Effects of glucose ingestion or glucose infusion on fuel substrate kinetics during prolonged exercise

To determine if bypassing both intestinal absorption and hepatic glucose uptake by intravenous glucose infusion might increase the rate of muscle glucose oxidation above 1 g · min^–1, ten endurance-trained subjects were studied during 125 min of cycling at 70% of peak oxygen uptake (VO_{2 peak}). During exercise the subjects ingested either a 15 g · 100 ml^–1 U-¹⁴C labelled glucose solution or H₂O labelled with a U-¹⁴C glucose tracer for the determination of the rates of plasma glucose oxidation (Rox) and exogenous carbohydrate (CHO) oxidation from plasma¹⁴C glucose and¹⁴CO₂ specific activities, and respiratory gas exchange. Simultaneously, 2-³H glucose was infused at a constant rate to measure glucose turnover, while unlabelled glucose (25% dextrose) was infused into those subjects not ingesting glucose to maintain plasma glucose concentration at 5 mmol · l^–1. Despite similar plasma glucose concentrations [ingestion 5.3 (SEM 0.13) mmol · l^–1; infusion 5.0 (0.09) mmol · l^–1], compared to glucose infusion, CHO ingestion significantly increased plasma insulin concentrations [12.9 (1.0) vs 4.8 (0.5) mU · l^–1;P<0.05], raised total Rox values [9.5 (1.2) vs 6.2 (0.7) mmol · 125 min^–1 kg fat free mass^–1 (FFM);P<0.05] and rates of CHO oxidation [37.2 (2.8)vs 24.1 (3.9) mmol · 125 min^–1 kg FFM^–1;P<0.05]. An increased reliance on CHO metabolism with CHO ingestion was associated with a decrease in fat oxidation. Whereas the contribution from fat oxidation to energy production increased to 51 (10)% with glucose infusion, it only reached 18 (4)% with glucose ingestion (P<0.05). Despite these differences in plasma insulin concentration and rates of fat oxidation, the rates of glucose oxidation by muscle were similar after 125 min of exercise for both trials [ingestion 93 (8); infusion 85 (5) mol · min^–1 kg FFM^–1], suggesting that peak rates of muscle glucose oxidation were primarily dependent on blood glucose concentration which, in turn, regulated the hepatic appearance of ingested CHO.     

Oxidation of exogenous carbohydrate during prolonged exercise: the effects of the carbohydrate type and its concentration

Summary We studied rates of exogenous carbohydrate (CHO) oxidation during 90 min of cycling exercise in trained cyclists exercising at 70% of maximal oxygen consumption (VO_2max) when they ingested glucose, sucrose, or glucose polymer solutions at concentrations of 7.5%, 10% or 15%. Drinks were labelled with [U-¹⁴C]glucose or sucrose and were ingested at a rate of 100 ml · 10 min^–1. Rates of oxidation of the ingested CHO were calculated from the specific radio-activity of the labelled CHO, expired¹⁴CO₂ and carbon dioxide output (VCO₂). Total CHO oxidation, determined from oxygen consumption andVCO₂ was not influenced by CHO type or concentration. Gastric emptying (P=0.01) and the rate of exogenous CHO oxidation (P=0.028) was greatest for the glucose polymer solutions, and least for glucose. Although gastric emptying (P=0.006) decreased with increasing CHO concentration, CHO delivery to the intestine and exogenous CHO oxidation increased linearly with increasing CHO concentration. The percentage of the CHO delivered to the intestine that was oxidized ranged from 30.0% for 7.5% CHO to 38.1% for 15% CHO. Our results indicated that the rate of gastric emptying for CHO was not controlled to provide a constant rate of energy delivery as is commonly believed and that factors subsequent to gastric emptying limit the rate of exogenous CHO oxidation from the ingested solution.     

