Here are a couple resources:
Fluid intake is a balance between fluid deficit, which has been shown in laboratory studies to result in a decrease in work capacity and to contribute to the development of hyperthermia, and fluid overload that can result in exercise-associated hyponatremia (EAH) (11,13,24,34,69,73,81). In 1969, Wyndham and Strydom published a report on runners who ran Sugar’s Marathon in Johannesburg, South Africa, on two fall days in 1969, which helped shape fluid consumption recommendations for the next 30 years (18,101). In the study, there was a linear correlation between weight loss and increasing rectal temperatures at the finish of the marathon. The authors noted a rise in temperature to >102°F occurred more frequently in individuals with >3% loss in body weight and felt the increase in temperature could lead to heat stroke (although none of the runners developed heat stroke). They concluded that thirst was not a good gauge of hydration status and stated, “The ideal regimen of water drinking is to take about 300 ml every 20 minutes or so. This should start right at the beginning of the race.” Of interest is that the runner with the greatest degree of dehydration and rectal temperature won the race on both days.
Results from more recent studies have refuted Wyndham and Strydom’s concerns that thirst is not a good indicator for hydration status and that >3% weight loss from evaporation of sweat will result in heat stroke. Looking at competitors in the South African Ironman, Sharwood et al. (87) showed that pre- to postrace changes in body weight were not related to postrace rectal temperatures. Some of the ultraendurance triathletes they studied sustained a 6% loss in body weight without a resultant increase in medical problems (87,88). In addition, body weight loss of 3% in the Ironman race did not lead to thermoregulatory failure (64).
In half-marathon runners and ultramarathoners, body mass alone may not be an accurate indicator of hydration status. Tam et al. (94) looked at runners of 21.1- and 52-km races and found they lost body mass (from a combination of substrate utilization, sweat evaporation, and insensible fluid loss) but overall gained total body water while preserving serum sodium and potassium and only had a small change in serum osmolality. Despite not drinking as much as sweat losses, they attributed the increase in total body water from release of water stored with glycogen that was released during muscle glycogen metabolism and possibly from fluid stores in the GI tract. Kao et al . (56) evaluated body weight changes before, at 4-h intervals during, and immediately after 12- and 24-h ultramarathons and found a positive correlation between weight loss and performance in the 24-h race.
Athletes should have a basic hydration plan prior to entering competition that they have developed and tested during their training sessions. The 2002 International Marathon Medical Directors Association guidelines, although targeted at marathon runners, are a good starting point for athletes in ultraendurance events (68). These advise that marathon runners drink approximately 400 to 800 mL·h−1, with increased rates for the faster, heavier runners, competing in warm conditions, but no more than 800 mL·h−1. Although faster competitors often finish with more significant dehydration, they often have greater experience and may be able to safely tolerate a higher degree of dehydration without suffering negative medical consequences (68,74,101). Pre- and postworkout weights can help athletes assess fluid needs for specific training loads, and monitoring urine color (aim for light yellow) and frequency can help athletes predict needs during events. On race day, athletes need to be flexible in their fluid plans to account for race day environmental conditions.
Historically, exercise-induced heat stroke was believed to be solely attributed to dehydration due to inhibited sweat evaporation secondary to decreased cutaneous blood flow with inhibited energy dissipation through evaporation. In addition, laboratory and dessert studies (not studies in actual racers) showed high fatigability in individuals with >2% dehydration thought to be due to decreased cardiac output — the rate limiting step for oxygen delivery to working muscles in most athletes (37).
Wyndham and Strydom’s (54) landmark study in 1969 evaluated marathon runners in race conditions. Several runners including the winner had dangerously high rectal temperature, according to the authors, and the rectal temperature correlated with the degree of dehydration. They interpreted their data as a warning to drink more during exercise instead of a possible alternative theory that those who can sustain greater dehydration levels and higher rectal temperatures will win the race.
For years to follow, athletes were told that thirst was not a reliable indicator of hydration status, and they needed to drink continuously. Consequently, hundreds of athletes imbibed high quantities of fluids before, during, and after racing resulting in excessive free fluid lowering serum sodium concentration with resultant exercise-associated hyponatremia (EAH). Exertional hyponatremia can result in pulmonary edema and, in more severe cases, brain swelling, and death (24).
In 2003, Dr. Tim Noakes and the International Marathon Medical Directors Association presented a hydration plan with a different interpretation of the evidence (37). They said runners (and we can extrapolate to triathletes) should drink ad libitum — as thirst dictates. They recommended an approximate guide of 400 to 800 mL·h−1 with increased rates for faster or larger athletes especially in warm environmental conditions and less fluids for smaller, slower athletes or those in colder environmental condition.
While severe dehydration with exercise can result in an increase in core body temperature, the heat burden is based on the athlete’s metabolic rate and can be lowered by slowing pace. There has been no compelling evidence to suggest a change in these well thought out recommendations.
More recent studies have only added support for the ad libitum fluid plan. Beis et al. (5) used retrospective video analysis of 10 elite marathoners in 13 city marathons evaluating footage from the cameraman on a motorcycle following the lead pack and showed that while a variation of fluid intake occurred, most stayed within the 400 to 800 mL·h−1 recommendation, and one of the winners had almost 10% dehydration. Wall et al. (51) had 10 well-trained cyclists perform a 2-h laboratory submaximum training session of biking and walking to produce 3% dehydration. Afterward, the athletes received blinded postexercise intravenous rehydration to return them to euhydration, 2%, or 3% dehydration. Subsequently, a 25-km time trial with a fan to simulate environmental cycling conditions did not result in a significant difference between the groups in time to completion, wattage produced, or rating of perceived exertion.
Currently, competitive cyclists are experimenting with controlled dehydration while climbing on the bike to lower overall body weight in an effort to lower wattage/kg ratio but not enough to decrease cardiac output. This strategy could provide some benefits for triathletes competing on hilly courses provided they could still handle the postbike run.
We advise triathletes to determine their individual hydration plan by initially starting with 400 to 800 mL·h−1 and listening to their thirst. Additional information on fluid needs can be gained by following the position article on exercise and fluid replacement (44) and obtaining preexercise and postexercise weights to determine sweat loss for individual workouts in specific ambient conditions to a get a sense of one’s individual sweat rate. In more experienced triathletes, the plan can be altered once the athlete determines in what hydration state he or she functions best. Like all plans, it will need to be altered based on environmental and personal conditions on each individual day.