiovascular disease risk factors among Japanese women. Eur J Clin Nutr 58, 1472–1478.

    CAS  Article  Google Scholar 

  2. Anderson GH, Catherine NL, Woodend DM, Wolever TM (2002). Inverse association between the effect of carbohydrates on blood glucose and subsequent short-term food intake in young men. Am J Clin Nutr 76, 1023–1030.

    CAS  Article  Google Scholar 

  3. Anderson GH, Woodend D (2003). Effect of glycemic carbohydrates on short-term satiety and food intake. Nutr Rev 61, S17–S26.

    Article  Google Scholar 

  4. Brand-Miller J, Hayne S, Petocz P, Colagiuri S (2003b). Low-glycemic index diets in the management of diabetes: a meta-analysis of randomized controlled trials. Diabetes Care 26, 2261–2267.

    Article  Google Scholar 

  5. Brand-Miller JC, Holt SH, Pawlak DB, McMillan J (2002). Glycemic index and obesity. Am J Clin Nutr 76, 281S–285S.

    CAS  Article  Google Scholar 

  6. Brand-Miller JC, Thomas M, Swan V, Ahmad ZI, Petocz P, Colagiuri S (2003c). Physiological validation of the concept of glycemic load in lean young adults. J Nutr 133, 2728–2732.

    CAS  Article  Google Scholar 

  7. Brand-Miller JC, Wolever TM (2005). The use of glycaemic index tables to predict glycaemic index of breakfast meals. Br J Nutr 94, 133–134.

    CAS  Article  Google Scholar 

  8. Brand-Miller J, Wolever TM, Foster-Powell K, Colagiuri S (2003a). The New Glucose Revolution. Marlowe & Company: New York.

    Google Scholar 

  9. Brouns F, Bjorck I, Frayn KN, Gibbs AL, Lang V, Slama G et al. (2005). Glycaemic index methodology. Nutr Res Rev 18, 145–171.

    CAS  Article  Google Scholar 

  10. Collier G, O'Dea K (1983). The effect of coingestion of fat on the glucose, insulin, and gastric inhibitory polypeptide responses to carbohydrate and protein. Am J Clin Nutr 37, 941–944.

    CAS  Article  Google Scholar 

  11. Coulston AM, Hollenbeck CB, Reaven GM (1984). Utility of studies measuring glucose and insulin responses to various carbohydrate-containing foods. Am J Clin Nutr 39, 163–167.

    CAS  Article  Google Scholar 

  12. Doi K, Matsuura M, Kawara A, Baba S (1979). Treatment of diabetes with glucomannan (konjac mannan). Lancet 1, 987–988.

    CAS  Article  Google Scholar 

  13. FAO/WHO (1998). Carbohydrates in human nutrition. Report of a Joint FAO/WHO Expert Consultation. FAO Food Nutr Pap 66, 1–140.

    Google Scholar 

  14. Flint A, Moller BK, Raben A, Pedersen D, Tetens I, Holst JJ et al. (2004). The use of glycaemic index tables to predict glycaemic index of composite breakfast meals. Br J Nutr 91, 979–989.

    CAS  Article  Google Scholar 

  15. Flood A, Subar AF, Hull SG, Zimmerman TP, Jenkins DJ, Schatzkin A (2006). Methodology for adding glycemic load values to the National Cancer Institute Diet History Questionnaire database. J Am Diet Assoc 106, 393–402.

    Article  Google Scholar 

  16. Ford ES, Liu S (2001). Glycemic index and serum high-density lipoprotein cholesterol concentration among US adults. Arch Intern Med 161, 572–576.

    CAS  Article  Google Scholar 

  17. Foster-Powell K, Holt SH, Brand-Miller JC (2002). International table of glycemic index and glycemic load values: 2002. Am J Clin Nutr 76, 5–56.

    CAS  Article  Google Scholar 

  18. Freeman J (2005). The glycemic index debate: does the type of carbohydrate really matter? Diabet Forecast 58, 11.

    Google Scholar 

  19. Frost G, Leeds A, Trew G, Margara R, Dornhorst A (1998). Insulin sensitivity in women at risk of coronary heart disease and the effect of a low glycemic diet. Metabolism 47, 1245–1251.

    CAS  Article  Google Scholar 

  20. Goldstein DE, Little RR, Lorenz RA, Malone JI, Nathan DM, Peterson CM (2004). Tests of glycemia in diabetes. Diabetes care 27 (Suppl 1), S91–S93.

