| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research Communication |
1 From the Department of Pediatrics, Harvard Medical School, and the Department of Medicine, Children's Hospital, Boston (MAP); the Division of Epidemiology (DRJ and MDG), the General Clinical Research Center (SKR), and the Departments of Food Science and Nutrition (JLS), Medicine (ERS), and Family Practice (JJP), University of Minnesota, Minneapolis; and the Institute of Nutrition Research, University of Oslo (DRJ).
2 Supported in part by grant MO1-RR00400 from the National Center for Research Resources; grant T32 HL07779 from the National Heart, Lung, and Blood Institute; a research award from the Division of Epidemiology, University of Minnesota; and a gift from General Mills, Inc, Golden Valley, MN.
3 MA Pereira, Children's Hospital, Division of Endocrinology, 333 Longwood Avenue, 6th floor, Boston, MA 02115. E-mail: mark.pereira{at}channing.harvard.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: We tested the hypothesis that whole-grain consumption improves insulin sensitivity in overweight and obese adults.
Design: This controlled experiment compared insulin sensitivity between diets (55% carbohydrate, 30% fat) including 6–10 servings/d of breakfast cereal, bread, rice, pasta, muffins, cookies, and snacks of either whole or refined grains. Total energy needs were estimated to maintain body weight. Eleven overweight or obese [body mass index (in kg/m2): 27–36] hyperinsulinemic adults aged 25–56 y participated in a randomized crossover design. At the end of each 6-wk diet period, the subjects consumed 355 mL (12 oz) of a liquid mixed meal, and blood samples were taken over 2 h. The next day a euglycemic hyperinsulinemic clamp test was administered.
Results: Fasting insulin was 10% lower during consumption of the whole-grain than during consumption of the refined-grain diet (mean difference: -15 ± 5.5 pmol/L; P = 0.03). After the whole-grain diet, the area under the 2-h insulin curve tended to be lower (-8832 pmol·min/L; 95% CI: -18720, 1062) than after the refined-grain diet. The rate of glucose infusion during the final 30 min of the clamp test was higher after the whole-grain diet (0.07 x 10-4 mmol·kg-1·min-1 per pmol/L; 95% CI: 0.003 x 10-4, 0.144 x 10-4).
Conclusion: Insulin sensitivity may be an important mechanism whereby whole-grain foods reduce the risk of type 2 diabetes and heart disease.
Key Words: Carbohydrate • diet • whole grains • nutrition • insulin • hyperinsulinemia • type 2 diabetes • insulin response
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
25% of energy consumption in the United States (1); however, an estimated 95% of grain available for human consumption is refined (S Gerrior, personal communication, 1997; 2). During the refining process grains are stripped of their bran and germ, thereby depleting many biologically active nutrients and constituents, including fiber, antioxidants, minerals, and phytoestrogens. When consumed close to their natural form, with the bran and germ present, either intact or pulverized, in breakfast cereals and bread products or as brown rice and other whole grains, their nutrients and other constituents may act synergistically to lower the risk of chronic diseases (3,4). Although most epidemiologic studies combine grains with other carbohydrates and therefore have not differentiated whole- from refined-grain foods, the observation of an inverse association between fiber from grains (cereal fiber) or whole-grain foods and the risk of developing ischemic heart disease is consistent (2,5–13), relatively strong (20–50% reduction in risk), independent of other lifestyle factors and body weight, and biologically plausible. A few studies observed inverse associations between cereal fiber or whole-grain foods and the risk of type 2 diabetes in men and women (14–17). However, the possibility of residual bias in the epidemiologic data cannot be ruled out. To make the inference that whole-grain intake causally reduces risk, elucidation of biological mechanisms is needed. Given that insulin resistance increases the risk of type 2 diabetes and cardiovascular disease (18–20), insulin sensitivity may be one important mechanism through which whole-grain consumption confers protection. Studies have reported fasting insulin, a good measure of insulin resistance in epidemiologic studies (21), to be lower in individuals reporting higher dietary fiber intakes, after adjustment for other lifestyle and dietary factors (22–25). Although ingestion of whole- and refined-grain flour results in a similar acute increase in blood glucose (high glycemic index), certain minimally processed whole-grain foods—such as oats, barley, bran-based cereals, and bulgar—have a more favorable, moderate glycemic index (26), owing to factors such as larger particle size, high ratios of bran or germ to endosperm, presence of viscous soluble fibers, and resistant starch content. We previously hypothesized that the consumption of whole-grain foods correlates inversely with fasting insulin concentrations, independent of other potential confounding factors. We supported this hypothesis in an observational study of young black and white adults (27).
