If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
This study characterized the glucose and insulin responses to temporal alterations in meal frequency, and alterations in the macronutrient composition.
Methods
Eight subjects underwent three separate 12-h meal tests: three high carbohydrate (3CHO) meals, 6 high carbohydrate meals (6CHO), 6 high-protein meals (6HP). Blood samples were taken at 15-min intervals. Integrated area under the curve (AUC) concentrations for glucose and plasma insulin were determined (total, 4-h, and 2-h blocks) for each meal condition.
Results
Baseline glucose and insulin values were not different between study days. Peak glucose levels were highest on the 3CHO day; however the 12 h glucose AUC was higher during the 6CHO condition (p = 0.029) than 3CHO condition, with no difference in the insulin response. The 6HP condition resulted in a decreased glucose AUC (p = 0.004) and insulin AUC (p = 0.012) compared to 6CHO.
Conclusions
In non-obese individuals, glucose levels remained elevated throughout the day with frequent CHO meals compared to 3CHO meals, without any differences in the insulin levels. Increasing the protein content of frequent meals attenuated both the glucose and insulin response. These findings of elevated glucose levels throughout the day warrant further research, particularly in overweight and obese individuals with and without type 2 diabetes.
Experimental studies have demonstrated beneficial anthropometric and metabolic adaptations with increased meal frequency, including decreased body weight
Previous studies have examined the fasting hormonal and metabolic responses to alterations in meal frequency with meal tests or oral glucose tolerance tests following dietary acclimatization periods ranging from two
Jenkins and colleagues reported that when healthy individuals consumed a nibbling diet for 2 weeks (17 small meals per day), they produced ∼28% less insulin in response to a subsequent test meal than during an isocaloric gorging diet (3 meals per day).
demonstrated that fasting glucose levels were increased, while insulin levels decreased in healthy individuals who consumed meals as 9 snacks every 2 h for 2 weeks compared to eating 3 meals at 7 h intervals. Using an 8 wk cross-over design, Carlson et al.
noted no changes in fasting levels of insulin, leptin, and ghrelin with alterations in meal frequency (1 meal/day vs. 3 meals/day), but those eating a single meal per day had exhibited elevated fasting glucose levels. Changes in the insulin response following acute changes in meal ingestion patterns are also reported.
have demonstrated that there was no difference in the total glucose area under the curve (AUC) in response to the ingestion of five small, high carbohydrate (70% CHO) meals as compared to one large, isocaloric meal. The serum insulin concentrations rose to higher levels with a single meal than compared to the first of the small pre-load meals, and insulin levels were higher following the fifth meal compared to the single meal at the same time point into the trial. Only one study to date has followed the glucose and insulin levels during frequent eating past 5 h. Solomon et al.
demonstrated no difference in the total insulin AUC in response to high carbohydrate (64% CHO) foods consumed as either two or twelve meals when studied over an 8 h period.
Confounding the meal frequency research is meal composition, as composition will impact postprandial glucose and insulin responses. The substitution of protein for carbohydrate has been shown to decrease glucose and insulin responses in both the short-
Consideration of meal frequency on glucose and insulin excursions is critical as the lay press and many clinicians promote increased meal frequency for weight loss or glucose control often with no consideration of meal composition.
Therefore, the purpose of this study was to establish the day-long glucose and insulin patterns of response to meals provided either three or six times per day, and of varying macronutrient composition. In contrast to previous research, this study used frequent blood sampling to track glucose and insulin concentrations to three and six subsequent nutrient ingestions. It was hypothesized that infrequent high carbohydrate (3CHO) meals would result in a greater total insulin response, while frequent, high carbohydrate (6CHO) meals would result in lower peak insulin levels but more sustained insulin concentrations. We anticipated that frequent, high-protein (6HP) meals would provide a lesser glucose insulin response than the 6CHO meals.
2. Subjects and methods
Participants in this study included eight young (18–35 years old), healthy, non-obese individuals (4 men, 4 women). Exclusion criteria included glucose intolerance, weight loss or gain in the prior 3 months, or current medications that could alter glucose tolerance. Participants were homogenous in their training status; training no more than five times per week. Female subjects were tested within the first eight days of their menstrual cycle to minimize the effect of estrogen on glucose tolerance. All participants were required to complete an informed consent document, approved by the Syracuse University Institutional Review Board prior to taking part in this study.
