The study goals were to (1) establish the variability in postprandial glucose control in healthy young people consuming a mixed meal and, then (2) determine the acute and residual impact of a single exercise bout on postprandial glucose control. In study 1, 18 people completed two similar mixed meal trials and an intravenous glucose tolerance test (IVGTT). There were strong test-retest correlations for the post-meal area under the curve (AUC) for glucose, insulin, and Cpeptide (
The recent increase in obesity and metabolic disorders in young people highlights the need for more effective lifestyle programs to address current and future disease risk [
Less than half of adults and children in the United States meet current physical activity recommendations [
To assess the reliability of the mixed meal test, a preliminary study was conducted with 18 people who completed two meal tests under similar conditions and an intravenous glucose tolerance test (IVGTT) for comparison. The group consisted of 14 females and 4 males who were 13–27 years old (3 children), 11 of whom were considered normal weight and 7 who were overweight based on body mass index (BMI) criteria. Four of the participants were recreationally active; the others were not regularly engaged in sports or exercise more than twice per week. The group was selected to be modestly diverse for age, body composition, and habitual physical activity so that the study outcomes would be generalizable. However, the participants could not have metabolic or other health conditions, or be using medications that would interfere with their safety or study outcomes.
Each participant completed an initial screening visit and three separate outpatient assessments of glucose tolerance/insulin sensitivity. The screening visit began with attainment of informed written consent from adults and consent and assent from each child and their parents in accordance with the university Institutional Review Board, which approved the study. Following a medical exam, body composition was measured using dual energy X-ray absorptiometry (DEXA, Lunar iDXA, GE-Healthcare, Fairfield, CT). The two meal tests and the IVGTT test were performed on three separate mornings at least one week apart at the University of Oklahoma General Clinical Research Center (GCRC). Before each trial, participants were instructed to maintain their normal activity pattern and to follow a consistent mixed diet for 3 days. Daily ambulatory activity during waking hours was recorded for 4 days before each trial with an accelerometer worn above the ankle (StepWatch 3, OrthoCare Innovations, Mountlake Terrace, WA). This monitor records the number of steps each minute and has high reliability and validity [
On the morning of each trial, the participants reported to the GCRC at 07:00 AM following a 10-hour overnight fast. After the participant was quietly settled in a supine position, resting energy expenditure (REE) was measured for 30 minutes on both meal trials using an indirect calorimetry system with a flow-through canopy placed over the head (TrueOne 2400, ParvoMedics, Sandy, UT). An intravenous catheter was then placed in a forearm vein and kept patent with saline infusion for serial blood sampling. The mixed meal, consisting of a chocolate shake made from milk powder, milk cream, and chocolate syrup (2803 kJ, 45/40/15% of energy from carbohydrate/fat/protein, resp.), was consumed within 5 minutes. The start of the meal was designated as time 0 minutes. Blood collections were performed at −15 and −2 minutes (averaged and presented as the 0 minute fasting value), and again at 10, 20, 30, 40, 60, 90, 120, 150, and 180 minutes after the meal. The indirect calorimetry measurement was repeated for the last 20 minutes of each hour after the meal, with the final 15 minutes of each measurement used for analyses. The IVGTT trial was similar, starting with 30 minutes of supine rest, but without REE measurement. Intravenous catheters were placed in both arms for infusion and blood draw, respectively. After baseline blood collections at −15 and −2 minutes, glucose was infused at 0.3 g/kg body mass at 0 minutes and insulin at 0.12 pmol/kg body mass at 10 minutes [
To assess the effect of a single exercise session on meal glucose tolerance, 11 young adults (7 women, 4 men) ages 20–30 years old were recruited from the local community. Participants were eligible if they had not been regularly engaged in organized sports or structured exercise programs for the previous 3 months and were not performing vigorous activity more than 30 minutes per session more than twice per week. Physical activity history was initially assessed by questionnaire and objectively measured prior to each trial with step monitors as in Study 1.
