Water replacement strategy? Leave it to Nature.

At four o’clock in the morning well-trained and heat-A dog (puppy Jack Russel) drinking water flowing from a bottle to a man's handadapted soldiers from the South African National Defence Force started 50 kilometres route-marсh in full equipment (45 kg) . It was a hot day and temperature reached 37.5 C at the peak. After thirteen hours of exercise twenty years old soldier was found unconscious just around 2 km from the finish line. His body temperature was 42.6 C. He was unresponsive with his eyes rolled back in his head; he made incomprehensible noises, had difficulty breathing and was exhibiting tonic-clonic movements. All these conditions are associated with the heatstroke. Despite all efforts to save him, soldier passed away in the hospital on the next day.

This tragic evenеt described in “The Physician and Sportsmedicine” (Nolte, Hew-Butler, Noakes, & Duvenage, 2015), unfortunately is not highly unusual. From time to time similar fatal incidents happen during sport and military exercises, especially in extreme environment conditions. However some circumstances made this case exceptional. In that ill-fated exercise, soldier had been monitored by group of South African sport scientists. They collected data on his (and his peers) fluid intake, core temperature, speed / distance and his hydration status. I just want to specify that scientists did not design the drill and data were not available straightway, otherwise they could stop the soldier and prevent his death. After close examination of all data and circumstances, scientists suggested that the reason of death was a heatstroke which was provoked by hot weather, extreme workload and … overdrinking! This combination looks so illogical that it is hard to believe that it might be possible. An aggressive fluid replacement is considered by many athletes and coaches as one of the main measures which can prevent heatstroke. So, how may be possible the opposite? In presented article I am going to discuss current knowledge about fluid replacement, review different strategies and give my opinion on practical issues.

Role of water in exercise.

Water is the main ingredient of the human body. We consist of around 60 % of it. Two thirds of total body water is inside body cells whereas one third is extracellular fluid. In its turn, extracellular water consists of water between body’s cells (interstitial water) and blood. Theses volumes are not static and they exchange fluid and solutes between them. Water is essential for biochemical reactions, thermoregulation, transport materials, it provides volumes for tissues and cells, etc. We can survive without it only for a few days (Sawka & Coyle, 1999).

Physical efforts, especially in the hot environment, put significant physiological stress on our body. To cope with this challenge we need to maintain homeostasis inside acceptable limits. There are three main issues during an exercise which are closely related to water: thermoregulation, cardio-vascular efficiency and electrolytes concentration.


Humans can operate only inside particular range of body temperature. Similar to mechanical engines, human body creates heat during its work which is increasing dramatically during exercise. This heat should be removed otherwise system will be overheated and stop functioning. Body water provides natural reservoir for heat absorption and mechanism for its removal.

Blood absorbs heat from working muscles and moves it to the border of the system – skin where heat can be dissipated into environment. Another way is even more effective during intense work – sweating. Sweat is produced by special glands which absorb water from blood and interstitial space and pump it to the surface of the skin where this water evaporates providing great cooling effect.

Cardio-vascular efficiency

During exercise blood transport function, which is delivering oxygen, hormones and energy substances to the working tissues as well as removing metabolites and CO2 from them, increases dramatically. This requires that the amount of water, circulating in the cardiovascular system, is maintained at level required for optimal pressure. If blood return to heart drops, which may happen as the result of sweat’s losses and/or increasing blood flow to the skin, blood pressure may drop as well. In order to maintain pressure heart starts to pump in higher rates, however decreasing stroke volume leads to impaired efficiency of cardio-vascular system.

Electrolyte’s concentration

Very important parameter of homeostasis is solutes concentrations in the blood, interstitial fluid and inside body cells. It is crucial for maintenance osmotic pressure, ionic and water balances. These concentrations have to be kept inside acceptable physiological borders otherwise homeostasis and consequently body functions will be disturbed. This may happen in exercise due to fluid and/or electrolytes losses.


Dehydration is a process of losing water. It may be more significant during exercise. First “victim” may be blood. I already said about dropping pressure. Another problem from insufficient blood volume is competition between peripheral (skin) and central (working muscles) supply systems for resources. Shortage of blood for the former leads to overheating, whereas shortage for latter causes drops in the work efficiency. There are some other issues. Reduction in the blood volume may give signal for decreasing sweat production hence impairs cooling. More viscous blood causes diminishing in penetration through the small capillaries thus decreasing oxygen and energy supply. It is generally accepted that dehydration may seriously disturb homeostasis and impair performance.

Is changes in body weight reliable indicator of dehydration?

