![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
,4
* Department of Food Science and Nutrition, Colleges of Human Ecology and Agricultural, Food, and Environmental Sciences, St. Paul, MN 55108;
Division of Biostatistics, School of Public Health, Minneapolis, MN 55414;
** Minneapolis VA Medical Center, Minneapolis, MN 55417, and Department of Medicine, Minneapolis, MN 55414; and
Department of Family Practice and Community Health, Medical School, University of Minnesota, Minneapolis, MN 55414
4To whom correspondence should be addressed. E-mail: keena001{at}umn.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • dietary fiber • viscous fiber • cholecystokinin • cellulose • glucose • humans
The ability of certain fibers to improve glucose tolerance and insulin metabolism has long been an area of interest for diabetes mellitus, obesity and cardiovascular disease prevention. One reason for this interest is the relationship of insulin to Syndrome X and lipid metabolism (1,2).
The effect of fiber on postprandial responses elicited by single test meals with or without a fiber has been studied most frequently. Earlier studies indicated that the amount of total fiber was inversely related to the magnitude of the insulin response evoked by the fiber-containing meal (3,4). Later studies, comparing the effects of fibers from different sources in both healthy subjects (5–7) and individuals with type 2 diabetes (6,8), reported lower postprandial plasma glucose and insulin concentrations after a fiber-containing meal or food, than after an oral glucose load or a very low fiber meal. In addition to acknowledging the importance of fiber viscosity, some of the investigators noted an inverse linear relationship between the amount of mixed linkage ß-glucans present in a meal and the glucose area under the curve (AUC) response, as well as the magnitude of the plasma glucose peak (7,8). However, several recent studies have not confirmed the significant effect of viscous fibers on fasting and postprandial concentrations of glucose and insulin (9,10), leaving the question of the beneficial effects of fiber on glucose tolerance still unanswered. Fewer studies have tested the longer-term effects of dietary fiber on glucose and insulin metabolism, and the evidence is even more ambiguous. Furthermore, the practical utility of viscous fibers as hypoglycemic agents is often limited by the gastrointestinal side effects associated with increased consumption and related to their fermentability.
Cholecystokinin (CCK) is a peptide hormone and a neurotransmitter rapidly secreted by cells in the upper part of the small intestine after the ingestion of food. Amino acids, peptones and fatty acids appear to be the most potent stimulators of CCK secretion, whereas it is only weakly stimulated by glucose via a mechanism that is poorly understood. CCK regulates gut motility, gall bladder contraction and pancreatic enzyme secretion. Additionally, it potentiates the amino acid–induced release of insulin and glucagon (11) and slows gastric emptying, thereby enhancing digestion and absorption of nutrients (12) and regulating glucose homeostasis (13). Observations such as these have led some to hypothesize that CCK might mediate the postprandial glycemic and insulinemic responses to viscous fibers (9). Additionally, studies have pointed to a role of CCK in inducing postprandial satiety and the possibility that viscous fibers mediate this mechanism (14,15). However very few studies have evaluated the relationship between fiber and CCK release, especially after a long-term fiber intervention, and their findings have been largely inconsistent.
The purpose of the present study was to evaluate the acute and long-term effects of a viscous insoluble fiber (solubilized cellulose, SC) on the postprandial glucose and CCK responses of hypercholesterolemic men and women to mixed liquid test meals. Solubilized cellulose was chosen because it is nonfermentable, highly viscous, and inexpensive and is derived from a widely abundant fiber. Because no other studies of SC have been conducted, the choice of study design and intervention duration was based on existing trials testing the hypoglycemic effect of viscous fibers.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
Healthy men and women with moderate hypercholesterolemia between the ages of 20 and 75 y were recruited through radio, campus newspaper and electronic newsletter advertisements as well as flyers around the University of Minnesota campus. Interested individuals completed a telephone questionnaire. All were questioned regarding unstable vascular disease, use of hypolipidemic agents within the last 30 d or regularly, diabetes mellitus, uncontrolled hypothyroidism, history of active peptic ulcers, liver or gall bladder disease, major intestinal surgery, malabsorption, or stenosis of the gastrointestinal tract, chronic use of antacids or bulk laxatives. Additional exclusion criteria included fasting triacylglycerol concentration > 4.52 mmol/L (400 mg/dL), excessive alcohol use, BMI > 35 kg/m2 and soluble fiber intake > 6 g/d as estimated by a baseline dietary questionnaire. Women who were pregnant, planned to become pregnant or received an unstable dose of hormone replacement therapy were also excluded from the study.
More than eighty individuals met the inclusion criteria and attended an hour-long diet class, led by a registered dietitian, introducing the American Heart Association (AHA) Step I diet and providing educational materials that explained appropriate diet choices based on the diet (16). Subjects followed the diet for a minimum of six weeks before intervention. Dietary counseling was available throughout the study. At the end of this 6-wk dietary adaptation period, a baseline lipid screen was performed to confirm that enrolled participants still met the lipid inclusion criteria; subjects with total cholesterol (TC) > 5.68 mmol/L or LDL cholesterol > 3.36 mmol/L were enrolled in the study. Thirty-four subjects were eligible to participate. Seven of the subjects were hypertensive and had been receiving treatment for at least 6 mo before randomization. Results of a routine biochemical panel were normal for all participants. The study protocol specified that subjects maintain their baseline body weight and exercise habits, as well as comply with the Step I diet throughout the study. The trial was approved by the University of Minnesota Institutional Review Board Human Subjects Committee and all subjects provided written, informed consent after the study procedures had been fully explained to them. Subjects were financially compensated for their time. Randomization to the SC group or the Control group occurred after all subjects had undergone the Acute postprandial study and was accomplished using a fixed allocation randomization scheme with permuted blocks of four and six.
