INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE ABILITY TO SUSTAIN prolonged aerobic exercise is determined to a large extent by fuel availability. It has been shown that maintenance of euglycemia and carbohydrate oxidation late in exercise can delay fatigue, which implies that carbohydrate intake before and/or during exercise may be a key to prolonging the duration of aerobic exercise activities (3, 7). However, data on metabolic and exercise performance measures associated with preexercise dietary carbohydrate intake are equivocal. Previous studies have shown a decrease (20), no effect (9, 12, 16, 19), or an increase (4, 15) in glycogen utilization when carbohydrate is ingested 30-60 min before initiation of aerobic exercise activities. Data on exercise performance after preexercise carbohydrate meals are equally unclear. Investigators have reported decreased (13), unchanged (9), or enhanced (7, 27, 29, 31) exercise performance after preexercise carbohydrate meals. Some studies have been limited by the lack of control of previous physical activity and dietary intake, both of which may have a profound effect on the metabolic and performance outcomes.

Furthermore, studies examining the effects of preexercise carbohydrate intake on performance and metabolic responses during exercise have used foods and/or fluids that have a relatively high glycemic index (GI), i.e., a glucose response close to 100 when indexed against glucose or white bread standard. Consequently, qualitative and quantitative differences in blood glucose responses may have been overlooked for some types and sources of carbohydrate. Some studies examining the effects of preexercise carbohydrate feeding with different GIs have shown a positive effect on physiological parameters, with the low-GI food resulting in favorable substrate levels during exercise (29, 30). However, the effects of dietary fiber and the influence of the GI of a meal on exercise performance require further investigation under conditions of controlled diet and activity levels.

The purpose of this study was to determine the effects of two moderate-GI breakfast cereals with different amounts of dietary fiber, combined with complex and simple carbohydrate, on the metabolic response to prolonged moderate-intensity exercise. The influence of preexercise feedings of these cereals on exercise time to exhaustion was also determined. We hypothesized that breakfast cereals with added dietary fiber, particularly in the form of soluble fiber, would produce a reduced glycemic and insulinemic response. The lower glycemic response would provide longer-lasting glucose availability during submaximal exercise, whereas a reduced insulinemic response might decrease the rate of glycogen utilization. A more sustained glucose availability could maintain carbohydrate oxidation when glycogen stores are low, thus leading to enhanced exercise capacity.

	SUBJECTS AND METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects. Six recreationally active women of college age volunteered to participate in the study (Table 1). The study was approved by the Institutional Review Board for Human Subjects, and all participants signed an informed consent in accordance with the Pennsylvania State University guidelines for the protection of human subjects. Initial screening included an evaluation of exercise training and menstrual and smoking status. Each subject performed an incremental semirecumbent cycle ergometer test to determine peak O₂ consumption (O_{2 peak}). A standard 75-g oral glucose tolerance test was performed to verify normal glucose tolerance. Body density was determined by hydrostatic weighing after an overnight fast according to the method of Akers and Buskirk (1). Underwater weight was determined using electronic load cells. Residual lung volume was determined during immersion by open-circuit nitrogen washout, and percent body fat was estimated using the equation of Siri (28). Height was measured to the nearest 0.1 cm without shoes. Body weight was measured to the nearest 0.1 kg with the subject wearing underclothing and a hospital gown.

Residency and dietary control. To control physical activity and dietary intake, subjects lived in the General Clinical Research Center for 2 consecutive days and 3 nights before each trial. During days 1 and 2 the diet consisted of normal foods and beverages, with 60% of energy from carbohydrate, 25% from fat, and 15% from protein. On day 2 the subjects consumed all their food before 8:00 PM. On day 3, after collection of baseline data and 45 min before the start of exercise, the subjects ate one of two test meals containing 75 g of available carbohydrate. A third control trial was performed with 300 ml of water alone (Con). The test meals consisted of sweetened whole-grain rolled oats (SRO) or sweetened whole-oat flour (SOF) and 300 ml of water. The meals were similar in carbohydrate, fat, and protein content but differed in dietary and soluble fiber content. In addition, viscosity measurements have shown that, on an equal weight basis, an SOF meal generates ~50% less viscosity than an SRO meal, and the GI for these foods is ~60-70 (unpublished observations). The composition of the SRO meal included 75 g of available carbohydrate, 4.5 g of fat, 9.1 g of protein, 6.8 g of dietary fiber, and 2.3 g of soluble fiber. The SOF meal contained 75 g of available carbohydrate, 4.7 g of fat, 9.4 g of protein, 3.1 g of dietary fiber, and 1.6 g of soluble fiber. Meals and Con were provided in a random order, with only one investigator aware of the test meal for each trial.

