Betaglucare.info

Four-Week Low-Glycemic Index Breakfast With a Modest Amount of Soluble
Fibers in Type 2 Diabetic Men
Morvarid Kabir, Jean-Michel Oppert, Hubert Vidal, Francoise Bruzzo, Caroline Fiquet, Pierre Wursch, Low-glycemic index diets are associated with a wide range of benefits when followed on a chronic basis. The chronic effects,
however, of the substitution of 1 meal per day are not well known in diabetic subjects. Therefore, we aimed to evaluate
whether the chronic use of a low-glycemic index breakfast (low-GIB) rich in low-GI carbohydrates and a modest amount of
soluble fibers could have an effect on lipemia at a subsequent lunch, and improve glucose and lipid metabolism in men with
type 2 diabetes. A total of 13 men with type 2 diabetes were randomly allocated in a double-blind cross-over design to a
4-week daily intake of a low-GI versus a high-GI breakfast separated by a 15-day washout interval. The low-GI breakfast was
composed of whole grain bread and muesli containing 3 g ␤-glucan from oats. Low-GIB induced lower postprandial plasma
glucose peaks than the high-GIB at the beginning (baseline, P < .001) and after the 4-week intake (P < .001). The incremental
area under the plasma glucose curve was also lower (P < .001, P < .01, baseline, and 4 weeks, respectively). There was no
effect on fasting plasma glucose, insulin, fructosamine, or glycosylated hemoglobin (HbA ). Fasting plasma cholesterol, as
well as the incremental area under the cholesterol curve, were lower (P < .03, P < .02) after the 4-week low-GIB period than
after the high-GIB period. Apolipoprotein B (apo B) was also decreased by the 4-week low-GIB. There was no effect of the
low-GI breakfast on triacylglycerol excursions or glucose and insulin responses at the second meal. The high-GIB, however,
tended to decrease the amount of mRNA of leptin in abdominal adipose tissue, but had no effect on peroxisome proliferator-
activated receptor ␥ (PPAR␥) and cholesterylester transfer protein (CETP) mRNA amounts. In conclusion, the intake of a
low-GI breakfast containing a modest amount (3 g) of ␤-glucan for 4 weeks allowed good glycemic control and induced low
plasma cholesterol levels in men with type 2 diabetes. The decrease in plasma cholesterol associated with low-GI breakfast
intake may reduce the risk of developing cardiovascular complications in subjects with type 2 diabetes.
Copyright 2002, Elsevier Science (USA). All rights reserved.
CONCERNS ABOUT USING high-carbohydrate diets in lowingamealisrecognizedtobeaccentuatedbycarbohydrate-
diabetes1 because of adverse effects on triglycerides and induced hyperinsulinemia.20 Thus, decreased plasma glucose high-density lipoprotein-cholesterol levels,2 are overcome by excursions and improved insulin sensitivity following a low-GI recommending carbohydrates that give low postprandial mealeaten at breakfast might enhance tolerance for carbohy- plasma glucose responses.3,4 For over half a century, it has been postulated that the increase in blood glucose was less pro- The acute effects of low- or high-GI breakfasts have been nounced after the consumption of starchy foods than after the evaluated in normal healthy subjects. Few studies have evalu- consumption of foods containing simple carbohydrates.
