(5) showed that muscle glycogen content after 2 hr of cycling at. 70% of V0,peak ... Twelve well-trained male endurance cyclists were recruited to participate in this investigation after being ..... 26:1772-1777, 1967. 2. Bergstrom, J. .... 24. Sherman, W.M., D.L. Costill, W.J. Fink, F.C. Hagerman, L.E. Armstrong, and T.F.. Murray.
International Journal of Sport Nutrition, 1995, 5, 25-36 O 1995 Human Kinetics Publishers, Inc.
The Effects of Carbohydrate Loading on Muscle Glycogen Content and Cycling Performance Laurie H.G. Rauch, Ian Rodger, Gary R. Wilson, Judy D. Belonje, Steven C. Dennis, Timothy D. Noakes, and John A. Hawley This study compared the effects of supplementing the normal diets of 8 endurance-trained cyclists with additional carbohydrate (CHO), in the form of potato starch, for 3 days on muscle glycogen utilization and performance during a 3-hr cycle ride. On two occasions prior to the trial, the subjects ingested in random order either their normal CHO intake of 6.15 f 0.23 gl kg body masslday or a high-CHO diet of 10.52 f 0.57 g/kg body masslday. The trial consisted of 2 hr of cycling at -75% of V02peak with five 60-s sprints at 100% VOzpeak at 20-min intervals, followed by a60-min performance ride. Increasing CHO intake by 72 9% for 3 days prior to the trial elevated preexercise muscle glycogen contents, improved power output, and extended the distance covered in 1 hr. Muscle glycogen contents were similar at the end of the 3-hr trial, indicating a greater utilization of glycogen when subjects were CHO loaded, which may have been responsible for their improved cycling performance.
+
Key Words: diet, potato starch
Three to four days prior to marathon or ultramarathon races many endurance athletes reduce their training volume and ingest a high (>7 g/kg body masslday) carbohydrate (CHO) diet. The "glycogen-loading" regimen typically increases the resting muscle glycogen content from -100 to >I40 mmol/kg wet wt and extends the time to exhaustion during prolonged (>3 hr) submaximal exercise (3). Conversely, dietary and exercise manipulations that reduce preexercise glycogen stores are associated with an impaired ability to continue exercise at a given percentage of peak oxygen uptake (V0,peak) (3). More recent studies have shown that while CHO loading may increase endurance at a given work rate, it does not improve the speed that can be maintained during high-intensity (>70% of V02peak) exercise lasting either 2 hr (9) or 80 min (25). However, athletic performance 52 hr is unlikely to be limited The authors are with the Bioenergetics of Exercise Research Unit of the Medical Research Council and the University of Cape Town, Department of Physiology, University of Cape Town Medical School, Observatory 7925, South Africa.
26 / Rauch, Rodger, Wilson, et a/.
by the availability of muscle glycogen; starting muscle glycogen stores are more likely to influence athletic performance in more prolonged (23 hr) exercise. Bosch et al. (5) showed that muscle glycogen content after 2 hr of cycling at 70% of V0,peak declined to 89 mmolkg wet wt in glycogen-loaded subjects compared to 44 mmolkg in subjects who were nonloaded. Accordingly, the primary aim of this study was to compare the effects of initial muscle glycogen concentration on the speed attained during the last hour of a 3-hr high-intensity cycle ride involving a series of sprints, to mimic the way cyclists would typically race (20). A further purpose of this investigation was to determine whether a single CHO supplement, potato starch, could be successfully employed for glycogen loading without producing the gastrointestinal discomfort that athletes often experience after consuming the large quantities of pasta, rice, and bread required to achieve sufficient CHO intake (17).