Fuel kinetics during intense running and cycling when fed carbohydrate

On two occasions, six well-trained, male competitive triathletes performed, in random order, two experimental trials consisting of either a timed ride to exhaustion on a cycle ergometer or a run to exhaustion on a motor-driven treadmill at 80% of their respective peak cycling and peak running oxygen (VO_2max) uptakes. At the start of exercise, subjects drank 250 ml of a 15 g·100 ml^–1 w/v [U-¹⁴C]glucose solution and, thereafter, 150 ml of the same solution every 15 min. Despite identical metabolic rates [VO₂ 3.51 (0.06) vs 3.51 (0.10) 1·min^–1; values are mean (SEM) for the cycling and running trials, respectively], exercise times to exhaustion were significantly longer during cycling than running [96 (14) vs 63 (11) min; P < 0.05]. The superior cycling than running endurance was not associated with any differences in either the rate of blood glucose oxidation [3.8 (0.1) vs 3.9 (0.4) mmol· min^–1], or the rate of ingested glucose oxidation [2.0 (0.1) vs 1.7 (0.2) mmol· min^–1] at the last common time point (40 min) before exhaustion, despite higher blood glucose concentrations at exhaustion during running than cycling [7.0 (0.9) vs 5.8 (0.5) mmol·1^–1; P < 0.05]. However, the final rate of total carbohydrate (CHO) oxidation was significantly greater during cycling than running [24.0 (0.8) vs 21.7 (1.4) mmol C₆·min^–1; P < 0.01]. At exhaustion, the estimated contribution to energy production from muscle glycogen had declined to similar extents in both cycling and running [68 (3) vs 65 (5)%]. These differences between the rates of total CHO oxidation and blood glucose oxidation suggest that the direct and/or indirect (via lactate) oxidation of muscle glycogen was greater in cycling than running.     

Single and choice reaction time during prolonged exercise in trained subjects: influence of carbohydrate availability

The aim of the present study was to examine the effects of prolonged exercise at the ventilatory threshold and carbohydrate ingestion on single (SRT) and choice (CRT) reaction time. Eight well-trained triathletes completed three testing sessions within a 3-week period. Maximal oxygen uptake was determined in the first test, whereas the second and the third sessions were composed of a 100-min run (treadmill 15 min, overground 70 min, treadmill 15 min) performed at the velocity associated with the ventilatory threshold. During these submaximal tests, the subjects ingested (in random order) 8 ml·kg^–1 body weight of either a placebo (Pl) or 5.5% carbohydrate (CHO) solution prior to the first submaximal run and 2 ml·kg^–1 body weight every 15 min after that. The cognitive tasks were performed before and after exercise for CRT, and before, during each submaximal run and after exercise for SRT. Furthermore, at the end of each submaximal test subjects were asked to report their rating of perceived exertion (RPE). Results showed a significant positive effect of CHO ingestion on RPE and CRT performance at the end of exercise, while no effect of exercise duration was found in the Pl condition. After a 100-min run, during the CHO condition, CRT mean (SD) group values decreased from 688.5 (51) ms to 654 (63) ms, while during the Pl condition, RPE mean group values increased from 11 (2) to 16 (1.02) and CRT mean values remained stable [688 (104) ms vs 676 (73.4) ms, P>0.05]. No similar effect was observed for SRT. These results suggest that CHO-electrolyte ingestion during a100-min run results in an improvement in the complex cognitive performance measured at the end of that run. Electronic Publication     

