Dehydration Myths, By Dr. Sandra Fowkes Godek

The following article was submitted to us by Dr. Sandra Fowkes Godek. Dr. Godek is the director of the HEAT Institute and a professor of sports medicine at West Chester University of Pennsylvania. She is an expert in thermoregulation, hydration, exercise-associated hyponatremia and exertional heat stroke, and has conducted research in these areas with NFL, NHL, NBA & collegiate sports teams. Tim Noakes also cites her research in his history of hyponatremia, Waterlogged.


Fluid and electrolyte balance logically and by the numbers.

When high school certified athletic trainers were asked the following question “what is the most important predisposing factor to exertional heat stroke” they responded in order; dehydration (2.6 rank), high humidity (3.4 rank), high ambient temperature (4.3 rank), acclimatization (5.1 rank), physical fitness (5.7 rank), history of heat illness (6.1 rank) and exercise intensity (6.2 rank). Similarly, when I posed the same question to 50 first semester freshmen athletic training students I received a unanimous and emphatic response, dehydration! Furthermore, on the first day of class in both my junior (35 fifth semester athletic training major students) and senior (36 seventh semester athletic training major students) handed out a survey to with just one question:

Which of the following (history of previous heat illness, genetics, sleep deprivation, aerobic fitness, body size, drugs or supplements, dehydration, physical exertion unmatched to physical fitness, underlying viral or bacterial illness, male gender, exercise intensity, lack of acclimatization, clothing/equipment worn, and body composition) is an important “intrinsic” predisposing factor in exertional heat stroke?

I gave three instructions, 1) do not put your name on the paper, 2) intrinsic means nothing environmental like heat or humidity, and 3) please rank the top 5 from most important (1) to 5th most important (5).  The results were disheartening at best. Of the 71 upper level athletic training students surveyed, 63 ranked dehydration in the top 5 with 35 (51%) ranking it as the MOST important determinant of exertional heat stroke.  My only comfort comes with knowing that exertional heat illness and hyponatremia are not covered in depth until mid-semester of the senior year so none of the students has been correctly taught otherwise.

When these students were asked “why do you think that?” they responded “everyone tells you that dehydration causes heat illnesses”… including teachers, high school certified athletic trainers, coaches, parents, the media, doctors etc. As well-educated medical professionals, athletic trainers need to ask ourselves why so many of us have bought into this notion. Clinicians must be able to differentiate between peer-reviewed research and propaganda.  If your information about thermoregulation or fluid and electrolyte balance comes primarily from commercials, magazine advertisements, NATA News, pamphlets provided to you in exhibits at sports medicine meetings or from your mailbox then you need to consider whether, as an allied health professional, these sources are trustworthy in helping you make informed decisions about issues related to the provision of health care to athletes.  Although I say that, it is also true that the propaganda, which is largely funded by the sports drink industry, also finds its way into peer-reviewed medical journals far too often.

Ask yourself this question; does a modest level of body mass loss (usually referred to as dehydration) cause hyperthermia? The answer may well be no, but let’s not blame the laboratory research that linked dehydration to hyperthermia. Important discrepancies exist between classic laboratory studies (that support this notion) and more recent field investigations yielding different results.  In order to receive IRB approval for a laboratory study where body temperature might be affected, the usual cutoff for core temperature is 39.5 to 40ºC (103 – 104º F).  In a 2006 study conducted by researchers in Singapore, 18 runners agreed to have their core temperatures continuously monitored during a competitive marathon. Seventeen of the runners had a core temperature > 103 ºF which would exclude them from continuing to exercise in the vast majority of laboratories including my own.  Ten of the 18 runners reached a core temperature ≥ 104ºF which would preclude them from continuing in any laboratory in the world.  How then, can we expect to get useful clinical information about exercise associated hyperthermia if experimental studies cannot duplicate real-life situations in sports?

