The ESSA/SDA Conference – Tim Noakes and the Hydration Debate - Part 1
As mentioned in the last post the recent ESSA/SDA Conference featured many memorable presentations in the sports nutrition field, but probably the most talked about was from Professor Tim Noakes. You may have heard of Tim from his books such as The Lore of Running and Waterlogged, or may be aware of his outspoken stance on hydration science. I’ve read the majority of Tim’s studies over the past few years, but it was the first time I’ve had the chance to see him present it all in one journey. To do this topic justice I’ve decided to review Tim’s presentation and arguments over two separate posts.
The Great Hydration Debate
For those of you unfamiliar with Tim’s stance of hydration, it goes something like this – thirst is the body’s way of regulating the level of fluid and electrolytes in the blood, and no amount of sweat losses will reduce performance provided fluid is available and the athlete drinks when thirsty. Quite a different stance to the usual dogma promoted by most sports scientists and dietitians - that a sweat loss of around 2% or greater of body weight will reduce performance.
So who’s right and wrong in this debate? Or does the reality lie somewhere in the middle? To examine this in detail let’s have a look at the arguments put forward by Tim at the conference, and the evidence that he provided to support them.
Argument #1 – Most studies claiming to study “dehydration” fail to measure the TBW and so are unable to draw appropriate conclusions
To investigate this claim, I went back to a review article published in 2003 1, which is the main piece of work on which the American College of Sports Medicine (ACSM) fluid replacement guidelines in endurance sports are based. The review article in question collates 13 studies on hydration and performance. Having gone back and looked at nine of these studies (the others were published prior to 1960 and are difficult to get hold of), Tim is correct that none measured Total Body Water (TBW), including two studies of which Noakes himself was an author.
Despite the lack of measurement of this in the vast majority of hydration studies, there are studies that have investigated the relationship between body weight loss from before to after exercise and the change in TBW. One such study, published in 2009, measured the change in body weight and TBW in a group of both men and women who were deliberately given either too much, not enough or just enough fluid to prevent body weight loss during over two hours of running intervals 2. They found that about 60% of the variation in TBW was explained by change in body weight. This suggests that measuring the change in body weight will at best only give a rough guide to TBW losses, supporting Noake’s argument for measuring TBW.
Despite the lack of TBW measurements, five of 13 studies in the 2003 review did measure blood plasma volume, and two measured plasma osmolarity (the concentration or dilution of solutes in the blood). This raises the question of what is it exactly that the brain monitors and adjusts performance to protect – Total Body Water (TBW), plasma volume, osmolarity or sodium concentration? Exercise has effects on all of these, but research to date suggests that the perception of thirst may be more sensitive to changes in blood osmolarity and sodium compared to volume 3. There are studies that look at altering osmolarity whilst keeping plasma volume stable by intravenously infusing small volumes of highly concentrated sodium fluids. But to my knowledge this has never been done to assess the independent effects of osmolarity or plasma volume (or TBW) on performance.
The relationship between the change in body mass and change in TBW. 60% of the variation in TBW was explained by variation in body mass. Source: Eur J Appl Physiol (2009) 105:959–967
Argument #2 – There is no evidence that "dehydration" causes anything other than the development of thirst
This is an interesting argument, and Tim showed data showing no evidence from studies where fluid is withheld from the “dehydrated” group that there were any health consequences from this. Rather, the athletes who have fluid withheld simply feel a greater perception of effort for the same level of actual effort, and one of two things happen:
- In fixed intensity, variable distance trials (Time To Exhaustion, TTE) the subjects who were not allowed any fluid cease exercise earlier and therefore cover less distance, but are not seen to exhibit any evidence of heat exhaustion or any other health concerns. Thus the subjects cease exercise before any physical harm occurs
- In fixed distance, variable intensity trials (Time Trials, TT) the subjects pace themselves slower and complete the distance in a slower time (or for a slower average power output).
Note that this occurs in studies where fluid is withheld (ie. participants are not allowed to drink even if they wanted to) – this becomes important when we look at the effect of fluid loss and consumption on performance.
The often mentioned “consequence” of dehydration is an increase in core body temperature. But it should be noted that in most lab studies of hydration the participants do not experience the same cooling effect that occurs outdoors due to a lack of wind resistance in the lab. Noakes’ group showed this in a 2005 study where participants cycled in 33oC heat with a consistent fluid intake whilst experiencing either 0%, 10%, 100% or 150% of expected outdoor wind resistance 4. When experiencing 100% or above wind resistance (as would occur outdoors) core temperature was significantly lower. In the study participants ceased riding either after two hours or when rectal temperature exceeded 40oC. Every single participant reached 40oC when riding with no wind resistance and had to stop exercising early. Only one participant made it to 40oC after the full two hours of riding with 100% of outdoor wind resistance.