As in Study 1, the initial screening visit began with attainment of informed written consent, followed by a medical history and exam, and body composition measurement using DEXA. A submaximal treadmill walking test was performed to establish steady-state relationships among walking velocity, heart rate and oxygen uptake. Five-minute stages were performed at velocities ranging from 4.0–6.4 km/h at 0% grade. Similarly, submaximal and maximal responses were measured during an incremental workload test to volitional exhaustion on a stationary bicycle (Lode Corival, Groningen, The Netherlands). Three-minute stages were performed at 25, 50, and 75 watts, followed by increments of 15–20 watts/minute until fatigue. Finally, participants played the boxing game on the Nintendo Wii Sports (Nintendo of America, Redmond, WA) interactive video game system for about 10 minutes. During exercise, heart rate was continuously recorded with a chest-strap monitor (Polar Electro USA, Lake Success, NY) interfaced with an expired gas analysis system (Ultima Cardio2, MedGraphics, St. Paul, MN) as previously described [
Each participant returned to the GCRC between 07:00 and 07:30 AM following a 10-hour overnight fast for three morning trials conducted at least one week apart. On one trial, the participants performed no vigorous exercise for at least 3 days prior to the visit (No Ex trial). On a second trial, they completed a 45 minute bout of moderate intensity aerobic exercise in the afternoon, approximately 17 hours prior to the meal test (Prior Day Ex trial). On a third trial, the same exercise session was performed in the morning, after completing the baseline measures and approximately 30 minutes before the meal test (Same Day Ex trial). Thus, the tests were designed to measure the acute (Same Day Ex) and residual (Prior Day Ex) effects of a single exercise session in relation to the habitually low physical activity lifestyle pattern (No Ex) of these participants. Trial order was randomized. The exercise sessions were comprised of 15 minutes each of walking on the treadmill, stationary cycling, and video game boxing. This exercise protocol was selected to be appropriate and fun for novice exercisers, while involving multiple muscle groups. Walking and cycling workloads were adjusted to elicit 75% HRpeak. During boxing the participants were instructed to remain actively engaged in the game. On the Prior Day Ex trial, the exercise session was performed in the afternoon, with a time delay of 16.7 ± 0.1 hours between the end of the exercise and the start of the meal the following morning. On the Same Day Ex trial, the exercise was performed after the resting measurements were completed, with a time delay of 26 ± 2 minutes between the end of the exercise and the start of the meal. Blood collection and REE followed the same schedule as in Study 1.
All blood samples were separated into plasma or serum and stored at −70°C until analysis. Plasma glucose was measured by the glucose oxidase method (2300STAT Plus, Yellow Springs Instruments, Yellow Springs, OH). Serum insulin and C-peptide were measured using Elisa kits from Millipore (St. Louis, MO). Nonesterified fatty acids (NEFAs) were measured in serum with an enzymatic colorimetric assay (Wako Chemicals, Richmond, VA). The postmeal peak and area under the curve (AUC) for each of these outcomes was used for analyses. The glucose and insulin values from the meal tests were also used to calculate the whole body insulin sensitivity index (ISI) described by Matsuda [
For Study 1 the within-subject coefficient of variation (CV) and intraclass correlation for the primary outcomes were calculated according to Hopkins [
Participant characteristics for both studies are presented in Table
Participant characteristics.
Study 1 | Study 2 | |
---|---|---|
Age, y | 24 ± 4 | 26 ± 3 |
Body mass, kg | 72.9 ± 3.9 | 65.8 ± 9.0 |
Height, m | 1.70 ± 0.09 | 1.70 ± 0.10 |
BMI, kg/m2 | 25.0 ± 1.1 | 22.8 ± 0.9 |
Body fat, kg | 24.2 ± 11.4 | 19.3 ± 3.7 |
Body fat, % | 32.4 ± 2.0 | 29.8 ± 6.5 |
Lean mass, kg | 45.3 ± 8.1 | 43.6 ± 8.7 |
Peak bike power, watts | n/a | 145 ± 43 |
Peak VO2, mL/kg/min | n/a | 26.4 ± 5.9 |
Peak heart rate, beats/min | n/a | 179 ± 15 |
Total cholesterol, mmol/L | 3.98 ± 0.99 | 4.03 ± 0.90 |
HDL cholesterol, mmol/L | 1.21 ± 0.37 | 1.11 ± 0.33 |
Triglycerides, mmol/L | 0.74 ± 0.39 | 0.89 ± 0.31 |
C-reactive protein, nmol/L | 18.4 ± 19.9 | 11.0 ± 5.2 |
Glucose, mmol/L | 4.7 ± 0.2 | 4.8 ± 0.5 |
Insulin, pmol/L | 41 ± 28 | 41 ± 27 |
Systolic blood pressure, mmHg | 109 ± 9 | 112 ± 10 |
Diastolic blood pressure, mmHg | 68 ± 8 | 62 ± 7 |
Values are mean ± SD for 14 females and 4 males in Study 1 and 7 females and 5 males in Study 2. Body composition determined by DEXA. Peak exercise responses were measured in Study 2 during a bicycle test to volitional exhaustion (not performed in Study 1). Blood test results are from a fasting sample collected during the first meal test in Study 1 and the No Ex trial in Study 2.
Due to missing data, step activity results were available for only 15 of the participants. The monitors were worn for 14.5 ± 0.6 hours/day for 3.4 ± 0.2 days, recording an average of 10,280 ± 569 steps/day. The amount of time with no or low (<30 steps/minute) activity was 86 ± 1%. There was no difference in activity volume or pattern across trials.
