Nutritional Strategies of Mountain Marathon Competitors An Observational Study

Transcription

1 International Journal of Sport Nutrition and Exercise Metabolism, 2005, 15, Human Kinetics Publishers, Inc. Nutritional Strategies of Mountain Marathon Competitors An Observational Study Heather R. Clark, Margo E. Barker, and Bernard M. Corfe Mountain marathons are 2-d, self-supported adventure races, during which competitors must carry all nutritional requirements to sustain athletic effort. This requires a compromise between the energy required to perform and the weight penalty of carrying it. We have undertaken a nutritional survey of event competitors in the UK using a questionnaire-based approach and have monitored dehydration during the event. We found that competitors in longerdistance classes (> 50 km) carry significantly less mass of food, which is more energy dense, but that the calorific value is lower than that of competitors in shorter classes. Carbohydrate and protein consumption both positively associated with performance. Competitors became progressively dehydrated throughout the event. Counterintuitively, the better-performing subjects became the most dehydrated. Competitors at all distances should make more effort to rehydrate during breaks in the event. Competitors at shorter distances could choose more energy-dense foods to reduce weight penalty. Key Words: carbohydrate, protein, hydration, ultra-endurance, running Ultra-endurance sport has been defined as continuous sporting activity, of a competitive nature or at a comparable intensity, lasting more than three hours (25). Within this definition are mountain marathons 2-d, self-supported events which combine fell running and orienteering over steep, uneven terrain. These events are popular across Europe and are forerunners of adventure racing. Nutritional strategies of ultra-endurance athletes depend on several factors, including the type, intensity, and duration of exercise, total energy expenditure, time for recovery, dietary preference of the athlete, and whether the event is supported (5). The nutritional strategies of an athlete for a multi-day, unsupported event are likely to be very different from those of an athlete preparing for a single-day or supported event. Burke (6) observed that elite endurance cyclists relied on preand post-effort meals to achieve nutritional goals, although it was not determined whether the strategy was optimal. Other studies have identified a nibbling pattern, characterized by frequent eating and drinking of small amounts while exercising (22). There has been considerable research into the use of the major macronutrients in training and sustaining different types of athletic effort. There is strong evidence that carbohydrate ingestion before and during exercise over 1 h improves work capacity and performance through increased glucose oxidation and enhanced water The authors are with the Human Nutrition Unit, Division of Clinical Sciences North, University of Sheffield, Northern General Hospital, Herries Road, Sheffield, S5 7AU England. 160