Exercise protocol. The exercise trials were performed on a semirecumbent cycle ergometer at an intensity that corresponded to 60% of O_{2 peak}. Each subject exercised to exhaustion, defined as the time at which the pedaling revolutions per minute declined to <90% of the designated cadence. Subjects received a monetary incentive of $250 for the first 90 min of exercise and $1/min for each subsequent minute that they continued to ride. Exercise time was recorded to the nearest second. During the exercise trial, each subject drank at least 250 ml of water per 0.5 h to ensure adequate hydration status (22). Heart rate was monitored continuously by means of radiotelemetry (UNIQ, Computer Instruments).

Blood analyses. On the morning of the test day an indwelling Teflon catheter was placed in an antecubital vein for blood sampling. Blood samples were drawn before the test meal, at 15, 30, and 45 min after the meal, and at 30-min intervals during exercise until exhaustion. Plasma glucose concentration was measured immediately by the glucose oxidase method (Beckman Instruments, Fullerton, CA). Blood samples for hormone and substrate measurements were centrifuged at 4°C and stored at 70°C for subsequent analysis. Samples for insulin were assayed in duplicate by a double-antibody radioimmunoassay (24). Blood samples for epinephrine and norepinephrine determination were collected in tubes containing reduced glutathione and ethylene glycol-bis(-aminoethyl ether)-N,N,N,N-tetraacetic acid. The samples were analyzed by high-performance liquid chromatography (Waters) using electrochemical detection based on a modification of the method described by Hjemdahl et al. (17). Serum free fatty acids (FFA) were determined using an enzymatic colorimetric procedure (NEFA C kit, Wako Chemicals, Dallas, TX). Glycerol concentrations were measured according to a modification of the method of McGowan et al. (23) using an enzymatic colorimetric procedure [triglyceride (GPO-Trinder) procedure 337, Sigma Chemical, St. Louis, MO].

Muscle analyses. Muscle biopsies were performed using the needle biopsy procedure, as previously described (10, 11). Tissue was taken from the vastus lateralis muscle of one leg before consumption of the test meal or Con and from the opposite leg immediately after the subject reached exhaustion. The samples were immediately frozen in liquid nitrogen for subsequent analysis of total muscle glycogen content (26). A second piece was prepared for histochemical analysis of muscle fiber type. The muscle sample was mounted in tragacanth gum and quickly frozen in isopentane cooled in liquid nitrogen. All muscle samples were stored at 70°C. Total muscle glycogen content was determined in duplicate on freeze-dried samples. Samples were weighed, acid hydrolyzed, and neutralized with NaOH, and the glucose concentration in each hydrolysate was measured by enzymatic fluorometry (21). A total of 15 serial cross sections (10 µm) of each muscle sample were cut at 20°C in a cryostat microtome (Cryocut 1800, Leica) and mounted on slides. The sections were dried at room temperature and stained for myosin adenosinetriphosphatase (preincubation at pH 4.3 and 4.6). An average of 300 muscle fibers were identified as types I and II on the basis of the adenosinetriphosphatase stain.

Exercise energy expenditure. Exercise O₂ consumption (O₂), CO₂ production, and respiratory exchange ratios (RER) were determined by indirect calorimetry during the first 30 min of exercise and at the time points that corresponded to blood sampling during the remaining period of each exercise trial. Gas volumes were measured with a dry gas meter (Parkinson-Cowan). Concentrations of O₂ and CO₂ were measured on an electrochemical O₂ analyzer (model S-3A, Applied Electrochemistry) and infrared CO₂ analyzer (model LB-2, Beckman), respectively. Carbohydrate oxidation was calculated from O₂ and RER data.

Statistics. Values are means ± SE. Differences between dependent variables were examined with repeated-measures analysis of variance. Specific mean differences were identified with a Newman-Keuls post hoc test. The  level for statistical significance was set at 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Exercise performance. Subjects exercised at a similar percentage of O_{2 peak} in all three trials: 60.5 ± 1.4, 60.1 ± 1.4, and 59.9 ± 0.8% for SRO, SOF, and Con trials, respectively. O₂ remained constant throughout each exercise bout and at the point of exhaustion was not different between any of the trials (Table 2). RER were similar for all trials during the first 60 min of exercise. At 90 and 120 min of exercise, RER was significantly higher (P < 0.05) for both test meals than for the Con trial, suggesting a greater reliance on carbohydrate as an energy source (Table 2). From 120 min to exhaustion there were no differences in RER values between any of the trials. Heart rate remained steady throughout the exercise bout, and there were no differences between trials or between trials at the point of exhaustion (Table 2).