Measuring blood and urine parameters are scientific ways to determine hydration status. This is not always practical however. Detecting changes in body weight is the simplest method to estimate water loss during exercise. Generally it is reliable, however not without some limitations. One of them is obvious; it is not only water which is being spent during exercise. It is fuel (mostly fat and glycogen) as well. Amount of fuel which is burnt doesn’t contribute a lot to the total weight changes in the low energy demanding events, nevertheless may be significant in the endurance and ultra-endurance competitions. Another issue with the assessing water loss through the weight measurements is that we lose not only free water during exercise. Some weight loss comes from metabolic water which is a product of fat (1.3 gram of water per 1 gram) and carbohydrates (0.6 gram of water per 1 gram) oxidation (Nicholas Tam & Noakes, 2013). Additional amount of non-free water  comes from glycogen-bound fluid which is around 3-4 gram of water per gram. Contribution of non-free water to the weight loss may be around 2 kg during marathon (Pastene, Germain, Allevard, Gharib, & Lacour, 1996; Nicholas Tam & Noakes, 2013). Possibly, this loss doesn’t require immediate replacement and may be restored together with glycogen restoration after exercise. It seems that weight measurement is practical but not precise method which can overestimate free water loss.

How much water we can lose?

Currently one of the most respected body in sport science – American College of Sports Medicine recommends 2% body weight limit as acceptable water loss (Sawka et al., 2007). However many researches argue that this number came mostly from laboratory based and not properly designed studies. Often participants were tested without airflow and were dehydrated artificially, by restricting water before and/or during exercise or by administrating diuretics. This is not kind of dehydration which happens (and actually is pretty common) in real life where athletes usually have free access to water thus their dehydration is “voluntary” (Nicholas Tam & Noakes, 2013).

One of the strongest arguments against 2% guidance is that dehydration, which exceeds this limit, actually does not impair performance (Goulet, 2011). In opposite, very often winners of endurance and ultra-endurance events are among the most dehydrated athletes in those races. If we measure dehydration as percentage of the body weight loss, they can be dehydrated up to 10 % of their body mass (Beis, Wright-Whyte, Fudge, Noakes, & Pitsiladis, 2012). Nevertheless, well trained and heat adapted athletes, are pretty good in defending their homeostasis. Despite significant losses of water with sweating, their blood volume which, as was mentioned before, is a very important parameter, may remain constant throughout the exercise due to redistribution of water from interstitial spaces (N. Tam, Nolte, & Noakes, 2011). Additionally, they lose much less electrolytes in their sweat than untrained persons, thus their osmotic balance remains inside acceptable limits.

Is dehydration our main enemy in exercise?

It is a common belief not only among athletes and fitness enthusiasts but among coaches and sport scientists as well that even mild dehydration is a very dangerous. However not all apprehensions, connected with dehydration, are scientifically well-grounded. For instance, it is wide spread belief that fluid loss is a main reason for exertional heatstroke (EHS) which is associated with the exercise. Heatstroke is a very dangerous, life-threatening condition which annually causes a number of fatal incidents in sport and military exercises. It develops very rapidly. Look how this happened recently with John Brownlee in Mexico https://www.youtube.com/watch?v=CS0GkCfljqk. After that, with the famous British sense of humour, Jonny twitted from the hospital that maybe he looked like a drunk but it was the opposite. What is the opposite? Most likely he meant dehydration. However many scientists don’t support the idea that dehydration is the main reason of EHS. They argue that exercise intensity combined with the extreme weather conditions are the main causes. Interestingly, Alistair Brownlee, Jonny brother who carried him through the finish line, blamed his sibling’s pacing strategy in that incident. The two-times Olympic champion said: “I wish the flipping idiot had just paced it right and won the race. He could have jogged the last 2km.” Actually the same conclusion was made by the winner of the race, South-African Henri Schoeman: “I knew it was going to be warm out there. I just tried to stay calm throughout the whole race. I was just running with the Brownlees and trying not to exert too much and then I saw Jonny pull ahead on that third lap. So it is just so unfortunate that he collapsed at the end…”

Interestingly, EHS more often occurs in relatively short and intense exercises in hot and humid weather rather than in long endurance events. This is due to inability of thermoregulation system to cope with the challenges of intensity and environment, though athlete still may have enough body fluid. Look at another striking example of EHS, which was happened in the US –USSR track and field match in July 1959, during Cold War https://www.youtube.com/watch?v=HecRmHa2W8A. Two athletes out of four (one American and one Russian) had severe heatstroke during 10000 m race. This footage is in Russian but you can understand everything without translation. It was relatively short distance compare to Jonny’s triathlon race, so athletes could not be severe dehydrated, however there were really hard weather conditions (33 C, 88% humidity) and extremely high rate of exertion based on enormous motivation to win. In conclusion: dehydration is unlikely the main reason for EHS.