ated the chronic effect of these breakfasts in type 2 diabetic Starchy foods have been recognized as the main candidate for reducing glycemic and insulinemic responses. However, coin- In this perspective, therefore, we aimed to evaluate the cidentalwith recommendations to increase the intake of starchy effects of a low-GI breakfast on both glucose and lipid metab- foods has been the recognition that the glycemic responses to olism in men with type 2 diabetes. We aimed also to evaluate all starches are not the same and that starches are not inter- the effects of a low-GI breakfast on hyperlipidemia at a sub- changeable.5,6 The glycemic index (GI) of foods has, thus, been sequent lunch. Furthermore, we determined the expression of established and used to classify foods according to their blood some lipid-related enzymes: cholesterylester transfer protein (CETP), leptin, and peroxisome proliferator-activated receptor Although the use of low-GI carbohydrates in the diet of ␥ (PPAR␥), because in a previous study from our laboratory, a patients with type 2 diabetes is still debated,7,8 epidemiologic similar diet for rats was found to decrease the satietogenic studies demonstrated that the GI of the diet may be an impor- factor, leptin, as well as some lipid-related enzymes.23 tant factor in preventing non–insulin-dependent diabetes.9,10Beneficialeffects of such a diet have been demonstrated indiabetic and normal subjects when the diet is followed during From the INSERM U341 Department of Diabetes, A.P. Hoˆtel-Dieu the 3 meals of the day on a short- or long-term basis.11-15 Hospital, EA 3502 University Paris, VI, Paris, France; INSERM U449, For practicalpurposes, we aimed to change onl Lyon, France; and the Nestle Product Technology Center, Orbe, Swit- during the day. Breakfast was chosen because it is frequently found that at this meal, diabetic patients require more insulin Submitted July 13, 2001; accepted January 2, 2002. than for an isocaloric midday or evening meal.16,17 More- Supported by grants from the NationalInstitute of Heal over, some patients are hyperglycemic throughout the day, MedicalResearch (INSERM), Pierre and Marie Curie University, andNestle Center, Orbe, Switzerland. while others are mainly hyperglycemic in the morning (partic- Address reprint requests to Salwa W. Rizkalla, MD, PhD, Depart- ularly after breakfast) and less hyperglycemic at noon and ment of Diabetes, INSERM U341, Hoˆtel-Dieu Hospital, 1, place du postlunch.18 It has been suggested that insulin sensitivity and Parvis Notre Dame, 75004 Paris, France. glucose responses during the second meal may be related to the Copyright 2002, Elsevier Science (USA). All rights reserved. effect from the preceding carbohydrate challenge.19 Moreover, the magnitude of the increase in plasma triacylglycerols fol- Metabolism, Vol 51, No 7 (July), 2002: pp 819-826 Table 1. Clinical Characteristics of the Subjects at the Time of Screening
NOTE. Order: randomization of the subjects to start with a high-GI dietary period then a low-GI dietary-period (H/L) or to start with a low-GI then a high-GI dietary period (L/H).
Abbreviations: BW, body weight; BMI, body mass index; FBG, fasting blood glucose; TG, triglycerides.
Before entering into the study, the subjects had been seen on a regular basis (at least every 6 months) in our department. All were well A totalof 13 men with type 2 diabetes were recruited from patients attending the outpatient clinic of the Department of Diabetes of Hoˆtel- educated, especially concerning the type and quantity of foods they Dieu Hospital. Patients were selected on the basis of having fasting should consume. Each subject entered a run-in period of 1 month.
plasma glucose of 7.7 to 13.0 mmol/L, glycosylated hemoglobin Patients received individualcounseling by a dietitian concerning di- (HbA ) 6.5% to 11%, and plasma triacylglycerols lower than 3 etary food intake. In addition, counseling sessions (in small groups) mmol/L. Due to known plasma glucose and insulin variations by were conducted during the run-in period. During this period, we asked estrogen and progesterone modifications, women were excluded from them to follow their usual diet more strictly. Patients were recom- mended to consume 55% of their caloric intake as carbohydrate, 15% The clinical and biological characteristics of these subjects are given as protein, and 30% as lipids. Dietary intake was prescribed individu- in Table 1. Patients with abnormal renal, hepatic, and thyroid functions ally according to data obtained from dietary questionnaires to maintain as determined by physical examination, blood cell count, and standard the initialcaloric intake and nutrient proportions constant throughout blood biochemical profile were excluded. Similarly, patients suffering from gastrointestinaldisorders were prohibited from entering the study.
Before the beginning of each treatment period, dietary question- Twelve patients were taking oral antidiabetic agents (sulfonylurea naires were obtained again (baseline data, recall technique) to and/or metformin) and 1 patient on an antidiabetic dietary regimen maintain the initialcaloric intake and nutrient proportions constant alone. One patient was receiving an ␣-glucosidase inhibitor (acarbose).
throughout the study. To assess compliance with the dietary recom- He was asked to stop the treatment 4 weeks before the beginning of the mendations, patients were asked to keep a food diary to be com- study. None of the patients were being or had been treated with insulin.
pleted on the last 7 days of each treatment period. Household Four patients were being treated for hypertension with ␤-blockers, measuring cups or spoons and photographed food pictures were angiotensin-converting enzyme (ACE) inhibitors, antidiuretics, and/or used to quantify proportion sizes of foods eaten. When each subject returned his records at the end of each 7-day period, the dietitian Despite the fact that sulfonylureas and metformin have very different checked the contents of the records and clarified any ambiguous mechanisms of action, the observed results might not be due to differ- information with the subject. These records were analyzed using the ent therapies: all patients continued and kept constant all therapies computer program Profile Dossier V3 software (Audit Conseil en throughout the whole study period. Moreover, the results of the low-GI Informatique Me´dicale, Bourges, France) whose dietary database is period were compared with those of the high-GI period in the same made up of 400 foods or groups of foods representative of the patient with the same accompanied treatments.