Test meals and experimental treatment.
The AHA Step I Diet (16), which all subjects consumed during the dietary adaptation period and throughout the study, provides <30% of energy as fat, <10% of energy as saturated fatty acids, 55% of energy as carbohydrate, 15% of energy as protein and <300 mg cholesterol daily.
The dietary fiber under investigation was produced at the University of Minnesota from cellulose derived from agricultural waste materials, namely, corn stalks and husks (Bert-Mar Farms, Osseo, WI) (17). The physical modification necessary to solubilize the cellulose and thus increase its viscosity involved a recently improved patented manufacturing process (18,19). The SC gel used was 74.5% cellulose, 23% lignin and 1.5% hemicellulose, as determined by GC. The highly viscous hydrocolloidal SC gel was cohomogenized with the liquid meals and fruit-flavored drinks at a desired concentration, using standard equipment to produce the modified test meal and research beverages.
The liquid test meals (502 g) administered on visits 1 and 8 (FF-B and FF-E) consisted of a chocolate-flavored shake providing 2.09 MJ, 80 g carbohydrate, 12 g fat, 18 g protein and 50% of the recommended daily intake for 24 vitamins and minerals (Ensure, Ross Product Division, Abbott Laboratories, Columbus, OH). The test meal consumed on visit 2 (SC-B) was identical to the FF-B and FF-E meals but contained 5 g of solubilized cellulose and had a slightly larger volume to account for the addition of the fiber gel (529 g). Because it had a more viscous consistency than the unmodified test meal, it was referred to as "the pudding meal" to facilitate subjects’ lack of awareness of treatment.
During the 6-wk intervention period, subjects in the SC group received weekly supplies of a low energy fruit-flavored beverage, providing 2.5 g fiber/240 mL along with 209 kJ, 14 g carbohydrate, 0 g fat and protein, 110 mg sodium and 30 mg potassium (Fierce Melon Gatorade, the Quaker Oats Company, Chicago, IL). Subjects in the Control group received the same amount of beverage not containing the fiber. The beverages were packaged in identical opaque plastic cups with snap-on lids. All subjects were instructed to drink one 240-mL cup of the beverage twice daily before, with or after their two largest meals for 6 wk. Subjects recorded the time and amount of beverage consumed on daily calendars throughout the study. Compliance was assessed by review of these calendars at each follow-up visit and by counting unopened cups at the end of the study. Subjects remained unaware of the treatments received as judged by an informal interview at the end of the study.
Study and analytical procedures.
Subjects fasted for 12–16 h before all three meal-testing periods. An intravenous saline lock catheter was placed in the antecubital vein and a baseline (-5 min) blood sample was drawn for determination of insulin, glucose and CCK concentrations. The subjects then had 2 min to consume the test meal. A blood sample was drawn immediately at the completion of the drink (0 min) and additional samples were drawn at 10, 20, 30, 45, 60, 90, 120 and 180 min for CCK. Samples for measurement of plasma glucose concentration were drawn at 30, 60, 90, 120 and 180 min after the consumption of the meal. During the testing period, subjects refrained from food intake but had free access to water.
Blood for glucose analysis was collected in lithium heparin-coated Vacutainer tubes (Becton Dickinson, Rutherford, NJ), centrifuged at 4°C (1800 x g, 10 min) and refrigerated at 2°C until assayed. Plasma glucose was determined in the Acute Care Laboratory at Fairview-University Medical Center, Minneapolis, MN using the glucose oxidase colorimetric method and a Vitros 950 Analyzer (Johnson and Johnson Clinical Diagnostics, Rochester, NY). Blood for insulin analysis was collected in serum Vacutainer tubes (Becton Dickinson), promptly centrifuged at 4°C (1800 x g, 10 min); the serum was divided into aliquots and stored at -20°C until assayed. Insulin was measured with a competitive double antibody RIA specific to human insulin and with <0.2% cross-reactivity with proinsulin (LINCO Research, St. Charles, MO). For CCK analysis, blood was collected in chilled EDTA-coated Vacutainer tubes (Becton Dickinson) to which 0.385 TIU (500 KIU, Kallikrein Inactivation units) of aprotinin [9 g/L (150 mmol/L) NaCl and 9 g/L benzyl alcohol solution] (Sigma-Aldrich, St. Louis, MO) per milliliter of whole blood had been added with a disposable 1-mL syringe. Samples were kept on ice during collection, promptly centrifuged at 4°C (1800 x g, 10 min), divided into aliquots, flash frozen on dry ice and stored at -70°C until assayed. Plasma samples for CCK determination underwent extraction with 950 g/L ethanol and were evaporated using a Centrivap concentrator (Labconco, Kansas City, MO) (20). Extraction recovery was controlled for during each extraction and was estimated to be 64.2%. Analysis was performed using a highly sensitive and specific competitive double antibody RIA with cross-reactivity to gastrin < 0.5%, using a 125I-labeled CCK-8 (Alpco Diagnostics, Windham, NH) (20). Radioactivity was counted in a Cobra II Auto-Gamma counter, model 5010 for 3 min (Packard Instruments, Downers Grove, IL). To reduce the effects of interassay variability, all samples from each subject were analyzed in duplicate in the same assay, along with standardized controls. Intra-assay variability based on the counts was 1.3% and interassay variability was 5.8 and 4.8% for controls with low (
2.0 pmol/L) and high (9.7 pmol/L) CCK concentration, respectively. CCK values below the lowest standard of 0.195 pmol/L were arbitrarily assigned a value of 0.1 pmol/L before all sample concentrations were corrected for extraction recovery.