Substrate and hormone measures. There was a significant increase in plasma glucose and insulin after the two test meals were ingested compared with the Con trial (Fig. 1). The glucose response was elevated for the first 30 min after the meal (P < 0.05) but was not significantly higher than Con at 45 min. Forty-five minutes after the meals, insulin concentrations were still significantly higher (P < 0.05) than during the Con trial (Fig. 2). The insulin response 45 min after the meal was significantly lower (P < 0.05) for the SRO than for the SOF meal. Areas under the insulin and glucose response curves were not significantly different between the two test meals.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Plasma glucose concentrations at rest and throughout exercise in response to meals [sweetened rolled oats (SRO) and sweetened oat flour (SOF)] or control (Con) trials. EXH, exhaustion. * Significantly different from Con, P < 0.05.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Plasma insulin concentrations at rest and throughout exercise in response to SRO, SOF, and Con trials. * Significantly different from Con, P < 0.05; f signficantly different from SOF, P < 0.05.

Muscle glycogen and carbohydrate oxidation. Premeal muscle glycogen concentrations were similar for all three trials: 553 ± 97, 500 ± 52, and 442 ± 92 mmol/kg dry wt for the SRO, SOF, and Con trials, respectively. At the end of each exercise bout there was a significant depletion of glycogen in the vastus lateralis muscle: 153 ± 56, 193 ± 43, and 152 ± 30 mmol/kg dry wt for the SRO, SOF, and Con trials, respectively. Total muscle glycogen utilization was not different for any of the trials: 266 ± 13, 251 ± 12, and 225 ± 8 mmol/kg dry wt for the SRO, SOF, and Con trials, respectively. Total carbohydrate oxidation was not significantly different among trials: 529 ± 40, 489 ± 32, and 440 ± 29 g for the SRO, SOF, and Con trials, respectively. Likewise, there were no significant differences in carbohydrate oxidation rates at the end of exercise among the three trials: 1.94 ± 0.08, 1.90 ± 0.08, and 1.89 ± 0.08 g/min for the SRO, SOF, and Con trials, respectively.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The metabolic and performance benefits resulting from preexercise nutritional supplementation in the hour before exercise are unclear (5). Data from this study show that ingestion of a breakfast cereal with a moderate GI 45 min before exercise can increase exercise time to exhaustion. Many endurance sports begin in the early morning to reduce the physiological stresses resulting from extreme environmental conditions including heat and humidity. Early morning exercise forces participants to address questions related to whether, when, or what they should eat before the event. The selection of appropriate nutrition to optimize energy storage and energy availability during the event is an important variable that can influence performance and can be controlled by the athlete. In the present study the extended exercise time after the SRO meal was observed in healthy young women who were not highly trained athletes, and thus the data have an application beyond elite athletic performers to participants in recreational endurance activities such as cycling, backpacking, or cross-country skiing. The data extend previous reports that indicate that the ingestion of low-glycemic carbohydrate foods before exercise is associated with enhanced exercise performance (29).

The amount of energy stored as glycogen is of great significance, because depletion of muscle glycogen during exercise coincides with muscular fatigue (6). Nutritional supplementation with carbohydrate before exercise has been shown to have an unpredictable effect on exercise performance and muscle fatigue (12, 13, 15, 16, 25, 27). Foster et al. (13) demonstrated that consumption of 75 g of glucose 45 min before relatively high-intensity exercise (84% O_{2 peak}) was associated with decreased time to exhaustion compared with a mixed meal or control. The decreased performance was attributed to an increased rate of carbohydrate utilization and decreased rate of lipid mobilization after carbohydrate ingestion. More recently, a number of studies (6, 25, 27, 31) have shown that exercise performance may be enhanced by carbohydrate feedings that prevent a hypoglycemic response and increase availability of glucose to the working muscles.

Unlike many of the reports on preexercise nutritional supplementation, the preexercise meals in the present study were whole foods that contained various amounts of dietary fiber. Jenkins et al. (18) ascribed a ranking, known as the GI, to foods on the basis of a comparison of the glucose response of the test food with the glucose response after the ingestion of 50 g of glucose. Theoretically, a meal with a reduced GI will result in a slower release of glucose from the gut to the circulating blood, increase glucose availability late in the exercise period, and thus delay the development of hypoglycemia. Although both test meals had equal amounts of carbohydrate, the SRO meal had characteristics that are generally associated with lowering the GI, i.e., greater soluble fiber content and increased viscosity. Also, the method used to process the grain can affect the glycemic response. Golay et al. (14) demonstrated that, when beans were processed by different methods into two physical forms, one maintaining the integrity of the bean cell and the other rupturing the cell, the insulinemic response was significantly lower after the undamaged beans were eaten. Thus, processing may also help explain some of the differences observed between the SRO and SOF exercise response. Although there was no difference in performance time between the test meals, performance was improved only after the SRO meal compared with the Con trial. From the data, we suggest that there may be an optimal combination of carbohydrate, dietary fiber, viscosity, and GI that results in maximizing exercise performance.