French diet. French food contents were obtained from CiqualRep- The purpose, nature, and potentialrisks of the study were explained, ertory.24 There was approximately 10% of missing values. If a food and a written informed consent was obtained from all participants. The described by a subject was not in the database, the ingredients of ethics committee of Hoˆtel-Dieu Hospital approved the experimental recipes or manufacturer’s information were used to complete coding The patients were free-living and attended the department once/ week. They were asked to maintain their usualhabitualdiet throughout The patients were randomly allocated in a cross-over design to 2 the study apart from the breakfasts that were provided according to the periods of 4 weeks with a daily breakfast consisting of either low- or experimentalperiod, being low- or high-GI. During the washout, period high-GI carbohydrates with a modest amount of ␤-glucan. The 2 patients followed their habitual diets throughout the day even during nutritionalperiods were separated by a 15-day washout interval.
LOW-GLYCEMIC INDEX BREAKFAST AND TYPE 2 DIABETES Table 2. Composition of Low- and High-GIB
Abbreviations: Q, quantity of each product; prot, protein; L, lipids; E, energy; CHO, available carbohydrates.
*Cereal based on extruded oat bran concentrate, apple and fructose in the low-GIB period. In the high-GI period, cereals formed of whole wheat †Bread was pumpernickel in the low-GIB period and wholemeal bread (wheat flour) in the high-GIB period.
lipoprotein (HDL) and apolipoprotein (apo) A1 and B were also mea-sured.
The size of the mealwas fixed prior to the study and based on each patient’s dietary record taken before the study. Treatment foods for breakfast were provided to the subjects and prescribed to meet 20% ofdaily energy requirements. The composition of the 2 experimental Patients were asked to maintain a constant lifestyle throughout the breakfasts is shown in Table 2. In the high-glycemic index breakfast study. Physicalactivity was assessed by recallquestionnaires. The kind (GIB) period (calculated total GI, 64%), the cereal used was whole of activity and its frequency, as well as the mode and duration of wheat grains (Weetabix, Burton Latimer, Kettering, Northants, UK; GI, transportation to and from work, were questioned. Patients’ physical 69%), whereas the bread used was whole meal bread (wheat flour) with activity remained constant during the study.
a GI of 69%. In the low-GIB period (calculated total GI, 40%), thecerealused was based on extruded oat bran concentrate, apple, and fructose (muesli containing 3 g ␤-glucan from oats, offered by Nestle, Plasma glucose was measured by the glucose-oxidase method with a Orbe, Switzerland; GI, 41%). The bread used was pumpernickel (GI, glucose analyzer (Beckman, Fullerton, CA). Plasma insulin was deter- 41%). The breads were prepared and donated by Jackson, Eury, France.
mined by radioimmunoassay. The antiserum used in the test showed a In the high-GIB, we have used breads and cereals with only a mean cross-reactivity of 100% with human insulin. Triglycerides and free GI of 64% and not higher. Breads and cereals with higher glycemic fatty acids were determined by using Biome´rieux kits (Marcy-l’Etoile, indices have lower fiber content than those used in the low-GIB. In an France). Total, free, and esterified cholesterol were also measured using attempt to keep constant the amount of fibers in the 2 breakfasts, we Labintest kits (Aix-en-Provence, France). Apo AI and apo B were used high-GI cereals and breads with a fiber content comparable to determined by an immunochemicalassay with Behring kits (Mauburg, Germany). Samples were taken in triplicate for the measurements of Breads and cereals used were analyzed for fat, protein, and dietary plasma glucose and insulin; for other measurements, single samples fiber using standard Association of Analytical Communities (AOAC) methods, with available carbohydrate calculated by difference. ␤-glu-can in the cereal products was analyzed following the method described Gene Expression of CETP, Ob, and PPAR␥ RNA preparation. The RNA from adipose tissue (about 80 to 100 mg of frozen tissue) was obtained by using the Rneasy totalRNA kit (Qiagen, Courtaboeuf, France) as previously indicated.26 The absorp- At the beginning and the end of each nutritionalperiod, subjects tion ratio 260 to 280 nm was between 1.7 and 2.0 for all preparations.
were hospitalized from 7:30 AM to 4:00 PM after an overnight fast. A RNA integrity was verified on agarose gelelectrophoresis. Average sample of abdominal subcutaneous adipose tissue was obtained by yields were similar in samples obtained after the low- (17 Ϯ 3 ␮g of needle biopsy using a 14-gauge needle and a 30-mL syringe under local RNA/g/tissue) and the high-GI breakfast (16 Ϯ 4 ␮g of RNA/g/tissue).