Metabolic effects of testosterone during prolonged physical exercise and fasting

Summary Previous studies have shown a decrease in plasma testosterone during prolonged physical exercise and 72 h fasting in rats.To determine whether this hormonal change has an influence upon energy metabolism, two experiments were carried out, in which the plasma levels of testosterone were elevated during prolonged physical exercise and fasting in male wistar rats.The effects of acute and chronic increases in the levels of circulating testosterone were studied, on the one hand after human chorionic gonadotropin (H.C.G.) injection, and on the other by prolonged testosterone perfusion with an osmotic minipump. Blood and tissue sampling were performed to evaluate blood glucose, alanine, and lactate, and tissue glycogen. The results in fed and rest control rats showed no changes in blood parameters under the effect of hypertestosteronemia but there was an increase in muscle glycogen after testosterone perfusion.In 72 h fasted rats both types of hypertestosteronemia were associated with a decrease in blood alanine and lactate ranging from 25% to 35%. Only testosterone perfusion was associated with higher concentrations of muscle glycogen.After 7 h of treadmill running, testosterone perfusion and H.C.G. injection induced a 35% decrease in blood alanine and a slight decrease in blood glucose, with no change in other parameters.Whereas an elevation in the level of testosterone can induce muscle glycogen compensation in the fed resting state, it cannot counteract the exhaustion of muscle glycogen during running.     

Hormonal responses during a prolonged military field exercise with variable exercise intensity

The purpose of the present study was to test the hypothesis that the magnitude of hormonal concentration alterations during a prolonged military field exercise with constant energy intake (EI) is influenced by changes in energy deficit (ED) induced by varying the exercise intensity. Basal serum hormone concentrations were measured in a group of healthy young male volunteers (n = 7) during a 20-day field exercise. During the first week of the exercise, the average ED was 4,000 kcal/day (P-I), in the second week only 450 kcal/day (P-II), and in the last week 1,000 kcal/day (P-III). During the first 5 days of the field exercise, significant increases in cortisol (COR, +32%) and growth hormone (GH, +616%) concentrations were observed, while insulin (INS, −70%), total testosterone (TES, −27%), free testosterone (TES_free, −26%) decreased. However, after these initial responses, COR and GH returned to the pre-exercise level by the beginning of P-II. Also TES and TES_free recovered to the pre-exercise level by the beginning of P-III, and INS by the end of P-III. The concentration of TES (+29%) increased above the pre-exercise level by the beginning of P-III. Serum thyroxin (T₄) concentration was significantly lesser (−12%) and urine urea concentration significantly higher (+78%) after the field exercise than before it. Therefore, it can be concluded that the lower levels of ED in the second and third phase (ED <1,000 kcal/day) allowed recovery of hormonal changes observed in the first phase with ED much greater than 1000 kcal/day.     

Psychomotor performance during prolonged exercise above and below the blood lactate threshold

Previous investigations from this laboratory have demonstrated that during graded exercise with exercise intensities increasing every 3 min until exhaustion the multiple choice reaction time (RT) decreased until the intensity exceeded the lactate threshold (LT) by approximately 25%, and then rapidly increased. The aim of this study was to follow up changes in RT during prolonged exercise at constant intensities above and below LT and to relate these changes to changes in venous blood lactate [La⁻]_b, and plasma catecholamine [CA]_pl concentration responses to the exercise. For this purpose eight young soccer players exercised for 20-min on a cycle ergometer at 10% above LT, and nine exercised for 60 min at an intensity 30% below LT. During both tests RT, heart rate (HR), as well as [La⁻]_b, and [CA]_pl were measured. Above LT, RT decreased from the 5th min until the end of exercise, whilst HR, [La⁻]_b, and [CA]_pl increased progressively. Significant inverse correlations were ascertained between RT and plasma adrenaline (r = − 0.651) and noradrenaline concentrations (r = − 0.678). During exercise below LT, RT decreased up to approximately 40 min, then it reached a nadir, and stabilized at this level. This was accompanied by only small changes in [La⁻]_b and [CA]_pl. The present findings would indicate that young athletes are able to maintain for a relatively long time, or even increase, their psychomotor performance during endurance exercise both below and above the LT. Accepted: 23 June 1997     

The significance of an increased RQ after sucrose ingestion during prolonged aerobic exercise