A similar problem exists for those who study MTBI.  In our field work using intestinal temperature sensors we have routinely documented core temperatures of between 103 and 105º F in numerous football players during preseason with no adverse effects or performance decrements. Why should we believe that this is not “normal” in athletes under certain conditions?  Additionally, in the study on marathon runners the authors concluded that “Core temperature responses demonstrated no significant relationship to absolute Δ mass ….or % dehydration”. Their subjects’ body mass losses ranged from 0.9% to 3.9% as is normal, and none suffered a heat illness. We have reported the same in college and professional football players and professional ice hockey players.


The other unsettling issue with laboratory (and even recent field research published in JAT) is that researchers commonly pre-dehydrate subjects prior to experimental trials so that they can induce the results they want. Many misconceptions exist in the ACSM position stand on fluid replacement which was largely copied by the authors of the NATA’s fluid replacement position statement. One of the most glaring ones is in the beginning when the authors outwardly choose to lump both hypohydration (a steady state of low body water after all fluid shifts have occurred) and dehydration (the process of losing body water when fluid shifts are still happening), and call them both “dehydration”.

The research is pretty clear that they are not the same when it comes to thermoregulation but combining hypohydration and dehydration allowed them to cite all papers related to either in support of the dehydration myths. This is, in my opinion, a blatant attempt to mislead the lay readers and the medical community. Some researchers use methodology that is clearly erroneous and clinically irrelevant just to support the unfounded claim that dehydration has a serious effect on body temperature during exercise. One such mythology involves bringing subjects in the day before the actual experimental trial and dehydrating them, and then fluid restricting them for 18 to 22 hours so that they show up to the exercise trial severely hypohydrated.  To top it off, these researchers then also fluid restricted the runners during the exercise and made them run at an intensity that their brains likely told them was not in their best interest. This will most certainly affect core temperature during exercise just for the simple fact that body water absorbs heat and they had significantly less of it. To top it all off, some researchers then refer to these subjects as being “dehydrated” which is again a blatant and successful attempt to mislead everyone. What athletic trainer in the world would fluid restrict their athletes during a Friday practice, not let them drink anything for the next 22 hrs and then fluid restrict them during the game on Saturday?  It is this type of research that is so often cited in fluid replacement position papers in support of the drink, drink, drink dogma.

Clinicians need to understand that euhydration (a normal state of hydration) is not determined by a single body mass measurement but rather, is defined as a body mass where serum osmolality (Sosm), the osmotic concentration of the serum, is maintained between 280 and 295 mOsm/kg.  Concurrently, urine osmolality between 400 and 800 mOsm/kg reflects “normal hydration”.  Sosm above 295mOsm/kg indicates a state of hypohydration (underhydration) whereas Sosm below 280 mOsm/kg represents a state of hyperhydration (over-hydration). This is so for good reason. Sosm of 280 mOsm/kg marks the point at which arginine vasopressin (antidiuretic hormone) is inhibited thus stimulating diuresis, and the release of naturetic peptides from the heart, brain and kidneys facilitates sodium excretion promoting free water clearance. Conversely, a Sosm of 295mOsm/kg stimulates thirst and the release of arginine vasopressin prompting drinking and water reabsorption.

In actuality, thirst and the release of arginine vasopressin occurs when Sosm reaches 287 mOsm/kg.   In addition, aldosterone is released from the adrenal cortex stimulating sodium retention. Varying degrees of both drinking and hormone modulation occur continuously to maintain Sosm within this range, and body weight fluctuates accordingly, both of which are normal and expected at rest and during exercise.

If we do the math with Sosm anchoring each end of the spectrum of normal hydration, 280 mOsm/kg (maximal normal hydration) and 295 mOsm/kg (minimal normal hydration) consider the following examples using a body mass with Sosm in the middle (287.5 mOsm/kg):  Body mass normally fluctuates between 78.7 kg (173 lb) and 81.3 kg (179 lb) in a 80 kg male, 64.1 kg (141 lb) and 65.9 kg (145 lb) in a 65 kg female, 29.5 kg (65 lb) and 30.5 kg (67 lb) in a 30 kg child, and between 147.5 kg (324.5 lb) and 152.5 kg (335.5 lb) in a 150 kg football lineman, all reflecting euhydration.