anesthesia with xylocaine 10% without adrenaline. The tissue obtained Total RNA was stored at – 80°C for less than 1 month before analysis.
was rapidly frozen in liquid nitrogen and stored at -80°C. Then, during Quantification of target mRNAs. Human PPAR␥, leptin, and CETP the same day, an indwelling cannula was inserted into an antecubital mRNAs were quantified by reverse transcriptase (RT)-competitive vein. This cannula served for the hourly withdrawal of blood samples polymerase chain reaction (PCR), which consists of coamplification of during the metabolic profile. At 9:00 AM, each subject consumed a target cdNA with known amounts of a specific DNA competitor mol- breakfast (low- or high-GI) that was the same as during the chronic ecule added in the same PCR tube.27 The RT reactions were performed period. At 12:00 PM, a standard mealwas served providing 45% of from 0.1 ␮g totalRNA/g/tissue in the presence of 1 of the specific energy as carbohydrate, 40% as lipids, and 15% as proteins. The antisense primers and 2.5 U of a thermostable RT (Tth; Promega, standard mealwas of the same quantity and composition for allof the Charbonnie`res, France) as previously described.27 Fluorescent dye- subjects at all times. Blood samples were collected in the fasting state labeled sense primers were used in the PCR, and the amplified products at time 0 and then hourly during the 7 hours of metabolic profile. Blood were separated and analyzed on an ALFexpress DNA sequencer (Phar- samples were centrifuged and plasma was frozen (-20°C) for further macia, Uppsala, Sweden) using the Fragment Manager software (Phar- measurements of plasma glucose, insulin, and lipids (triglycerides, macia).28 Absence of contamination with genomic DNA was checked cholesterol, and free fatty acids). At time 0, HbA , high-density by omitting RT in the reaction. The construction of the competitor DNA molecules, the sequence of the primers, and the validation of the macronutrients were unchanged. Concomitantly, body weight RT-competitive PCR assay for PPAR28 and leptin26,27 mRNAs have was comparable after the 2 periods of low- and high-GIBs The analyses take into account the design of the experiment, the type, and the distribution of the variables.14,29 For continuous variables with normal distribution, a multiple analysis of variance followed by a fected by the chronic changes in the type of breakfast (Table 3).
post hoc test (LSD) was used using CSS statisticalpackage (StatSoft, However, there were some differences in the plasma glucose Tulsa, OK). The main factors considered in the analysis were the and insulin profiles between the 2 breakfasts at the beginning following: treatment (with 2 levels: low-GIB and high-GIB), time (with and the end of the chronic periods. With the high-GIB, plasma 2 levels: baseline and 4-week diet), and order (with 2 levels). The mean glucose increased more rapidly to give high peaks in the value at the end of each diet minus the baseline value before each diet beginning (baseline data, high-GIB v low-GIB, P Ͻ .001) and was used as the basis of a test of different carryover effects between the at the end (4 weeks, P Ͻ .001) of the nutritionalperiod. Plasma 2 diets. For continuous variables with normal distribution, a test for insulin peaks showed the same profiles (but not significantly different carryover effects at the 10% level was used. If the test was notsignificant, a t test for different treatment effects at the 5% level was different). After these postprandial peaks, blood glucose con- used. If the carryover test was significant, only data from the first centrations decreased with the high-GI starch breakfast dipping dietary period were used in comparisons of treatment effects. If the below the fasting values in some patients (results not shown).
usualassumption for t test did not hold or if the data were on an ordinal The area under the glucose curve after breakfast was signifi- scale, the Mann-Whitney U test replaced the t test. Results are ex- cantly greater for the high-GIB than after the low-GIB at the beginning and the end of the nutritionalperiod (P Ͻ .01) asshown in Table 3. These results validate the use of 2 breakfasts with different postprandial plasma glucose responses. Therewas no significant difference in plasma glucose and insulin Patients followed the 2 dietary periods without any difficulty.
excursions after lunch during the the 2 nutritional periods.