Weekly body weight measurements without shoes were taken on an electronic scale (Scale-Tronix model 5005 Stand-On-Scale). Height was measured in the beginning of the study for verification of the subjects’ BMI using a wall-mounted stadiometer.
Participants were asked not to change their habitual levels of physical activity during the study. They all completed a lifestyle questionnaire before and after intervention, answering questions about smoking habits and rating their exercise frequency on a 4-point scale with 1 being "never," 2 = "rarely," 3 = "once or twice weekly" and 4 = "three or more times per week." The selected number was used to calculate a mean exercise frequency score for each group at baseline and study end. Additionally, the percentage of subjects who exercised at least once weekly (exercise status) was calculated for within- and between-group comparisons. Adverse events related to the test product, medication and overall health changes were solicited by open-ended questions at each weekly clinic visit during the 6-wk intervention period. Moreover, subjects completed a side effect questionnaire at baseline and study end, rating the occurrence and frequency of eleven gastrointestinal symptoms on a hierarchical 5-point scale from 0 (none) to 4 (extremely), with 2 being "moderately." A side effect score could be calculated from the questionnaire as a multiple of the number and frequency of side effects and could be between 0 and 44.
To assess changes in dietary intake, three 3-d food records (2 weekdays and 1 weekend day) were collected from all subjects at baseline (visit 2), at the end of wk 3 (visit 5) and after the last week of intervention (visit 8). A registered dietitian reviewed all food records for completeness. All records were analyzed using the Nutrition Data System for Research, version 4.02, database 30 (Nutrition Coordinating Center, University of Minnesota, Minneapolis, MN). For each record, the mean intake of total energy, the percentage of energy from total, saturated, monounsaturated and polyunsaturated fat, carbohydrate, protein and alcohol; the intake of total and soluble fiber, cholesterol, total trans fatty acids, the ratio of polyunsaturated to saturated fat, as well as the intake of sodium, potassium, calcium and magnesium were calculated. The mean of all of these variables for each of the three 3-d food records was determined for each individual and used in subsequent statistical analyses. The 5 g of fiber that each of the subjects in the SC group consumed daily in the research beverages was not entered as part of subjects’ dietary intake due to the lack of an appropriate fiber substitute and a desire to monitor only "background" diet.
Statistical analyses.
The primary outcome measure was change in the glucose and CCK AUC calculated by the trapezoid rule (21). Peak glucose and peak CCK concentrations were selected as the first highest postprandial values during a testing period, immediately preceding a lower concentration. For these endpoints, within-subject changes were compared by paired t test; comparisons between treatment groups and genders were by two-sample t tests. Time to peak (minutes from time 0 to time of highest concentration in plasma, i.e., peak) medians for glucose and insulin were compared within subjects by Wilcoxon signed-rank test, and were compared between groups by the Wilcoxon two-sample test. A 
2 test was used to compare categorical variables at baseline.
All data are presented as means ± SD, unless otherwise stated and the probability level at which differences were considered significant was P < 0.05; P < 0.2 was considered a trend. All analyses were performed using SAS 6.12 (SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
There were no differences in the baseline characteristics of the individuals participating in the initial acute study and those who underwent the longer-term study (Table 1). Data about menopausal status or menstrual cycle distribution of the female participants were not collected. Of the participants, 91% were Caucasian and 15.6% were students. Body weight did not change during the study. Additionally, participants in both groups maintained their exercise status and frequency, as well as smoking habits.
|
|
Glucose and CCK endpoints in response to the FF-B and SC-B meals are shown in Table 3. The glucose AUC was not different between the meals, although it tended to be lower after the SC meal (P = 0.22). Both meals increased CCK from very low fasting concentrations of 0.5 pmol/L to a mean peak of 11.1 pmol/L after the FF-B meal and 6.2 pmol/L after the SC-B meal. These CCK peaks were significantly different from each other (P = 0.01). The CCK AUC tended to be lower after the SC-B meal (P = 0.08) (Fig. 2). The BMI at baseline was inversely correlated with the change (SC-B - FF-B) in the CCK peak concentration (r = -0.391, P = 0.02).
|
|
|
|
|
|
Long-term study of adaptive changes
The second study hypothesis was that FF-B to FF-E changes would differ in the SC and Control groups. The two groups had a similar response to the FF-B meal (Table 6) and, despite the different interventions, did not differ in insulin, glucose, and CCK endpoints in response to the FF-E meal. Fasting plasma glucose concentration decreased in the SC group and increased in the Control group, but these mean changes did not differ significantly between the groups (P = 0.13).
|
Side effects. At baseline, 81% of subjects in the SC group and 75% of subjects in the Control group reported at least one of the 11 gastrointestinal side effects. Both the frequency and number of side effects decreased slightly and similarly in both groups during the intervention. At the end of the intervention, 75% of subjects in the SC group and 56% of subjects in the Control group reported having experienced at least one of the 11 side effects (P = 0.26). Subjects in the SC group reported twice the number and frequency of side effects as subjects in the Control group, but the mean was 2.8 (out of 11) (P = 0.14) and the mean side effect score was 4.1 (out of 44) (P = 0.17). Subjects in the SC group seemed to report more side effects related to the lower gastrointestinal tract (e.g., flatulence) and satiety (e.g., feeling of fullness in the stomach). No serious adverse effects related to treatment were reported in either group during any of the five visits throughout the 6-wk intervention.