Materials and Methods Subjects and Preliminary Testing
Twelve well-trained male endurance cyclists were recruited to participate in this investigation after being fully informed of the nature and risks and having given their written consent. All had been involved in endurance training for 3-5 years and competed regularly in local and national cycle races. Unfortunately, 2 of the 12 subjects experienced such severe gastrointestinal distress with CHO loading that they failed to complete the trial, while another 2, who also complained of diarrhea, had to be excluded because their starting muscle glycogen contents were not different before and after CHO loading (114 vs. 100 and 133 vs. 138 mmolkg, respectively). The characteristics of the 8 subjects who successfully completed all phases of the investigation were as follows: age = 22.4 k 0.6 years; mass = 71.3 k 1.4 kg; V0,peak = 4.72 f 0.15 Llmin, or 66.3 f 1.3 ml/kg/min; peak sustained power output (PPO) = 376 rt 12 W; the ratio of PPO to mass = 5.28 f 0.11 Wkg; and peak heart rate (HR,,,,) = 191 k 2 beatslmin. The PPO, V02peak,and HR,, values were obtained from maximal exercise tests performed on an electronically braked cycle ergometer (Lode, Gronigen, Holland), modified to the configuration of a racing bicycle with adjustable saddle heights and handlebar positions. In this test, the subjects started cycling at an exercise intensity equivalent to 3.33 W k g body mass for 150 s; thereafter, the work rate was increased by 50 W for the next two 150-s workloads and then by 25 W every 150 s until the subjects were exhausted (12). PPO was defined as the highest exercise intensity the subject completed plus the proportion of the final exercise intensity that he sustained, as described by Kuipers et al. (16). V0,peak data were calculated from each subject's PPO (12) and were used to determine the work rate correspondingto -75% of V02peak, for use in the experimental trials. HR,, values were recorded with a Polar heart rate monitor (Polar Electro OY, Kempele, Finland). In addition to completing the maximal exercise test, all subjects completed a familiarization ride on a Kingcycle cycle simulator (Kingcycle Ltd., High Wycombe, Bucks, U.K.). This ergometer allows subjects to ride their own racing bicycles. Bicycles were attached by the front fork to the frame of the simulator,
Carbohydrate Loading / 27
with the bottom bracket resting on an adjustable support arm. By raising or lowering the bottom bracket support arm,we could adjust the rolling resistance of the rear tire on an air-braked flywheel to match the resistance experienced by a 70-kg cyclist on a level road. From the rolling resistance and the output of a photo-optic sensor monitoring the revolutions of the flywheel, an ~ ~ ~ - c o m ~ a t i b l e computer calculated the power output (W) that would be generated by a 70-kg cyclist riding at that speed (kmhr) on level terrain and the distance (km) that would be covered. Power outputs at given speeds were calculated using the following equation: W = 0.000136 RPS3+ 1.09 RPS, where W is watts and RPS is the revolutions per second on the flywheel. ~djustmentsof the tension on the bottom bracket to produce a rolling resistance of a 70-kg cyclist were achieved by a series of "rundown" calibrations before each trial, during which the subject accelerated to a work rate of -300 W and then immediately stopped pedalling, while remaining seated on the bike in the time-trial position. These calibrations were repeated until the computer display indicated that the slowing of the flywheel matched a reference power decay curve for a 70-kg cyclist. Preliminary data on the test-to-test reliability of the Kingcycle cycle simulator have shown that the mean and standard deviation of the coefficient of variation of the time taken for 5 subjects who each completed three 40-km time trials were 1.11 0.58% (G.S. Palmer, unpublished observations).
+
Dietary Analysis
Prior to the experimental trials, a 3-day dietary record, including 1 day of a weekend, was obtained from each subject. Subjects were given precise written and verbal instructions on how to record all fluid and food consumed and were requested to note the frequency of feedings and any ingestion of extra vitamin or mineral supplements. The nutritional composition of each subject's diet was determined by a commercial computer program (Food Finder Diet Analysis, Medtech, Tygerberg, Cape Town, South Africa) to estimate the subject's normal dietary CHO intake and the quantity of additional CHO ingestion required to supercompensate the muscle glycogen stores (9, 24). Dietary Manipulation and Experimental Trials
All subjects completed a random order of two experimental trials, separated by a minimum of 4 days, during which they either ingested their normal diets (Norm trial) or CHO loaded (Load trial) for 3 days prior to exercise testing. CHO loading was achieved by having subjects supplement their habitual diets with potato starch (Roquette Freres, Lestrem, France). This starch was derived from raw, spray-dried, whole potatoes and was 78% CHO by weight. In an attempt to induce a range of elevated preexercise muscle glycogen contents, the CHO intakes of the subjects were adjusted to between 8 and 12 g of CHO/kg body masslday for the 3 days prior to the Load trial. This was achieved by randomly supplementing half the subjects' habitual diets with -3 g starchkglday and the other half with -5 g starch/kg/day. The potato starch was made more palatable by adding 20 g starch/100 ml chocolate-flavored milkshake mix. In order to reduce gastrointestinal disturbances, subjects were instructed to ingest the starch mix in small amounts (not
28
/ Rauch, Rodger, Wilson, et a/.
