| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Medical Dietetics Division, School of Allied Medical Professions, College of Medicine and Public Health (M.L.A.)
Department of Human Nutrition, College of Human Ecology (S.R.H.)
The Ohio State University, Ross Products Division, Abbott Laboratories (J.C., B.W.W.) Columbus, Ohio
Address reprint requests to: Steven R. Hertzler, PhD, RD, LD, Assistant Professor, Department of Human Nutrition, College of Human Ecology, The Ohio State University, 325 Campbell Hall, 1787 Neil Avenue, Columbus, OH 43210-1295. E-mail: hertzler.4{at}osu.edu
 |
ABSTRACT |
|---|
 TOPFOOTNOTES ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSCONCLUSIONREFERENCES |
|---|
Methods: Healthy adult subjects (N = 32) received 50 g of carbohydrate from GCD or a rapidly digested maltodextrin (MD) in a double-masked, randomized, crossover design. Plasma glucose (fingerstick) and serum insulin (venous) concentrations were measured at baseline and at 15, 30, 45, 60, 90, 120, 150, and 180 min postprandially. Breath hydrogen excretion was monitored hourly for 8 h postprandially. The severity of gastrointestinal symptoms (nausea, cramping, distension, flatulence) was rated by the subjects on a ranked scale for two 24-h periods postprandially.
Results: The mean baseline-adjusted peak plasma glucose concentration was 47% lower (P < 0.001), and the mean baseline-adjusted peak serum insulin concentration was decreased by 45% (P < 0.001) after subjects consumed GCD compared with MD. Positive incremental area under the curve (0–120 min) was reduced 45% for plasma glucose and 49% for serum insulin by GCD compared with MD (P < 0.001 in each case). There were no differences between GCD and MD in the proportion of positive breath hydrogen tests and both carbohydrates were equally well tolerated.
Conclusions: GCD effectively lowers postprandial glycemia and insulinemia compared with MD, without resulting in appreciable carbohydrate malabsorption or gastrointestinal intolerance.
Key words: gamma-cyclodextrin, glycemia, insulinemia, breath hydrogen
|
INTRODUCTION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSCONCLUSIONREFERENCES |
|---|
Cyclodextrins are cyclic oligosaccharides consisting of 6 or more glucose units linked by 
-1,4 bonds. The three most prominent are alpha-, beta-, and gamma-cyclodextrin, which contain 6, 7, and 8 glucose units, respectively. Because the inside of the ring is more lipophilic than the outside, cyclodextrins have been used as carriers, stabilizers, and emulsifiers in foods [9]. In 1999, gamma-cyclodextrin (GCD) was granted GRAS (Generally Recognized As Safe) status by the US Food and Drug Administration [10]. The ability of salivary and pancreatic amylases to hydrolyze cyclodextrins is variable. In vitro studies have demonstrated that GCD, unlike alpha- and beta-cyclodextrin, is hydrolyzable, albeit slowly, by porcine and human amylases [11]. Spears et al. [12] found that GCD is nearly completely digested in the small intestine of dogs cannulated at the ileum. To our knowledge, there is only one study in which GCD has been fed to humans [13]. This study reported that an 8-g dose of GCD caused similar levels of gastrointestinal symptoms as compared with an equal dose of maltodextrin (MD). However, the digestibility and glycemic/insulinemic responses to GCD were not evaluated.
The low glycemic response to many carbohydrates is due to the presence of high amounts of dietary fiber or resistance to digestion in the small intestine [14]. It is rarer to identify a carbohydrate that is slowly, but still fully, digested in the small intestine. Identification of such a carbohydrate could have important applications in the dietary management of diabetes (avoidance of night-time hypoglycemia), sports nutrition (sustained release of glucose for endurance events), and the management of glycogen storage disorders (reduction in fasting hypoglycemia).