The purpose of the current study was to test the hypothesis that a healthful diet that includes whole-grain rather than refined-grain foods improves insulin sensitivity independent of body weight change in overweight hyperinsulinemic individuals.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subject recruitment
The subjects were recruited from the University of Minnesota and its surrounding community. Responders were screened via a telephone interview according to the following inclusion criteria: 1) age between 21 and 65 y; 2) body mass index (in kg/m2) 26–36; 3) body weight fluctuation over the past 6 mo of <10%; 4) not currently smoking cigarettes; 5) consuming 
2 alcoholic beverages/d; 6) free of diabetes, cancer, cardiovascular disease, and other chronic clinical conditions; 7) not taking medications that would affect glucose, insulin, lipids, or blood pressure; 8) not engaging in a high level of physical activity; 9) not following a special diet (eg, vegetarian); 10) not allergic to any foods; and 11) not planning to change dietary habits, increase physical activity, change body weight, move out of town, or take a lengthy vacation during the time of the study. Those satisfying these criteria were invited to visit the GCRC for measurement of height, weight, blood pressure, and fasting glucose and insulin. They were invited to participate in the study if their fasting (
12 h) blood glucose was normal (<6.1 mmol/L) and their fasting insulin was elevated above the 75th percentile of the distribution on the basis of epidemiologic studies (90 pmol/L). Six men and 6 women were enrolled in the study. One man did not complete the study because of an illness (unrelated to diet) that developed during the washout period before the second diet. The recruitment and study procedures were approved by the Institutional Review Board of the Human Subjects Committee at the University of Minnesota. All subjects read and signed consent forms before enrollment in the study, and they were remunerated on completion of the study. See Table 1
for subject characteristics.
|
80% were wheat and the remainder oats, rice, corn, barley, and rye. The diets were fed at an energy level to maintain body weight according to the Harris-Benedict prediction equation adjusted for activity level (28). A sample 1-d menu for the refined-grain and whole-grain diets is shown in Table 2.
|
. The lower total availability of energy as carbohydrate, owing to the lower ratio of starch to fiber in whole grains than in refined grains, was estimated to result in the provision of 250 kJ/d less by the whole-grain than the refined-grain diet. The macronutrient contents of the 2 diets were similar, whereas the total fiber content of the whole-grain diet was 1.2 g/MJ higher than that of the refined-grain diet. Of the micronutrients, magnesium and vitamin E were somewhat higher in the whole-grain than in the refined-grain diet.
|
12 h. The subjects were asked to avoid strenuous exercise for 24 h before each clinic visit. Height to the nearest 0.5 cm and weight to the nearest 0.1 kg were measured with a stadiometer and a digital scale, respectively, with the subjects dressed in light clothes and no shoes. With the subjects seated quietly, blood was drawn from an antecubital vein into evacuated tubes containing a gel serum separator. At both 6-wk clinic visits, the subjects were administered a nongrain liquid mixed meal (Ensure; Abbott Laboratories, Abbott Park, IL) that provided 50% of energy as carbohydrate, 30% as fat, and 20% as protein. The subjects consumed 355 mL within 5 min after the fasting blood draw, and a butterfly catheter was used to draw a 5-mL blood sample at 15, 30, 60, 90, and 120 min after consumption. The subjects reclined for the duration of the test. The area under the 2-h insulin curve after the liquid mixed meal was computed with the use of the trapezoidal rule for uneven time intervals (30).