Summary Four well-trained male subjects worked for periods of 6 h on bicycle ergometers at work loads requiring about 47% of their maximal aerobic capacity. In one series of studies they received only water; in a second series they received 100 g of sucrose containing 100 c U-C¹⁴-labelled sucrose at the beginning of the fourth hour of work. In a third series of experiments, the same subjects received 100 g of non-labelled sucrose at the beginning of the fourth hour. During the experiment without U-C¹⁴-labelled sucrose, blood samples were withdrawn and analysed for glucose, lactate and pyruvate content. Data from C¹⁴O₂ recovery in expired air showed a good correlation with the amount of carbohydrate oxidized during the sucrose experiment. Peak values for the respiratory exchange ratio showed the same time response as those observed for the C¹⁴O₂ in the expired air. It is concluded that the observed rise in RQ after sucrose ingestion, under the conditions studied, is of metabolic origin, resulting from a complete conversion of pyruvate to CO₂.     

Exogenous 13C glucose oxidation during exercise: North American vs Western European studies

Summary The purpose of this study was to test the hypothesis that the well-documented changes in background ¹³C enrichment of expired CO₂ observed in response to exercise and carbohydrate ingestion, in subjects living on a North American diet, are not present in subjects living on a Western European diet. The experimental protocol used by Pirnay et al. in 1977 and by Krzentowski et al. in 1984 in subjects living on a Western European diet (4 h of exercise on a treadmill at 50% VO_2max with ingestion of 100 g of glucose in 400 ml of water) was duplicated as closely as possible in six subjects living on a North American diet. The actual amounts of exogenous glucose oxidized, computed with a high artificial ¹³C enrichment of glucose (+189.7 ¹³C PDB-1) which allows one to neglect the 1–2 changes in ¹³C background, were [mean (SEM)] 54.7 (5.4) and 84.2 (3.4) g over 2 h and 4 h of exercise, respectively. These values compare well with data computed by Pirnay et al. [56.6 (13.1) and 94.9 (4.2) g] and by Krzentowski et al. [55.0 (6.2) and 88.0 (4.5) g] using a natural enrichment of glucose (–11.21 and –10.63 ¹³C PDB-1, respectively) assuming no change in ¹³C background in their Western European subjects. Under the same assumption and using a natural enrichment of glucose (–11.30 ¹³C PDB-1) the oxidation of exogenous glucose was overestimated by 30–40% in our North American subjects. This result indicates that because of a lower input of ¹³C in their diet, the difference between the isotopic composition of carbohydrate and fat stores are smaller, and changes in ¹³C background are small or absent in response to moderate workload in Western European subjects, when compared to their North American counterparts.     

Caffeine modifies blood glucose availability during prolonged low-intensity exercise in individuals with type-2 diabetes

Objective:

The study investigated the effect of supplementation with maltodextrin (CHO) alone or associated to caffeine during exercise in T2DM subjects.

Methods:

Pilot study, using eight subjects with T2DM, aged 55±10 years, received CHO (1 g/kg) or caffeine (1.5 mg/kg) alone or associated before exercise protocol. The exercise was executed at 40% heart rate (HR) reserve for 40 min, with 10-min recovery. Blood pressure (BP) and perceived exertion scale (Borg) were checked every 2 min. Blood glucose (BG) was checked every 10 min. For statistical analysis, ANOVA test was used and the value was considered statistically significant at p <0.05.

Results:

The results showed that BP and HR did not change significantly among all treatments. Caffeine promoted a significant reduction in BG of 75 mg/dL (65%, p <0.05) during 40 min of exercise protocol compared to all groups.

Conclusion:

Supplementation with 1.5 mg/kg of caffeine reduces BG concentration during prolonged exercise in T2DM patients.     