2 Nutrition for Self-Supported Endurance Races 161 absorption (16, 25). The American College of Sports Medicine recommends an intake of 30 to 60 g carbohydrate/hour during prolonged exercise (17). Studies of highly trained endurance athletes including runners, cyclists, and triathletes have shown consistent performance benefits related to carbohydrate ingestion (30, 23, 7, 21). Colombani et al. (12) reported the nutrient intake during the ultra-endurance Swiss Gigathlon. Energy balance was negative overall; energy intake corresponded to 45% of energy expenditure. The nutritional focus was on high carbohydrate intakes, which allowed for high rates of carbohydrate oxidation. In contrast, Saris et al. (27) recorded the food intake and energy expenditure of 5 cyclists throughout the 22-d Tour de France. Mean energy intake was 24.7 MJ with the highest daily intake 32.4 MJ. Mean energy expenditure was 25.4 MJ with the highest daily expenditure 32.7 MJ: 62% of total energy came from carbohydrate. It might be inferred that significant energy deficits are sustainable in short multi-day events, but not over the course of a 3-wk race. The post-exercise recovery period is important for athletes competing on consecutive days. Glycogen resynthesis is optimal in the first 2 h after exercise (6, 14, 17). Consumption of high glycemic index (GI) carbohydrates is recommended for the most rapid increase in muscle glycogen (8, 9). This effect was demonstrated when 5 trained cyclists were fed high-carbohydrate diets after exercising to deplete muscle glycogen (8). The increase in muscle glycogen after 24 h was greater for cyclists consuming high GI foods than those consuming low GI foods. Alternative nutritional strategies might be beneficial in some situations. Brown (5) recommended high-fat diets for exercise in which energy expenditure is high and recovery time is limited, and for events in which athletes carry their own food. Fat releases double the energy per unit weight of carbohydrate. No consistent performance benefit has been shown in endurance athletes following a high-fat diet (29, 24). Protein is utilized as a fuel during prolonged exercise, albeit to a lesser extent than carbohydrate and fat. Carraro et al. (10) assessed the rates of muscle protein turnover in normal volunteers after 4 h of submaximal aerobic exercise and again after 4 h of recovery. Although muscle protein breakdown was stimulated during exercise, the overall depletion of muscle mass was not significant because muscle protein synthesis was stimulated in recovery. Fielding and Parkington (15) suggested that athletes should consume a meal rich in amino acids soon after endurance exercise to enhance protein accretion. Observational studies of fluid intake, across a wide range of sports, have shown that athletes typically replace 30 to 70% of the sweat lost during exercise (7). Laboratory studies have indicated that fluid intakes of > 1.0 L/h should be possible during exercise (4). Consumption of large volumes of fluid while running, however, is often impractical and can cause gastrointestinal discomfort. Horswill (20) suggested 450 to 1200 ml/h as a general recommendation for fluid replacement during exercise. During the 1998 Swiss Gigathlon, median fluid intake was 560 ml/h, which is within this range (12). In contrast, during the 1988 Swiss Alpine Marathon, mean fluid intake was 398 ml/h for men and 302 ml/h for women (26). Mountain marathons represent a distinct subclass of ultra-endurance adventure racing. The events require athletes to be completely self-sufficient for 36 h. Athletes must therefore carry sufficient and appropriate food, together with a specified minimum amount of clothing and camping equipment, to sustain athletic effort,

3 162 Clark et al. recover overnight, and then repeat the effort without incurring a significant weight penalty. Total weight carried is a major concern elite teams often carry less than 4 kg in total for 36 h. Consequently food intake might be restricted through performance concerns. Liquid is heavy to carry so mountain marathon runners often rely on natural water sources, which could be irregular and unreliable in summer. We have undertaken an observational study of nutritional strategies of athletes competing in mountain marathon events in the UK during the 2003 season. Hydration was monitored. Critically, the data are linked to a measure of performance. Study Design Methods This cross-sectional study was designed to provide descriptive information on macronutrient intake and hydration status of ultra-endurance athletes competing in 4 mountain marathon events in the UK in summer and autumn These measures were then related to performance. Subjects A total of 4190 athletes competed at these 4 mountain marathon events, of which 90 volunteered for this study. There were various levels of subject participation: a) completion of a weighed food/fluid inventory; b) provision of urine samples; and c) completion of a questionnaire. The event organizers were approached via and permission was obtained to recruit volunteers. All competitors received an from the event organizers, directing them to a web page and encouraging them to participate. Volunteers were asked to register their interest by telephone or . A wide range of ages is represented in the total sample (Figure 1). Ultraendurance events of this type are popular with both older and younger athletes. The predominance of male subjects (84%) reflects the gender balance in these events. Athletes competing in every class are represented in the sample (Figure 2). Figure 1 Age distribution of sample.

4 Nutrition for Self-Supported Endurance Races 163 Figure 2 Class distribution of sample. Weighed Food/Fluid Inventory Subjects were asked to record their entire food intake over the 2-d event and to compile a detailed food inventory at home prior to the event. The food inventory comprised a detailed description of the foods and the weight (in grams) of each food item. Food consumed before the start of the event was not included. After the event, any leftover food was weighed and net consumption calculated. Soehnle digital scales (Murrhardt, Germany) (accurate to 1g and weighing up to 2 kg) were provided at the event for this purpose. Water consumption data was gathered by questionnaire retrospectively; it was estimated by subjects to the nearest half-liter per hour. Sports drinks were measured as for foodstuffs. Macronutrient intakes were analyzed using dietary software (NetWISP version 2.0, Tinuviel Software, Warrington, England). Macronutrient intake was expressed absolutely (g over the 2-d event) and as a percent of total energy and per kg of bodyweight, per hour of athletic effort. Urine Sampling Hydration status was assessed using the urine color method. This method has been validated against other measures of hydration, including urine specific gravity, urine osmolality, and various hematologic indices (2, 28). Subjects were asked to provide two 20 ml midstream urine samples as close as possible before the start and after the finish each day. Each urine sample was matched to the closest color on a urine color chart (1) illustrating a spectrum of possible urine colors, ranging from very pale yellow (color 1) to brownish green (color 8). Urine color was assessed immediately on receipt of the samples and always in a well-lit area. Questionnaire Subjects were asked to complete a questionnaire at the overnight camp. This was self-administered in summer, but researcher-administered in autumn because of