The 2 breakfasts were well tolerated, without any complaints or As shown in Tables 4 and 5, a 4-week consumption of the side effects. According to self-report, subjects’ lifestyles re- low-GIB induced a 10% decrease (total relative difference) in mained unchanged throughout the entire study. There was no fasting totalcholesterol(P Ͻ .03). Fasting cholesterol before effect of crossover design (low- or high-GIB) or diet by period the 4-week low-GIB period was found to be slightly higher (but interaction for any of the parameters studied.
stillin the normalrange) than before the high-GI dietary period,despite the cross-over design of the study. Consequently, it is more physiologic to compare statistically the differences during Results of the 7-day records at the end of the 2 nutritional each period (baseline data-4-week data, high-GIB v low-GIB) periods demonstrated that daily intakes of total energy and rather than the effect at the end of the 4-week period. This Table 3. HbA , Plasma Glucose, and Insulin Concentrations at Baseline and After 4 Weeks of Low- and High-GIB
NOTE. Data are mean Ϯ SEM (n ϭ 13).
Abbreviation: NS, not significant.
*Results during the 1-day metabolic profile taken just before (data in the fasting state) or after the respective breakfast.
†Real chronic dietary effect: the delta (4-week data-baseline data) during the high-GI dietary period was compared with the delta during the aSignificant acute dietary effect after a single breakfast at the beginning of the chronic dietary period (baseline data, high-GI v low-GI).
bSignificant effect after a single breakfast at the end of the chronic dietary periods (4-week data), high-GI v low-GI. ‡P Ͻ .001, §P Ͻ .01. Theseresults validated the fact that the 2 breakfasts had different glycemic responses at the beginning (a) and the end of the study (b) AUC is expressed LOW-GLYCEMIC INDEX BREAKFAST AND TYPE 2 DIABETES Table 4. Circulating Lipid and Lipoprotein Concentrations at Baseline and After 4 Weeks of Low- and High GIB
NOTE. Data are mean Ϯ SEM (n ϭ 13).
Abbreviation: NS, not significant.
*Responses to a single low-GI or high-GI breakfast during the 1-day meiabolic profile before (baseline data) or after the 4-week dietary periods †Real chronic dietary effect: the delta (4-week data -baseline data) during the high-GI dietary period was compared with the delta during the comparison reflects correctly the effect of the dietary interven- terol response curves during the 7-hour profile day were also tions and eliminates the influence of nonspecific baseline vari- lower (P Ͻ .02 and Ͻ .04, respectively) after the period with ations. The same situation was found for mean cholesterol and the low-GIB than after the high-GIB (Table 4).
mean free cholesterol after a single low-GIB during the 1-day There was no significant difference in the fasting levels or metabolic profile, before the beginning of the chronic period the area under the curves for triacylglycerols and free fatty (baseline data). This finding might be due simply to initially acids (baseline-4 week) after breakfasts. Similarly, there was slightly higher fasting levels. Again, it was the delta (baseline no detectable effect of the low-GIB on triacylglycerol and free data-4-week data) that was compared taking into consideration fatty acid excursions at the lunch meal. Apo B was found to be the order of the 2 interventions. The changes (baseline-4 week) lower (P Ͻ .03) after 4 weeks of the low-GIB compared with in the incrementalarea under the plasma totaland free choles- the high-GIB (basal– 4 week) as shown in Table 4.
Table 5. Individual Changes in Circulating Total Cholesterol Concentrations at Baseline and After 4 Weeks of Low- and High-GIB
NOTE. Total cholesterol values are expressed as mmol/L. Results during the 1-day metabolic profiles taken just before (data in the fasting state) the respective breakfasts. Different letters in the same row indicate significant difference (P Ͻ .03).
PPAR␥, Leptin, and CETP mRNAs well as low cholesterol levels. In the present study, the fiber content of the 2 breakfasts was the same. Consequently, the successfully performed for 6 patients after the 2 dietary periods.
quantity of fiber could not be implicated in the observed low There was no difference in the amount of PPAR␥ (16.8 Ϯ 4 v postprandial glycemic responses found or in the low choles- 17.3 Ϯ 3.7 amol/␮g totalRNA) or CETP (2.1 Ϯ 0.5 v 2.3 Ϯ 0.6 terol levels induced after 4 weeks of such a breakfast.
amol/␮g totalRNA) mRNAs at the end of the low-GIB and the However, the low-GIB contained 3 g of ␤-glucan. Jenkins et high-GIB periods, respectively. The RNAm of ob, however, al34 showed that the increase in postprandialglucose and showed a trend toward a decrease in most of the subjects (n ϭ insulin concentrations were reduced after meals containing 6) when the end of the high-GIB period was compared with the viscous soluble fibers. Guar, psyllium, ␤-glucan, and pectin end of the low-GIB period (3.5 Ϯ 1.0 v 1.3 Ϯ 0.7 amol/␮g total were allfound to flatten postprandialglycemia more consis- RNA, respectively), (amol ϭ attomol ϭ 10-18mol).
tently than wheat bran and other insoluble particulate fi-bers.35 The decrease of cholesterol levels by the addition ofsoluble fibers present in oat bran was found in many studies.