more than 250 m1/2 hr) throughout the day. Apart from the addition of starch supplement, subjects did not change the composition or quantity of food they normally ate. This was done to minimize the dietary intervention that subjects had to undergo. During both trials dietary records were recorded to aid compliance. On the morning of the trials, subjects came to the laboratory 3 hr after a standardized breakfast that was similar in content and composition to that which they would normally ingest before competition (7040 g of CHO). While the subjects rested, a percutaneous muscle biopsy was taken from the vastus lateralis of the right leg, according to the technique of Bergstrom (2) as modified by Evans et al. (8). At this time, an incision was also made in the left leg for the immediate (within 3 min) postexercise muscle biopsy, and a Jelco l&gauge cannula (Critikon, Halfway House, TVL, South Africa) was inserted into a forearm vein for blood sampling. The subject's bicycle was then mounted onto the Kingcycle and a rundown calibration was conducted, as described previously, for the final 1-hr performance ride. After the calibration of the Kingcycle, the subjects began a 5-min self-paced warm-up on an electronically braked ergometer and then proceeded to cycle for 2 hr at an exercise intensity equal to 65% of PPO (244 f 8 W, -75% of V02peak).After 20, 40, 60, 80, and 100 min of the 2-hr submaximal ride, subjects performed a 60-s sprint at the maximal work rate (376 f 12 W) they attained during the maximal test, and then the workload was reduced to 0 W with subjects still turning the pedals for 60 s before continuing submaximal exercise. These sprints were designed to mimic the way cyclists would typically race (20) and to promote muscle glycogen depletion prior to the subsequent performance ride. For the performance ride, the subjects transferred (within 60 s) from the ergometer to the Kingcycle and maintained as high an exercise intensity as possible for 1 hr. In this ride, the only feedback to the subjects was the elapsed time; that is, they were kept blind as to distance covered, speed attained, and heart rate achieved during the ride. In order to offset the potential negative effects of dehydration and to prevent possible hypoglycemia, subjects ingested 750 ml/hr of an 8 g/100 ml short-chain glucose polymer solution (Energade Marathon, Bromor Foods, Salt River, Cape Town) during the first 2 hr of exercise. However, only water was available to drink, ad libitum, in the 1-hr performance ride. CHO-containing solutions were not provided during the performance rides because cyclists voluntarily ingest variable volumes of fluid during exercise. It was felt that making all subjects drink the same volume might interfere with their ability to concentrate on cycling as hard as possible. Analytical Techniques
Venous blood samples (3 ml) were collected into ice-cold tubes containing potassium oxalate and sodium fluoride at rest, at 30-min intervals during the 2-hr ride, and at the end of the performance ride. After the trial, the blood samples were centrifuged for 10 min at 2,500 revlmin in a refrigerated centrifuge (0 "C), and the supernatant was stored at -20 "C for subsequent analyses of plasma glucose and lactate concentration. Plasma glucose concentration was measured with an automated glucose analyzer (Beckman Glucose Analyser 2, Fullerton, CA, U.S.A.), which uses the glucose oxidase assay (14). Plasma lactate concentrations were measured with a standard enzymatic spectrophotometrictechnique (1 1).
Carbohydrate Loading / 29
Muscle glycogen content in the frozen muscle samples was determined in duplicate by the method of Passoneau and Lauderdale (21). The coefficientof variation for assays was 6% for duplicate glycogen assays of a single piece of muscle and 4% of the glycogen content of three separate pieces of the same muscle biopsy. Statistical Analysis
All results are expressed as mean f SEM. The statistical significance of the effects of dietary CHO on muscle glycogen content, power output, and the distance attained during the performance ride was assessed with a paired Student's t test. A p value of