Effects of glucose ingestion at the onset of moderate-intensity, prolonged exercise in women as compared to men

To test glucose tolerance during exercise, the effects of oral glucose ingestion (0.5 g · kg⁻¹) on plasma glucose and hormonal responses (insulin, catecholamines) were investigated in 11 women [mean (SEM) age 21.6 (1.3) years] and 10 men [22.0 (0.3) years] during cycle ergometer exercise (30 min at 60% maximum oxygen consumption, V˙O_2max). The two groups exhibited similar V˙O_2max values, when expressed per kg of lean body mass. Venous blood samples (5 ml) were withdrawn immediately before the exercise, during the exercise (at 3, 5, 10, 15 and 30 min) and at the 30th min of the recovery period. Glucose was ingested orally between the 2nd and the 3rd min of the exercise. As compared to men, plasma glucose concentrations were lower in women during exercise (P < 0.05 at 3, 15 and 30 min) and at the 30th min of the recovery period (P < 0.001), while plasma insulin concentrations were higher in women during exercise (P < 0.05 at 3, 15 and 30 min). The ratio of the area under the curve for glucose over the area under the curve for insulin was lower in women during exercise (P < 0.0002). A linear relationship between glucose and insulin concentrations was found only for women during exercise (r = 0.615, P < 0.0001). No gender difference was observed for the catecholamine concentration during exercise. In conclusion, this study postulates that an oral glucose load given at the onset of a prolonged and moderate exercise bout induced lesser plasma glucose and greater insulin concentrations in women as compared to men. These data argue in favour of a greater glucose tolerance in women during exercise. Accepted: 5 June 1999     

Metabolic and circulatory responses to the ingestion of glucose polymer and glucose/electrolyte solutions during exercise in man

Summary Six men exercised on a cycle ergometer for 60 min on two occasions one week apart, at 68±3% of . On one occasion, a dilute glucose/electrolyte solution (E: osmolality 310 mosmol · kg^–1, glucose content 200 mmol·l^–1) was given orally at a rate of 100 ml every 10 min, beginning immediately prior to exercise. On the other occasion, a glucose polymer solution (P: osmolality 630 mosmol · kg^–1, glucose content equivalent to 916 mmol · l^–1) was given at the same rate. Blood samples were obtained from a superficial forearm vein immediately prior to exercise and at 15-min intervals during exercise; further samples were obtained at 15-min intervals for 60 min at rest following exercise. Heart rate and rectal temperature were measured at 5-min intervals during exercise.Blood glucose concentration was not different between the two tests during exercise, but rose to a peak of 8.7±1.2 mmol · l^–1 (mean±SD) at 30 min post-exercise when P was drunk. Blood glucose remained unchanged during and after exercise when E was drunk. Plasma insulin levels were unchanged during exercise and were the same on both trials, but again a sharp rise in plasma insulin concentration was seen after exercise when P was drunk. The rate of carbohydrate oxidation during exercise, as calculated from and the respiratory exchange ratio, was not different between the two tests. A fall in plasma volume, calculated from changes in haematocrit and haemoglobin concentration, occurred after 15 mins of exercise: the fall was of the same magnitude (9%) at this point on both tests, but thereafter plasma volume was significantly lower with P than with E for the remainder of the exercise period and throughout recovery. Serum osmolality increased during exercise (p<0.05) on the P trial, but was unchanged on the E trial. Heart rate was higher (p<0.05) during the last 20 min of exercise on the P trial.These results suggest that the carbohydrate consumed during the P trial was not available to the working muscles during exercise, and was probably not emptied from the stomach and absorbed to any significant extent until exercise stopped. The differences in plasma volume and osmolality between the two trials are consistent with the net movement of water into the gut which is known to occur at rest when solutions of high osmolality are taken. In more prolonged exercise, this effective dehydration may impair performance.     