Another problem, for which dehydration is routinely blamed, is muscle cramps. Here again it is not so simple. Connected with severe dehydration disturbance in ionic balance may be one of the reasons for cramps. However more common enemies are: previous history of crumping, pre-exercise muscle damage and high intensity load during exercise (Hoffman & Stuempfle, 2015).

I would like to specify that I do not argue against undesirability of dehydration. Severe water loss may result in disturbing metabolism, to contribute to EHS and muscle crumps. Moreover, mentioned above statistics about endurance events winners, does not prove cause-effect relationship. Hence we cannot say that to win a marathon you should lose more weight than other competitors or should drink less. Perhaps the winners often lose significant weight due to their higher rate of sweat loss though they could drink a lot (Beis, et al., 2012). However I want to stress that hydration status is just one of the interacting variables in complexly regulated homeostasis during exercise and we cannot define universal limits of dehydration, especially if they will be based exclusively on body weight changes.


Whereas problems from water poisoning are well-known in medicine for a long time, it received much less attention among sport professionals. For a few decades to drink a lot of water was considered as panacea in sport. However when, not least due to prevailing recommendation “to drink as much as you can tolerate” and aggressive advertisement from sport beverages producers, amount of incidents connected with overhydration increases (Noakes, 2010), sport scientist started to be more careful in their recommendations.

So what can be the problems when athlete drinks too much during exercise?

Probably the most harmless is weight gain. You don’t want to carry additional kilo for many kilometres during the race or accelerate and decelerate with it in sport game. Of course that does not influence outcome too much but in a high level sport every detail matters. Another issue with drinking too often during the race is that it takes time and interrupts a pace.

Gastric inconvenience is probably more serious problem because it is difficult to concentrate on competition when it is something wrong with your stomach. Sometimes athlete may even vomiting or to have diarrhoea from excessive fluid consumption.

However the most dangerous consequence of hyperhydration is hyponatremia. Hyponatremia is low concentration of Na in the blood. It may be caused by severe Na loss or dilution of blood with excessive fluid. Hyponatremia due to sweating is rare because sweat contains lower concentration of Na than blood, thus sweating basically increases blood osmolality. Excessive fluid consumption may critically decrease Na concentration in blood, especially if person exercising for a long time with a low sweating rates. What can happen next is very dangerous. In order to maintain necessary balance between Na concentration in blood and inside body cells, excessive water moves from the blood into the cells and actually drown them causing oedema. Particularly it is dangerous when this happens in the brain. (Figure 1).untitled11Figure 1. Excessive water in the blood cases osmotic pressure to move it through the cell’s membrane into the brain.


So, what was the possible explanation for the soldier death?

Chain of events:

1. High rate of fluid intake in the particular intervals: for unknown reason soldier consumed excessive amount of water in 25-30 (4L) and 40-45 (4.7 L) km of the route-march

2. That might lead to hyponatremia and brain swelling.

3. Swelling might compress hypothalamus and disturb thermoregulation.

4. Disturbed thermoregulation, accentuated by extremely hot weather, resulted in the severe heatstroke.


Whereas excessive fluid intake provides no advantages in exercise, it may cause inconvenience, serious gastric problems and even death from hyponatremia.

Fluid replacement strategies

1. Water restriction during exercise

When I started to learn football in 1970-th, our coaches did not allow us to drink during a training session and matches. They explained that by increasing heart strain, weight gain and gastric inconvenience. They told us that learning to tolerate thirst develops your willpower. It was “cool” not to drink. This notion was typical for that time (see (Noakes, 2010 ) for historical review about fluid replacement strategies) and it was not a good idea. Besides the dangers of dehydration which can arise from sever fluid loss, probably the more immediate consequence of fluid restriction is impairment in performance due to thirst. I will talk in more details about thirst later, now I just want to note that feeling of thirst tells our brain that there is not enough fluid in our body and we need to stop any activity that may lead to further dehydration. Fighting against thirst possibly takes additional resources from the brain. In the experiment, subjects, who were not feeling thirsty despite being dehydrated, performed at the same level with those, who were hydrated normally, whereas subjects with the stronger thirst’s feeling performed worse (Edmonds, Crombie, & Gardner, 2013).