5 164 Clark et al. colder conditions. The questionnaire was designed to collect information on a) selfreported height and weight; b) the subject s level of experience of ultra-endurance events, measured by number of events entered and number of years competing; and c) choices and patterns of food and fluid consumption during events. Performance Performance was calculated as a percentage of the winning time in the same class. This method produced values >100% (i.e., always greater than the winning time). The reciprocal percentage was calculated to give values less than 100%: 200 (individual time / winning time 100) Score classes were scored as a proportion of the winner s time and hence the values were always less than the winning number of points. Statistical Analysis SPSS version 11.0 (SPSS, Inc., Chicago, IL) was used for statistical analysis of the data set. Urine color was recoded into 3 groups; well hydrated (colors 1 to 3), moderately hydrated (colors 4 to 5) and severely dehydrated (colors 6 to 7). Class was also categorized into 3 groups; longer distances (50 km or longer), medium distances (40 to 49 km), and shorter distances (30 to 39 km or shorter). Means and standard errors are reported for all macronutrient intake variables, height, weight, energy density, and performance, which were all normally distributed. Total running time and total weight of food/fluid carried were not normally distributed. One-way ANOVA was used to test the effects of class length and age on nutrient intake. A Kruskal-Wallis test was used to investigate the association between total weight of food/fluid carried and class length. A chi-squared test was used to investigate the association between fluid consumption and class length. Pearson s and Spearman s rank correlation coefficients were used to investigate the associations between various continuous variables. Multiple linear regression was used to control for class length in models of performance. Wilcoxon signed ranks were used to compare urine color at different stages. Results were considered significant when P < Ethical Considerations All volunteers gave written informed consent prior to participation. The protocol was approved by North Sheffield Local Research Ethics Committee. Food and Fluid Consumption Results There was considerable between-subject variation in nutrient intake. The mean nutrient intake (range) was (11.42 to 36.94) kj kg -1 h -1 for total energy, 0.89 (range: 0.53 to 1.76) g kg -1 h -1 for carbohydrate, 0.10 (0.03 to 0.20) g kg -1 h -1 for protein and 0.11 (0.03 to 0.28) g kg -1 h -1 for fat. Percentage mean macronutri-

6 Nutrition for Self-Supported Endurance Races 165 ent contribution to total energy intakes were: 54.9 (24.7 to 79.3) % carbohydrate, 13.5 (6.2 to 26.5) % protein and 29.9 (13.6 to 46.6) % fat. Nutrient intake varied with class length, i.e., distance covered during the event (Table 1). Subjects competing in longer classes consumed significantly less total energy (kj kg -1 h -1 ), less protein (g kg -1 h -1 ), and less fat (g kg -1 h -1 ) than those competing in shorter classes (P = 0.033, 0.005, and 0.007, respectively). Carbohydrate intake did not differ significantly between classes, but again the lowest intake was in the longest classes. Subjects competing in longer classes consumed significantly more carbohydrate from sports drinks (as a percent of total carbohydrate) and more energy from sports drinks (as a percent of total energy) than those competing in the shortest classes (P = and 0.025, respectively). Subjects in the medium-length classes, however, consumed the greatest proportion of total carbohydrate and total energy from sports drinks. Total weight of food/fluid carried also varied with class length, i.e., distance covered (Table 1). Subjects competing in shorter classes carried a significantly greater mass of food/fluid than those competing in longer classes (P = 0.035). Energy density was negatively correlated with total weight of food/fluid carried (ρ = 0.802, P = 0.000) and positively correlated with distance covered (r = 0.436, P = 0.001). Table 2 shows the statistics from regression models with performance as the outcome variable and length of class and macronutrient intake as predictor variables. Performance was positively associated with energy intake (g kg -1 h -1 ), carbohydrate intake (g kg -1 h -1 ) and protein intake (g kg -1 h -1 ) (P = 0.009, 0.023, and 0.019, respectively). Table 1 Nutrient Data According to Distance Covered Short Medium Long classes classes classes Class length n = 20 n = 32 n = 6 Total energy MJ Total carbohydrate (g) Total protein (g) Total fat (g) Energy (kj kg -1 h -1 ) Carbohydrate (g kg -1 h -1 ) Protein (g kg -1 h -1 ) Fat (g kg -1 h -1 ) % energy from carbohydrate % energy from protein % energy from fat % energy from sports drinks % carbs from sports drinks Energy density (kj/g) Total food/fluid carried (g)* Note. Values are means, except where *indicates medians; other factors indicated by italics.