In a meta-analysis of the literature of clinical trials of free The replacement of a high-GIB by a low-GIB for 4 weeks living subjects,36 reduction in cholesterol levels was found in type 2 diabetic patients in the present study lowered in subjects who had initially high blood cholesterol levels postprandial plasma glucose peaks, as well as plasma glu- (Ͼ 5.9 mmol), particularly when a daily dose of 3 g soluble cose and insulin responses. The effect of a low-GIB was fiber was used. The reduction was on average 6%. A single found after the consumption of the breakfast meal(during dose of oat bran cerealcontaining approximately 4g ␤-glu- the 1-day profile) at the beginning of the study, as well as at can taken by hypercholesterolemic patients for 3 weeks the end of the 4-week nutritionalperiod. These daily dietary produced an 11% decrease in total serum cholesterol.32 A changes over 4 weeks was insufficient to improve either lack of an effect, however, was reported by Lovegrove et or fructosamine, but were quite enough to improve al37 in a UK population. In our study with only 3 g ␤-glucan, a 10% reduction of cholesterol was found after a 4-week The present study demonstrated for the first time reduced intake. ␤-glucan might decrease cholesterol levels by in- cholesterol levels by 4 weeks of a low-GI meal taken once a creasing fecal steroid secretion or short chain volatile fatty day in the morning in normocholesterolemic subjects. This acid production.38 Whatever the mechanism might be, the decrease in plasma cholesterol levels (-10%) might be due to decrease in cholesterol found in our study was due likely to changing the type of carbohydrates (in bread and cereals) from the association of ␤-glucan with low-GI carbohydrates in the high-GI to low-GI during the breakfast meal for 4 weeks. In the literature, there is an important body of evidence in support of Changing breakfast from a high-GI to a low-GI had no impact on postprandial plasma glucose, insulin, or triacylglyc- during the 3 meals, but not only once at breakfast, in subjects erol excursions after the second slightly hyperlipidemic meal with type 2 diabetes and dyslipidemia.11,14,30,31 However, few (lunch). Several factors influence triacylglycerol in the post- studies demonstrated positive effects by changing only the prandialstate. Improving insulin sensitivity was demonstrated breakfast composition. In hypercholesterolemic type IIa pa- to decrease postprandial triacylglycerol levels,39 whereas in- tients, replacing a conventionalcontinentalbreakfast by a sin- creasing insulin levels induced by carbohydrates was found to gle oat bran cereal (muesli) for 3 weeks induced a decrease in accentuate postprandial triacylglycerol excursions.20 In the present study, the fact that insulin peaks were not significantly lipoprotein (LDL)-cholesterol levels.32 Golay et al,21 however, decreased by the low-GIB might explain the absence of any found that switching from standard cereals (cornflakes) to slow effect on postprandial triacylglycerol levels. This might be also release starch cereals (muesli) at breakfast for only 2 weeks due to the absence of any effect of the low-GIB on insulin improved carbohydrate metabolism and reduced insulin re- resistance or chronic glucose control.
quirements in type 2 insulin-treated diabetic patients, but no Since high-GI diets and subsequent postprandial plasma glu- effect on cholesterol levels was detected. A 2-week dietary cose responses might predispose to cardiovascular complica- period might be insufficient to decrease plasma cholesterol as tions, as well as other metabolic diseases, such as obesity,40,41 in our study. Furthermore, in normalsubjects, a high carbohy- we also aimed to evaluate the effect of this type of breakfast on drate/low-GIB meal reduced free fatty acid responses to break- the expression of some genes implicated in lipid metabolism. In fast when compared with a low-carbohydrate breakfast. The a previous study from our laboratory, a chronic period of investigators did not mention any effect on cholesterol levels.33 high-GI diet was able to decrease both plasma leptin and ob Recently, in type 2 diabetic subjects, the same group22 did not gene expression in epididymaladipose tissue in normalrats find an effect on plasma cholesterol levels after 6 months of before any detectable modification in weight gain or adipose low-GIB cereal compared with high-GIB cereal. The difference tissue mass.23 This decrease was considered as a marker that between this study and the present one might be due to the predicts further weight gain.42 In the present study, however, 1 incorporation of more than 1 type of low-GI food in the mealper day of a low-GI diet in patients with type 2 diabetes breakfast (cereals and bread) with a modest amount of soluble led to a trend towards a decrease of ob gene expression that did not reach significant values. Differences of results between rat Increasing the fiber content of the diet whether soluble or and man might be due to the difference in quantity of low-GI not might lead to low postprandial glycemic responses, as food consumed during the day and/or the presence of diabetes.