On an introductory visit, each participant's habitual dietary intake and meal frequency, as well as general health and physical activity levels were recorded through the use of questionnaires. Body composition was assessed using air-displacement plethysmography (BODPOD system, Life Measurement, Inc. Concorde, CA, USA) according to manufacturer's specifications. Subsequently, each participant reported to the Human Performance Lab on three separate occasions for 12 h of meal testing, beginning at 0700 h. Participants were fasted and free from caffeine consumption for 12 h, and free from alcohol consumption and exercise for at least 24 h.
Upon arrival to the lab on testing days, participants had a catheter inserted in their antecubital vein by a registered nurse. Baseline blood samples were drawn prior to the ingestion of the first meal. Participants received their meal conditions in a randomized, counterbalanced order, blinded to meal composition. Liquid meals were chosen for this experimental protocol, as previous work has shown that alterations in the macronutrient composition of liquids does not affect gastric emptying rate.
The energy-matched meal conditions consisted of 6276 kJ (15% protein (PRO), 65% carbohydrate (CHO), 20% fat (FAT)), consumed in evenly spaced intervals as either 3 large meals (3CHO; ∼2093 kJ/meal) or 6 small meals (6CHO; ∼1046 kJ/meal). Wegmans Nutritional Beverage (Wegmans, Rochester, NY, USA) was used for this study and consisted primarily of sucrose and corn syrup (CHO), soy, and whey (PRO). The high-protein condition (45% PRO, 35% CHO, 20% FAT) was consumed as 6 meals (6HP; ∼1046 kJ/meal), over the 12-h period. The additional protein added in this condition consisted of Pro Complex whey protein and branched-chain amino acids which also contains a small amount of fat (∼2 g) sufficient to balance the fat component of the dietary conditions at 20% (Pro Complex, Optimum Nutrition Inc., Aurora, IL, USA). All of the subjects participated in quiet, sedentary activities including reading, studying, and watching movies, with minor physical activity (e.g. walking to restroom).
Blood samples were obtained every 15 min, with a small fraction used to measure whole blood glucose status (YSI 2300 STAT PLUS glucose–lactate analyzer, YSI Incorporated, Yellow Springs, OH, USA). The remaining blood was separated by centrifuge and stored at −80 °C until subsequent analysis. Plasma samples for every half hour of testing were assayed in duplicate for insulin concentrations using a commercially available 125I radioimmunoassay assay (Diagnostics Products Corporation-DPC; Los Angeles, CA, USA). The insulin interassay coefficient of variation was 11%, and the intraassay coefficient of variation was 7.6%.
Integrated area under the curve (AUC; 12 h) for insulin and glucose across meal conditions were calculated as the area above baseline using the trapezoidal method (GraphPad Prism version 3.00 for Windows, GraphPad Software, San Diego, California, USA). In addition, 2 h AUC blocks for the 6CHO and 6HP conditions were calculated, as well as 4 h AUC blocks in each of these conditions. A one-way ANOVA with repeated measures was used to assess the differences in the 12 h AUC for glucose and insulin concentrations between meal conditions. A two-way ANOVA with repeated measures (meal condition – 6CHO vs. 6HP; time – 2 h blocks) was utilized to establish changes in AUC over time with subsequent meals and study day. Significance levels in all statistical tests was accepted at α = 0.05. Statistical analyses were performed with SPSS for Windows, version 16.0 (SPSS Inc., Chicago, USA), and all data are reported as mean ± standard deviation.