The primary objective of this study was to determine the postprandial glycemic response of healthy adult subjects to GCD. Secondary objectives were to determine the effects of GCD on 1) postprandial insulinemic response, 2) postprandial breath hydrogen response, and 3) subjective gastrointestinal tolerance symptoms and stool frequency and consistency in healthy adult subjects.
|
SUBJECTS AND METHODS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSCONCLUSIONREFERENCES |
|---|
6.10 mmol/L; no previous diagnosis of diabetes mellitus or other metabolic or gastrointestinal diseases; no infection, surgery, or corticosteroid treatment within the past 3 months or antibiotic therapy within the past 3 weeks; hydrogen producer and produces more hydrogen than methane. Out of 65 subjects that went through the screening process, 35 met all eligibility criteria and were randomized into the study. Of these subjects, 32 completed all study visits and were deemed evaluable for statistical analysis. All three subjects that exited from the study prematurely did so before the first treatment visit. One female began taking antibiotics and therefore no longer met eligibility criteria, one male exited due to employment conflicts, and one male was lost to follow up. The 32 evaluable subjects had a mean age of 24.97 ± 0.64 y (range 21–36 y), a mean weight of 67.8 ± 2.0 kg, a mean BMI of 23.2 ± 0.3 kg/m2, and a mean fasting plasma glucose concentration of 4.73 ± 0.10 mmol/L. The subject sample consisted of 11 (34%) males, 21 (66%) females, 3 (9%) African American, 7 (22%) Asian, and 22 (69%) white. The study was reviewed and approved by the Western Institutional Review Board in Olympia, WA. Informed consent was obtained from all subjects prior to starting the study.
Feeding Protocol
A randomized, double-masked, crossover design, with a 4–14 day washout period between the two treatments (GCD and MD) was employed for this study. Randomization was carried out using sealed envelopes that contained the treatment sequence for each subject. The study beverages were developed at Ross Products Division, Abbott Laboratories and shipped to the study site in 240 mL cans labeled with a code number corresponding to either GCD or MD. The composition of the study products (per 240 mL can) was as follows: 25 g either MD or GCD, 52 mg sodium, 56 mg potassium, 86 mg chloride, and 308 mg ascorbic acid. At each of the two treatment visits, subjects received 480 mL (2 cans) of the study product that contained either 50 g of MD or 50 g of GCD served in opaque, covered cups. The 50 g carbohydrate dose was chosen for each of the treatment visits because this is the amount of carbohydrate typically given in published glycemic response studies, and is also the carbohydrate dose utilized in determining the glycemic index of a food [16].
Before each of the two treatment visits, subjects consumed 
150 g of carbohydrate each day for the three days prior to each treatment [17] and refrained from exercise the day before each treatment to ensure that subjects had adequate glycogen stores. Three-day food records kept by the subjects were analyzed for carbohydrate content prior to commencement of each treatment. The evening before each treatment, subjects consumed a low-residue dinner consisting of one can (240 mL, 1507 kJ) of Ensure Plus® (Ross Products Division, Abbott Laboratories, Columbus, OH) with additional Ensure® Nutrition and Energy bars (963 kJ each) to provide one-third of each subject’s daily caloric requirement, as estimated by the Harris-Benedict equation [18] multiplied by an activity factor of 1.3 [19]. Subjects stopped consuming regular foods and beverages at 1600 h and consumed the Ensure® meal between 1600 and 2100 h. No foods or beverages except water were consumed after 2100 h. Subjects fasted overnight for a total of 10–16 h.
The morning of each treatment, subjects consumed 120–180 mL of water upon waking, and arrived at the study site between 0700 and 0730 h. Upon completing a 30-min rest period, the subjects’ weight, blood pressure, temperature, heart rate, and respiratory rate were measured, and a catheter was then placed in an antecubital vein of each subject by a registered nurse. Baseline blood samples were collected by finger-prick (glucose analysis) and from the catheter (insulin analysis). Immediately following the collection of baseline data, the subjects consumed the study product within 10 min, and the timing for postprandial samples began (after the first sip of the study beverage).
Plasma Glucose and Serum Insulin Analysis
Finger-prick capillary blood samples and venous blood samples were collected at baseline and at 15, 30, 45, 60, 90, 120, 150, and 180 min after the start of the study product consumption. Subjects were limited to the consumption of 240 mL of water during the 3-h blood collection period. Approximately 10 mL of normal saline was injected into the vein after each blood draw in order to keep the line open and to replace the fluid lost from the blood collected.
