Consideration of the validity of glycemic index using blood glucose and insulin levels and breath hydrogen excretion in healthy subjects

Article Outline

• Abstract
• 1. Introduction
• 2. Materials and methods
  • 2.1. Subjects
  • 2.2. Test foods
  • 2.3. Experimental protocol
  • 2.4. Analysis of blood glucose, insulin and breath hydrogen
  • 2.5. Calculation and statistical analysis
  • 2.6. Ethics
• 3. Results
  • 3.1. Cultivar variations in the GI and II
    • 3.1.1. GI and II of the four cultivars of white rice
    • 3.1.2. GI and II of the three cultivars of steamed potato
    • 3.1.3. GI and II of the three brands of noodle and others
  • 3.2. Correlation between GI and II
  • 3.3. Comparison of AUCs of blood glucose and insulin elevation between 2h and 3h after the ingestion of test foods
  • 3.4. Interaction among AUC-8h of breath hydrogen excretion, AUC-3h-glucose and AUC-3h-insulin of test foods
• 4. Discussions
• Acknowledgements
• References
• Copyright

Abstract 

Aim

Although glycemic index (GI) is very important in choosing appropriate foods for patients with diabetes mellitus, GI itself does not provide sufficient information for choosing adequate foods. The validity of GI is considered by measurement of blood glucose and insulin levels, and breath hydrogen excretion, testing several cultivars in the same type food.

Methods

Twelve females, 23.8y participated in this within-subject, repeated-measures study. To clarify variations in GI in inter-cultivars of various foods, we examined four white rice and three potato cultivars and three noodle brands. Starchy-foods with a glucose equivalent of 50g were repeatedly and randomly given to each subject. Blood and end-expiration were collected at selected periods.

Results

The significant difference of GI and insulinemic index (II) was not observed among the four white rice cultivars, though II of one cultivar were smaller than those of other three white rice cultivars. GI of three potato cultivars was relatively small, but the range of II was very large among three cultivars. Moreover, GI did not correspond to II among three noodle brands. AUC-3h-glucose and AUC-3h-insulin scores of white rice and noodle were significantly larger than those for 2h. The amount of breath hydrogen excretion showed a negative correlation to GI of tested foods.

Conclusions

We should recognize that rare foods in which GI does not correspond to II exist in the cultivar of foods used for diet therapy of diabetes mellitus. We propose the addition of other information such as II and breath hydrogen excretion of selected foods.

Keywords: Validity, Glycemic index, Blood glucose, Blood insulin, Breath hydrogen excretion

Back to Article Outline

1. Introduction 

The glycemic index (GI) is an indicator of the early response of postprandial blood glucose, and is currently used to choose suitable foods in diet therapy for diabetes mellitus. It is true that GI contributes to the prevention and the recovery of diabetes mellitus and metabolic syndromes. However, GI itself seems not to provide sufficient information for choosing adequate foods in diet therapy, though it is an interesting concept to use GI in this way to help patients with diabetes mellitus or other metabolic syndromes [1], [2], [3]. Indeed, new concepts such as glycemic load, glycemic impact, glycemic glucose equivalent and carbohydrate counting have recently been proposed to enhance the use of the GI [4], [5], [6].

In order to influence systemic metabolism, it is more important for patients to control postprandial blood insulin, rather than blood glucose, because the patients have to suppress final insulin secretion from the pancreas [7]. However, there are some foods and combinatorial diets that do not show a low insulinemic index (II) despite their low GI [8], [9], [10], [11], [12]. Furthermore, although foods with abundant fat and protein that are low in carbohydrates surely have a low GI, they provide high available energy. The nutritional components of these non-starchy foods are not presumed from GI. Information besides GI can provide more suitable diet therapy for diabetes mellitus or other metabolic syndromes. GI is calculated based on the area under the curve (AUC) for incremental blood glucose until 2h after the ingestion of food [13], [14], but the AUC until 3h or more after ingestion may more accurately reflect the relevant features of the food. This is because there are some foods where the ratio of AUC of blood glucose versus that of 50g of glucose for 2h is different from the ratio of AUC of blood glucose versus that of 50g of glucose for 3h.