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Effect of design choice on findings
Meal type. The test meal used in this study contained carbohydrate, fat and protein, and previously had been shown to stimulate both glucose and CCK release (27). A mixed meal was chosen because of an interest in the ability of SC to modify the glucose and CCK responses to a mixture of nutrients. Pure glucose loads increase postprandial hyperglycemia to a greater extent than mixed meals. Thus, the presence of protein and fat in the meals may have obscured or reduced any differences in plasma glucose response attributable to the fiber (28). The incremental CCK responses were higher than those reported in the literature after meals of similar macronutrient composition (9), but this is consistent with previous evidence that plasma CCK response is higher after a liquid meal than after a solid meal (29,30).
Dose of SC. The 5-g dose of SC may not have been adequate to cause the hypothesized changes. Guévin et al. (31) showed that the overall quantity of fiber can be important. In a study of hypertriglyceridemic individuals with type 2 diabetes, the addition of 20 but not 10 g of fiber to a mixed meal resulted in lowered glucose and insulin AUC. High concentrations of SC yield a material with viscous, nectar-like consistency. At 2.5 g/240 mL, the study beverages were still palatable and easily ingested with regular meals. To observe a more pronounced effect on glucose metabolism, an increased amount of SC may be necessary.
Patient population. The variability observed in the present study in many of the endpoints was high and may have made it more difficult to detect significant changes in response to the fiber. The subjects were hypercholesterolemic and most were overweight, whereas most other studies were conducted in normal-weight, healthy individuals. Additionally, our participants spanned a wide age range (19–63 y). MacIntosh et al. (32) reported increased CCK concentrations associated with aging in men. Both baseline values and the rise in CCK during lipid infusions were significantly different in the older (65–80 y) compared with the younger (20–43 y) men in that study.
Diet adaptation period. Dietary changes in the present study may have also played a role in the results obtained. Statistical analyses in this study showed that the two groups were not comparable with regard to their background intake of total energy, percentage of energy from total, saturated and monounsaturated fat, total trans fat, as well as total and soluble dietary fiber. These differences existed despite the fact that both groups fell within desirable ranges for all macronutrients and total cholesterol as recommended by the AHA Step I diet (16). It is likely that the differences in these dietary components between the groups at baseline, wk 3 and study end confounded our ability to separate the effect of SC from that of background diet.
Time course of data collection. Measurement of plasma CCK for only 3 h postprandially may have been problematic. Bourdon et al. (9) suggested that a prolonged elevation of CCK might be the best indicator of an interaction between viscous fibers and the peptide.
Studies of the effect of fiber on glucose
There have been several studies with a design comparable to that of the present study. Kestin et al. (33) compared the glucose responses of 24 middle-aged, mildly hypercholesterolemic men after consumption of diets containing 11.8 g wheat, rice or oat bran for 4 wk each. The glucose and insulin responses to mixed meal tests, containing whole foods and conducted at the end of each diet period, did not differ from those measured at baseline (33). The meals were not tested acutely and there were no washout periods between the three diet periods. Hallfrisch et al. (34) evaluated the changes in glucose and insulin as a result of a 5-wk intake of 1 and 10% ß-glucan extracts. Glucose tolerance was determined after intake of a pure glucose solution at baseline and glucose plus each of the oat extracts at the end of the respective diet period. Fasting glucose concentrations were unchanged but postprandial glucose responses were significantly reduced after consumption of both extract-containing solutions (34). Background diet was fully controlled, but again, there was no washout period between the experimental treatments. Despite the positive findings, the authors could not determine why these results occurred because the responses to the oat extract-containing solutions represented both adaptation and acute response and it was impossible to separate the two (34). Finally, Hanai et al. (35) reported significant decreases in the glucose AUC response to an oral glucose load consumed after a 6-mo intervention with 5 g/d of soluble corn bran hemicellulose compared with a control period. The study involved obese, healthy and glucose-intolerant individuals, but only the glucose-intolerant subjects experienced the significant decreases in the glucose and insulin AUC; the rest of the subjects experienced only a lowering of fasting glucose concentration (35). The study however, had several methodological weaknesses in terms of randomization and data analysis.
Reasons for discrepancies in findings concerning glucose
Patient population. Jenkins and colleagues (36) hypothesized that the effect of fiber on glucose tolerance in normal subjects may be less pronounced than its effect in individuals with diabetes. Data from long-term studies in both type 1 and type 2 diabetes have consistently shown improvements in fasting and postprandial blood glucose concentrations and overall glycemic control (37–41). On the other hand, results from studies in healthy individuals have shown little or no effect of fiber on glucose tolerance, utilization and/or production (42–44).