7 166 Clark et al. Table 2 Linear Regression Models for Performance Regression Standard Adjusted coefficient b error t P r 2 Class length Energy (kj kg -1 h -1 ) Class length Carbohydrate (kj kg -1 h -1 ) Class length Protein (kj kg -1 h -1 ) Class length Urine color end day Class length Energy (kj kg -1 h -1 ) Urine color end day Nutrient intake also varied with subject age (Table 3). Older subjects consumed significantly less total energy (kcal kg -1 h -1 ) and less carbohydrate (g kg -1 h -1 ) than younger subjects (P = and 0.001, respectively). Older subjects ingested significantly less carbohydrate from sports drinks (as a percent of total carbohydrate) and less energy from sports drinks (as a percent of total energy) than younger subjects (P = and 0.030, respectively). Performance was negatively correlated with subject age (r = 0.281, P = 0.005). There was a wide range of use of specialized products such as sports energy drinks and high-calorie dehydrated meals, ranging from 0 to 77% energy derived from such products. Traditional/domestic products chosen included cereal bars, flapjacks, jelly babies, cous cous, pasta, noodles, hot chocolate, and custard. Sports drinks were used by 81% of subjects. There was no difference in energy intake (per kg of body weight) between subjects who consumed sports drinks during the event and those who did not (Table 4). Subjects who consumed sports drinks during the event (n = 50) consumed a greater proportion of carbohydrate (as a percent of total energy, P = 0.010) and a lesser proportion of fat energy (P = 0.034). Nutrient intakes (g kg -1 h -1 ) did not differ by use of sports drinks. Hydration Status Over three-quarters (79%) of subjects (n = 88) estimated their fluid consumption while running to be 0.5 L/hour. Subjects competing in longer classes drank

8 Nutrition for Self-Supported Endurance Races 167 Table 3 Nutrient Data According to Subject Age Age (y) n = 8 n = 23 n = 19 n = 7 n = 1 Total energy (MJ) Total carbohydrate (g) Total protein (g) Total fat (g) Energy (kj kg -1 h -1 ) Carbohydrate (g kg -1 h -1 ) Protein (g kg -1 h -1 ) Fat (g kg -1 h -1 ) % energy as carbohydrate % energy as protein % energy as fat % energy from sports drinks % carbs from sports drinks Performance (%) Energy density (kj/g) Note. Values are means; performance data shown in italics. Table 4 Nutrient Data According to Sports Drink Consumption Sports drinks Sports drinks used n = 50 not used n = 6 Total energy (kcal) Total carbohydrate (g) Total protein (g) Total fat (g) Energy (kj kg -1 h -1 ) Carbohydrate (g kg -1 h -1 ) Protein (g kg -1 h -1 ) Fat (g kg -1 h -1 ) % energy as carbohydrate % energy as protein % energy as fat Note. Values are means. significantly (P = 0.000) more fluid per hour while running than subjects competing in shorter classes (Table 5). Figure 3 illustrates the extent of dehydration at different stages of the events. Urine color was significantly darker at the end of day one than at the start of day