    PubMed  Google Scholar 

  21. Granfeldt Y, Liljeberg H, Drews A, Newman R, Bjorck I (1994). Glucose and insulin responses to barley products: influence of food structure and amylose-amylopectin ratio. Am J Clin Nutr 59, 1075–1082.

    CAS  Article  Google Scholar 

  22. Granfeldt Y, Hagander B, Bjorck I (1995). Metabolic responses to starch in oat and wheat products. On the importance of food structure, incomplete gelatinization or presence of viscous dietary fibre. Eur J Clin Nutr 49, 189–199.

    CAS  PubMed  Google Scholar 

  23. Hanefeld M, Cagatay M, Petrowitsch T, Neuser D, Petzinna D, Rupp M (2004). Acarbose reduces the risk for myocardial infarction in type 2 diabetic patients: meta-analysis of seven long-term studies. Eur Heart J 25, 10–16.

    CAS  Article  Google Scholar 

  24. Heilbronn LK, Noakes M, Clifton PM (2002). The effect of high- and low-glycemic index energy restricted diets on plasma lipid and glucose profiles in type 2 diabetic subjects with varying glycemic control. J Am Coll Nutr 21, 120–127.

    CAS  Article  Google Scholar 

  25. Henry CJ, Lightowler HJ, Kendall FL, Storey M (2006). The impact of the addition of toppings/fillings on the glycaemic response to commonly consumed carbohydrate foods. Eur J Clin Nutr 60, 763–769.

    CAS  Article  Google Scholar 

  26. Henry CJ, Lightowler HJ, Strik CM, Storey M (2005). Glycaemic index values for commercially available potatoes in Great Britain. Br J Nutr 94, 917–921.

    CAS  Article  Google Scholar 

  27. Hodge AM, English DR, O'Dea K, Giles GG (2004). Glycemic index and dietary fiber and the risk of type 2 diabetes. Diabetes Care 27, 2701–2706.

    Article  Google Scholar 

  28. Holt SH, Miller JC, Petocz P (1997). An insulin index of foods: the insulin demand generated by 1000-kJ portions of common foods. Am J Clin Nutr 66, 1264–1276.

    CAS  Article  Google Scholar 

  29. Holt SH, Miller JC, Petocz P, Farmakalidis E (1995). A satiety index of common foods. Eur J Clin Nutr 49, 675–690.

    CAS  PubMed  Google Scholar 

  30. Holt SH, Brand Miller JC, Petocz P (1996). Inter-relationships among postprandial satiety, glucose and insulin responses and changes in subsequent food intake. Eur J Clin Nutr 50, 788–797.

    CAS  Google Scholar 

  31. Jenkins DJ, Goff DV, Leeds AR, Alberti KG, Wolever TM, Gassull MA et al. (1976). Unabsorbable carbohydrates and diabetes: decreased post-prandial hyperglycaemia. Lancet 2, 172–174.

    CAS  Article  Google Scholar 

  32. Jenkins DJ, Kendall CW, Augustin LS, Franceschi S, Hamidi M, Marchie A et al. (2002). Glycemic index: overview of implications in health and disease. Am J Clin Nutr 76, 266S–273S.

    CAS  Article  Google Scholar 

  33. Jenkins DJ, Wolever TM, Buckley G, Lam KY, Giudici S, Kalmusky J et al. (1988). Low-glycemic index starchy foods in the diabetic diet. Am J Clin Nutr 48, 248–254.

    CAS  Article  Google Scholar 

  34. Jenkins DJ, Wolever TM, Jenkins AL, Giordano C, Giudici S, Thompson LU et al. (1986). Low-glycemic response to traditionally processed wheat and rye products: bulgur and pumpernickel bread. Am J Clin Nutr 43, 516–520.

    CAS  Article  Google Scholar 

  35. Jenkins DJ, Wolever TM, Jenkins AL, Josse RG, Wong GS (1984). The glycaemic response to carbohydrate foods. Lancet 2, 388–391.

    CAS  Article  Google Scholar 

  36. Jenkins DJ, Wolever TM, Jenkins AL, Thorne MJ, Lee R, Kalmusky J et al. (1983). The glycaemic index of foods tested in diabetic patients: a new basis for carbohydrate exchange favouring the use of legumes. Diabetologia 24, 257–264.

    CAS  Article  Google Scholar 

  37. Jenkins DJ, Wolever TM, Taylor RH, Barker H, Fielden H, Baldwin JM et al. (1981). Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 34, 362–366.

    CAS  Article  Google Scholar 

  38. Kelly S, Frost G, Whittaker V, Summerbell C (2004). Low-glycaemic index diets for coronary heart disease. Cochrane Database Syst Rev CD004467.