Experimental protocol. Gastric emptying in early type 2 diabetes is increased (45), whereas in the later stages of the disease, slowed gastric emptying is common (46). Abnormalities in CCK secretion have also been reported in type 2 diabetics (47), including a lower peak plasma CCK compared with normal individuals (48). Participants in the present study were not diabetic; however, data from the larger study indicated that individuals in the SC group had significantly lower insulin sensitivity and a significantly higher acute insulin response to glucose than individuals in the Control group, presenting the possibility of confounding.
Studies of the effect of fiber on CCK
Unlike the effect of viscous fibers on glucose tolerance, only a few studies have evaluated the effects of fiber on CCK metabolism. An acute study in nine obese, middle-aged individuals, eight of whom were women, found no modification in the postprandial release of CCK after ingestion of an egg-sandwich meal containing either 15 g of pectin or placebo (methylcellulose) (49). Another group evaluated the effect of six high-carbohydrate solid test meals on the CCK response in seven healthy men in their mid-20s (50). CCK AUC responses did not differ among the meals. However, significant inverse correlations between CCK AUC and glucose and insulin AUC were reported (50). Generally, the higher the total dietary fiber content of a meal, the higher the CCK AUC and the lower the glycemic index. Another trial tested two liquid formula diets with or without a low viscosity hydrolyzed guar gum at a concentration of 21 g/L, compared with a self-selected solid food diet in a group of 12 men with a mean age of 29 y in a 14-d crossover study (30). There were no significant differences between the effects of the fiber-free and fiber-containing formulas on CCK peak, time to peak values and gastric emptying as estimated by orocecal time (30). More recently, Heini et al. (51) evaluated the efficacy of 20 g/d of the same hydrolyzed guar fiber incorporated into a liquid formula on weight loss and CCK postprandial concentrations in a randomized, double-blind, crossover study of 25 obese middle-aged women. Fasting CCK concentrations were not different, but the fiber-supplemented formula significantly increased postprandial CCK. The CCK peak concentration also occurred earlier after the fiber-containing meal than after the control meal (15 min vs. 30 min) (51).
Most studies have not observed a significant effect of fiber on CCK postprandial release or a decrease in the magnitude of the CCK peak as was observed in the present study. In the two studies that did report a significant change in CCK AUC, the response was increased after an acute intake of the investigated fiber. Even though subjects in the study of Heini et al. (51) consumed the fiber for 1 wk, the increased CCK AUC reported was in response to the intake of a liquid meal that contained the fiber, making it difficult to discriminate between the acute and chronic effects of the fiber. It is apparent that more long-term studies, as well as acute experiments with standardized protocols, testing a variety of fibers and measuring a variety of CCK-related endpoints, including gastric emptying, are required before conclusions can be drawn about the effect of dietary fiber on CCK metabolism.
In the present study, plasma CCK increased rapidly and usually reached a peak at 10 min. That the increase in CCK preceded the elevation in plasma glucose (mean peak at 30 min) is consistent with most studies of postprandial CCK metabolism (13,48). Furthermore, it highlights the physiologic roles of CCK in glucose homeostasis. It has been hypothesized that the increase in CCK preceding the increase in glucose may be an indication that CCK sensitizes or primes the pancreatic ß-cells to induce sufficient insulin response to subsequent direct glucose stimulation (48).
Reasons for discrepancies in CCK findings. Our observation that the CCK response to the liquid mixed meal in many individuals exhibited more than one prominent peak is consistent with a similar report of an acute study of the effects of carbohydrate foods on CCK, glucose and insulin metabolism in healthy young men (50). However, it was surprising to note multiple peaks in the postprandial glucose response. In a paper describing the postprandial changes in plasma lipids in response to a fat-rich meal, Cohn et al. (52) reported biphasic and triphasic triacylglycerol responses in the 22 study subjects. Their surprising observation was not associated with age or gender, nor was it correlated with fasting lipid levels (52). This, and findings from the present study, suggest that the postprandial patterns of incretin release are highly variable, and that presentation of mean data may disguise important trends. The high interindividual variability may be primarily responsible for the inconsistent findings concerning the effect of fiber on CCK and glucose metabolism reported in the literature.
The gender differences observed in the present study were somewhat surprising. Most studies of CCK metabolism have involved men rather than women. This is likely due to a desire to avoid the cyclic variations in estrogen and progesterone, which may influence appetite signals, such as CCK (53). There are four reports of significant gender differences in CCK and/or glucose concentration. One study, available only in abstract form, evaluated the effect of fat availability on postprandial satiety (54). Two others were studies of the effect of meals of varying macronutrient composition on glucose and insulin (55,56). The most recent of these studies reported a significant increase in the mean CCK response to a fiber-containing, low fat meal in women but not men (14). Despite the lack of significant sex differences, women in the present study tended to experience an increase in all CCK variables evaluated except for CCK AUC, and it may be hypothesized that had it not been for the four outliers in the men who experienced a notable drop in CCK measures, the men may not have had a change in CCK after the consumption of the fiber-containing test meal (Fig. 3).