Glucose ingestion before and during exercise does not enhance performance of daily repeated endurance exercise

Summary The effect of glucose (Glc) ingestion before and during daily, repeated, prolonged exercise on metabolism and performance was tested. Seven young, healthy males performed cycling exercise in two series, with 1 month interval. Each exercise series consisted of 1 h/day on 3 successive days. On the 3rd day, exercise was continued until exhaustion. The intensity was 73.4 (7.7) % [mean (SD)] of maximal oxygen uptake ( ). Glucose (Glc) or placebo (P) drink was ingested 15 min before the start, and at 15 and 45 min of each daily exercise. The total amount of Glc ingested was 43.1 (4.2) g. During exercise, blood Glc concentrations were significantly higher (P<0.05) when Glc was ingested than when P was ingested [Glc 5.14 (0.32) and P 4.12 (4.17) mmol · 1^–1 at exhaustion]. However, Glc ingestion did not improve performance time to exhaustion [Glc 92.05 (29.55) and P 98.07 (27.33) min]. Free fatty acid concentrations were significantly lower when Glc was ingested than when P was ingested [Glc 0.63 (0.21) and P 1.39 (0.46) mmol · l^–1 at exhaustion]. There were no significant differences in exercise heart rate, , respiratory exchange ratio, blood lactate concentrations or rating of perceived exertion between the conditions nor were there any significant differences in these parameters on different days of exercise. It seems that ingestion of small amounts of Glc does not increase the metabolism of carbohydrate or improve the performance of intensive endurance exercise of poorly trained subjects, even when the exercise is repeated daily.     

Metabolic changes induced by combined prolonged exercise and low-calorie intake in man

Summary Thirteen middle-aged women and 10 men walked 344 km during 7 days. The daily walking distances were 57, 53, 67, 53, 41, 36, and 37 km at an average speed of 3.5 km·h^–1. During the hike the subjects drank water, mineral drinks, and juices ad libitum. Except for some natural products, no food intake was allowed. During the hike the body weight and serum protein concentration of the subjects decreased by about 7%, on average. Serum triglyceride and total cholesterol decreased drastically, about 30–40% during the hike, but HDL-cholesterol showed a tendency to increase, giving a 40% increment in HDL/total cholesterol ratio. Serum free fatty acids rose 1.5–2 times above the starting level. Serum glucose and evening insulin levels decreased significantly during the hike. Serum cortisol in evening samples after the daily walking and plasma norepinephrine concentrations were significantly increased, reflecting the immediate daily response to the combined fasting and walking. Serum testosterone levels decreased in men but not in women, indicating the involvement of the LH-testis pathway in the decrease obtained. Serum ASAT activity rose to about three times the starting level during the hike, whereas -GT activity gradually decreased. These marked metabolic changes caused by combined fasting and several days exercise were in many respects (as in cholesterol, HDL/total cholesterol ratio or testosterone levels) more pronounced than those earlier reported to be caused by exercise or fasting alone.     

Human adipose tissue blood flow during prolonged exercise II

Subcutaneous and perirenal adipose tissue blood flows (ATBF) were measured by the¹³³Xe washout method before, during and after 4 h exercise on a bicycle ergometer. The load corresponded to about 50% of max (i.e. about 1.7l/min). Subcutaneous and perirenal ATBF increased at an average to 3–400 and 700% of their initial control values respectively. In six of nine measuring sites ATBF remained increased in the hour following work. During work plasma glycerol concentrations increased 8 fold. The core temperature increased 0.9°C, skin temperature did not change significantly. During passive elevation of body temperature (core temperature +1.5°C; skin temperature +3°C) neither subcutaneous ATBF nor plasma glycerol concentrations changed significantly. It is concluded that the increase in subcutaneous ATBF during exercise is not a reaction to increased body temperature.     