On the morning after each 6-wk clinic visit, a 180-min euglycemic hyperinsulinemic clamp test (31) was administered. The subjects were infused with insulin at a rate of 6.0 or 12.0 pmol·kg-1·min-1. Euglycemia was maintained by the infusion of 20% dextrose. Samples for glucose (analyzed immediately in whole blood) were taken before, at baseline, and thereafter every 5 min during the clamp. Samples for serum insulin were taken before, at baseline, and every 20 min during the clamp and, as described below, were processed and stored for later analysis. Because of technical difficulties, clamps were not completed for 1 man and 1 woman. Insulin sensitivity was computed as the average rate of glucose infusion over the final 30 min (M value = mmol glucose·kg body wt-1·min-1) divided by the average serum insulin concentration (pmol/L) over the final 40 min of the test.
Sample processing and laboratory procedures
Within 30 min of phlebotomy, whole-blood samples were centrifuged at 2800 x g for 10 min at 5°C, and 0.5- to 1.0-mL samples of serum or plasma were pipetted into polyethylene cryovials. Samples were stored at -70°C for 7 mo. Laboratory results at all time points for all subjects were performed in batch within the same assay. Serum insulin was measured at the Diabetes Institute for Immunology and Transplantation, University of Minnesota, with a radioimmunoassay specific to human insulin and with <0.2% cross-reactivity with proinsulin and a within-assay CV of 3% (Linco Research, St Louis). Serum glucose was measured at the University Hospital General Chemistry Laboratory with the use of a thin-film adaptation of a glucose oxidase enzymatic, spectrophotometric procedure with a Vitros analyzer (Johnson & Johnson Clinical Diagnostics, Inc, Rochester, NY) that has a within-assay CV of 1.4%.
Questionnaire
The Health Habits and History Questionnaire (32) was administered before the baseline examination to assess the subjects' usual dietary intake of major food groups and nutrients. A daily log—listing nonprotocol food eaten and protocol food not eaten, unusual symptoms, medication use, satiety, bowel movements, and physical activity—was administered every evening during each treatment period when the subjects arrived at the clinic to eat dinner. The subjects responded on a 5-point Likert scale (with 1 being not hungry at all and 5 being extremely hungry) to the question, "How hungry were you between meals over the past 24 h?," and to a questionnaire about side effects (with 1 being no symptoms and 5 being severe symptoms) for 44 items, including gastrointestinal symptoms and general symptoms of acute or chronic illnesses.
Statistical analysis
The dependent variables were satiety, body weight, fasting insulin, the homeostasis model for insulin resistance, the M value from the clamp, and the area under the 2-h curve after the mixed meal. The dependent variable for satiety was the average of each of the three 2-wk intervals, with no baseline satiety score available. The homeostasis model for insulin resistance was computed according to Matthews et al (33), fasting glucose (mmol/L) x fasting insulin (pmol/L)/22.5, as an additional way of estimating insulin sensitivity. SAS software version 6.12 was used for all statistical analyses (34). We first used the PROC MIXED program to perform repeated-measures regression of the baseline value of body weight on period, finding that body weight was 85.4 in period 1 and 86.5 in period 2 (SE for difference: 0.57). Similarly, baseline fasting insulin was 155 in period 1 and 160 in period 2 (SE for difference: 11.5), and the homeostasis model for insulin resistance was 6.1 in period 1 and 6.6 in period 2 (SE for difference: 0.60). Given the within-person variability in starting values before the 2 diet periods, we adjusted the follow-up values by subtracting their period-specific baseline values. We next evaluated treatment differences at baseline, computing similar regressions of baseline body weight, fasting insulin, and the homeostasis model for insulin resistance with treatment (whole-grain compared with refined-grain diet) and period (1 compared with 2).