9 168 Clark et al. Table 5 Fluid Consumption Estimates According to Distance Covered (% in parentheses) Class length Short classes Medium classes Long classes Fluid consumption n = 34 n = 42 n = L/h or less 32 (94.1) 35 (83.3) 4 (33.3) ~1.0 L/h 2 (5.9) 7 (16.7) 5 (41.7) ~1.5 L/h 0 (0) 0 (0) 3 (25.0) one (P = 0.000) and significantly darker at the end of day two than at the start of day two (P = 0.013). The change in urine color on day two was significantly greater than the change in urine color on day one (P = 0.030). Therefore, athletes dehydrated while running on both days, but more so on the second day. Urine color was darker at the end of day two than at the end of day one, although this was not significant (P = 0.059). Over one-quarter (27%) of subjects experienced symptoms of dehydration during previous events, including cramps, headaches, and feeling faint.urine color was significantly paler at the start of day two than at the end of day one (P = 0.002). Therefore, athletes did rehydrate to some extent at the overnight camp. Not all athletes, however, rehydrated adequately during this recovery period. Figure 3 shows that some athletes remained dehydrated at the start of day two. Urine color at the end of day two was a predictor of performance (P = 0.002) after controlling for class length (Table 2). Therefore, the athletes who performed best overall became more dehydrated on day two. The association between urine color at the end of day two and performance remained significant (P = 0.001) after controlling for total energy (Table 2). Discussion Planning an optimum diet and fluid intake should be part of every athlete s strategy for performance (18). The principal aim of this study was to investigate nutritional strategies of self-supported ultra-endurance athletes and link these findings to performance. This study focused exclusively on 2-d, unsupported mountain marathon events. Although dietary composition varied substantially, mean carbohydrate intake was less than recommended (60 to 70%) for ultra-endurance athletes (29). Mean protein intake was within the recommended range (10 to 15% energy) for nonathletes. Mean fat intake was at the top end of the recommended range (20 to 33%) for nonathletes (17). Brown (5) recommended high-fat diets for this type of ultra-endurance event and it appears that athletes might be choosing high-fat rather than high-carbohydrate foods. This could have implications for performance, as the benefits of high-fat diets are unproven (29, 24).

10 Nutrition for Self-Supported Endurance Races 169 Figure 3 Urine color at various stages of the event. Dark filled bars represent severely dehydrated subjects, pale filled bars represent moderately dehydrated subjects, unfilled (white) bars represent well-hydrated subjects. The data show subjects became progressively and significantly dehydrated across the course of the event. There were clear differences in nutritional strategies according to the distances the athletes were running. Athletes competing in the longest classes (Lowe Alpine Mountain Marathon elite and A classes; Karrimor International Mountain Marathon elite, A, and long score classes) had the lowest nutrient intakes overall and carried lighter, more energy-dense food than those competing in shorter classes. The weight penalty associated with carrying the additional calories might have been an important factor in pragmatic decisions governing food choice. The observed increase in energy density with class length did not compensate entirely for the reduced mass of food. Food intake prior to events was not recorded, but anecdotally it was apparent that many athletes had carbohydrate-loaded beforehand. The exact timing of food intake during the event was not recorded as the result of practical constraints, so it is not clear which foods were consumed to sustain athletic effort and which were consumed during the recovery period. The type of food consumed showed considerable variation. Some athletes survived almost entirely on high-energy snack food whereas others ate more substantial meals. Some teams cooked 2 large meals at the overnight camp to maximize energy stores for the second day. Specialized, high-energy foods and fluids were used by most athletes to varying extents. Some athletes relied on sports drinks as their main fuel source, while others consumed much less or none at all. It was clear that older athletes relied much less on sports drinks as a source of energy and carbohydrate compared with younger athletes. Use of sports drinks did not affect absolute nutrient intakes, but did affect dietary composition shifting the balance further towards carbohydrate. A striking finding of this study was the distinct and progressive dehydration that occurred during day one and day two of the mountain marathon events. Only 13% of subjects were well hydrated at the end of their event, which supports the

