Time for the American College of Sports Medicine to acknowledge that humans, like all other earthly creatures, do not need to be told how much to drink during exercise

In the light of our response to Dr Murray and our suggestion that there should be greater transparency about the nature (if any) of the financial interactions among Gatorade, the Gatorade Sports Science Institute (GSSI) and leading exercise scientists and sports physicians, we find it interesting that Dr Roberts, a former President of the American College of Sports Medicine (ACSM), should feel it necessary to stress that he draws no personal financial benefit from the sports drink industry (other than a single moment of weakness). Dr Roberts' credentials are indeed impeccable; he is an honourable and independent sports physician who has made substantive contributions to the understanding of, especially, exercise‐associated collapse.¹ But his reference to the fractured fairy tale is misguided.

Like Dr Murray, Dr Roberts accuses us of misrepresenting the ACSM guidelines by selectively quoting from those guidelines. The point, as made in our response to Dr Murray, is that when the GSSI began to advertise the ACSM guidelines, it conveniently forgot to include some of the qualifications included in those guidelines and that might have reduced the extent to which athletes in the USA began to overdrink in the 1990s.

Thus, the message that the sporting public received was not that athletes should “consume the maximal amount of fluids during exercise that can be tolerated … or up to a rate equal to that lost in sweat”, but that they should simply drink “as much as tolerable”. Or, in the text of the Gatorade advertisement published in the New York Runner in January/February 2002, “at least [our emphasis] 40 oz of fluid every hour (ie, 1200 ml/h) or your performance could suffer”.

Had the same message that was being conveyed in South Africa and New Zealand also been given to US athletes, one wonders what effect this might have had on the incidence of exercise‐associated hyponatraemia (EAH) in the USA. In the 2002 Boston Marathon, the incidence of EAH was 13%,² with one death.³ By contrast, we have yet to encounter the first case of EAH in 42 km marathon runners in either South Africa or New Zealand, and the incidence of EAH in ultra‐endurance events such as the 226 km New Zealand⁴ or South African Ironman Triathlons⁵ or the 90 km Comrades Marathon footrace in South Africa⁶ is uncommon, despite the fact that these events last for up to 17 and 12 h, respectively.

The crucial difference between the USA and South Africa and New Zealand is that, by our efforts in our own countries, we were able to ensure that the safe drinking advice—that is, drinking to thirst⁷,⁸—became the accepted standard. These uncomfortable facts are part of the published record.

The ACSM failed in this matter because it made no effort to reverse the erroneous message that was being presented by the GSSI (as fully detailed in our companion response to Dr Murray). Elsewhere, Noakes has detailed criticisms about the nature of the scientific evidence used by the ACSM in the development of their 1996 guidelines,⁹,¹⁰ about which he was clearly ignorant when he first reviewed those guidelines 11 years ago.

Dr Roberts also believes that the “exercise associated hyponatremia [EAH] case is not closed”. Although he is welcome to his personal opinion, he should perhaps acknowledge that there is an International Consensus Statement on the aetiology of EAH that reviews all the published evidence and is evidence based.¹¹ Instead, Dr Roberts considers that an earlier South African study published in 1971¹² shows that EAH can develop in athletes who lose weight during exercise. If true, this suggests that large unreplaced sodium losses in sweat must contribute to the development of EAH. This is indeed the argument that Dr Murray and the GSSI have been advancing since 1993.¹³

This conclusion is, of course, industry friendly because it suggests that the ingestion of an electrolyte‐containing drink might prevent EAH. However, the International Consensus Statement¹¹ could find no published scientific evidence to support this theory. Only in those who overdrink because they have been advised not to lose any weight during exercise is there some evidence that the ingestion of an electrolyte‐containing sports drink may marginally reduce the overdrinking‐induced fall in the serum sodium concentration ([Na⁺]).¹⁴,¹⁵,¹⁶,¹⁷ However, the avoidance of overdrinking remains the best option to prevent EAH.¹⁸

In our recent collaborative international study of 2135 athletes in whom body weight changes during and blood [Na⁺] after exercise were measured, we scoured the world literature to include every report in which all these variables had been accurately measured.¹⁸ That study included data from an English study published in 1967¹⁹ and those from two other early South African studies from the 1970's.²⁰,²¹ The data of our South African colleagues, Dancaster and Whereat,¹² from the 1970 90 km Comrades Marathon, naturally attracted our attention. However, we concluded that the data were sufficiently inconsistent and could not be included. Dr Roberts seems not to have noticed that 18 of 35 (51%) pre‐race (Na⁺) measurements were ⩽136 mmol/l. The lowest pre‐race value was 126 mmol/l, with another value of 130 mmol/l. As serum [Na⁺] this low cannot exist in healthy asymptomatic athletes before a 90 km ultramarathon race, we concluded that these serum [Na⁺] measurements could not be believed. Of 30 values measured after the race, only 9 (30%) were ⩽136 mmol/l, indicating that even when incorrectly measured, low serum [Na⁺] became less likely when athletes lost weight in the course of a 90 km running race. For these reasons, we did not include these data in our final analysis.

Perhaps if Dr Roberts wishes to conclude that the “exercise associated hyponatremia [EAH] case is not closed”, he would be better advised to quote from the largest yet published study¹⁸ rather than from a single, unreliable study on a small group of subjects. That larger study concluded that three factors contribute to EAH—namely, (1) overdrinking, (2) a failure of the normal suppression of anti‐diuretic hormone secretion during prolonged exercise in those who overdrink and (3) either inappropriate inactivation of circulating Na⁺ or a failure of the mobilisation of osmotically inactive sodium stores. Furthermore, that study¹⁸ showed that the probability of developing the serious encephalopathy caused by EAH was 45% in those who gained >4% of their body weight during exercise. However, no subject who lost >6% of body weight during exercise had a post‐race serum [Na⁺] <135 mmol/l, disproving Dr Roberts' conclusion, based on the unsound study of Dancaster and Whereat.¹²

Interestingly, the conclusion of the International Consensus Statement is that EAH is not caused by dehydration and abnormal sodium losses.¹¹ We are at a loss to understand why Dr Roberts finds it difficult to accept this conclusion which is based on all the published literature.

Finally, Dr Roberts argues that to “replace only your sweat losses remains the best advice today”. However, this statement is not evidence based.⁷,⁹,¹⁰ Indeed, there is no evidence that the human body was designed to protect its body weight by replacing sweat losses during exercise. Rather, the regulated variable is plasma osmolality.⁸ Furthermore, plasma osmolality is regulated by fluid intake in response to changes in the perceptions of thirst.⁸

Thus, if athletes drink only according to the dictates of thirst as we have advocated,⁷ they will maximise their performances¹⁰ without any evidence that they risk ill‐health.⁷

The real fairy tale is that humans must be instructed on how to drink during exercise.⁹ Rather, if they would just “listen to their bodies” and to those internal homoeostatic controls developed over millions of years through the process of biological selection, they would do just fine.

We eagerly anticipate the day when the ACSM will acknowledge that, like all other creatures in the animal kingdom, humans really do not need to be told when and how much they should drink.