For the main analyses, we performed repeated-measures regression with the use of follow-up values of several dependent variables to examine the effects of treatment (whole-grain compared with refined-grain diet), time (wk), and treatment x time interaction, adjusting for the baseline value of the dependent variable, period (1 compared with 2), and cohort [the first group of 4 subjects (participated August through mid-December) compared with the second group of 7 subjects (participated October– November and January–February)]. Additional models included period x treatment interaction to assess whether the crossover design assumption was violated. The period x time interaction was not significant in any model; the main effect of period was included in all final models. Sex and cohort had no bearing on the results and were therefore not included in the final models. In the baseline analysis, SEs are computed within person, after removing the overall mean difference between the baseline values of the 2 diet periods. Findings during the active diet periods are presented as the mean (±SE) treatment difference (computed from variation within person) observed over the baseline-adjusted values for the 3 follow-up time points (weeks 2, 4, and 6). The M value from the clamp and the area under the 2-h insulin curve after the mixed meal were only measured at the 6-wk time point. For these measures the mean differences between the 2 treatments and their 95% CIs and P values were computed. A P value <0.05 was considered statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Treatment compliance
The investigators or research dietitians at the GCRC often joined the participants for dinner to encourage study compliance and good relationships among the participants and research team. As reported on the daily logs and as analyzed by the NDS System, compliance with the diets was very good. The energy content of food not eaten during the two 6-wk diets did not differ significantly between the whole-grain (159 ± 75 kJ/d) and refined-grain (222 ± 126 kJ/d) diets. The same was true of extra (nonprotocol) food from sources other than the treatment diets (whole-grain diet: 46 ± 25 kJ/d; refined-grain diet: 63 ± 33 kJ/d).
Body weight and satiety
Body weight was not significantly different during the follow-up periods with the whole-grain (84.8 ± 0.29 kg) compared with the refined-grain (85.1 ± 0.29 kg) diet (Figure 1). The diets were not hypoenergenic; therefore, subjects were not very hungry between meals. However, in comparison with the refined-grain diet, there was a tendency for the subjects to be less hungry between meals with the whole-grain diet (P = 0.08; data not shown). Self-reported physical activity was similar during the 2 diets (data not shown).
|
). Adjustment for the change in body weight (mean difference in insulin concentration: -13 ± 5.5 pmol/L; P = 0.02) and physical activity (mean difference in insulin concentration: -14.0 ± 5.5 pmol/L; P = 0.01) did not explain this significant difference in fasting insulin between the 2 diets. There was a tendency for fasting glucose to be lower with the whole-grain diet (5.2 ± 0.08 mmol/L) than with the refined-grain diet (5.3 ± 0.08 mmol/L), although the difference was not significant. The homeostasis model also showed that insulin resistance was lower with the whole-grain (5.4 ± 0.18 U) than with the refined-grain (6.2 ± 0.18 U) diet (Figure 3).
|
|
).
|
Side effects
Unusual symptoms were reported more often on the daily questionnaire during the refined-grain than during the whole-grain diet period, and the difference appeared to be primarily due to common upper respiratory illnesses in 2 individuals and to chronic low-grade abdominal pain in another subject. Therefore, the incidence of prescription medication use was higher with the refined-grain than with the whole-grain diet. Dry cough (1.4 ± 0.13 compared with 1.1 ± 0.13) and sweating (1.3 ± 0.13 compared with 1.0 ± 0.13) tended to be scored significantly higher on the Likert scale with the refined-grain diet. Bowel movements were more frequent with the whole-grain than with the refined-grain diet (1.8 ± 0.17 compared with 1.4 ± 0.17 movements/d; P < 0.001). No significant differences between treatments were apparent for heartburn, indigestion, diarrhea, or loose stools.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Diet composition has a major influence on insulin secretion (36,37). Although the postprandial rise in serum insulin is greater after carbohydrate ingestion than it is after ingestion of fat or protein, there is a broad range in the glycemic and insulinemic responses to ingested carbohydrate, depending on such factors as the source, structure, composition, processing, and preparation of the foods (38–40). Fiber may attenuate the glycemic response to ingested carbohydrate through its physical action in the gut, where it tends to slow the absorption of nutrients (41,42). Reduced serum glucose concentrations decrease the amount of insulin needed to clear the glucose load; over time, the reduced ambient insulin concentrations may result in an up-regulation of cell surface insulin receptors, thereby increasing insulin sensitivity (43). Although most whole-grain flours have glycemic indexes that are similar to those of refined-grain flours (26), because of their small particle size, we previously observed that habitual consumers of primarily whole-grain flour products had lower fasting insulin concentrations than did habitual consumers of refined-grain flour (27). The high concentration of fiber and indigestible carbohydrate in many whole-grain foods may be fermented by indigenous bacteria in the large intestine, thereby producing short-chain fatty acids that may enter the portal circulation (3,4,44,45). There is evidence that hepatocytes may, when exposed to an increase in short-chain fatty acids, increase glucose oxidation, decrease fatty acid release, and increase insulin clearance—an environment conducive to enhanced insulin sensitivity (44,46,47). A whole-grain, high-fiber diet could therefore enhance insulin receptor sensitivity through a chronic lowering of the overall dietary glycemic index and related insulin secretion (40,48–50) as well as through short-chain fatty acid production, leading to enhanced hepatic glucose oxidation and insulin clearance (43,44,46).