  39. Liljeberg H, Granfeldt Y, Bjorck I (1992). Metabolic responses to starch in bread containing intact kernels versus milled flour. Eur J Clin Nutr 46, 561–575.

    CAS  PubMed  Google Scholar 

  40. Liu S, Manson JE, Stampfer MJ, Holmes MD, Hu FB, Hankinson SE et al. (2001). Dietary glycemic load assessed by food-frequency questionnaire in relation to plasma high-density lipoprotein cholesterol and fasting plasma triacylglycerols in postmenopausal women. Am J Clin Nutr 73, 560–566.

    CAS  Article  Google Scholar 

  41. Ma Y, Li Y, Chiriboga DE, Olendzki BC, Hebert JR, Li W et al. (2006). Association between carbohydrate intake and serum lipids. J Am Coll Nutr 25, 155–163.

    CAS  Article  Google Scholar 

  42. Mann J (2007). Dietary carbohydrate: relationship to cardiovascular disease and disorders of carbohydrate metabolism. Eur J Clin Nutr 61 (Suppl 1), S100–S111.

    CAS  Article  Google Scholar 

  43. McAuley K, Mann J (2006). Thematic review series: patient-oriented research. Nutritional determinants of insulin resistance. J Lipid Res 47, 1668–1676.

    CAS  Article  Google Scholar 

  44. Mettler S, Lamprecht-Rusca F, Stoffel-Kurt N, Wenk C, Colombani PC (2007). The influence of the subjects' training state on the glycemic index. Eur J Clin Nutr 61, 19–24.

    CAS  Article  Google Scholar 

  45. Miller CK, Gabbay RA, Dillon J, Apgar J, Miller D (2006). The effect of three snack bars on glycemic response in healthy adults. J Am Diet Assoc 106, 745–748.

    Article  Google Scholar 

  46. Murakami K, Sasaki S, Takahashi Y, Okubo H, Hosoi Y, Horiguchi H et al. (2006). Dietary glycemic index and load in relation to metabolic risk factors in Japanese female farmers with traditional dietary habits. Am J Clin Nutr 83, 1161–1169.

    CAS  Article  Google Scholar 

  47. Neuhouser ML, Tinker LF, Thomson C, Caan B, Horn LV, Snetselaar L et al. (2006). Development of a glycemic index database for food frequency questionnaires used in epidemiologic studies. J Nutr 136, 1604–1609.

    CAS  Article  Google Scholar 

  48. Nuttall FQ, Mooradian AD, Gannon MC, Billington C, Krezowski P (1984). Effect of protein ingestion on the glucose and insulin response to a standardized oral glucose load. Diabetes Care 7, 465–470.

    CAS  Article  Google Scholar 

  49. Opperman AM, Venter CS, Oosthuizen W, Thompson RL, Vorster HH (2004). Meta-analysis of the health effects of using the glycaemic index in meal-planning. Br J Nutr 92, 367–381.

    CAS  Article  Google Scholar 

  50. Pi-Sunyer FX (2002). Glycemic index and disease. Am J Clin Nutr 76, 290S–298S.

    CAS  Article  Google Scholar 

  51. Read NW, Welch IM, Austen CJ, Barnish C, Bartlett CE, Baxter AJ et al. (1986). Swallowing food without chewing; a simple way to reduce postprandial glycaemia. Br J Nutr 55, 43–47.

    CAS  Article  Google Scholar 

  52. Salmeron J, Ascherio A, Rimm EB, Colditz GA, Spiegelman D, Jenkins DJ et al. (1997a). Dietary fiber, glycemic load, and risk of NIDDM in men. Diabetes Care 20, 545–550.

    CAS  Article  Google Scholar 

  53. Salmeron J, Manson JE, Stampfer MJ, Colditz GA, Wing AL, Willett WC (1997b). Dietary fiber, glycemic load, and risk of non-insulin-dependent diabetes mellitus in women. JAMA 277, 472–477.

    CAS  Article  Google Scholar 

  54. Schenk S, Davidson CJ, Zderic TW, Byerley LO, Coyle EF (2003). Different glycemic indexes of breakfast cereals are not due to glucose entry into blood but to glucose removal by tissue. Am J Clin Nutr 78, 742–748.

    CAS  Article  Google Scholar 

  55. Schulze MB, Liu S, Rimm EB, Manson JE, Willett WC, Hu FB (2004). Glycemic index, glycemic load, and dietary fiber intake and incidence of type 2 diabetes in younger and middle-aged women. Am J Clin Nutr 80, 348–356.