Findings from the present study indicate that solubilized cellulose decreases the postprandial CCK peak after a meal and it might have an acute effect on postprandial glucose and CCK responses. The importance of these findings is not clear. In addition to the many potentially confounding factors discussed above, the multitude of components linked to the plant cell wall, which may induce or modulate the metabolic effects of fiber in general (e.g., phytic acid, plant sterols, saponins), make it even more difficult to interpret the effects of fiber-rich meals or long-term fiber supplementation. The fact that the fiber investigated in this study was free of such compounds makes it a useful tool in future attempts to identify and discriminate between the effects of purified fiber and those of fiber-rich foods on postprandial metabolism. Evaluating the effect of SC in different populations and at different doses is essential before any conclusions concerning its utility in human diets can be drawn.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Supported by FiberStar, Willmar, MN, General Clinical Research Center grant no. M01-RR00400 from the National Center for Research Resources, National Institutes of Health and funds from the American Diabetes Association.
3 Present address: General Mills, Inc., Minneapolis, MN.
5 Abbreviations used: AHA, American Heart Association; AUC, area under the curve; CCK, cholecystokinin; FF-B, fiber-free baseline meal; FF-E, fiber-free meal at study end; SC, solubilized cellulose; SC-B, solubilized cellulose-containing baseline meal.
Manuscript received 5 November 2002. Initial review completed 1 January 2003. Revision accepted 7 April 2003.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Reaven, G. M. (1988) Role of insulin resistance in human disease. Diabetes 37:1595-1607.[Abstract]
2. Heaton, K. W. (1979) Fibre, satiety and insulin—a new approach to overnutrition and obesity. Heaton, K. W. eds. Dietary Fibre: Current Developments of Importance to Health 1979 John Libbey & Co. London, UK. .
3. Haber, G. B., Heaton, K. W., Murphy, D. & Burroughs, L. F. (1977) Depletion and disruption of dietary fibre; effects on satiety, plasma-glucose, and serum-insulin. Lancet 2:679-682.[Medline]
4. Potter, J. G., Coffman, K. P., Reid, R. L., Krall, J. M. & Albrink, M J (1981) Effect of test meals of varying dietary fiber content on plasma insulin and glucose response. Am. J. Clin. Nutr. 34:328-334.
5. Braaten, J. T., Wood, P. J., Scott, F. W., Riedel, K. D., Poste, L. M. & Collins, M. W. (1991) Oat gum lowers glucose and insulin after an oral glucose load. Am. J. Clin. Nutr. 53:1425-1430.
6. Braaten, J. T., Scott, F. W., Wood, P. J., Riedel, K. D., Wolynetz, M. S., Brulé, D. & Collins, M. W. (1993) High ß-glucan oat bran and oat gum reduce postprandial blood glucose and insulin in subjects with and without type 2 diabetes. Diabetes Med. 11:312-318.
7. Wood, P. J., Braaten, J. T., Scott, F. W., Riedel, K. D., Wolynetz, M. S. & Collins, M. W. (1994) Effect of dose and modification of viscous properties of oat gum on plasma glucose and insulin following an oral glucose load. Br. J. Nutr. 72:731-743.[Medline]
8. Tappy, L., Gügolz, E. & Würsch, P. (1996) Effects of breakfast cereals containing various amounts of ß-glucan fibers on plasma glucose and insulin responses in NIDDM subjects. Diabetes Care 19:831-834.[Abstract]
9. Bourdon, I., Yokoyama, W., Davis, P., Hudson, C., Backus, R., Richter, D., Knuckles, B. & Schneeman, B. O. (1999) Postprandial lipid, glucose, insulin, and cholecystokinin responses in men fed barley pasta enriched with ß-glucan. Am. J. Clin. Nutr. 69:55-63.
10. Bourdon, I., Olson, B., Backus, R., Richter, B. D., Davis, P. A. & Schneeman, B. O. (2001) Beans, as a source of dietary fiber, increase cholecystokinin and apolipoprotein B48 response to test meals in men. J. Nutr. 131:1485-1490.
11. Rushakoff, R. J., Goldfine, I. D., Carter, J. D. & Liddle, R. A. (1987) Physiological concentrations of cholecystokinin stimulate amino acid-induced insulin release in humans. J. Clin. Endocrinol. Metab. 65:395-401.[Abstract]
12. Liddle, R. A., Morita, E. T., Conrad, C. K. & Williams, J. A. (1986) Regulation of gastric emptying in humans by cholecystokinin. J. Clin. Investig. 77:992-996.
13. Liddle, R. A., Rushakoff, R. J., Morita, E. T., Beccaria, L., Carter, J. D. & Goldfine, I. D. (1988) Physiological role for cholecystokinin in reducing postprandial hyperglycemia in humans. J. Clin. Investig. 81:1675-1681.
14. Burton-Freeman, B., Davis, P. A. & Schneeman, B. O. (2002) Plasma cholecystokinin is associated with subjective measures of satiety in women. Am. J. Clin. Nutr. 76:659-667.
15. Burton-Freeman, B. & Schneeman, B. (1996) Lipid infused into the duodenum of rats at varied rates influences food intake and body weight gain. J. Nutr. 126:2934-2939.
16. American Heart Association (1988) Dietary Treatment of Hypercholesterolemia. A Handbook for Counselors 1988 American Heart Association Dallas, TX.
17. Ruan, R., Lin, Y., Zhang, J., Addis, P. & Chen, P. (1996) Structure-function relationships of highly refined cellulose made from agricultural fibrous residues. Appl. Eng. Agric. 12:465-468.