2. As much as you can tolerate

Probably it is the worst choice in my opinion. I already discussed the danger of hyponatremia. Though fatal cases from this state are rare among athletes, they occur regularly. At present time, significant percent of sportsmen are developing hyponatremia condition during exercise, hence may be exposed to danger. For example, in Boston marathon, 13% of analysed after race runners (64 athletes) had hyponatremia, whereas three runners had this condition at critical level (Almond et al., 2005). Other disadvantages of this method included additional weight and gastric discomfort. Currently most governing bodies in sport and military (e.g. American College of Sport Medicine) revised their position stands and lowered their water consumption recommendations.

3. Universal guidance

These recommendations are based on general calculations about how much fluid average person may lose in particular weather condition, during particular exercises and intensities. Because it is too general and based on hypothetical “average person” this guidance is not suitable for athletes. There are myriad nuances in individual abilities, weather conditions, clothing and exercise intensities which are very important and should be taken into account. Advantage from universal guidance is that it makes possible planning for fluid supply during sport and military exercises, as well as for people working in extremely hot environment.

4. Drinking as thirst dictates

Thirst is physiological mechanism which is designed to prevent dehydration. It reacts on rise in blood osmolality, hypovolemia (decrease in blood volume) and some other factors. Supporters of this strategy argue that thirst is sensitive enough to detect small changes in concentration of solutes in blood and person becomes thirsty before he/she lost 2% of their weight. Experienced athletes have better thirst sensitivity. On the other hand, opponents claim that in high intensity exercise with significant sweat loss, thirst may fail to detect changes in time (Armstrong, Johnson, & Bergeron). Additionally, thirst is influenced by other subjectively perceived factors included but not limited by: mouth sensation, beverages and food content and temperature, stomach fullness and even psychological stress. Thirst may have different individual thresholds and in some environmental conditions people less sensitive to it. Another consideration may be what is exactly : “drinking to thirst”? As correctly pointed out (Armstrong, et al.) this may be: to drink to avoid feeling thirsty or drinking only when feeling thirsty.

In conclusion: drinking to thirst, in my opinion, is physiologically based advice. However, it may be biased by many factors and probably is too vague in some situations.

5. Individual plan, based on personal characteristics and experience.

This method looks reasonable for me. However there are two problems with its realisation.

One is practical issues. How can I scientifically advise how much to drink on practice? For example I have to advise football team on their fluid replacement regime during a game. To be better prepared and be able to provide players with detailed plan, I have to make a fluid loss dossier on twenty players. These files should include individual peculiarities in rate of fluid loss in different weather conditions and for different game intensities. It would be a hard work. I can measure player’s hydration status before the game, to make sure that they are hydrated, to make some assumptions about influence of weather conditions and give players the schedules for fluid replacement. However this won’t be precise, because I cannot predict exact fluid loss since I don’t know how much and how intense player is going to run.

Second problem is theoretical. I am not convinced about 2% rule. Even if I will be able to predict exact sweat loss for players, I am not sure that this will allow me to give them what this method claims – individual, scientifically based advice. I am not sure how much from their sweat loss they have to compensate immediately and how much later. Sweat loss may include metabolic water, glycogen bound water and, as some researchers suggest, fluid reserve from the gut. All these fluids do not require immediate replacement and their actual contribution to sweat loss varies among individuals and conditions. My advice, again, will not be precise.

6. Ad libitum

Ad libitum is drinking as much as you want and whenever you want. More exact it can be called self-determined intake (Hoffman, Cotter, Goulet, & Laursen). It is a plausible strategy. It includes thirst as physiological indicator of fluid demand as well as individual experience, preferences and exercise characteristics. In overwhelming majority situations drinking ad libitum allows to keep water loss inside acceptable limits. However we cannot claim that this is unbiased method. Even feeling of thirst may be influenced by person’s fluid replacement’s culture and traditions. If, for example, person grew up in tradition of water restriction during exercise he/she became used to the feeling of thirst. Then probably a mild thirst doesn’t give them a strong signal for fluid replacement thus they do not start to drink. And this non-drinking behaviour is ad libitum because actually nobody prevents them from drinking. Another person, who is strongly influenced by fluid replacement hysteria, may develop feeling of thirst even without physiological needs and start to drink excessively. And it is ad libitum as well because nobody makes him/her to drink. Nobody made that soldier to drink so huge amount of water. Why he did that? We don’t know for sure but maybe he was influenced by military guidance which was based on out of date knowledge.

What we can do on practice?