The GI score is low when the carbohydrate in foods is digested slowly and then absorbed in the small intestine [15], or are not digested by the inhibition to digestive enzymes and are then transferred to the lower intestine [16]. When carbohydrates are not digested and are fermented by intestinal microbes in the large intestine, hydrogen is produced and excreted in the breath [17], [18], [19]. Digestible carbohydrate, which rapidly increases blood glucose after ingestion, does not produce hydrogen, while non-digestible carbohydrate, which does not increase blood glucose, produces spontaneously breath hydrogen [20], [21]. Therefore, it is considered that blood glucose and insulin levels have an inverse relation to breath hydrogen excretion. Breath hydrogen excretion was used in the present study, as a new parameter to enhance the data provided by blood glucose and insulin levels.

GI scores have been reported for approximately 1300 foods in GI data bases [5], but data on both blood glucose and insulin levels are limited. Additionally, to the best of our knowledge, it is the variation in the GI depending on different inter-cultivars or inter-brands of the same class of food has been insufficiently discussed, even though there are many cultivars of foods such as rice and potatoes. The aim of the present study was to examine and to consider the validity of GI of different varieties of same type food by measuring insulin levels, and breath hydrogen excretion in healthy subjects.

Back to Article Outline

2. Materials and methods 

2.1. Subjects 

Twelve healthy females, aged 23.8±3.7y and with a body mass index of 20.3±1.9kg/m² participated in this study. The exclusion criteria were a history of diabetes, carbohydrate malabsorption, obesity or pulmonary disease. The average and standard deviation (SD) of fasting blood glucose levels of subjects was 83±4.3mg/100mL. The subjects had not taken antibiotics or laxatives for at least 2-week prior to the experiment. All subjects provided their informed consent to participate in the study.

2.2. Test foods 

To clarify variations in the GI in the inter-cultivar of various foods, the present test substances were four cultivars of white rice, three cultivars of potato and three brands of noodle (Table 1). White rice-1 (had already cooked and packed with a plastic container) was purchased from Sato Foods Co. (Niigata, Japan) and white rice-2 (Nihonbare), white rice-3 (Hinohikari) and white rice-4 (Koshihikari) were cooked using an electric rice cooker (SR-JTM15; Matsushita Electric Industrial Co. Ltd., Osaka, Japan). The three potato cultivars (Nishiyutaka, Ainoaka, Dejima) were cut to the size of about 1.3cm³ and then cooked using a microwave oven (NE-S300F; Matsushita Electric Industrial Co. Ltd., Osaka, Japan) for approximately 90s, and the three noodle brands (Shimabara-udon, Goto-udon, and Katokichi-udon) were cooked according to the manufacturers’ instructions using a cookpot. White bread and sponge cake (Castilla, made with wheat flour) were purchased from a nearby grocery store. To ingest available carbohydrates with a glucose equivalent of 50g, 135–150g of cooked white rice and 284g of steamed potato were given to each subject and were eaten without other dishes, and 170–172g of cooked (boiled) noodles were eaten with the arranged soup. White bread (107g) and sponge cake (80g) were also eaten by each subject. We determined the portion size based on the available carbohydrate that is published in “Standard tables of food composition in Japan, fifth revised edition” or that is analyzed by Japan Food Research Laboratories (Tokyo, Japan). Glucose (50g) was dissolved in 150mL of warmed water and then the solution was drunk adding approximately 100mL of warmed water by each subject.

Table 1. Nutritive components of test foods.

TDF: total dietary fibers.