18. Ruan, R., Chen, L., Johnson, E., Addis, P. B. & Xu, L. (1998) Cellulose fiber based compositions and film and the process for their manufacture. U.S. Patent 5,817,381.
19. Lundberg, B. M. (2000) Development of a New Process to Make Highly Refined Cellulose 2000 Department of Agricultural Engineering, University of Minnesota Twin Cities St. Paul, MN.
20. Rehfeld, J. E. (1998) Accurate measurement of cholecystokinin in plasma. Clin. Chem. 44:991-1001.
21. Matthews, J.N.S., Altman, D. G., Campbell, M. J. & Royston, P. (1990) Analysis of serial measurements in medical research. Br. Med. J. 300:230-235.
22. The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus (1997) Report of the Expert Committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 20:1183-1197.[Medline]
23. Holt, S., Carter, D. C., Heading, R. C., Prescott, L. F. & Tothill, P. (1979) Effect of gel fibre on gastric emptying and absorption of glucose and paracetamol. Lancet 1:636-639.[Medline]
24. Schwartz, S. E., Levine, R. A., Singh, A., Scheidecker, J. R. & Track, N. S. (1982) Sustained pectin ingestion delays gastric emptying. Gastroenterology 83:812-817.[Medline]
25. Reiser, S. & Lewis, C. G. (1986) Effect of the type of dietary carbohydrate on small intestinal functions. Prog. Biochem. Pharmacol. 21:135-159.[Medline]
26. Marciani, L., Gowland, P. A., Spiller, R. C., Manoj, P., Moore, R. J., Young, P., Al-Sahab, S., Bush, D., Wright, J. & Fillery-Travis, A. J. (2000) Gastric response to increased meal viscosity assessed by echo-planar magnetic resonance imaging in humans. J. Nutr. 130:122-127.
27. Fried, M., Schwizer, W., Beglinger, C., Keller, U., Jansen, J. B. & Lamers, C. B. (1991) Physiological role of cholecystokinin on postprandial insulin secretion and gastric meal emptying in man. Studies with the cholecystokinin receptor antagonist loxiglumide. Diabetologia 34:721-726.[Medline]
28. Bornet, F.R.J., Costagliola, D., Rizkalla, S. W., Blayo, A., Fontvieille, A., Haardt, M., Letanoux, M., Tchobroutsky, G. & Slama, G. (1987) Insulinemic and glycemic indexes of six starch-rich foods taken alone and in a mixed meal by type 2 diabetics. Am. J. Clin. Nutr. 45:588-595.
29. Mössner, J., Grumann, M., Zeeh, J. & Fischbach, W. (1992) Influence of various nutrients and their mode of application on plasma cholecystokinin (CCK) bioactivity. Clin. Investig. 70:125-129.[Medline]
30. Meier, R., Beglinger, C., Schneider, H., Rowedder, A. & Gyr, K. (1993) Effect of a liquid diet with and without soluble fiber supplementation on intestinal transit and cholecystokinin release in volunteers. J. Parenter. Enteral Nutr. 17:231-235.[Abstract]
31. Guévin, N., Jacques, H., Nadeau, A. & Galibois, I. (1996) Postprandial glucose, insulin, and lipid responses to four meals containing unpurified dietary fiber in non-insulin-dependent diabetes mellitus (NIDDM), hypertriglyceridemic subjects. J. Am. Coll. Nutr. 15:389-396.[Abstract]
32. MacIntosh, C. G., Andrews, J. M., Jones, K. L., Wishart, J. M., Morris, H. A., Jansen, J.B.M.J., Morley, J. E., Horowitz, M. & Chapman, I. M. (1999) Effects of age on concentrations of plasma cholecystokinin, glucagon-like peptide, and peptide YY and their relation to appetite and pyloric motility. Am. J. Clin. Nutr. 69:999-1006.
33. Kestin, M., Moss, R., Clifton, P. M. & Nestel, P. J. (1990) Comparative effects of three cereal brans on plasma lipids, blood pressure, and glucose metabolism in mildly hypercholesterolemic men. Am. J. Clin. Nutr. 52:661-666.
34. Hallfrisch, J., Scholfield, D. J. & Behall, K. M. (1995) Diets containing soluble oat extracts improve glucose and insulin responses of moderately hypercholesterolemic men and women. Am. J. Clin. Nutr. 61:379-384.
35. Hanai, H., Ikuma, M., Sato, Y., Iida, T., Hosoda, Y., Matsushita, I., Nogaki, A., Yamada, M. & Kaneko, E. (1997) Long-term effects of water-soluble corn bran hemicellulose on glucose tolerance in obese and non-obese patients: improved insulin sensitivity and glucose metabolism in obese subjects. Biosci. Biotechnol. Biochem. 61:1358-1361.[Medline]
36. Jenkins, D.J.A., Leeds, A. R., Houston, H., Hinks, L., Alberti, K.G.M.M. & Cummings, J. H. (1977) Carbohydrate tolerance in man after 6 weeks of pectin administration. Proc. Nutr. Soc. 36:60A(abs.).[Medline]
37. Simpson, H.C.R., Lousley, S., Geekie, M., Simpson, R. W., Carter, R. D., Hockaday, T.D.R. & Mann, J. I. (1981) A high carbohydrate leguminous fibre diet improves all aspects of diabetic control. Lancet 1:1-5.[Medline]
38. Hagander, B., Asp, N.-G., Efendíc, S., Nilsson-Ehle, P. & Scherstén, B. (1988) Dietary fiber decreases fasting blood glucose levels and plasma LDL concentration in noninsulin-dependent diabetes mellitus patients. Am. J. Clin. Nutr. 47:852-858.