LOW-GLYCEMIC INDEX BREAKFAST AND TYPE 2 DIABETES The PPAR␥, which is one of the key messengers responsible Using cereals and breads of low-GI carbohydrates and with for the translation of nutritional and pharmacologic stimuli into modest doses of soluble fibers would be of benefit to patients changes in gene expression and differentiation pathways, did with type 2 diabetes. Intervention to reduce plasma lipids and not change during the nutritionalmodifications in our study. A postprandial glycemia in patients with diabetes could reduce third gene implicated in the cholesterol ester transfer from LDL the risk of patients developing atherosclerotic heart diseases.
to very–low-density lipoprotein (VLDL)-cholesterol is the Dietary intervention might be one of the major approaches in CETP, which plays a crucial role in modifying lipoprotein patients with diabetes and might be usefulin normalnondia- patterns, as well as LDL size distribution. The changes in total cholesterol in the present study were not associated withchanges in either HDL-cholesterol or CETP gene. Thus, in thepresent study, changing 1 mealper day (breakfast) from a high to low-GI for 4 weeks in non–insulin-dependent diabetes mel- We express our gratitude to Professor B. Guy-Grand (Nutrition litus patients was not sufficient to modulate genes implicated in Department, Hoˆtel-Dieu Hospital) for the opportunity to perform lipid and lipoprotein measurements in his laboratory.
1. Fukagawa NK, Anderson JW, Hageman G, et al: High-carbohy- 17. Vlaachokosta F, Piper C, Cleason R, et al: Dietary carbohydrate, drate, high fiber diets increase peripheral insulin sensitivity in healthy a Big Mac, and insulin requirements in type I diabetes. Diabetes Care young and old adults. Am J Clin Nutr 52:524-528, 1990 2. Perrotti N, Santoro D, Genovose S, et al: Effects of digestible 18. Guillausseau P: Monitoring of metabolic control in patients with carbohydrates on glucose control in insulindependent patients. Diabe- non insulin-dependent diabetes mellitus on oral hypoglycemic agents: Value of evening blood glucose determination. Diabet Med 14:798- 3. Ha T, Lean M: Recommendations for the nutritionalmanage- ment of patients with diabetes mellitus. Eur J Clin Nutr 52:467-481, 19. Liljeberg H, Akerberg A, Bjorck I: Effect of the glycemic index and content of indigestible carbohydrates of cereal-based breakfast 4. Wolever T: Dietary recommendations for diabetes: High car- meals on glucose tolerance at lunch in healthy subjects. Am J Clin Nutr bohydrate or high monounsaturated fat? Nutrition Today 34:73-77, 20. Chen Y-K, Coulston A, Zhou M-K, et al: Why do low-fat 5. Crapo P, Reaven G, Olefsky J: Postprandial plasma glucose and high-carbohydrate diets accentuate postprandiall insulin responses to different complex carbohydrates. Diabetes 26: 21. Golay A, Koellreutter B, Bloise D, et al: The effect of muesli or 6. Bornet F, Fontvieille A, Rizkalla S, et al: Insulin and glycemic cornflakes at breakfast on carbohydrate metabolism in type 2 diabetic responses in healthy humans to native starches processed in different patients. Diabetes Res Clin Pract 15:135-142, 1992 ways: Correlation with in vitro alpha-amylase hydrolysis. Am J Clin 22. Tsihlias E, Gibbs A, McBurney M, et al: Comparison of high and low-glycemic-index breakfast cereals with monounsaturated fat in 7. Brand-Miller J, Colagiuri S, Foster-Powell K: The glycemic the long-term dietary management of type 2 diabetes. Am J Clin Nutr index is easy and works in practice. Diabetes Care 20:1628-1629, 1997 8. Coulston A, Reaven G: Much ado about (almost) nothing. Dia- 23. Kabir M, Guerre-Millo M, Laromigiere M, et al: Negative regulation of leptin by chronic high glycemic index starch diet. Me- 9. Salmeron J, Ascherio A, Rimm E, et al: Dietary fiber, glycemic load, and risk of NIDDM in men. Diabetes Care 20:545-550, 1997 24. Feinberg M, Favier J, Ireland-Rippert J: Table de Composition- 10. Meyer K, Kashi L, Jacobs D, et al: Carbohydrates, dietary fiber, Table ciqual. INRA, Paris, France, Lavoisier, 1991 and incident type 2 diabetes in older women. Am J Clin Nutr 71:921- 25. McCleary B, McCleary B: Importance of enzyme purity and activity in the measurement of totaldietary fiber contents. J AOAC Int 11. Brand J, Colagiuri S, Crossman S, et al: Low glycemic index foods improve long-term glycemic control in NIDDM. Diabetes 14:95-101, 1991 26. VidalH, Auboeuf D, DevVos P, et al: The expression of ob gene 12. Fontvieille A, Rizkalla S, Penfornis A, et al: The use of low is not acutely regulmated by insulin and fasting in human abdominal glycemic index foods improves metabolic control of diabetic patients subcutaneous adipose tissue. J Clin Invest 98:251-255, 1996 27. Auboeuf D, VidalH: The use of the reverse transcription- 13. Behall K, Howe J: Effect of long-term consumption of amylose competitive polymerase chain reaction to investigate the in vivo regu- vs amylopectin starch on metabolic variables in human subjects. Am J lation of gene expression in small tissue samples. Anal Biochem 14. Jarvi A, Karlstrom B, Granfeldt Y, et al: Improved glycemic 28. Auboeuf D, Rieusser J, Fajas L, et al: Tissue distribution and control and lipid profile and normalized fibrinolytic activity on a quantification of the expression of mRNAs of peroxisome prolif- low-glycemic index diet in type 2 diabetic patients. Diabetes Care erator-activated receptors and liver X receptor-alpha in humans. No alteration in adipose tissue of obese and NIDDM patients. Diabetes 15. Ludwig D, Majzoub J, AL-Zahrani A, et al: High glyce- mic index foods, overeating and obesity. Pediatrics 103:E261-266, 29. Jones B, Kenward M: Monographs on statistics and applied probability, in Design and Analysis of Cross-over Trials. London, UK, 16. Nestler J, Gebhart S, Blackard W: Failure of a midnocturnal insulin infusion to suppress the increased insulin need for breakfast in 30. Wolever T, Jenkins D, Vuksan V, et al: Beneficial effect of low insulin dependent diabetic patients. Diabetes 33:266-270, 1984 glycemic index diet in type 2 diabetes. Diabet Med 9:451-458, 1992 31. Frost G, Leeds A, Dore´ J, et al: Glycemic index as a determinant 37. Lovegrove J, Clohessy A, Milon H, et al: Modest doses of of serum HDL-cholesterol concentration. Lancet 353:1045-1048, 1999 B-glucan do not reduce concentrations of potentially atherogenic li- 32. Bartram P, Gerlach S, Scheppach W, et al: Effects of single oat poproteins. Am J Clin Nutr 72:49-55, 2000 bran cereal breakfast on serum cholesterol, lipoproteins, and apoli- 38. Illman R, Topping D: Effects of dietary oat bran on fecal steroid poproteins in patients with hyperlipoproteinemia Type IIa. JPEN 16: excretion, plasma volatile fatty acids and lipid synthesis in rats. Nutr 33. Wolever T, Bentum-Williams A, Jenkins D: Physiologic mod- 39. Jeppesen J, Zhou M, Chen Y, et al: Effect of metformin on ulation of plasma FFA concentrations by diet: Metabolic implications postprandial lipemia in patients with fairly to poorly controlled in non-diabetic subjects. Diabetes Care 18:962-970, 1995 34. Jenkins D, Wolever T, Leeds A, et al: Dietary fibers, fiber 40. Jacobs D, Meyer K, Kushi L, et al: Whole grain intake may reduce analogues and glucose tolerance: Importance of viscosity. BMJ 1:1392- the risk of ischemic heart disease death in postmenopausalwomen: The Iowa Women Health Study. Am J Clin Nutr 68:248-257, 1998 35. Anderson J, Allgood L, Turner J, et al: Effects of psyllium on 41. Morris K, Zemel MB: Glycemic index, cardiovascular disease glucose and serum lipid responses in men with type 2 diabetes and hypercholesterolemia. Am J Clin Nutr 70:466-473, 1999 42. Ravussin E, Pratley R, Maffei M, et al: Relatively low plasma 36. Ripsin C, Keenan J, Jacobs D, et al: Oat products and lipid leptin concentrations precede weight gain in Pima Indians. Nat Med lowering:A meta-analysis. JAMA 267:3317-3325, 1992

Source: https://www.betaglucare.info/download/06_Kabir_2002.pdf