Oxidation of [13C]glucose ingested before and/or during prolonged exercise

Ingestion of glucose before exercise results in a transient increase in plasma insulin concentrations. We hypothesized that if glucose was also ingested during the exercise period the elevated plasma insulin concentration could increase exogenous glucose oxidation. The oxidation rate of glucose ingested 30 min before (50 g) and/or during (110 or 160 g in fractionated doses) exercise [120 min; 67.3 (1.2)% maximal O₂ uptake] was studied on six young male subjects, using ¹³C-labelling. Ingestion of glucose before exercise significantly increased plasma insulin concentration [from 196 (45) to 415 (57) pmol l^–1] but the value returned to pre-exercise level within the first 30 min of exercise in spite of a continuous increase in plasma glucose concentration. Ingestion of glucose 30 min before exercise did not increase the oxidation of exogenous glucose between minutes 30 and 60 during the exercise period [0.36 (0.03) vs 0.30 (0.02) g min^–1, when placebo or unlabelled glucose was ingested respectively]. Over the last 90 min of exercise, when glucose was ingested only during exercise, 49.2 (3.1) g [0.55 (0.04) g min^–1) was oxidized, while when it was ingested both before and during exercise, 65.7 (4.6) g [0.73 (0.05) g min^–1] was oxidized [26.7 (2.1) g of the 50 g ingested before exercise but only 39.0 (2.4) g of the 110 g ingested during the exercise period]. Thus, ingestion of glucose 30 min before the beginning of exercise did not enhance the oxidation rate of exogenous glucose ingested during the exercise period, although the total amount of exogenous glucose oxidized was larger than when ingested only during the exercise period.     

Effects of prolonged physical exercise and fasting upon plasma testosterone level in rats

Summary Prolonged physical exercise and fasting in male rats were studied to determine the effect of these two treatments on plasma testosterone level. Blood and tissue samples were drawn after 1 h, 3 h, 5 h, and 7 h treadmill running, and after 24 h, 48 h, and 72 h of fasting. Both treatments resulted in a significant fall in plasma testosterone, plasma luteinizing hormone (LH), plasma Insulin (IRI) and in liver and muscle glycogen stores. In the course of these two treatments the injection of a supra maximal dose of Human Chorionic Gonadotropin (HCG) produced a rise in plasma testosterone similar to that in control rats. This indicates that the decrease of plasma LH may be responsible for the decrease in plasma testosterone, which is time-related with the decrease in glycogen stores. The possible metabolic role of the decrease in plasma testosterone is discussed.     

The effects of a high carbohydrate diet on postprandial energy expenditure during exercise in rats

Summary Whether or not a high intake of carbohydrate increases postprandial energy expenditure during exercise was studied in rats. The rats were meal-fed regularly twice a day (0800–0900 hours and 1800–1900 hours) on either a high carbohydrate (CHO) (carbohydrate/fat/protein = 70/5/25, % of energy) or high fat (FAT) (35/40/25) diet for 12 days. On the final day of the experiment, all of the rats in each dietary group were fed an evening meal containing equal amounts of energy (420 kJ · kg^–1 body mass). After the meal, they were divided into three subgroups: pre-exercise control (PC), exercise (EX), and resting control (RC). The PC-CHO and PC-FAT groups were sacrificed at 2030 hours. The EX-CHO and EX-FAT groups were given a period of 3-h swimming, and then sacrificed at 2330 hours. The RC-CHO and RC-FAT groups rested after the meal and were sacrificed at 2330 hours. Total energy expenditure during the period 1.5 h from the commencement of exercise was higher in EX-CHO than in EX-FAT. The respiratory exchange ratio was also higher in EX-CHO than in EX-FAT, suggesting enhanced carbohydrate oxidation in the former. Compared with both PC-FAT and RC-FAT, the liver glycogen content of EX-FAT rats was significantly decreased by exercise. On the other hand, the liver glycogen content of both EX-CHO and RC-CHO was higher than that of PC-CHO rats. The glycogen content of soleus muscle of EX-FAT was slightly decreased during exercise, however, that of EX-CHO increased significantly. Thus postprandial energy expenditure during exercise was higher in the rats fed the CHO diet than in those fed the FAT diet, which could have been related to the increase of both liver and muscle glycogen storage during exercise in the former.