39. Anderson, J. W., Zeigler, J. A., Deakins, D. A., Floore, T. L., Dillon, D. W., Wood, C. L., Oeltgen, P. R. & Whitley, R. J. (1991) Metabolic effects of high-carbohydrate, high-fiber diets for insulin-dependent diabetic individuals. Am. J. Clin. Nutr. 54:936-943.
40. Vuksan, V., Jenkins, D.J.A., Spadafora, P., Sievenpiper, J. L., Owen, R., Vidgen, E., Brighenti, F., Josse, R., Leiter, L. A. & Bruce-Thompson, C. (1999) Konjac-mannan (glucomannan) improves glycemia and other associated risk factors for coronary heart disease in type 2 diabetes: a randomized controlled metabolic trial. Diabetes Care 22:913-919.[Abstract]
41. Chandalia, M., Garg, A., Lutjohann, D., vonBergmann, K., Grundy, S. M. & Brinkley, L. J. (2000) Beneficial effects of high dietary fiber intake in patients with type 2 diabetes mellitus. N. Engl. J. Med. 342:1392-1398.
42. Nestel, P. J., Nolan, C., Bazelmans, J. & Cook, R. (1984) Effects of a high-starch diet with low or high fiber content on postabsorptive glucose utilization and glucose production in normal subjects. Diabetes Care 7:207-210.[Abstract]
43. Hollenbeck, C. B., Coulston, A. M. & Reaven, G. M. (1986) To what extent does increased dietary fiber improve glucose and lipid metabolism in patients with noninsulin-dependent diabetes mellitus (NIDDM)?. Am. J. Clin. Nutr. 43:16-24.
44. Behall, K. M. (1990) Effect of soluble fibers on plasma lipids, glucose tolerance and mineral balance. Furda, I. Brine, C. J. eds. New Developments in Dietary Fiber 1990 Plenum Press New York, NY. .
45. Phillips, W. T., Schwartz, J. G. & McMahan, C. A. (1991) Rapid gastric emptying in patients with early non-insulin dependent diabetes mellitus. N. Engl. J. Med. 324:130-131.[Medline]
46. Loo, F. D., Palmer, D. W., Soergel, K. H., Kalbfleisch, J. H. & Wood, C. M. (1984) Gastric emptying in patients with diabetes mellitus. Gastroenterology 86:485-494.[Medline]
47. Rushakoff, R. A., Goldfine, I. D., Beccaria, L. J., Mathur, A., Brand, R. J. & Liddle, R. A. (1993) Reduced postprandial cholecystokinin (CCK) secretion inpatients with noninsulin-dependent diabetes mellitus: evidence for a role for CCK in regulating postprandial hyperglycemia. J. Clin. Endocrinol. Metab. 76:489-493.[Abstract]
48. Hasegawa, H., Shirohara, H., Okabayashi, Y., Nakamura, T., Fujii, M., Koide, M. & Otsuki, M. (1996) Oral glucose ingestion stimulates cholecystokinin release in normal subjects and patients with non-insulin-dependent diabetes mellitus. Metabolism 45:196-202.[Medline]
49. DiLorenzo, C., Williams, C. M., Hajnal, F. & Valenzuela, J. E. (1988) Pectin delays gastric emptying and increases satiety in obese subjects. Gastroenterology 95:1211-1215.[Medline]
50. Holt, S., Brand, J., Soveny, C. & Hansky, J. (1992) Relationship of satiety to postprandial glycaemic, insulin and cholecystokinin responses. Appetite 18:129-141.[Medline]
51. Heini, A. F., Lara-Castro, C., Schneider, H., Kirk, K. A., Considine, R. V. & Weinsier, R. L. (1998) Effect of hydrolyzed guar fiber on fasting and postprandial satiety and satiety hormones: a double-blind, placebo-controlled trial during controlled weight loss. Int. J. Obes. 22:906-909.
52. Cohn, J. S., McNamara, J. R., Cohn, S. D., Ordovas, J. M. & Schaefer, E. J. (1988) Postprandial plasma lipoprotein changes in human subjects of different ages. J. Lipid Res. 29:469-479.[Abstract]
53. Lyons, P. M., Truswell, A. S., Mira, M., Vizzard, J. & Abraham, S. F. (1989) Regulation of food intake in the ovulatory phase of the menstrual cycle. Am. J. Clin. Nutr. 49:1164-1168.
54. Burton-Freeman, B. M., Davis, P. A. & Schneeman, B. O. (1998) Postprandial satiety: the effect of fat availability in meals. FASEB J. 12:A650(abs.).
55. Nuttall, F. Q., Gannon, M. C., Wald, J. L. & Ahmed, M. (1985) Plasma glucose and insulin profiles in normal subjects ingesting diets of varying carbohydrate, fat, and protein content. J. Am. Coll. Nutr. 4:437-450.[Abstract]
56. Holtschlag, D. J., Gannon, M. C. & Nuttall, F. Q. (1998) State-space models of insulin and glucose responses to diets of varying nutrient content in men and women. J. Appl. Physiol. 85:935-945.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