Other suspected mechanisms that might explain the effect of whole grains on insulin sensitivity include the metabolic effects of certain micronutrients, such as magnesium and vitamin E (24,51–55). Our previous observation of an inverse dose-response association between self-reported whole-grain consumption and fasting insulin was found to be partially explained by body mass index, dietary magnesium, and dietary fiber, although no single mediator accounted for all of the association (27). In addition to these nutrients, whole grains contain a plethora of other nutrients and constituents (eg, antioxidants, phenolic compounds, and phytoestrogens), which have potential independent or synergistic biological effects (3,4) that need to be elucidated in future studies.
Fukagawa et al (56) reported significant decreases in fasting insulin and glucose and increases in the rate of glucose disposal in healthy adults after a high-carbohydrate, high-fiber diet. However, because 60% of the fiber-rich foods in this diet came from nongrain foods, it was not possible to determine how much of the effect was due to whole grains. A supplement study performed by Keenan et al (50) in a clinical outpatient population of subjects with hypertension, impaired fasting glucose, or both showed beneficial effects on postload insulin and blood pressure with whole-oat cereal but not with a refined rice cereal. Chandalia et al (57) found significant beneficial effects on glucose and insulin in a diabetic population who increased their dietary fiber intake to 25 g soluble and 25 g insoluble fiber daily. In this population, reductions were observed in mean daily blood glucose (0.7 mmol/L) and urinary glucose (0.007 mmol) and in 24-h glucose (10%) and insulin (12%) areas under the curve. However, in the current study, most of the fiber in the whole-grain foods was insoluble, suggesting that soluble fiber may not be necessary to improve insulin sensitivity and raising the possibility that other nutrients and compounds in whole-grain foods may have important biological effects.
Although there was a suggestion of higher satiety between meals with the whole-grain diet than with the refined-grain diet, the subjects rarely reported feeling hungry with either of these euenergetic diets, and our analysis of the daily logs showed no indication of differential compliance. Note that the data in these logs were self-reported and may have masked true differences. Studies have generally provided support for an inverse association between whole grains and body weight (5,13,27,58,59). In the Coronary Artery Risk Development in Young Adults Study, we found inverse associations between dietary fiber and 10-y body weight gain in young black and white adults (22). The difference in 10-y weight gain was 3.6 kg (8 lb) for low compared with high quintiles of dietary fiber intake, and it was independent of dietary fat, total energy intake, physical activity, and many other possibly confounding factors. Although this study did not include sources of fiber in its analysis, whole-grain intake made an important contribution to dietary fiber (60).
Taken together, the findings from the current study are consistent with, and supportive of, epidemiologic evidence showing inverse associations between consumption of cereal fiber or whole grains and type 2 diabetes (14–17), cardiovascular disease (2,5,6,8–13,61), and total mortality (2,5). Insulin resistance may be a common antecedent to each of these chronic diseases, because it is known to markedly increase the risk of type 2 diabetes (62,63) and may be the cornerstone of a metabolic syndrome involving impaired fasting glucose, obesity, hypertension, dyslipidemia, and hypofibrinolysis (64).
In conclusion, whole-grain foods may have favorable effects on insulin sensitivity over a period of 6 wk in overweight and obese adults. These effects may reduce the risk of type 2 diabetes and ischemic heart disease. Although larger and longer controlled trials are needed to confirm these findings and elucidate the mechanisms involved, the epidemiologic and clinical evidence is sufficient to encourage increased consumption of whole-grain foods.