3. Results
All of the subjects were in their early twenties (23.6 ± 3.4 y), with an average BMI of 24.4 ± 2.6 kg/m2, with small differences noted between the sexes. Male participants had a lower % body fat than females (12.1 ± 7.6% vs. 30.7 ± 5.7%, P < 0.01). A normal range of glucose tolerance was established for all participants during the first glucose response to the 3CHO condition, where the fasting glucose levels were (70.9 ± 14.5 mg/dL) and the mean 2 h blood glucose level was (67.2 ± 12.6 mg/dL). Dietary records showed that subjects consumed energy-containing foods on 3.5 ± 0.7 occasions per day, with no significant differences between sexes. Subjects performed an average of 4.5 ± 1.8 sessions of light to moderate physical activity per week, with no significant differences between males and females. No differences were noted between sexes, or with regard to habitual meal frequency in the glucose or insulin responses to the meal stimuli.
Baseline glucose and insulin values were similar across the three study days. The pattern of response for glucose and insulin levels is in Fig. 1. These subjects demonstrated significantly increased blood glucose levels during the 6CHO (710.0 ± 251.0 mmol/L∗min) compared to both the 3CHO (522.7 ± 99.3 mmol/L∗min; p = 0.029) and 6HP (442.1 ± 121.0 mmol/L∗min; p = 0.008) meal conditions when assessed as 12 h AUC (Fig. 2a). The 12 h insulin AUC were higher in both the 3CHO (103,974.0 ± 32,123.9 pmol/L∗min; p = 0.05) and 6CHO (80,807.9 ± 27,266.0 pmol/L∗min; p = 0.012) conditions compared to the 6HP (54,373.0 ± 16,115.7 pmol/L∗min) condition (Fig. 2b). Similarly, the 12-hour insulin AUC in the 6CHO condition resulted in a lower insulin release than the isocaloric 3CHO condition (p = 0.05). When divided into 4 h blocks, glucose AUC was higher in the 6CHO condition (p = 0.026) compared to the 3CHO condition, although no difference was found in the insulin values. The 6HP condition had the smallest insulin response to meal ingestion (p = 0.018 vs. 3CHO; p = 0.007 vs. 6CHO). Finally, when comparing the 2 h net AUC values between the 6CHO and 6HP conditions, we observed a main effect for condition, such that the 6CHO treatment resulted in higher blood glucose levels than 6HP (p = 0.004; Fig. 3a), paralleled by a lower insulin response during the 6HP condition (p = 0.012 vs. 6CHO; Fig. 3b). Interestingly during the third and fourth 2 h blocks of these frequent meal conditions, there was a decrease in the insulin AUC (0700–0900 > 1100–1300 p = 0.004; 0700–0900 > 1300–1500 p = 0.015; Fig. 3b), compared to the initial 2 h block.
Fig. 1The pattern of response for glucose and insulin levels throughout the day. The arrows indicate the times of the meals - ↓3 meals; 6 meals.
Fig. 2Integrated 12 h area und the curve for glucose and plasma insulin concentrations. ∗P < 0.05 vs. 3CHO; †P < 0.05 vs. 6CHO; ‡P < 0.05 vs. 3CHO. CHO – high carbohydrate, HP – high protein.
Fig. 3Integrated 2 h area under the curve for glucose and insulin levels for the 6 meal/day. ∗P < 0.001 6CHO vs. 6HP; †P < 0.01 vs. 0700–0900 > 1100–1300; ‡P < 0.05 vs. 0700–0900. CHO – high carbohydrate, HP – high protein.
The present study is one of the first to investigate glucose and insulin excursions in response to altered meal frequency and macronutrient composition in healthy young adults over a 12 h period. Our primary finding is that consumption of 6 frequent meals in 12 h resulted in higher blood glucose levels over the course of the day than the consumption of 3 meals, although there was no difference in the insulin response between these two conditions. During frequent meal conditions (2 h AUC), an attenuated insulin response was noted with subsequent meal consumption in the middle of the day (1100 h and 1300 h), despite the fact that glucose AUC did not change significantly. Similar to previous reports a higher protein intake with frequent meal consumption decreases both the insulin and glucose response as compared to a meal higher in CHO.