Few studies have examined the effects of altering the GI of foods provided before exercise. Thomas et al. (29) were among the first to report a longer endurance time after a lentil meal with a low GI (GI = 29) than after a potato meal with high GI. The improvement was attributed to the attenuated glucose and insulin response to the low-GI meal before exercise and to the maintenance of plasma glucose during exercise. However, there was no significant difference in time to exhaustion among lentils, glucose, and water. A subsequent study by Thomas et al. (30) suggested a positive correlation between the GI of a meal and glucose availability during exercise, such that a 10-unit difference in GI yielded a plasma glucose difference of 0.2 mmol/l at the end of exercise. Our data show no differences in glucose concentrations late in exercise. Indeed, the plasma glucose concentrations were slightly lower at exhaustion after the SRO meal than in the SOF and Con trials. However, the SRO meal was associated with a longer time to exhaustion, which suggests that there was sufficient glucose to provide energy to the working muscle during the extended exercise time.

In the present study both test meals resulted in significant hyperglycemia and hyperinsulinemia before the exercise bout. However, the insulin response was significantly lower after the SRO meal than after the SOF meal. Hyperinsulinemia persisted for the first 30 min of exercise during the SOF trial, although insulin concentrations were somewhat higher for both meal trials until ~90 min of exercise. At the onset of exercise there was a sharp drop in plasma glucose during the SRO and SOF trials. The drop may be attributed to a combination of increased insulin-mediated glucose disposal due to hyperinsulinemia, increased exercise-mediated glucose disposal, and suppression of hepatic glucose production (2, 8). We did not observe the levels of hypoglycemia reported in previous studies (4, 13). The magnitude of the hypoglycemia is individually regulated by the pancreatic insulin response to the glucose load. The subjects in the present study did not demonstrate a marked pancreatic insulin response and, consequently, did not experience hypoglycemia. Both test meals did, however, blunt the lipolytic response to exercise, and plasma FFA concentrations were significantly reduced for the first 60 min of exercise during the SRO trial and the first 90 min of exercise during the SOF trial. Because the blunted lipolytic response persisted for less time during the SRO trial, there appeared to be greater fat utilization, an observation that is supported by the observed trends in RER data (Table 2). It is possible that greater fat utilization during the SRO trial conserved carbohydrate for the later stages of exercise and may have been responsible for the enhanced performance associated with this trial.

Carbohydrate ingestion before exercise has been shown to have a variable effect on glycogen utilization (4, 12, 16, 19, 20). Costill et al. (4) demonstrated that, when FFA concentrations are increased by heparin, muscle glycogen utilization is reduced. However, when glucose (75 g) was ingested 45 min before exercise, hyperinsulinemia coupled with enhanced insulin sensitivity and decreased hepatic glucose production caused a rapid fall in blood glucose concentrations to a hypoglycemic state, which persisted until the end of exercise and caused a significantly greater rate of glycogen utilization. Data from the present study do not directly support the hypothesis that a preexercise carbohydrate meal can reduce glycogen utilization. However, muscle biopsies were obtained at the end of exercise when the subjects were exhausted and glycogen was depleted. Because the duration of exercise was greater for the SRO trial and some of the energy for muscle contraction would have been derived from muscle glycogen, it is possible that, if biopsies had been obtained at the same time point toward the end of exercise for each trial, we may have observed differences in glycogen concentration. Improvements in exercise performance after preexercise carbohydrate supplementation have also been attributed to increased carbohydrate oxidation late in exercise due to a sparing of muscle glycogen and/or a greater contribution of exogenous glucose to fuel use during the exercise bout (7, 27, 31). We did not observe a significantly greater total carbohydrate oxidation during the exercise bout, although there was a 20% difference in total carbohydrate oxidation between the SRO and Con trials.

In summary, we have observed a significant improvement in exercise time when a moderate-GI meal is ingested 45 min before cycle ergometer exercise. The metabolic response to the meal and during exercise did not provide clear evidence regarding the mechanisms associated with the improved performance. Possibly, an attenuated insulin response and reduction in the antilipolytic suppression of FFA release may have influenced substrate oxidation during the first half of the exercise period and may have spared endogenous glucose. When these factors are combined with the possible delayed emptying of ingested glucose from the gut, there may have been a glucose-sparing effect that facilitated sustained energy production toward the end of exercise. These data support the ingestion of a small meal with a high dietary fiber content and moderate GI before prolonged aerobic exercise.