Plasma glucose was measured using the AccuChek Advantage Blood Glucose Monitoring System® (Roche Diagnostics, Indianapolis, IN). This system measures glucose in whole blood but is calibrated to plasma glucose [20]. The accuracy of the AccuChek® meter was evaluated in a previous study [21], and it was found to correlate well with a standard laboratory glucose analyzer (YSI 2300 Stat Plus, YSI Instruments, Yellow Springs, OH).
Venous blood samples (5–10 mL) were collected into serum separator tubes and allowed to clot for 15 min. After clotting, the samples were centrifuged at 1168 x g for 15 min at room temperature. The serum was aspirated off the top of the sample and stored at –20°C until analysis. Insulin was analyzed using DSL-10-1600 Insulin ELISA Kits (Diagnostic Systems Laboratories, Inc, Webster, TX). The sensitivity of this analysis is 1.87 pmol/L [22]. The intra-assay precision is 1.3–2.6% and the inter-assay precision is 5.2–6.2% [22].
Breath Hydrogen Analysis
Breath samples were collected at baseline and at hourly intervals for 8 h following consumption of the study beverage. After the study beverage was consumed, subjects resumed fasting until all the remaining breath samples were collected, with the exception of the consumption of up to two cans of Ensure Plus® (a fiber-free enteral formula) if desired after blood collection was completed. The amount and flavor of Ensure Plus, if consumed, remained the same during both treatments. Ensure Plus® does not contain fermentable fiber and results in negligible production of breath hydrogen [unpublished data]. Samples of end-alveolar air were collected into sealed Vacutainer® tubes (Becton Dickinson, Fisher Health Care, Houston, TX) using an AlveoSampler® mouthpiece (Quintron Instruments, Milwaukee, WI). Samples were extracted from the tubes using the Sample Xtractor® device (Quintron Instruments, Milwaukee, WI), and injected into a Microlyzer® Gas Analyzer, model SC (Quintron Instruments, Milwaukee, WI) for analysis via gas chromatography. End-alveolar air samples were analyzed for hydrogen, methane, and carbon dioxide concentrations. Observed hydrogen and methane concentrations were corrected for atmospheric contamination of alveolar air by normalizing the concentrations of observed carbon dioxide to 5.3 kPa, which is the partial pressure of carbon dioxide in alveolar air [23]. Changes in breath hydrogen concentration were determined by subtracting the lowest breath hydrogen concentration from the first three samples (0, 1, and 2 h) from all subsequent samples. The lowest value of the first three samples was used as the nadir value because some subjects have residual accumulation of hydrogen in the colon while sleeping. This hydrogen is excreted during the first few hours and therefore the 0-h value may not represent the true basal hydrogen concentration. An increase in the breath hydrogen concentration of 10 ppm from the baseline value was considered to be indicative of carbohydrate malabsorption [24].
Calculation of Area under the Curve (AUC)
Positive incremental area under the plasma glucose and serum insulin response curve for two and three hours after each treatment was calculated using the method of Wolever [25]. This method excludes any area that falls below the baseline concentration.
Intolerance Symptoms
Subjective gastrointestinal tolerance symptoms were collected for the first and second 24-h periods after the consumption of each study product. The intensity and frequency of nausea, abdominal cramping, abdominal distention, and flatulence were rated by subjects using a 10-cm visual analog rating scale (0 = absent, 10 = severe; 0 = usual, 10 = more than usual). Subjects made a single mark on a continuous horizontal line to indicate the severity of their symptoms. These scales were completed by the subjects outside of the laboratory.
Stool frequency and consistency information was also collected for the first and second 24-h periods after consumption of each study product. Subjects recorded the time of each bowel movement and stool consistency based upon a 5-point Likert scale (1 = hard/dry, 2 = hard/formed, 3 = soft/formed, 4 = soft/unformed, 5 = watery). This information was collected by the subjects outside of the laboratory.