2.3. Experimental protocol 

This study was performed using a within-subject, repeated-measures design, and the test solution and foods were given in random order with intervals of at least 1week. The period of total experiment was 2.5–3months and we carried out the experiment in the best season to be able to obtain the high quality foods constantly. The subjects ingested the test substances within a few minutes. The experimental protocol was carried out in accordance with our previous study [19], [21], [22]. After overnight fasting, the subject’s health status was examined, the food intakes for last 1week was interviewed, and blood pressure and pulse rate were measured, and then a baseline blood sample and end-expiratory gas were collected. After ingesting the test food, blood was collected from the fingertips using a heparinized capillary tube (hematocrit tube) at 15, 30, 45, 60, 90, 120, 150 and 180min, and 750mL of end-respiration after dead-space air was eliminated were collected at 1h intervals for 8h. The subjects consumed their usual diet and they were prohibited from ingesting foods containing non-digestible carbohydrates starting at 3days before each experimental day. We provided supper, which consisted of cooked paddy white rice, chicken, a small amount of cabbage, and soup on the day before each experiment. We have already evaluated that the breath hydrogen was not detected by the ingestion of these foods for supper. The subjects were prohibited from ingesting foods containing non-digestible carbohydrates starting at 3days before the experimental day. During the experiment, they were also prohibited from ingesting food or beverages except for water, as well as from sleeping or smoking. The subjects ingested an experimental meal which had been previously tested that breath hydrogen was not produced, after the final collection of blood, so the subjects would not feel hungry.

2.4. Analysis of blood glucose, insulin and breath hydrogen 

Blood samples were centrifuged at 13,750g (Kubota 3110 for hematocrit; Kubota Corp., Tokyo, Japan) for 5min at room temperature to separate plasma. Plasma glucose was measured in duplicate following Trinder’s method using glucose oxidase [23], and plasma insulin was measured using an enzyme-linked immunosorbent assay kit [24] (Seikagaku Corp., Kanagawa, Japan). Breath hydrogen was analyzed in duplicate, using a simple gas chromatograph (Breath Gas Analyzer TGA2000; Teramecs Co. Ltd., Kyoto, Japan).

2.5. Calculation and statistical analysis 

GI scores were calculated in accordance with calculation method in GI methodology [25], [14]. Incremental blood glucose AUC versus time were calculated using the trapezoidal rule with fasting values as the baseline, and the ratio of AUCs for 2h versus 50g of glucose ingestion was considered to be the GI. II was calculated using the same methods. The mean and SD of the incremental blood glucose and insulin from the baseline as well as the concentrations of breath hydrogen were calculated using all 12 subjects. The AUCs of blood glucose and insulin for 2h (AUC-2h-glucose and AUC-2h-insulin) and 3h (AUC-3h-glucose and AUC-3h-insulin) after ingestion were calculated, when AUC of 50g of glucose was 100. AUC of breath hydrogen for 8h after ingestion of test materials were calculated when AUC of 50g of glucose is 1. One-way analysis of variance and Tukey’s post hoc test were used to analyze the difference among cultivars of same type food, the difference between AUC-2h and AUC-3h of glucose and insulin were compared with paired Student’s t-test, and Pearson’s correlation was analyzed. A p-value of less than 0.05 was considered significant with two-sided analysis using SPSS ver.12 for Windows, Japan, (SPSS Inc., Tokyo, Japan).

2.6. Ethics 

The study protocol was approved by the ethical committee of University of Nagasaki, Siebold. All experiments were carried out in the Public Health Nutrition Laboratory of the Graduate School of Human Health Science, University of Nagasaki, Siebold.

Back to Article Outline

3. Results 

No subjects withdrew and all of them completely performed without side effects. Their health status was good through the full period of the experiment.

3.1. Cultivar variations in the GI and II 

Fig. 1 shows the curves of blood glucose and insulin and Table 2 shows a summary of the GI and II of the tested foods.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Curves of elevation of blood glucose and insulin after the ingestion of glucose and test foods. All subjects ingested the solution (150mL) containing 50g of glucose within 1min, and ate test foods containing available carbohydrate with a glucose equivalent of 50g with 150mL of water within 5min. About 150μL of blood was collected from the fingertip using a heparinized capillary tube at 15, 30, 45, 60, 90, 120, 150 and 180min. (A) Four white rice cultivars, (B) three potato cultivars, and (C) three noodle brands, respectively. Values are expressed as mean±SD. A significant difference was found between values with same lowercase letters at each time point, at p<0.05 by Tukey’s test.

Table 2. AUCs of blood glucose and insulin for 2h and 3h, difference in cultivar in rice, potato and noodle, and difference between AUC-2h and AUC-3h after the ingestion of test foods.

Foods	AUC-glucose			AUC-insulin
	2h (GI)	3h	2–3h⁎(p-value)	2h (II)	3h	2–3h⁎(p-value)
Glucose	100	100	–	100	100	–
Rice
Rice-1	71±25	86±28	<0.05	72±25	84±28	<0.05
Rice-2	69±28	82±34	<0.05	50±20	58±19	<0.05
Rice-3	74±23	82±24	<0.05	71±27	84±40	<0.05
Rice-4	75±14	88±17	<0.05	72±16	91±24	<0.05
Difference in cultivar (p-value)	N.S. (0.95)	N.S. (0.55)	–	N.S. (0.45)	N.S. (0.51)	–
Potato
Potato-1	64±15	65±17	N.S. (0.80)	75±10	79±31	<0.05
Potato-2	63±19	63±19	N.S. (0.27)	97±28	108±29	<0.05
Potato-3	54±17	52±17	<0.05	47±26	47±26	N.S. (0.29)
Difference in cultivar (p-value)	N.S. (0.12)	N.S. (0.28)	–	<0.05	<0.05	–
Noodle
Noodle-1	62±27	80±36	<0.05	22±19	27±20	<0.05
Noodle-2	38±15	49±20	<0.05	22±12	31±19	<0.05
Noodle-3	55±7	67±15	<0.05	46±17	60±29	<0.05
Difference in cultivar (p-value)	<0.05	<0.05	–	<0.05	N.S.(0.08)	–
White bread	58±25	59±15	<0.05	69±24	70±23	<0.05
Sponge cake	64±20	65±19	<0.05	52±26	53±24	<0.05

Fig. 1A shows blood glucose and insulin level curves, when four cultivars of white rice were ingested by 12 subjects. The mean of GI of the four white rice cultivars was 72±5 with little variation, and the significant difference of GI was not observed among four cultivars (Table 2). IIs of white rice-1, -3 and -4 were very similar, and corresponded to the GIs, while II of white rice-2 tended to be smaller than those of the other three white rice cultivars (Table 2). However, the significant difference of II was not observed among four white rice cultivars.

The GIs of the three potato cultivars were 64, 63 and 54, respectively (Table 2), and the range of variation was relatively small. However, the insulin level curves showed greatly different responses among the three potato cultivars (Fig. 1B-2). The range of II between potato-1 and potato-3 was 50, and II was significantly different among the three potato cultivars (p<0.05). These results demonstrate that there are rare foods in which II does not correspond to GI in potato cultivars.

The blood glucose concentrations of the three noodle brands were much lower than that of 50g of glucose until 60min after ingestion (Fig. 1C-1). The 38 of GI of noodle-2 was lower in comparison with those (55 and 62) of two other brands (Table 2), and the range of GI was significantly different among the three noodle brands (p<0.05). The 22 of II scores in two brands (noodle-1 and -2) were less than half of 46 of II of noodle-3. The II scores were significantly different among the three noodle brands (p<0.05).

GI of white bread was slightly smaller than that of sponge cake, but II of white bread was slightly larger than that of sponge cake. These values of GI were roughly similar to those of potato.

3.2. Correlation between GI and II 

Table 3 shows that GI and II of foods tested in this study were extrapolated on the regression equation between GI and II for 43 starchy foods, which the data from several papers were combined by Björck et al. [26]. Potato-2 and noodle-1 stand apart from the regression line (y=0.746858x+22.1164), though 10 of 12 foods existed on or near the regression line. The expected II from the regression equation, y=0.746858x+22.1164, was 69 for 63 of GI score of potato-2, but the measured value of II was 97 actually. The range between the calculated II and the measured II was 29 in potato-2. Also, the calculated value of II was 68 for 62 of GI of noodle-1, while the measured value of II was 22. The range between the calculated II and the measured II was 46 in noodle-1. These results suggest that some foods with an extreme value of II which does not correspond to GI exist rarely in common foods, and we must perform carefully in diet therapy of diabetes mellitus and metabolic syndromes.

Table 3. Glycemic index (GI) and insulinemic index (II) measured directly and calculated by regression equation on test foods.

3.3. Comparison of AUCs of blood glucose and insulin elevation between 2h and 3h after the ingestion of test foods 

The GI is regularly calculated on the basis of the AUC of incremental blood glucose until 2h after the ingestion of test materials versus the ingestion of 50g of glucose or other reference food. However, in order to compare the differences in the AUC between 2h and 3h after ingestion, the AUCs for 2h (AUC-2h-glucose and AUC-2h-insulin) and 3h (AUC-3h-glucose and AUC-3h-insulin) for both incremental blood glucose and insulin were calculated and compared with each other for some foods in the present study.

All AUC-3h-glucose of the four white rice cultivars significantly increased somewhat from their respective AUCs-2h-glucose (p<0.05) (Table 2). AUCs-3h-insulin also increased significantly from AUCs-2h-insulin among the four white rice cultivars. AUC-3h-glucose of potato-3 decreased significantly from AUC-2h-glucose (p<0.05), but those of the two other cultivars were similar. AUCs-3h-insulin was larger than AUCs-2h-insulin in two cultivars, potato-1 and potato-2. Both AUCs-3h-glucose and AUCs-3h-insulin significantly increased from the respective AUCs-2h-glucose and AUCs-2h-insulin in all three brands of noodle (p<0.05). Additionally, not only AUC-2h-glucose and AUC-3h-glucose, but also AUC-2h-insulin and AUC-3h-insulin were very similar for white bread and sponge cake.

These results demonstrate that AUC-3h-glucose and AUC-3h-insulin increase from those for 2h in some foods, though they are similar to their respective AUC-2h-glucose and AUC-2h-insulin in other tested foods.

3.4. Interaction among AUC-8h of breath hydrogen excretion, AUC-3h-glucose and AUC-3h-insulin of test foods 

When starchy foods containing 50g of glucose equivalent were ingested to test blood glucose response, hydrogen excretion was simultaneously measured at 0, 1, 2, 3, 4, 5, 6, 7 and 8h after the ingestion. The AUCs-8h of breath hydrogen excretion (AUCs-8h-H₂) was calculated versus 1 for the ingestion of 50g of glucose.

AUCs-8h-H₂ were relatively small (12–16), but significant, among the four white rice cultivars, and the values of white rice with higher GI were smaller than those (38 and 36) of potato-1 and potato-3 with the lower GI (Table 4). AUCs-8h-H₂ of potato-2 was lower than those of two other cultivars, but not significant. Though the GI scores of noodle-1 and noodle-3 were similar those of the three potato cultivars, the values (20 and 15) of AUCs-8h-H₂ in the noodle were smaller than those (38, 25 and 36) of the three potato cultivars. The Pearson correlation was negative between AUC-8h-H₂ and AUC-2h-glucose, but not significant (r=−0.53). However, the Pearson correlation between AUC-8h-H₂ and AUC-3h-glucose was significantly negative in all test foods (r=−0.81, p<0.01). In contrast, it was not significant between AUC-8h-H₂ and AUC-3h-insulin in any of the tested foods.

Table 4. Difference of AUC-8h-H₂ of after the ingestion of test foods among cultivar in rice, potato, and noodle.

Foods	AUC-8h-H2
Glucose	1.0
Rice
Rice-1	12.3±4.0
Rice-2	10.7±3.2
Rice-3	16.2±5.4
Rice-4	15.7±5.3
Difference in cultivar (p-value)	<0.05
Potato
Potato-1	38.4±13.0
Potato-2	24.9±5.9
Potato-3	36.4±20.1
Difference in cultivar (p-value)	N.S. (0.72)
Noodle
Noodle-1	19.5±7.6
Noodle-2	29.6±9.3
Noodle-3	15.0±7.6
Difference in brand (p-value)	<0.05
White bread	31.4±10.4
Sponge cake	25.9±3.7

Back to Article Outline

4. Discussions 

To examine the validity of the GI, the variations in the GI in response to various inter-cultivars or inter-brands of three kinds of food, we measured the responses of both postprandial blood glucose and insulin, the differences in the AUCs of glucose and insulin for 2h and 3h after the ingestion of test substances, and the relationship between the GI and breath hydrogen excretion time-dependently after the ingestion of several major foods in healthy subjects.

The range of GI and II scores was not significantly different among the four tested cultivars of white rice. But II scores were significantly different among the three potato cultivars (p<0.05). In contrast, both GI and II scores were significantly different among the three noodle cultivars (p<0.05). These results demonstrate that there are some foods that show significant variation in their GI scores depending on the cultivar, and that it is not appropriate to assign a single GI to a given food. Therefore, a separate GI should be determined for every cultivar of certain foods which include starchy foods such as cereals, potatoes, and pulses, and should be offered as information by which to choose suitable foods in diet therapy for diabetes mellitus.

The GI has been found to correspond statistically to the II in many foods [26]. However, the present study strongly suggests that we have to carefully choose foods in diet therapy for diabetes mellitus, because GI and II were greatly different in different varieties of same type food. Besides, when the GI and II obtained in the present study was extrapolated to the regression equation (y=0.746858x+22.1164) between GI and II for 43 starchy foods, by which the data from several papers was combined by Björck et al. [26], those of potato-2 and noodle-1 were outliers. This fact demonstrates that there are some foods where the blood glucose response dose not corresponds to that of insulin. There are many kinds of potato cultivar in the world, and a large number of people eat familiarly them. Therefore, GI and II data of foods which many cultivars have developed, should be accumulated to help the diet therapy of patients with diabetes mellitus.

Furthermore, GI and II of potato-2 were 63 and 97, respectively, while GI and II of noodle-1 was 62 and 22, respectively. The range of difference of II was more than four times, but GI was similar. Hence, these results suggest that the secretion of insulin may bring into [represent] a large difference of more than four times when the selection of a certain food is based on GI alone [8], [9], [10], [11], [12]. The discrepancy, by which II score is greatly different in foods with similar value of GI, causes confusion in the selection of suitable foods by patients with diabetes mellitus, because the diabetic patient finally has to save the secretion of insulin from the pancreas.

The GI is calculated based on AUC until 2h after ingestion in order to control early changes in postprandial blood glucose. The GI corresponded statistically to the II in all foods used in the present study. Nevertheless, the AUC-glucose until 3h or more after ingestion may more accurately reflect the bioavailability of foods. Both AUCs-3h-glucose and AUCs-3h-insulin of the four cultivars of white rice and the three brands of noodle increased significantly from AUCs-2h-glucose and AUCs-2h-insulin, respectively. But AUCs-3h-insulin alone increased from AUCs-2h-insulin in the two potato cultivars, though AUC-glucose did not change for 2h and 3h. Some foods whose AUC-3h-glucose is larger than AUC-2h-glucose appear to be slowly digested and absorbed in the small intestine, while some foods whose AUC-3h-glucose is similar to AUC-2h-glucose appear to be readily digested and absorbed, as well as glucose. It appears to be important in the diet therapy of diabetes mellitus that there are some starchy foods whose AUC-3h-glucose or AUC-3h-insulin increases from AUC-2h-glucose or AUC-2h-insulin. So, we need to transmit detailed information to diet therapy patients. The reason why the response of insulin is different between 2h and 3h after the ingestion is unknown, in spite of GI scores being similar in some foods.

The values of AUC-8h-H₂ were relatively similar in inter-cultivars of the same food type except for a few cultivars, but these differed between the foods used in the present study. They also varied from subject to subject in the ingestion of potato. This was caused by the contents of resistant starch or chemical constituents of starch in potato. The digestibility of carbohydrate or the movement in the gastrointestinal tract may differ from food to food. The Pearson correlation was negative between AUC-8h-H₂ and AUC-3h-glucose, but not AUC-2h-glucose, in all tested foods (r=−0.81, p<0.01), though breath hydrogen excretion differed greatly among the tested foods in the present study. The reason appears to concern that the response of insulin did not correspond to that of glucose in foods such as potato-2, noodle-1 and white rice-2. The results obtained may provide superior information for selecting food for diabetes mellitus patients who wish to control their postprandial blood glucose and insulin.

The excretion of breath hydrogen demonstrates that the carbohydrates ingested are transferred to the lower intestine and fermented by intestinal microbes. As a result, intestinal microflora is improved by the fermentation of carbohydrates and available energy, which is provided when the food ingested decreases. Because the starchy foods in which breath hydrogen excretion is numerous do not drastically increase blood glucose and insulin levels, they can be selected for diet therapy in diabetes mellitus. Therefore, the data accumulation of breath hydrogen excretion of starch foods appears to help in the selection of foods. In addition, the low GI foods may cause prebiotic effects as well as non-digestible oligosaccharides. AUCs-8h-H₂ of potato-1, potato-3, white bread and noodle-2 was high, and those of the four cultivars of white rice were low in the present study.

In conclusion, GI and II varied greatly among the different cultivars or brands of some foods in the present study, although they are similar in the cultivars or brands of many other foods. Although the GI for most starchy foods commonly showed a positive correlation to II, there are some foods where GI does not correspond to II. Although the amount of breath hydrogen excretion varied from cultivar to cultivar in some foods, it showed a negative correlation to AUC-3h-glucose of the tested foods. These results indicate that other information such as II, breath hydrogen excretion, AUC-3h-glucose and AUC-3-insulin added to GI scores will help the selection of suitable foods in diet therapy for type-2 diabetes mellitus.

Back to Article Outline

Acknowledgements 

The authors would like to thank the subjects who participated in this study. We also wish to thank the Nagasaki Agriculture and Forestry Experiment Station for providing some test substances. This study was supported in part by University funding for research to T.O.

Back to Article Outline

References 

1. Clement S, Braithwaite SS, Magee MF, Ahmann A, Smith EP, Schafer RG, et al. Management of diabetes and hyperglycemia in hospitals. Diab Care. 2004;27:553–597
   • View In Article
2. Riccardi G, Rivellese AA, Giacco R. Role of glycemic index and glycemic load in the healthy state, in prediabetes, and in diabetes. Am J Clin Nutr. 2008;87:269S–274S
   • View In Article
3. Vrolix R, van Meijl LE, Mensink RP. The metabolic syndrome in relation with the glycemic index and the glycemic load. Physiol Behav. 2008;94:293–299
   • View In Article
   • CrossRef
4. Bergenstal RM, Johnson M, Powers M, Wynne A, Vlajnic A, Hollander P. Adjust to target in type 2 diabetes: comparison of a simple algorithm with carbohydrate counting for adjustment of mealtime insulin glulisine. Diab Care. 2008;31:1305–1313
   • View In Article
5. Foster-Powell K, Holt SH, Brand-Miller JC. International table of glycemic index and glycemic load values. Am J Clin Nutr. 2002;76:5–56
   • View In Article
   • MEDLINE
6. Monro JA, Shaw M. Glycemic impact, glycemic glucose equivalents, glycemic index, and glycemic load: definitions, distinctions, and implications. Am J Clin Nutr. 2008;87:237S–243S
   • View In Article
7. Vander Weele DA. Insulin as a satiationg signal. In:  Smith GP editors. Satiation from gut to brain. New York: Oxford University Press; 1998;p. 198–216
   • View In Article
8. Brand-Miller J, Holt SH, de Jong V, Petocz P. Cocoa powder increases postprandial insulinemia in lean young adults. J Nutr. 2003;133:3149–3152
   • View In Article
   • MEDLINE
9. Gentilcore D, Chaikomin R, Jones KL, Russo A, Feinle-Bisset C, Wishart JM, et al. Effects of fat on gastric emptying of and the glycemic, insulin, and incretin responses to a carbohydrate meal in type 2 diabetes. J Clin Endocrinol Metab. 2006;91:2062–2067
   • View In Article
   • MEDLINE
   • CrossRef
10. Nilsson M, Stenberg M, Frid AH, Holst JJ, Björck IM. Glycemia and insulinemia in healthy subjects after lactose-equivalent meals of milk and other food proteins: the role of plasma amino acids and incretins. Am J Clin Nutr. 2004;80:1246–1253
    • View In Article
    • MEDLINE
11. Östman EM, Liljeberg EHG, Björck IM. Inconsistency between glycemic and insulinemic responses to regular and fermented milk products. Am J Clin Nutr. 2001;74:96–100
    • View In Article
    • MEDLINE
12. Shah M, Adams-Huet B, Brinkley L, Grundy SM, Garg A. Lipid, glycemic, and insulin responses to meals rich in saturated, cis-monounsaturated, and polyunsaturated (n−3 and n−6) fatty acids in subjects with type 2 diabetes. Diab Care. 2007;30:2993–2999
    • View In Article
13. Jenkins DJ, Wolever TM, Taylor RH, Barker H, Fielden H, Baldwin JM, et al. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr. 1981;34:362–366
    • View In Article
    • MEDLINE
14. Wolever TM, Jenkins DJ, Jenkins AL, Josse RG. The glycemic index: methodology and clinical implications. Am J Clin Nutr. 1991;54:846–854
    • View In Article
    • MEDLINE
15. Jenkins DJ, Kendall CW, Augustin LS, Franceschi S, Hamidi M, Marchie A, et al. Glycemic index: overview of implications in health and disease. Am J Clin Nutr. 2002;76:266S–273S
    • View In Article
    • MEDLINE
16. Nilsson A, Granfeldt Y, Ostman E, Preston T, Björck I. Effects of GI and content of indigestible carbohydrates of cereal-based evening meals on glucose tolerance at a subsequent standardised breakfast. Eur J Clin Nutr. 2006;60:1092–1099
    • View In Article
    • MEDLINE
    • CrossRef
17. Jenkins DJ, Gassull MA, Leeds AR, Metz G, Dilawari JB, Slavin B, et al. Effect of dietary fiber on complications of gastric surgery: prevention of postprandial hypoglycemia by pectin. Gastroenterology. 1977;73:215–217
    • View In Article
    • Abstract
18. Mudra M, Ercan-Fang N, Zhong L, Furne J, Levitt M. Influence of mulberry leaf extract on the blood glucose and breath hydrogen response to ingestion of 75g sucrose by type 2 diabetic and control subjects. Diab Care. 2007;30:1272–1274
    • View In Article
19. Oku T, Nakamura S. Digestion, absorption, fermentation, and metabolism of functional sugar substitutes and their available energy. Pure Appl Chem. 2002;74:1253–1261
    • View In Article
    • CrossRef
20. Levitt MD. Production and excretion of hydrogen gas in man. N Engl J Med. 1969;281:122–127
    • View In Article
    • MEDLINE
    • CrossRef
21. Oku T, Nakamura S. Comparison of digestibility and breath hydrogen gas excretion of fructo-oligosaccharide, galactosyl-sucrose, and isomalto-oligosaccharide in healthy human subjects. Eur J Clin Nutr. 2003;57:1150–1156
    • View In Article
    • MEDLINE
    • CrossRef
22. Oku T, Nakamura M, Takasugi A, Hashiguchi-Ishiguro M, Kennichi K, Nakamura S. Effects of cake made from whole soy powder on postprandial blood glucose and insulin levels in humans. Int J Food Sci Nutr. 2009;60(S4):224–231
    • View In Article
    • CrossRef
23. Trinder P. Determination of blood glucose using an oxidase-peroxidase system with a non-carcinogenic chromogen. J Clin Pathol. 1969;22:158–161
    • View In Article
    • MEDLINE
    • CrossRef
24. Liversy JH, Hodgkinson SC, Round HR, Donald RA. Effect of time, temperature and freezing on the stability of immunoreactive LH, FSH, TSH, growth hormone, prolactin and insulin in plasma. Clin Biochem. 1980;3:151–155
    • View In Article
    • MEDLINE
    • CrossRef
25. Brouns F, Bjorck I, Frayn KN, Gibbs AI, Lang V, Slama G, et al. Glycaemic index methodology. Nutr Res Rev. 2005;18:145–171
    • View In Article
    • CrossRef
26. Björck IM, Liljeberg EHG, Östman EM. Low glycaemic-index foods. Br J Nutr. 2000;83:S149–S155
    • View In Article