On the two meal trials, there were no differences in the fasting, peak, or time to peak values for glucose, insulin, C-peptide or fatty acids (not shown). The test-retest correlation between meal tests and the CV for the AUCs, respectively, were for glucose:
Insulin sensitivity in young healthy people in Study 1. (a) Correlation between the Matsuda whole body insulin sensitivity index (ISI, arbitrary units) measured during two identical mixed meal tests. (b) Correlation between the minimal model-derived estimate of insulin sensitivity (SI, units = 10−4/min ×
The meal test results were used to calculate the statistical power and sample sizes for future studies. With an expected test-retest correlation coefficient of
There was no difference in basal REE between meal studies (4.51 ± 0.23 versus 4.48 ± 0.22 kJ/min, resp., test-retest
All participants wore the step monitors, although after accounting for missing days the average recording time was 3.5 ± 0.2 days before each test, recording 7635 ± 559 steps over 13.0 ± 0.4 hours per day with no difference across trials. Time spent in no activity or low-intensity activity accounted for 89 ± 1% of daily monitoring time. During the Prior Day Ex and Same Day Ex trials, the average exercise HR was 76 ± 4% and 75 ± 4% of peak, respectively, with no difference between trials or across exercise modes (72, 77, and 78% HRpeak for treadmill, cycling, and boxing resp.). The estimated total energy expenditure during exercise was 1183 ± 41 kJ on both exercise trials.
As shown in Figure
Postmeal responses in glucose, insulin, C-peptide, and nonesterified fatty acids in Study 2. Values shown as mean ± SEM for 11 people. *No Ex different from Same Day Ex trial; †Prior Day Ex different from Same Day Ex,
Area under the curve for glucose, insulin, C-peptide, and fatty acids during the post meal period. Values shown as mean ± SEM for 11 people. *No Ex different from Same Day Ex trial; †Prior Day Ex different from Same Day Ex,
The average baseline REE (Figure
Postmeal responses in energy expenditure and fuel oxidation. Energy expenditure and the relative carbohydrate (CHO) oxidation were increased throughout the postmeal period relative to baseline but did not differ among trials.
The goals of this investigation were to establish the reliability of a mixed meal test for assessing glucose tolerance and insulin action, and to measure the acute and residual impact of a single session of endurance exercise on meal glucose tolerance in habitually sedentary, but healthy young adults. Results of the first study demonstrated acceptably high test-retest reliability for postprandial glycemic and insulinemic responses and strong correlation between the meal ISI and IVGTT SI, while the main finding of the second study was that meal glucose control was improved following a moderate intensity exercise session compared to a trial without prior exercise. However, the beneficial effect of exercise was evident only when completed within an hour before the mixed meal test but not when the participants exercised the day prior (~17 hours) to the meal. Thus, for habitually sedentary young adults, insulin sensitivity was acutely responsive to a volume and intensity of exercise that is consistent with current recommendations for adults. The finding that the improvement was transient, however, highlights the importance of engaging in frequent exercise to promote metabolic health.
The results of Study 1 demonstrated that the postprandial glucose and insulin responses to the mixed meal test had acceptably high reproducibility between tests, and validity when compared to the IVGTT. Thus, the mixed meal can be used to assess the impact of exercise with a less-intensive method than the IVGTT or euglycemic hyperinsulinemic clamp but is perhaps more physiological than the oral glucose tolerance test [
In Study 2, the exercise energy expenditure of ~1180 kJ during the 45-minute session was lower than typically used (~1250–3350 kJ) in prior single exercise session studies performed with adults [
In animals and humans insulin and noninsulin mediated pathways for glucose uptake in skeletal muscle are activated for several hours after exercise, although the specific pathways and how they are regulated have not been fully elucidated [
Additional factors responsible for variation among studies that measured postexercise insulin sensitivity may be the type and amount of nutrients consumed between the end of exercise and the measurement of insulin action, and the characteristics of the group under study, such as their age, anthropometric differences, medical or physical fitness status, and family history of diabetes. For example, recent work in humans and prior studies in rodents showed that the magnitude and/or duration of the postexercise stimulation of insulin sensitivity can be enhanced by consuming a low-carbohydrate diet and/or maintaining a short-term energy deficit [
In summary, the results of the present study demonstrate that young adults with low aerobic fitness and habitually low physical activity respond to a single moderate intensity bout of exercise with an acute improvement in insulin sensitivity when measured within 3 hours of exercise with a mixed meal test, but that this effect is no longer evident when measured 17 hours after exercise. Whether and how the magnitude of this response can be enhanced by modification of the exercise conditions, diet, or other manipulations is not yet known but could have important effects on the health of sedentary people. These results highlight the need to encourage young adults to engage in daily moderate-to-vigorous physical activity and the immediate beneficial impact of exercise on metabolic function.
The authors have nothing to declare.
This project was supported by Oklahoma Centers for Advancement of Science and Technology grant number HR07-156S to K. R. Short, and National Center for Research Resources Grants P20-RR024215 to K. R. Short and M01-RR14467 to the University of Oklahoma Health Sciences Center General Clinical Research Center. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health. The authors are grateful for the excellent nursing and dietary support provided by Ashley Roof, Tawney Pantoya, and others in the OUHSC GCRC, the data collection efforts of Karah Sanchez, Jennifer Tolbert, and Courtney Young, the clinical assistance of Ryan Brown and Evynn Boss, and most importantly, the children and adults who participated in these studies.