11 170 Clark et al. statement that ultra-endurance athletes do not consume sufficient fluids to replace sweat losses (28, 16). The hourly target of 1.0 to 2.0 L/h recommended by Clark et al. (11) was certainly not met by most subjects. Fluid intake was approximately 0.5 L/h for the majority of subjects, which is within the recommended range (20), but was evidently insufficient to replace sweat losses. There was a strong positive link between dehydration at the end of day two and performance. We hypothesize that this counterintuitive result can be attributed to the ability of top-level athletes to sustain performance under the specific conditions of this class of event, despite becoming dehydrated. There was evidence that athletes in the longer classes made more effort to rehydrate, perhaps aware that the increased dehydration resulting from more prolonged efforts would impair performance. There are several reasons why ultra-endurance athletes might not consume sufficient fluids while racing. Availability of water could be a key factor: mountain marathon runners tend to look for natural water sources, rather than carry heavy bottles of fluid. At summer events, streams are often low or dry. In autumn, many athletes commented that the stream water was so cold as to be unpalatable, with consumption causing gastric discomfort. Subjects were generally more dehydrated at the end of day two than at the end of day one, despite running a shorter distance on day two. This could be because athletes become less concerned about hydration towards the end of the event, as they race towards the finish. We also found, however, that athletes did not rehydrate adequately overnight and remained dehydrated (moderately or severely) at the start of day two. Given athletes have free access to water, it seems that this is a missed opportunity to rehydrate, which we speculate could improve overall performance. Thirst is an insufficient stimulus and indicator of dehydration (19) and fluid loss continues during the recovery period, so athletes have clearly underestimated how much fluid was required for recovery. Some subjects began their events in a dehydrated state. This could be explained by the early morning start times on day one (starts between 7:00 and 10:00 AM), so the earlier starters had little time to rehydrate after the overnight fasting period. Equally, some competitors regard the events as social as much as competitive and might have consumed alcohol in the 24 h prior to the start. Nevertheless, this is a key point at which many athletes could improve their nutritional strategy. Performance in mountain marathons is governed by a number of factors. Good navigational skill is central to outcome, as is general fitness, experience, and tolerance of challenging conditions. For a given athlete, having a suitable nutritional strategy might enhance performance, though it will also depend on these other factors. Our data indicate that athletes in the short and medium classes could reduce their weight burden further by carrying more energy-dense food. Along with most other studies in the area, we found strong correlation between carbohydrate consumption and performance, and also a strong link between protein consumption and performance, possibly reinforcing studies of the role of protein in recovery. We were surprised by the positive association between dehydration and performance and concluded that the more competitive athletes can tolerate severe dehydration. All athletes should put greater effort into rehydration at the overnight stop, however, where neither water nor time is limiting.