The average effect of stationary cycling on core body temperature when the participants experienced 0%, 10%, 100% and 150% of expected outdoor wind resistance. Exercise was ceased either after 2 hours or if rectal temperature exceeded 40oC. Source: Acta Physiol Scand 2005, 183, 241–255.
In actual races a similar pattern is observed. At the 2004 Ironman Western Australia core temperature was measured in 10 competitors (seven of which finished the race in less than 10 hours) on a day that reached a maximum of 26oC – it was found that core temperature did not increase above 39oC in any competitor (expect for one strange value of 40.5oC at T1), and there was no relationship between core temperature and finish time 5. At Ironman South Africa a total of 767 triathletes were measured in the 2001 and 2002 races (max. temp of 20.9oC) 6. They found a significant but weak correlation between rectal temperature at the finish and performance (and fluid loss), but no competitors suffered from hyperthermia (heat related illness).
Source: Br J Sports Med 2006;40:320–325.
In shorter events raced at higher intensity the core temperature is significantly greater, which makes sense considering the higher pace. In a half marathon in Singapore (average temperature 26.4oC) twenty five runners had their pace and core temperature constantly measured 7. Ten of them experienced core temperatures of greater than 40oC, but if anything higher core temperature reflected a faster pace than runners with lower temperatures. None of the runners experienced any health consequences of these high temperatures.
So why do we see athletes collapse exhausted at the end of an Ironman or ultramarathon? No doubt everyone has heard of the infamous Ironman “crawl”, where athletes stumble and collapse in sight of the finish line.
The consensus is that collapse at the end of an ultraendurance event is rarely the result of dehydration or hyperthermia, but the result of Exercise Associated Postural Hypotension (EAPH). EAPH is a rapid fall in blood pressure due to pooling of blood in the feet (and lack of blood to the brain) 8. It often occurs after the finish line when athletes stop running, but can also happen when an athlete slows because they’ve hit the wall (hypoglycaemia or low blood sugar).
As early as 1994 a study of ultramarathon runners who collapsed found that core body temperatures were around 38oC, well below the levels that would be required to cause problems to the health and performance of the athlete 9. Of the eight competitors who collapsed during the race, three were diagnosed with hypoglycaemia, three with gastroenteritis, one with angina and one with asthma.
From my look through the available evidence I would agree with Noakes that dehydration does not routinely result in health problems in an endurance race situation. Instead athletes pace themselves, which amongst other things prevents core body temperature rising to dangerous levels. This reduction in pace reduces performance but as a result prevents any health consequences.
Clearly severe dehydration will result in health problems - people lost in the desert or example will experience this. However remember that the crucial difference is that people lost in the desert run out of fluid, and cannot drink despite extreme thirst telling them to do so. In virtually all endurance sports at least some fluid will be available to athletes to drink when they feel the need. The critical question for athletes (covered in Part 2) is how much fluid needs to be consumed to prevent any loss of performance.
That's part one completed. In Part Two I'll take a look at the other two core arguments from Tim Noakes, and summarise how I interpret the research and what's the missing pieces of the puzzle.
1. Cheuvront S. et al (2003). Fluid Balance and Endurance Exercise Performance. Current Sports Medicine Reports, 2:202–208.
2. Baker L. et al (2009) Change in body mass accurately and reliably predicts change in body water after endurance exercise. Eur J Appl Physiol (2009) 105:959–967.
3. Stachenfeld N. (2008). Acute Effects of Sodium Ingestion on Thirst and Cardiovascular Function. Curr Sports Med Rep. 7(4 Suppl): S7–13.
4. Saunders A et al (2005). The effects of different air velocities on heat storage and body temperature in humans cycling in a hot, humid environment. Acta Physiol Scand 2005, 183: 241–255.
5. Laursen P et al (2006). Core temperature and hydration status during an Ironman Triathlon. Br J Sports Med 2006;40:320–325.
6. Sharwood K et al (2002). Weight Changes, Sodium Levels, and Performance in the South African Ironman Triathlon. Clin J Sport Med, 12:391–399.
7. Lee J et al (2010). Thermoregulation, pacing and fluid balance during mass participation distance running in a warm and humid environment. Eur J Appl Physiol (2010) 109:887–898.
8. Asplund C et al. (2011). Exercise-associated collapse: an evidence-based review and primer for clinicians. Br J Sports Med 2011;45:1157-1162.
9. Holtzhausen L et al (1994). Clinical and biochemical characteristics of collapsed ultra-marathon runners. Med Sci Sports Exerc. 1994 Sep;26(9):1095-101.