    CAS  Article  Google Scholar 

  56. Slyper A, Jurva J, Pleuss J, Hoffmann R, Gutterman D (2005). Influence of glycemic load on HDL cholesterol in youth. Am J Clin Nutr 81, 376–379.

    CAS  Article  Google Scholar 

  57. Sugiyama M, Tang AC, Wakaki Y, Koyama W (2003). Glycemic index of single and mixed meal foods among common Japanese foods with white rice as a reference food. Eur J Clin Nutr 57, 743–752.

    CAS  Article  Google Scholar 

  58. Suzuki H, Fukushima M, Okamoto S, Takahashi O, Shimbo T, Kurose T et al. (2005). Effects of thorough mastication on postprandial plasma glucose concentrations in nonobese Japanese subjects. Metabolism 54, 1593–1599.

    CAS  Article  Google Scholar 

  59. Tappy L, Gugolz E, Wursch P (1996). Effects of breakfast cereals containing various amounts of beta-glucan fibers on plasma glucose and insulin responses in NIDDM subjects. Diabetes Care 19, 831–834.

    CAS  Article  Google Scholar 

  60. van de Laar FA, Lucassen PL, Akkermans RP, van de Lisdonk EH, Rutten GE, van Weel C (2005). Alpha-glucosidase inhibitors for patients with type 2 diabetes: results from a Cochrane systematic review and meta-analysis. Diabetes Care 28, 154–163.

    CAS  Article  Google Scholar 

  61. Venn BJ, Wallace AJ, Monro JA, Perry T, Brown R, Frampton C et al. (2006). The glycemic load estimated from the glycemic index does not differ greatly from that measured using a standard curve in healthy volunteers. J Nutr 136, 1377–1381.

    CAS  Article  Google Scholar 

  62. Wolever TM, Jenkins DJ (1986). The use of the glycemic index in predicting the blood glucose response to mixed meals. Am J Clin Nutr 43, 167–172.

    CAS  Article  Google Scholar 

  63. Wolever TM, Vorster HH, Bjorck I, Brand-Miller J, Brighenti F, Mann JI et al. (2003). Determination of the glycaemic index of foods: interlaboratory study. Eur J Clin Nutr 57, 475–482.

    CAS  Article  Google Scholar 

  64. Wolever TM (1990). The glycemic index. World Rev Nutr Diet 62, 120–185.

    CAS  Article  Google Scholar 

  65. Wolever TM, Vuksan V, Eshuis H, Spadafora P, Peterson RD, Chao ES et al. (1991). Effect of method of administration of psyllium on glycemic response and carbohydrate digestibility. J Am Coll Nutr 10, 364–371.

    CAS  Article  Google Scholar 

  66. Wolever TM, Yang M, Zeng XY, Atkinson F, Brand-Miller JC (2006). Food glycemic index, as given in glycemic index tables, is a significant determinant of glycemic responses elicited by composite breakfast meals. Am J Clin Nutr 83, 1306–1312.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

We wish to thank Dr Jennie Brand-Miller, Professor Gary Frost, Professor Philip James, Professor Simin Liu, Professor Jim Mann, Dr Gabriele Riccardi, Dr M Robertson and Professor HH Vorster for their valuable comments.

Author information

Affiliations

Authors

Corresponding author

Correspondence to B J Venn.

Additional information

Conflict of interest

During the preparation and peer-review of this paper in 2006, the authors and peer-reviewers declared the following interests.

Authors Dr Bernard J Venn: None declared.

Dr Tim Green: Affiliated with GI Otago, a commercial glycemic index testing service.

Peer-reviewers

Dr Jennie Brand-Miller: Publishing books in the popular press: ‘The New Glucose Revolution Series’; Director of a University-based service for GI testing; Director of a not-for-profit food-labelling programme based on the GI.

Professor Gary Frost: None declared.

Professor Philip James: None declared.

Professor Simin Liu: None declared.

Professor Jim Mann: None declared.

Dr Gabriele Riccardi: None declared.

Dr M Robertson: Research Grant from National Chemical and Starch.

Professor HH Vorster: Member and Director of the Africa Unit for Transdisciplinary health Research (AUTHeR), Research grant from the South African Sugar Association.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Venn, B., Green, T. Glycemic index and glycemic load: measurement issues and their effect on diet–disease relationships. Eur J Clin Nutr 61, S122–S131 (2007). https://doi.org/10.1038/sj.ejcn.1602942

Download citation

Keywords

  • glycemic index
  • glycemic load
  • methodology
  • diet–disease relationships

Further reading