12 Nutrition for Self-Supported Endurance Races 171 Acknowledgments We would like to thank the event organizers: Martin Stone (Lowe Alpine Mountain Marathon), Charlotte Webb (Saunders Lakeland Mountain Marathon), Prof. Mike Parsons (Karrimor International Mountain Marathon), and Nigel Thomas (Mountain Navigational Challenge) for their support and encouragement with this study; Kim Proctor and Eleas Vikelis for their assistance with data collection. References 1. Armstrong, L.E. Performing in Extreme Environments. Champaign, IL: Human Kinetics Publishers, Armstrong, L.E., J.A. Herrera Soto, F.T. Hacker, D.J. Casa, S.A. Kavouras, and C.M. Maresh. Urinary indices during dehydration, exercise and rehydration. Int. J. Sport Nutr. 8: , Armstrong, L.E., C.M. Maresh, J.W. Castellani, M.F. Bergeron, R.W. Kenefick, K.E. LaGasse, and D. Riebe. Urinary indices of hydration status. Int. J. Sport Nutr. 4: , Brouns, F., and E. Kovacs. Functional drinks for athletes. Trends Food Sci. Tech. 8: , Brown, R.C. Nutrition for optimal performance during exercise: carbohydrate and fat. Curr. Sports Med. Rep. 1: , Burke, L.M. Nutrition for post-exercise recovery. Aust. J. Sci. Med. Sport 29:3-10, Burke, L.M. Nutritional practices of male and female endurance cyclists. Sports Med. 31: , Burke, L.M., G.R. Collier, and M. Hargreaves. Muscle glycogen storage after prolonged exercise: effects of the glycemic index of carbohydrate feedings. J. Appl. Physiol. 75: , Burke, L.M., G.R. Collier, and M. Hargreaves. Glycemic index a new tool in sport nutrition? Int. J. Sport Nutr. 8: , Carraro, F., C.A. Stuart, W.H. Hartl, J. Rosenblatt, and R.R. Wolfe. Effect of exercise and recovery on muscle protein synthesis in human subjects. Am. J. Physiol. 258:E470- E476, Clark, N., J. Tobin Jr., and C. Ellis. Feeding the ultra-endurance athlete: practical tips and a case study. J. Am. Diet. Assoc. 92: , Colombani, P.C., C. Mannhart, C. Wenk and W.O. Frey. Nutritional intake during a 244 km multisport ultraendurance race. Pakistan J. Nutr. 1: , Convertino, V.A., L.E. Armstrong, E.F. Coyle, G.W. Mack, M.N. Sawka, L.C. Senay Jr., and W.M. Sherman, American College of Sports Medicine position stand: exercise and fluid replacement. Med. Sci. Sports Exerc. 28:i-vii, Costill, D.L. Carbohydrate for athletic training and performance. Bol. Assoc. Med. P. R. 83: , Fielding, R.A. and J. Parkington. What are the dietary protein requirements of physically active individuals? New evidence on the effects of exercise on protein utilization during post-exercise recovery. Nutr. Clin. Care 5: , Galloway, S.D. Dehydration, rehydration, and exercise in the heat: rehydration strategies for athletic competition. Can. J. Appl. Physiol. 24: , 1999.

13 172 Clark et al. 17. Garrow, J.S., W.P.T. James and A. Ralph. Human Nutrition and Dietetics, 10th ed. Edinburgh: Churchill Livingstone, 2000, p Gonzalez-Gross, M., A. Gutierrez, J.L. Mesa, J. Ruiz-Ruiz, and M.J. Castillo. Nutrition in the sport practice: adaptation of the food guide pyramid to the characteristics of athletes diet. Arch. Latinoam. Nutr. 51: , Greenleaf, J.E. Problem: thirst, drinking behaviour, and involuntary dehydration. Med. Sci. Sports Exerc. 24: , Horswill, C.A. Effective fluid replacement. Int. J. Sport Nutr. 8: , Kimber, N.E., J.J. Ross, S.L. Mason, and D.B. Speedy. Energy balance during an ironman triathlon in male and female triathletes. Int. J. Sport Nutr. Exerc. Metab. 12:47-62, Kirsch, K.A. and H. von Ameln. Feeding patterns of endurance athletes. Eur. J. Appl. Physiol. Occup. Physiol. 47: , Millard-Stafford, M.L., P.B. Sparling, L.B. Rosskopf, and L.J. DiCarlo. Carbohydrateelectrolyte replacement improves distance running performance in the heat. Med. Sci. Sports Exerc. 24: , Peters, E.M. Nutritional aspects in ultra-endurance exercise. Curr. Opin. Clin. Nutr. Metab. Care 6: , Rehrer, N.J. Fluid and electrolyte balance in ultra-endurance sport. Sports Med. 31: , Rehrer, N.J., F. Brouns, E.J. Beckers, W.O. Frey, B. Villiger, C.J. Riddoch, P.P. Menheere, and W.H. Saris. Physiological changes and gastro-intestinal symptoms as a result of ultra-endurance running. Eur. J. Appl. Physiol. 64:1-8, Saris, W.H., M.A. van Erp-Baart, F. Brouns, K.P. Westerterp, and F. ten Hoor. Study on food intake and energy expenditure during extreme sustained exercise: the Tour de France. Int. J. Sports Med. 10 (suppl 1):S26-S31, Shirreffs, S.M. Markers of hydration status. J. Sports Med. Phys. Fitness. 40:80-84, Williams, C. Macronutrients and performance. J. Sports Sci. 13:S1-S10, Wright, D.A., W.M. Sherman, and A.R. Dernbach. Carbohydrate feedings before, during, or in combination improve cycling endurance performance. J. Appl. Physiol. 71: , 1991.