In my opinion, we can clearly exclude the first three methods. All prominent scientists on the field agree that we should avoid water restriction as well as its overconsumption. Basically, all agree that we cannot use one, universal guidance for everybody and individual approach should be implemented. The question is: what should be the basis of this approach. Should it be some scientific calculations or the natural signals, like thirst? The problem with the former is inability to avoid generalisation (e.g.2 % limit) and practical difficulties with comprehensive plan. The problem with the latter is possible bias from insensitivity, unrelated to exercise factors and vagueness.

Well, both sides of this discussion agree that thirst and individual experience should be taken into account. Another established fact is that many well trained and well accustomed to different conditions endurance athletes possess very strong homeostasis-defending adaptations and have optimal for them hydration strategy. Unlikely that they followed “scientifically” prescribed water replacement behaviour from their childhood. Many came from poor background and trained without scientific supervision at the beginning of their career. They developed their strategy naturally.

I think here is the answer. Taking into consideration that our body is a very complexly regulated system and a comprehensive scientific advice is not possible in this case, at least for now, I tend to delegate the decision about how much to drink to athletes themselves. They will acquire necessary knowledge and experience by being exposed to different conditions throughout their development. From years of the sport practice, exercising in heat and cold, with immediate water available and with not, athlete will derive necessary information and gradually develops suitable for him/her behaviour. Do not restrict young sportsmen from water, however, at the same time, do not impose particular drinking guidance on them. Leave it to Nature. Possibly scientists, in this case, cannot claim that their advice is better than natural mechanisms developed for millions years of evolution.


Almond, C. S., Shin, A. Y., Fortescue, E. B., Mannix, R. C., Wypij, D., Binstadt, B. A., . . . Greenes, D. S. (2005). Hyponatremia among runners in the Boston Marathon. N Engl J Med, 352(15), 1550-1556.

Armstrong, L. E., Johnson, E. C., & Bergeron, M. F. COUNTERVIEW: Is Drinking to Thirst Adequate to Appropriately Maintain Hydration Status During Prolonged Endurance Exercise? No: Wilderness Environ Med. 2016 Jun;27(2):195-8. doi: 10.1016/j.wem.2016.03.002.

Beis, L. Y., Wright-Whyte, M., Fudge, B., Noakes, T., & Pitsiladis, Y. P. (2012). Drinking behaviors of elite male runners during marathon competition. Clinical Journal of Sport Medicine, 22(3), 254-261.

Edmonds, C. J., Crombie, R., & Gardner, M. R. (2013). Subjective thirst moderates changes in speed of responding associated with water consumption. Front Hum Neurosci, 7(363).

Goulet, E. D. (2011). Effect of exercise-induced dehydration on time-trial exercise performance: a meta-analysis. British Journal of Sports Medicine, bjsports077966.

Hoffman, M. D., Cotter, J. D., Goulet, E. D., & Laursen, P. B. REBUTTAL from “Yes”: Wilderness Environ Med. 2016 Jun;27(2):198-200. doi: 10.1016/j.wem.2016.04.004.

Hoffman, M. D., & Stuempfle, K. J. (2015). Muscle Cramping During a 161-km Ultramarathon: Comparison of Characteristics of Those With and Without Cramping. Sports Medicine – Open, 1, 24. doi: 10.1186/s40798-015-0019-7

Noakes, T. D. (2010). Is drinking to thirst optimum? Ann Nutr Metab, 2, 9-17.

Nolte, H. W., Hew-Butler, T., Noakes, T. D., & Duvenage, C. S. (2015). Exercise-associated hyponatremic encephalopathy and exertional heatstroke in a soldier: high rates of fluid intake during exercise caused rather than prevented a fatal outcome. The Physician and sportsmedicine, 43(1), 93-98.

Pastene, J., Germain, M., Allevard, A. M., Gharib, C., & Lacour, J. R. (1996). Water balance during and after marathon running. Eur J Appl Physiol Occup Physiol, 73(1-2), 49-55.

Sawka, M. N., Burke, L. M., Eichner, E. R., Maughan, R. J., Montain, S. J., & Stachenfeld, N. S. (2007). American College of Sports Medicine position stand. Exercise and fluid replacement. Med Sci Sports Exerc, 39(2), 377-390.

Sawka, M. N., & Coyle, E. F. (1999). Influence of body water and blood volume on thermoregulation and exercise performance in the heat. Exerc Sport Sci Rev, 27, 167-218.

Tam, N., & Noakes, T. D. (2013). The quantification of body fluid allostasis during exercise. Sports Medicine, 43(12), 1289-1299.

Tam, N., Nolte, H. W., & Noakes, T. D. (2011). Changes in total body water content during running races of 21.1 km and 56 km in athletes drinking ad libitum. Clin J Sport Med, 21(3), 218-225.