So that’s a 6-lb normal fluctuation in an average sized male. a 4-lb normal fluctuation in an averaged sized female, a 2.2-lb normal fluctuation in a child and amazingly, an 11-lb normal fluctuation in a football offensive lineman.  Certainly, in a 24-hour time period we are generally at the lowest level of hydration in the morning after a 6 – 10 h fast, which is why people who are dieting weigh themselves at that time.  This just reflects the lowest level of normal hydration and is accompanied by the sensation of thirst and the release of arginine vasopressin.  So, most people upon awakening in the morning have something to drink, which serves to decrease their serum osmolality (and increase their body weight) due to fluid intake.  It all works pretty well.

Those who believe that “dehydration” is any body mass less than the mass at which Sosm is 280 mOsm/kg are misinformed. In the 80 kg male, the mass difference between 280 mOsm/kg and 295mOsm/kg is 2.7 liters (~ 3%).  If the athlete was well-hydrated and weighed 81.3 kg (179 lb) at the onset of exercise, losing 2.7 liters of sweat during exercise does not represent “dehydration” by 3%.  In actuality, the athlete is at the lower end of “normal” fluid balance (Sosm = 295 mOsm/kg).  The starting point for dehydration (or hypohydration) would not begin until Sosm rose above 295 mOsm/kg with a concurrent body mass less than 78.7 kg (173 lb). So if a 350 lb football offensive linemen was well-hydrated at the start of practice, he could lose 11 lbs and still be normally hydrated (at the lower end of euhydration) at the end of practice. 11 lbs! 

It is often assumed that the human body is unable to maintain sodium and fluid balance during and after exercise, when in fact, humans have very well developed feedback loops involving complex integration of the brain, heart and kidneys which continually modulate fluid homeostasis. Even medical professionals often disregard basic physiology thereby promoting the idea that athletes need to drink prior to being signaled by their brain to do so. In defense, it is likely that our consumer driven society is responsible for perpetuating these misguided beliefs. All athletic trainers and other professionals who work with athletes need to do, however, is ensure that cold water and appropriate rest breaks are available during exercise sessions for ad libitum drinking. In other words, let athletes drink when they are thirsty.  This will keep them adequately hydrated (< 2.5 – 3 % body mass loss which is beyond the mass at which Sosm is 295 mOsm/kg), without the potential for over-hydration promoting hyponatremia.

And what about the silly idea that everyone needs to drink eight 8-ounce glasses (~ 1.9 liters) of water each day, beyond everything else that they drink?  This would not come close to satisfying the fluids needs of a 350-lb football player, however, if consumed quickly would surely make my 7-year-old, 52-lb niece hyponatremic (serum sodium would fall below 127 mmol/l) if significant diuresis did not ensue.  Clinicians need to pay attention to what makes sense and not let propaganda drive clinical practice.

The answer to the original question “what is the most important predisposing factor to exertional heat stroke” is that it is complicated, multifaceted and still not well understood.  Most good research suggests that in someone who is not taking drugs or supplements that raise basal metabolic rate, the top 10 factors include a history of previous heat illness, genetics, sleep deprivation, physical exertion unmatched to physical fitness, underlying viral or bacterial illness, male gender, and lack of acclimatization.  Dehydration is most certainly not one of these 10 factors related to heat stroke – but, by all means, athletes should drink when they are thirsty!

Consider this – sleep when you are tired, eat when you are hungry, urinate when your bladder is full AND drink when you are thirsty!!!  There is a reason why we have a brain – maybe we should listen to it!!

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: