| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research Communication |
1 From the Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, and the Clinical Nutrition and Risk Factor Modification Centre, St Michael’s Hospital, Toronto.
2 Supported by the Canadian Diabetes Association and the International Olive Oil Council. 3 Reprints not available. Address correspondence to TMS Wolever, Department of Nutritional Sciences, University of Toronto, Toronto, Ontario, Canada M5S 3E2. E-mail: thomas.wolever{at}utoronto.ca.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: We studied whether these 2 dietary maneuvers have the same long-term effects on postprandial plasma glucose, insulin, triacylglycerol, and free fatty acid (FFA) concentrations in subjects with impaired glucose tolerance (IGT).
Design: Thirty-four subjects with IGT were randomly assigned to high-carbohydrate, high-GI (high-GI); high-carbohydrate, low-GI (low-GI); and low-carbohydrate, high–monounsaturated fatty acid (MUFA) diets for 4 mo. Plasma glucose, insulin, and FFAs were measured from 0800 to 1600 at baseline in response to high-GI meals (60% carbohydrate, GI = 61, GL = 63) and after 4 mo in response to meals representative of the study diet.
Results: Carbohydrate intake (% of energy), GI, and GL in the high-GI, low-GI, and MUFA groups (breakfast and lunch meals combined), respectively, were 60%, 61, and 63; 60%, 53, and 55; and 49%, 61, and 52. Compared with the change after 4 mo of the high-GI diet, both the low-GI and MUFA diets reduced 0–8-h mean plasma glucose concentrations by 0.35 mmol/L (P < 0.05). Mean plasma insulin was 
20% higher (P < 0.05) and FFAs 12% lower (P < 0.05) after the low-GI diet than after the high-GI diet, with no significant effect of MUFA. Changes in 0–8-h mean plasma triacylglycerols in the 3 treatment groups differed significantly: –0.14, 0.04, and 0.18 mmol/L, respectively, with the high-GI, MUFA, and low-GI diets.
Conclusions: In subjects with IGT, reducing the GI of the diet for 4 mo reduced postprandial plasma glucose by the same amount as did reducing carbohydrate intake. The 2 dietary maneuvers had different effects on postprandial plasma insulin, triacylglycerols, and FFAs.
Key Words: Glycemic index • glycemic load • randomized clinical trial • postprandial plasma glucose • postprandial plasma insulin • postprandial plasma fatty acids
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The glycemic load (GL) has been proposed as a measure of the overall blood glucose- and insulin-raising potential of the diet. A low GL is associated with a reduced risk of developing diabetes (11,12) and coronary heart disease (13). The GL is derived by multiplying the amount of carbohydrate consumed in the diet by its GI. Therefore, the GL can be reduced either by reducing carbohydrate intake or by reducing the dietary GI. Although both of these maneuvers reduce acute plasma glucose and insulin responses (14,15), they have different effects on postprandial plasma free fatty acids (FFAs). Replacing carbohydrate in a meal with an isoenergetic amount of monounsaturated fatty acids (MUFAs) increases acute postprandial FFA concentrations in both healthy (16) and diabetic (17) subjects. On the other hand, FFAs are reduced by lowering the meal GI with no change in carbohydrate content (16). Long-term reductions in dietary carbohydrate raise postprandial FFA concentrations by > 30% in subjects with type 2 diabetes (18). A high plasma FFA concentration is associated with an increased risk of developing diabetes (19), possibly via reduced insulin secretion (20,21) and reduced insulin action (22), factors involved in the pathogenesis of type 2 diabetes (23,24). High plasma FFA concentrations are also associated with dyslipidemia (25), hypertension (26), and increased risk of cardiovascular disease (27). This suggests that the long-term implications of reducing carbohydrate intake may not be the same as those of reducing the dietary GI, even though both of these maneuvers reduce the GL. Thus, the aim of this study was to determine the long-term effects of altering the source and amount of dietary carbohydrate on postprandial plasma glucose, insulin, and FFAs in subjects with impaired glucose tolerance (IGT), individuals who are at increased risk of cardiovascular disease (4) and diabetes (28).
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
7.8 mmol/L and < 11.1 mmol/L. Diagnostic criteria for diabetes changed after recruitment started (30). Of the 44 subjects identified as having IGT, 37 (84%) had normal fasting plasma glucose (< 6.1 mmol/L), 6 had impaired fasting glucose (6.1–6.9 mmol/L), and one, with fasting plasma glucose of 7.0 mmol/L, would now be considered to have diabetes. After the baseline data were collected, the subjects were randomly assigned to 1 of 3 diets for 4 mo: a high-carbohydrate, high-GI (high-GI) diet; a high-carbohydrate, low-GI (low-GI) diet; or a low-carbohydrate, high-MUFA (MUFA) diet. During the baseline period, two 3-d food records were used as a basis for individualized dietary advice. Diets were prescribed on an ad libitum basis. The aim was for the diets to be weight maintaining, with the high-carbohydrate diets containing 55% of energy from carbohydrate and 30% from fat and the MUFA diet containing 45% carbohydrate and 40% fat, of which one-half was MUFAs. Subjects assigned to the high- or low-GI diet, respectively, were asked to have at least one serving of a high- or low-GI food at each meal. Lists of high- or low-GI foods were provided to the subjects. In addition, the subjects in each group were given specific key foods to be consumed. Subjects in the high-GI group were given breakfast cereals, polished rice, instant potato, breads, and crackers. Subjects in the low-GI group were given breakfast cereals, bread, pasta, barley, parboiled rice, legumes, and instant soups. Subjects in the MUFA group were given olive oil and margarine made from unhydrogenated canola oil. Subjects were seen monthly for the collection of fasting blood samples, weight measurement, and consultation with the dietitian and to hand in the 3-d diet records and pick up study foods.
Nutrient intakes were calculated from the food records with an in-house program that used a database based on the US Department of Agriculture food-composition tables, with missing values for fiber and values for GI added; the GI of the diet was calculated as previously described (31,32). Carbohydrate is expressed as available carbohydrate (ie, total carbohydrate minus total dietary fiber) and GI is expressed with the GI of glucose = 100. The GL was calculated by multiplying the GI of the diet by the grams of available carbohydrate intake and adjusting for energy intake (12).
At baseline and during the last week of the study, the subjects came to the nutrition center after fasting overnight for an 8-h metabolic profile. After giving a fasting blood sample, the subjects ate a breakfast test meal and gave further blood samples 0.5, 1, 1.5, 2, 3, 4, and 5 h after starting to eat. Immediately after the 5-h sample, lunch was consumed, and further blood samples were obtained at 5.5, 6, 7, and 8 h. At baseline, all subjects ate the same breakfast and lunch meals, which represented a high-carbohydrate, high-GI diet. At the end of the study, the breakfast and lunch meals represented the composition of the subject’s study diet.
Blood samples were used for analysis of plasma glucose (model 2300 STAT glucose analyzer; Yellow Springs Instruments, Yellow Springs, OH), insulin (Insulin RIA; Pharmacia, Dorval, Quebec, Canada), FFAs (NEFA C, ACS-ACOD method; WAKO Chemicals USA, Richmond, VA), and triacylglycerols [Triacylglycerol (GPO-Trinder); Sigma Diagnostics, St Louis]. Fasting total cholesterol and triacylglycerol were measured enzymatically by using a Vitros Analyser 950 (Johnson & Johnson Clinical Diagnostics, Rochester, NY), with HDL cholesterol measured after precipitation of other lipoproteins with dextran sulfate and magnesium chloride. LDL cholesterol was calculated as total cholesterol - (HDL cholesterol + triacylglycerols/2.2) (only for triacylglycerol < 4.51 mmol/L). Glycated hemoglobin (Hb A1c) was measured by Diamat HPLC [Bio-Rad Laboratories (Canada) Ltd, Mississauga, Ontario, Canada].
Of 44 eligible subjects, 7 declined participation and 37 were randomly assigned by coin toss, with stratification for age, sex, and body mass index, to receive 1 of the 3 test diets. Two subjects in the high-GI group and 1 in the MUFA group dropped out before the study ended. One subject participated in 2 arms of the study (MUFA and high-GI diets). For comparison with baseline data from subjects with IGT, 8 lean and 7 obese but otherwise healthy control subjects underwent the metabolic profile.
The number of subjects studied was based on the power to detect the effects of the diets on insulin sensitivity (SI ) and insulin secretion measured by the minimal model method, the results of which are published elsewhere (33). We had previously found, by using the insulin suppression test, that acarbose significantly improved insulin sensitivity by 17% in a parallel design study involving only 10 placebo- and 8 acarbose-treated subjects with IGT (34). In healthy subjects, the CV of the intraindividual variation in SI was reported to be 14.4% (35). With this CV and 12 subjects in each group, there is 80% power to detect a difference in SI of 17% (STATMATE version 1.01; Graph Pad software, San Diego).
Data are presented as means ± SEMs. Differences in the plasma concentrations of glucose, insulin, FFAs, and triacylglycerols measured at baseline and 4 mo were examined by analysis of variance (ANOVA) for the main effects of diet and time of day (ie, 0–8 h) and the diet x time interaction. Significant main effects were confirmed after adjusting for baseline values by using the residuals from the linear regression of difference on baseline value (36). Differences between the lean, obese, and IGT subjects were assessed by ANOVA for the main effects of subject group and time of day and the diet x time interaction. Tukey’s test was used for post hoc comparisons of individual means to adjust for multiple comparisons. Incremental areas under the plasma glucose and insulin response curves (AUCs) after breakfast and lunch, ignoring the area beneath the premeal value, were calculated geometrically (10). The AUC after breakfast was calculated from the plasma glucose and insulin values from 0 to 5 h, and the AUC after lunch was calculated by using the 5–8-h values. The significance of differences between the AUC at baseline and that at 4 mo was determined by use of paired t tests with Bonferroni’s adjustment (n = 3 treatments) for multiple comparisons (37). All statistical tests were two-tailed with P 
0.05 taken as being significant.
All procedures were reviewed and approved by the St Michael’s Hospital Research Ethics Board, and all subjects gave informed consent.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and Figure 1). Subjects with IGT had higher mean 0–8-h plasma FFA concentrations than did both control groups, with the difference between the lean and obese control subjects being significant only at 6 h (Figure 1). Subjects with IGT had slightly but significantly higher 0–8-h mean triacylglycerol concentrations than did the obese control subjects, who, in turn, had significantly higher concentrations than did the lean control subjects (Figure 1). There were no significant differences in fasting metabolic variables at baseline between the subjects with IGT randomly assigned to the different dietary treatments. However, there were small but significant differences in 0–8-h mean glucose, insulin, FFA, and triacylglycerol concentrations (Table 2).
|
|
|
Composition of the diet on the metabolic profile days
The metabolic profile diet (combined breakfast and lunch) fed at baseline contained 59.5 ± 0.4% of energy as carbohydrate, 27.8 ± 0.4% of energy as fat, and 12.7 ± 0.1% of energy as protein with a diet GI of 61.1 ± 0.1 and a GL of 88.7 ± 0.7. At the end of the study, the composition of the metabolic profile diet for the high-GI group was not significantly different from that at baseline. In the low-GI group, the overall GI of the diet was reduced to 53.1 ± 0.4, with no significant change in the amounts of protein, fat, or carbohydrate. In the MUFA group, fat intake was increased by 11.8 ± 1.2% of energy, with a similar reduction in carbohydrate intake (-11.2 ± 1.2%) and no significant change in the GI of the diet (-0.2 ± 0.2). The GL of the metabolic profile diet was reduced by 13.0 ± 1.5% with the low-GI diet and by 18.6 ± 2.8% with the MUFA diet, with no significant change with the high-GI diet (-0.06 ± 0.58%). Representative menus for breakfast and lunch meals during the metabolic profile are shown in Table 3, and the composition of the metabolic profile breakfast and lunch meals actually consumed is shown in Table 4.
|
|
). Hb A1c increased significantly with the MUFA diet relative to the low-GI and high-GI diets (Table 5). The changes in fasting plasma glucose, insulin, and FFAs (Table 5) and in serum total and LDL cholesterol and triacylglycerols (data not shown) in the 3 treatment groups did not differ significantly from each other. The change in serum HDL cholesterol with the high-GI diet, 0.09 ± 0.04 mmol/L, differed significantly from that with the low-GI diet, -0.01 ± 0.03 mmol/L, with the change with the MUFA diet, 0.05 ± 0.03 mmol/L, being intermediate.
|
. There were significant main effects of diet and time of day for the difference in plasma glucose from baseline to 4 mo but no significant diet x time interaction. The effect of diet remained after adjustment for baseline values (P = 0.002). Mean plasma glucose decreased from baseline by 0.28 ± 0.10 mmol/L (3.8%) in the low-GI group (P = 0.056) and by 0.28 ± 0.10 mmol/L (4.0%) in the MUFA group (P = 0.0502), with an 0.07 ± 0.10-mmol/L (1%) increase in the high-GI group (NS). Relative to the change in the high-GI group, 0–8-h mean plasma glucose decreased significantly in both the low-GI and MUFA groups by 0.35 ± 0.10 mmol/L (P < 0.05; Figure 3).
|
|
There was a significant main effect of diet for the difference in plasma insulin concentration from baseline to 4 mo but no significant effect of time of day and no significant diet x time interaction (Figure 2
). The effect of diet remained after adjustment for baseline concentrations (P < 0.001). Mean 0–8-h plasma insulin decreased significantly from baseline by 35.2 ± 8.7 pmol/L (18%) in the high-GI (P = 0.006) and by 28.3 ± 8.4 (12%) in the MUFA group (P = 0.019), with a nonsignificant decrease in the low-GI group of 3.6 ± 11.9 pmol/L (2%; Figure 3). The change in mean plasma insulin in the low-GI group differed significantly from that in the high-GI group, but the change in the MUFA group did not differ significantly from that in the low- or high-GI groups (Figure 3). When an ANOVA was performed on the data for the individual time points, the difference in plasma insulin from baseline to 4 mo was significantly greater for the low-GI than for the high-GI group at 6 h (P < 0.05).
At 4 mo, the plasma insulin AUC after breakfast tended to be lower than at baseline in all 3 treatment groups, but the changes did not differ significantly from the respective baseline values nor from each other. The insulin AUC after lunch was 88 ± 14 pmol · min/L (24%; P < 0.001) less than at baseline in the high-GI group and 56 ± 17 pmol · min/L (15%; P = 0.021) less than at baseline in the MUFA group. However, in the low-GI group, the insulin AUC after lunch tended to be higher than at baseline by 19 ± 32 pmol · min/L (NS), and this change was significantly different from those in the high-GI and MUFA groups.
There were significant main effects of diet and time of day for the difference in plasma FFAs from baseline to 4 mo but no diet x time interaction (Figure 2). Mean 0–8-h plasma FFA concentrations fell significantly from baseline in the high-GI group by 0.043 ± 0.014 mEq/L (13%; P = 0.032), in the low-GI group by 0.094 ± 0.019 mEq/L (25%; P < 0.001), and in the MUFA group by 0.057 ± 0.018 mEq/L (17%; P = 0.027). The reduction in FFAs in the low-GI group was significantly greater than that in the high-GI group, whereas the change in the MUFA group did not differ significantly from that in either of the other treatment groups (Figure 3). After adjustment for baseline values, the difference in FFAs between the low-GI and high-GI groups became nonsignificant (P = 0.18). When an ANOVA was performed on the data for the individual time points, the difference from baseline to 4 mo in plasma FFAs at 6 h was significantly greater in the MUFA group than in the other 2 treatment groups.
There was a significant main effect of diet for the difference in plasma triacylglycerol from baseline to 4 mo, which remained after adjustment for the baseline value (P < 0.001). There was no significant effect of time of day and no diet x time interaction. Mean 0–8-h triacylglycerol increased significantly from baseline in the low-GI group by 0.18 ± 0.06 mmol/L (10%; P = 0.037), and this change was significantly different from the 0.04 ± 0.03-mmol/L (3%) increase in the MUFA group, which, in turn, differed significantly from the 0.14 ± 0.06-mmol/L (7%) reduction in the high-GI group (Figure 3). The changes in the MUFA and high-GI groups did not differ significantly from their respective baseline values.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Similar reductions in the GL with the low-GI and MUFA treatments, 13% and 16%, respectively, were associated with an identical reduction in mean plasma glucose, 0.35 mmol/L, which represents 4.7% of the baseline value. The change in glucose is smaller than what was achieved in shorter and more tightly controlled studies (3,39), but the dietary changes were also much smaller than those in other studies. The changes in the diet GL were modest, by intention, because we wished to study the effect of realistic dietary changes that could be adhered to by free-living subjects over a relatively long period of time. It could be argued that although the 0.35-mmol/L reduction in plasma glucose was statistically significant, it is too small to be biologically significant. However, cardiovascular disease risk increases continuously as blood glucose increases within the normal range (40). In addition, the results of animal studies have shown that increases in mean blood glucose < 1 mmol/L can markedly affect ß cell function via glucose toxicity (41,42).
The percentage reductions in the incremental blood glucose AUC after the low-GI and MUFA breakfast (21% and 17%, respectively) and lunch (17% and 13%) meals were of similar or greater magnitude than the percentage reductions in the GL (25%, 18%, 19%, and 2%). This does not support the American Diabetes Association position that "With regard to the glycemic effect of carbohydrates, the total amount of carbohydrate...is more important than the source or type" (43). Indeed, our results suggest that, in subjects with IGT consuming normal mixed meals, reducing the GL by altering carbohydrate source has the same effect on postprandial blood glucose as the same reduction in GL achieved by reducing carbohydrate intake.
There was no significant difference in Hb A1c between the low-GI and high-GI diets, despite a significant difference in mean plasma glucose. In addition, Hb A1c was significantly higher with the MUFA diet than with the high-GI diet, despite significantly lower mean plasma glucose. Studies of low-carbohydrate diets in subjects with diabetes have been unable to detect significant changes in glycated serum proteins (Hb A1c or fructosamine) despite substantial reductions in postprandial glucose (39,44,45). The lack of change in Hb A1c has been suggested to be due to insensitivity of the measurement (39) or the short time of the study (46). However, the present study was long enough to detect a change in Hb A1c, and it is difficult to explain a significantly higher Hb A1c in the face of significantly lower mean plasma glucose. Possible explanations are that postprandial plasma glucose concentrations measured under controlled conditions are not representative of plasma glucose concentrations during the rest of the study period under free-living situations. Alternatively, diet may influence the glycation of proteins by mechanisms independent of changes in plasma glucose, such as the rate of protein turnover or antioxidant status (47). The lack of concordance between changes in blood glucose and Hb A1c in this and other studies raises the question as to which is the more appropriate measure of long-term glycemic control.
The increase in plasma insulin with the low-GI diet and the lack of effect of the MUFA diet, relative to the change with the high-GI diet, were unexpected. However, relative to baseline, plasma insulin was maintained with the low-GI diet and fell significantly with the MUFA diet. Thus, the effects were due to a significant decrease in plasma insulin with the high-GI diet. This is not consistent with the results of short-term studies showing that foods that elicit high glycemic responses also elicit high insulin responses regardless of whether the portions fed contain equal amounts of carbohydrate (48) or equal amounts of energy (49). Our results are also not consistent with previous studies showing that plasma insulin is lower 2–6 wk after the consumption of a 40%-carbohydrate diet than after a 55–60%-carbohydrate diet in insulin-resistant women (3) or subjects with type 2 diabetes (39). However, our study lasted 4 mo, and it is possible that the reduction in plasma insulin with a high-GI diet represents the natural history of deteriorating ß cell function in persons with IGT (50). There is debate as to whether glucose toxicity or raised FFAs are responsible for this deterioration in ß cell function (51). Our results suggest that both factors may be important. The low-GI diet, which was associated with significantly improved ß cell function (33), was also associated with significant reductions in both plasma glucose and FFAs. However, the MUFA diet, which had no effect on ß cell function (33), was associated only with a reduction in plasma glucose.
Plasma FFAs were transiently higher after lunch with the MUFA diet, as expected, but the lack of effect on 0–8-h mean FFAs was unexpected. Lower postprandial insulin may lead to reduced suppression of FFA mobilization from adipose tissue. In addition, increased fat intake raises plasma FFAs directly because a proportion of the FFAs released from chylomicrons via the activity of lipoprotein lipase remain in the circulation (52). We previously showed that a 10% reduction in carbohydrate intake increased mean plasma FFAs by 30% in subjects with type 2 diabetes (18). The difference between the studies may be because subjects with IGT have less impaired fatty acid metabolism than do subjects with diabetes (53).
The low-carbohydrate diet did not reduce postprandial serum triacylglycerols, as was reported in shorter studies of postmenopausal women and subjects with type 2 diabetes (3,39). Indeed, postprandial triacylglycerol was 11% higher after 4 mo of the MUFA diet than after the high-GI diet. Because postprandial triacylglycerol was not measured at an intermediate time point, we do not know whether this represents a long-term adaptation to the diets or not. However, classic early studies in South African prisoners showed that the rise in serum triacylglycerols associated with a high-carbohydrate diet lasts for 3–6 mo (54).
The increase in triacylglycerols with the low-GI diet is consistent with the effect we observed of feeding a high-fiber cereal to subjects with type 2 diabetes (18), and may be related to increased acetate availability from colonic fermentation stimulated by the increased fiber intake (55). This could be seen as deleterious because high plasma triacylglycerols increase cardiovascular disease risk (56). However, increased triacylglycerol concentrations are usually associated with insulin resistance and raised plasma FFAs (52,53), and these factors may at least contribute to cardiovascular disease risk independently of serum triacylglycerols (6,27). The rise in postprandial triacylglycerols seen with the low-GI diet was not due to increased insulin resistance and was associated with reduced plasma FFAs and glucose. This suggests that the rise in triacylglycerols with the low-GI diet may not be associated with the same degree of cardiovascular disease risk associated with a similar rise in triacylglycerols due to insulin resistance.
Weight changes in this study were small; however, significantly more weight was lost with the high-GI diet than with either the low-GI or MUFA diet. This does not support the popular notion that reducing the GL, either by reducing the GI or by reducing carbohydrate intake, results in weight loss. In addition, changes in body weight are not associated with and therefore cannot explain the changes in glucose, insulin, FFAs, or triacylglycerols we observed.
We conclude, therefore, that in subjects with IGT, reducing the GL of the diet for 4 mo significantly reduces postprandial plasma glucose concentrations. The same reduction in plasma glucose was achieved by reducing the GI of the diet and by reducing carbohydrate intake. However, these 2 dietary maneuvers had different effects on postprandial plasma insulin, triacylglycerols, and FFAs.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Junyent, L. D Parnell, C.-Q. Lai, Y.-C. Lee, C. E Smith, D. K Arnett, M. Y Tsai, E. K Kabagambe, R. J Straka, M. Province, et al. Novel variants at KCTD10, MVK, and MMAB genes interact with dietary carbohydrates to modulate HDL-cholesterol concentrations in the Genetics of Lipid Lowering Drugs and Diet Network Study Am. J. Clinical Nutrition, September 1, 2009; 90(3): 686 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J de Souza, J. F Swain, L. J Appel, and F. M Sacks Alternatives for macronutrient intake and chronic disease: a comparison of the OmniHeart diets with popular diets and with dietary recommendations Am. J. Clinical Nutrition, July 1, 2008; 88(1): 1 - 11. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Lajous, M.-C. Boutron-Ruault, A. Fabre, F. Clavel-Chapelon, and I. Romieu Carbohydrate intake, glycemic index, glycemic load, and risk of postmenopausal breast cancer in a prospective study of French women Am. J. Clinical Nutrition, May 1, 2008; 87(5): 1384 - 1391. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. W Barclay, P. Petocz, J. McMillan-Price, V. M Flood, T. Prvan, P. Mitchell, and J. C Brand-Miller Glycemic index, glycemic load, and chronic disease risk--a meta-analysis of observational studies Am. J. Clinical Nutrition, March 1, 2008; 87(3): 627 - 637. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Valtuena, N. Pellegrini, D. Ardigo, D. Del Rio, F. Numeroso, F. Scazzina, L. Monti, I. Zavaroni, and F. Brighenti Dietary glycemic index and liver steatosis Am. J. Clinical Nutrition, July 1, 2006; 84(1): 136 - 142. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Blouet, F. Mariotti, D. Azzout-Marniche, C. Bos, V. Mathe, D. Tome, and J.-F. Huneau The Reduced Energy Intake of Rats Fed a High-Protein Low-Carbohydrate Diet Explains the Lower Fat Deposition, but Macronutrient Substitution Accounts for the Improved Glycemic Control J. Nutr., July 1, 2006; 136(7): 1849 - 1854. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Ma, Y. Li, D. E. Chiriboga, B. C. Olendzki, J. R. Hebert, W. Li, K. Leung, A. R. Hafner, and I. S. Ockene Association between Carbohydrate Intake and Serum Lipids. J. Am. Coll. Nutr., April 1, 2006; 25(2): 155 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. R Sahyoun, A. L Anderson, A. M Kanaya, P. Koh-Banerjee, S. B Kritchevsky, N. de Rekeneire, F. A Tylavsky, A. V Schwartz, J. S. Lee, and T. B Harris Dietary glycemic index and load, measures of glucose metabolism, and body fat distribution in older adults Am. J. Clinical Nutrition, September 1, 2005; 82(3): 547 - 552. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Hodge, D. R. English, K. O'Dea, and G. G. Giles Glycemic Index and Dietary Fiber and the Risk of Type 2 Diabetes Diabetes Care, November 1, 2004; 27(11): 2701 - 2706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Harbis, S. Perdreau, S. Vincent-Baudry, M. Charbonnier, M.-C. Bernard, D. Raccah, M. Senft, A.-M. Lorec, C. Defoort, H. Portugal, et al. Glycemic and insulinemic meal responses modulate postprandial hepatic and intestinal lipoprotein accumulation in obese, insulin-resistant subjects Am. J. Clinical Nutrition, October 1, 2004; 80(4): 896 - 902. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. F. Sheard, N. G. Clark, J. C. Brand-Miller, M. J. Franz, F. X. Pi-Sunyer, E. Mayer-Davis, K. Kulkarni, and P. Geil Dietary Carbohydrate (Amount and Type) in the Prevention and Management of Diabetes: A statement by the American Diabetes Association Diabetes Care, September 1, 2004; 27(9): 2266 - 2271. [Full Text] [PDF]
|
|
|
|
|
|
J. C Brand-Miller Postprandial glycemia, glycemic index, and the prevention of type 2 diabetes Am. J. Clinical Nutrition, August 1, 2004; 80(2): 243 - 244. [Full Text] [PDF]
|
|
|
|
|
|
M. B Schulze, S. Liu, E. B Rimm, J. E Manson, W. C Willett, and F. B Hu Glycemic index, glycemic load, and dietary fiber intake and incidence of type 2 diabetes in younger and middle-aged women Am. J. Clinical Nutrition, August 1, 2004; 80(2): 348 - 356. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. O. Scholl, X. Chen, C. S. Khoo, and C. Lenders The Dietary Glycemic Index during Pregnancy: Influence on Infant Birth Weight, Fetal Growth, and Biomarkers of Carbohydrate Metabolism Am. J. Epidemiol., March 1, 2004; 159(5): 467 - 474. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Ludwig Glycemic Load Comes of Age J. Nutr., September 1, 2003; 133(9): 2695 - 2696. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), and 8 lean control subjects (
). Breakfast was consumed at 0 h and lunch at 5 h. Error bars are not shown if they are smaller than the symbol or overlap other error bars. Means with different letters are significantly different, P < 0.05.