·Vmax
(·Vmax
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. S. Bassuk and J. E. Manson Lifestyle and Risk of Cardiovascular Disease and Type 2 Diabetes in Women: A Review of the Epidemiologic Evidence American Journal of Lifestyle Medicine, June 1, 2008; 2(3): 191 - 213. [Abstract] [PDF]
|
|
|
|
 |
|
 M. O. Weickert and A. F. H. Pfeiffer Metabolic Effects of Dietary Fiber Consumption and Prevention of Diabetes J. Nutr., March 1, 2008; 138(3): 439 - 442. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. Andon and J. W. Anderson State of the Art Reviews: The Oatmeal-Cholesterol Connection: 10 Years Later American Journal of Lifestyle Medicine, February 1, 2008; 2(1): 51 - 57. [Abstract] [PDF]
|
|
|
|
 |
|
 P. Newby, J. Maras, P. Bakun, D. Muller, L. Ferrucci, and K. L Tucker Intake of whole grains, refined grains, and cereal fiber measured with 7-d diet records and associations with risk factors for chronic disease Am. J. Clinical Nutrition, December 1, 2007; 86(6): 1745 - 1753. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. N Davis, K. E Alexander, E. E Ventura, L. A Kelly, C. J Lane, C. E Byrd-Williams, C. M Toledo-Corral, C. K Roberts, D. Spruijt-Metz, M. J Weigensberg, et al. Associations of dietary sugar and glycemic index with adiposity and insulin dynamics in overweight Latino youth Am. J. Clinical Nutrition, November 1, 2007; 86(5): 1331 - 1338. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. T. Fung, M. McCullough, R. M. van Dam, and F. B. Hu A Prospective Study of Overall Diet Quality and Risk of Type 2 Diabetes in Women Diabetes Care, July 1, 2007; 30(7): 1753 - 1757. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Andersson, S. Tengblad, B. Karlstrom, A. Kamal-Eldin, R. Landberg, S. Basu, P. Aman, and B. Vessby Whole-Grain Foods Do Not Affect Insulin Sensitivity or Markers of Lipid Peroxidation and Inflammation in Healthy, Moderately Overweight Subjects J. Nutr., June 1, 2007; 137(6): 1401 - 1407. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Rumawas, N. M. McKeown, G. Rogers, J. B. Meigs, P. W.F. Wilson, and P. F. Jacques Magnesium Intake Is Related to Improved Insulin Homeostasis in the Framingham Offspring Cohort J. Am. Coll. Nutr., December 1, 2006; 25(6): 486 - 492. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. van Dam, F. B. Hu, L. Rosenberg, S. Krishnan, and J. R. Palmer Dietary Calcium and Magnesium, Major Food Sources, and Risk of Type 2 Diabetes in U.S. Black Women Diabetes Care, October 1, 2006; 29(10): 2238 - 2243. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Qi, R. M. van Dam, S. Liu, M. Franz, C. Mantzoros, and F. B. Hu Whole-Grain, Bran, and Cereal Fiber Intakes and Markers of Systemic Inflammation in Diabetic Women Diabetes Care, February 1, 2006; 29(2): 207 - 211. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. R Sahyoun, P. F Jacques, X. L Zhang, W. Juan, and N. M McKeown Whole-grain intake is inversely associated with the metabolic syndrome and mortality in older adults Am. J. Clinical Nutrition, January 1, 2006; 83(1): 124 - 131. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Lairon, N. Arnault, S. Bertrais, R. Planells, E. Clero, S. Hercberg, and M.-C. Boutron-Ruault Dietary fiber intake and risk factors for cardiovascular disease in French adults Am. J. Clinical Nutrition, December 1, 2005; 82(6): 1185 - 1194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. N Davis, E. E Ventura, M. J Weigensberg, G. D. Ball, M. L Cruz, G. Q Shaibi, and M. I Goran The relation of sugar intake to {beta} cell function in overweight Latino children Am. J. Clinical Nutrition, November 1, 2005; 82(5): 1004 - 1010. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Esmaillzadeh, P. Mirmiran, and F. Azizi Whole-grain intake and the prevalence of hypertriglyceridemic waist phenotype in Tehranian adults Am. J. Clinical Nutrition, January 1, 2005; 81(1): 55 - 63. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W Anderson Whole grains and coronary heart disease: the whole kernel of truth Am. J. Clinical Nutrition, December 1, 2004; 80(6): 1459 - 1460. [Full Text] [PDF]
|
|
|
|
|
) or whole-grain (
) diet. There were no significant differences between treatment groups by repeated-measures regression.