Statistical Analysis
A power analysis conducted prior to the study indicated that at least 28 subjects (14 in each treatment sequence) would be needed to detect, with 85% power and P < 0.05, a 15% difference in the change from baseline for peak postprandial plasma glucose concentration (nQuery Advisor, Release 4.0).
Statistical analysis was performed on both evaluable and intent to treat data; however, only evaluable subject data are reported. Subjects with missing data for a variable at one or both periods were not included in the analysis for that variable. All data are reported as mean ± SEM. Significance was defined as P < 0.05 for treatment effects or P < 0.10 for sequence effects.
Two steps were carried out for each two-period, two-treatment crossover analysis. The two treatment sequences were first compared for a sequence effect, and then a treatment effect, using a two-sided t-test (or two-sided Wilcoxon Signed Rank test if data was deemed non-normal by the Shapiro Wilk test). If the sequence effect was significant, then the two treatments were compared using only the first period data by a two-sided t-test (or two-sided Wilcoxon Signed Rank test if data was non-normal). For comparison of the proportions of positive breath hydrogen tests with each treatment, the McNemar test for paired samples was used [26].
|
RESULTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSCONCLUSIONREFERENCES |
|---|
|
|
|
10 ppm increase above the basal fasting level) [24]. Similar results were obtained when a 20 ppm increase above fasting [27] criterion was applied. Three hours after subjects consumed GCD, breath hydrogen concentration increased 4 ppm (P = 0.012) compared with MD (Fig. 3). However, the 3-h breath hydrogen concentration for the GCD beverage was only about 2 ppm higher than the corresponding fasting value, an increase which would be insufficient to be categorized as carbohydrate malabsorption by either of the above breath hydrogen criteria.
|
|
|
DISCUSSION |
|---|
Beyond the low glycemic and insulinemic responses described above, it is noteworthy that even the large amount of GCD fed in this study (50 g) did not significantly elevate breath hydrogen excretion over an 8 h postprandial period. This finding suggests that GCD is slowly, but essentially fully, hydrolyzed in the small intestine. Lending further support to this contention is the lack of significant symptoms of flatulence or diarrhea for either 24 or 48 h postprandially. However, there are other possible explanations for the low breath hydrogen excretion from GCD. The first possibility, which is that the subjects have a non-hydrogen producing colonic microflora, was ruled out by our requirement that all subjects had to have a positive breath hydrogen response after lactulose administration. However, it is conceivable that the fermentation of malabsorbed GCD could occur without the generation of appreciable amounts of hydrogen, even in subjects who are hydrogen-producers via the lactulose breath hydrogen test. A study of beta-cyclodextrin administration (30-g dose) in ileostomists showed that 91–97% of the administered dose was recovered in the ileostomy effluent [28]. Further, the same study demonstrated that there was no significant breath hydrogen response to the same dose of beta-cyclodextrin in healthy subjects despite the fact that beta-cyclodextrin was not detected in the feces of these subjects. These findings suggest that malabsorbed beta-cyclodextrin is completely fermented in the colon without elevating breath hydrogen excretion. It is unknown if GCD might be metabolized in a similar way.
The present study demonstrated that GCD effectively reduced postprandial glycemia compared with a rapidly digested MD in healthy adults. People with diabetes mellitus (DM) may be able to safely ingest carbohydrates in the form of GCD without compromising glycemic control. Because of the many co-morbid conditions often suffered by people with DM, such as CVD, gastroparesis resulting from neuropathy, and renal insufficiency [29], a high carbohydrate diet may be more beneficial than a diet with high levels of fat or protein. Calories from GCD could be used to replace some of the calories from fat in medical nutritional products currently marketed for people with DM. Because it is completely digested, GCD will effectively provide needed calories to people requiring medical food supplements.
Treatment of hypoglycemic conditions may be another potential application of GCD. Raw cornstarch, another slowly digested carbohydrate, has been used with some success to treat nighttime hypoglycemia in children with type 1 DM [30]. Raw cornstarch has also been used in the treatment of glycogen storage disease, a condition in which abnormal glycogen storage and accumulation in the liver and other tissues results in a variety of complications including hypoglycemia. Because GCD appeared to retain its slow digestibility even after the heat sterilization of the clinical product fed in this study, this might constitute a definite advantage of GCD compared with raw cornstarch in heat-treated, liquid enteral products because, when cooked, cornstarch loses its structure and becomes rapidly digested [31]. The long-term stability of GCD in a liquid enteral product, however, is still unknown.