There has been considerable promotion both by the medical community and the lay press to consume 6 meals per day for weight loss or for glycemic control but our data indicate that the glucose AUC is ∼30% higher over the course of the day with a frequent high carbohydrate feeding than when consuming 3 meals per day. This could potentially have profound implications for individuals with glucose intolerance or those with type 2 diabetes, and should be studied further in this population. Further we observed that the total insulin response was not dramatically different between these two study days. While our results do not align with the reported responses to single meal protocols,
who reported that there were no differences in the 8 h insulin area under the curve between the isocaloric ingestion of 2 meals compared to 12 meals. A recent meta-analysis of the effectiveness of increasing meal frequency on reducing cardiovascular risk proposes that 6 or more ‘nibbling’ meals per day may in fact lower circulating cholesterol levels due to a reduction in insulin release when compared to a ‘gorging’ pattern,
similar to the reduced insulin response seen with subsequent small nutrient ingestions during our study.
When the meals were consumed with increased frequency whether high in protein or CHO, the insulin response in the middle of the study days was lower than that observed previously in the morning. This occurred despite the fact that the glucose levels were still elevated. This finding is in agreement with earlier work
Postprandial oscillatory patterns of blood glucose and insulin in NIDDM. Abnormal diurnal insulin secretion patterns and glucose homeostasis independent of obesity.
which demonstrated in healthy controls that insulin concentrations were higher after breakfast than after lunch. The reverse was observed in individuals with type 2 diabetes.
Though there is a paucity of research concerning alterations in the macronutrient profile of frequent meals, evidence points to the fact that substitution of dietary carbohydrate with protein during frequent meal consumption may have a beneficial effect on the insulin response. Whitley and colleagues
found an inverse relationship between the amount of carbohydrate in an isoenergetic meal and fat metabolism, which may reflect the outcome of a decreased insulin response to a reduced carbohydrate load. A long-term intervention study designed to induce weight loss found that a frequent (6-meal), high-protein (40% PRO, 40% CHO, 20% FAT) diet was more effective than either isocaloric frequent, high carbohydrate (15% PRO, 60% CHO, 25% FAT) or infrequent (3-meal), high-protein diet at reducing both total and abdominal fat mass.
The reduction in the insulin response found with frequent meal consumption, and particularly with high-protein content, reinforces the findings of previous studies. An increase in fat metabolism with frequent, high-protein meals has also been demonstrated previously, as even small increases in plasma insulin concentrations are known to suppress fat metabolism greatly.
In conclusion, in lean, healthy individuals an increase in meal frequency with a high carbohydrate meal results in augmented glucose levels throughout the day without a corresponding increase in insulin levels. Since elevated glucose levels are associated with pathological conditions, including impaired glucose tolerance
further research into the potential effects of increasing meal frequency, in overweight and obese individuals with and without type 2 diabetes is warranted. Additionally the substitution of dietary protein for dietary carbohydrate in this population resulted in an attenuated glucose and insulin response, and those individuals who are encouraged to eat more frequent meals should consume more dietary protein, particularly if controlling glucose levels is of interest.
Conflict of interest
The authors have nothing to disclose in the way of financial or personal relationship with other people or organizations that could inappropriately influence (bias) this work.
Statement of authorship
Michael Holmstrup: conception and design of the study, acquisition of data, or analysis and interpretation of data, drafting the article or revising it critically for important intellectual content, final approval of manuscript.
Christopher Owens: the conception and design of the study, acquisition of data, or analysis and interpretation of data, drafting the article, final approval of manuscript.
Timothy Fairchild: design of the study, data analysis and interpretation of data, critically revising the manuscript for intellectual content, final approval of manuscript.
Jill Kanaley: conception and design of the study, acquisition of data, or analysis and interpretation of data, revising the article critically for important intellectual content, final approval of manuscript.
Acknowledgements
This project was supported in part by NIH grant R21DK063179. Sincere thanks goes to Rose Kingbury, RN, NP for the placement of all of the catheters.
References
Cross A.T.
Babicz D.
Cushman L.F.
Snacking patterns among 1,800 adults and children.
Postprandial oscillatory patterns of blood glucose and insulin in NIDDM. Abnormal diurnal insulin secretion patterns and glucose homeostasis independent of obesity.
6 meals.
00054-5&id=gr1.jpg){kind=link}
00054-5&id=gr2.jpg){kind=link}
00054-5&id=gr3.jpg){kind=link}