Sports nutrition represents another potential application for GCD. Endurance athletes, who require a steady supply of glucose over an extended period of time, may benefit from a slowly digested carbohydrate such as GCD. The blunted insulin response to GCD may also improve the ability to mobilize fatty acids from the adipose tissue during exercise, allowing for conservation of muscle glycogen [32]. However, the rate of gastric emptying of GCD in humans has not yet been determined. Because slow gastric emptying might increase symptoms of nausea during exercise, it will be important to obtain more information on the gastric emptying rate of GCD before its possible incorporation into sports nutrition products.
Critically ill patients, both with and without DM, often suffer from hyperglycemia as a result of the acute phase metabolic response to trauma or injury [33]. Feeding these patients early is important due to their high caloric expenditures. However, this can prove to be immensely challenging for healthcare professionals, because providing the energy needed for their hypermetabolic state may often be done at the expense of glycemic control. Because poor glycemic control is known to inhibit wound healing [33], hyperglycemia is an important issue to address in this patient population. Because GCD results in decreased postprandial glycemia, it may have potential applications in the feeding of critically ill patients.
Potential applications of GCD may extend beyond the treatment of acute hyper- and hypoglycemic conditions. Low glycemic-index (GI) foods have promise in the treatment and prevention of diseases such as CVD, obesity, and metabolic syndrome [6]. Incorporation of low GI foods into the diet serves to decrease postprandial glycemia, prevent overstimulation of insulin secretion by pancreatic ß-cells, and lower the subsequent counterregulatory hormonal response resulting from rebound hypoglycemia [6]. Attenuating the postprandial and postabsorptive hormonal response to foods could potentially 1) increase satiety; 2) decrease fasting levels of insulin, glucose, and FFA; and 3) increase insulin sensitivity [6]. Over the long term, these changes may decrease the risk of many interrelated disease states associated with food intake and nutrient metabolism. Incorporating GCD into meal replacements, meal supplements, or snack foods could offer people with these diseases or at risk to develop these diseases an opportunity to lower the overall GI of their diet and take an active step in improving their health.
Incorporation of GCD into medical nutritional products for people with diabetes and other hyper- or hypoglycemic health conditions, as well as for people at risk of metabolic disease states such as CVD, obesity and metabolic syndrome, deserves strong consideration. Further research should be done to test the effects of GCD in these specific patient populations. Because this study examined the effect of a single 50 g bolus feeding of GCD, further research should look at the efficacy of this oligosaccharide when fed in smaller quantities over several feeding periods, and when fed continuously through a tube feeding. Researchers could also examine the metabolic effects of long-term GCD feeding. Another study such as this one may be designed to collect blood samples over a time period longer than three hours in order to elucidate how long it takes for glucose to return to fasting levels. GCD could also be tested in athletes to see if performance is improved. A satiety rating scale would be beneficial in future studies with GCD, especially in an obese population attempting to lose weight. Finally, determining the GI of GCD is necessary to understand how the glycemic response of GCD compares with other sources of carbohydrate.
|
CONCLUSION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSCONCLUSIONREFERENCES |
|---|
|
FOOTNOTES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSCONCLUSIONREFERENCES |
|---|
Received May 19, 2005. Accepted December 20, 2005.
|
REFERENCES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSCONCLUSIONREFERENCES |
|---|
rdb/opa046.html (Accessed 17 February2004
).
-cyclodextrin affect apparent digestibility and metabolism in healthy adult ileal cannulated dogs. J Nutr135
:1946
–1952,2005
.-cyclodextrin in humans. Food Addit Contam16
:313
–317,1999
.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or gamma-cyclodextrin (•) in a crossover design. Letters represent treatment differences for that time point (P < 0.001). Values are mean ± SEM, n = 31–32.
