Wednesday, June 22, 2011

Hydration and fluid consumption during endurance exercise – Does it make a difference?

Way back in October last year I discussed the various factors that lead to fatigue during exercise and therefore limit performance. Since this blog has looked at the role of carbohydrate in preventing hypoglycaemia, and the role of buffering supplements in preventing rising blood acidity during high intensity exercise. This time we’ll turn to the role of hydration and sporting performance, and have a look at the science in this area.



The Recommendation


Most sports nutrition text books and fact sheets will tell you that hydration is an important factor in maximising sporting performance. In 2010 the International Olympic Committee published their latest Consensus Statement on Sports Nutrition. The recommendations of the report were that “Athletes should be well hydrated before exercise and drink sufficient fluid during exercise to limit dehydration to less than about 2% of body mass.” A similar statement appears in the American College of Sports Medicine Position Stand on Exercise and Fluid Replacement from 2007. These statements are based on almost a century of research, starting off with desert hikers and soldiers in the 1930’s and progressing to athletes from a variety of abilities and sports since.



Are these recommendations based on a correct interpretation of the science?



However I think there’s one problem with the conclusions that were made in these two statements. When you go back and examine the studies closely, two things become clear:



1)      Almost all of these studies used a design with athletes were asked to exercise until exhaustion, and the difference in time-to-exhaustion between the well hydrated and the poorly hydrated groups is used to determine the effectiveness of replacing sweat losses during exercise.

2)      With very few exceptions these studies achieved dehydration by not allowing the athletes in the dehydrated group to consume ANY fluid. Those in the well hydrated group were generally asked to consume enough fluid to minimize the loss of body weight from sweating.



These two factors are important ones. Time-to-exhaustion testing is dramatically different to what occurs in real sporting events, and is now recognized as not a good way of studying exercise performance. Endurance sports almost always involve the athlete covering a certain distance in the shortest possible time, allowing the athlete to pace themselves however they see fit.



I also don’t know any endurance athlete (barring a major mishap) that would go into any race and not expect to drink for the entire event. Studies where fluid is completely withheld does not reflect the real world. Comparing good hydration to absolutely no hydration doesn’t make sense. A much more important question to study is whether you should deliberately drink more than your normal thirst instinct in order to completely replace sweat losses, or whether drinking according to thirst is sufficient to maxmise performance.



Alternative Research



So are there studies that represent what happens in the real world? In April of this year Eric Goulet from the University of Sherbrooke in Canada published a meta-analysis, a statistical analysis of all the available studies comparing athletes that didn’t replace sweat losses during exercise to those that did. Importantly he only included studies that measured performance by way of a lab-based “time trial”. That is the athlete was asked to cover a certain distance in the quickest possible time, just like in a race. The studies also had to be well controlled, so other factors like the amount of carbohydrate consumed didn’t interfere with the result. Amazingly after 80 years of research in the area only five studies were found that met these criteria, all using a cycling time trial as the performance measure. All five were published between 1995 and 2009.



The result of the meta-analysis – an average weight loss of 2.2% in the dehydrated groups compared to 0.4% weight loss in the well hydrated groups resulted in no difference in time trial power output whatsoever. It’s important to note however that four of the five studies in this analysis used performance tests of only around one hour – hardly endurance exercise. Only one study has evaluated this effect over a longer distance.



This study from the University of Cape Town in South Africa was published in 2009, and involved participants cycling an 80km time trial in the lab  (finishing time of just under 2hrs 10min). The interesting thing about this study is it’s one of only two that compare none or minimal fluid consumption with the participant’s own choice of fluid consumption (ad libitum drinking), as well as deliberate drinking to prevent weight lost from sweating. The study was also conducted in the heat (33oC), whereas many other studies were run in milder conditions.



The findings of this study make for very interesting reading. Participants who drank ad libitum improved their performance by 3% (or almost four minutes over 80km) compared to when not being allowed any fluid. But the really interesting finding was that drinking above ad libitum levels did not improve performance any more. In all there were six conditions in the trial – clients drinking no fluid, rinsing their mouth but not swallowing fluid, replacing 33% of sweat losses, 66% of sweat losses, 100% of sweat losses and ad libitum drinking (which resulted in 51% of sweat losses being replaced, and a 2.1% loss of weight over the duration of the time trial). The no fluid, mouth rinse and 33% groups performed worse compared to ad libitum, 66% and 100% fluid replacement.

Graph created from tabulated results in Dugas JP et al. Eur J Appl Physiol (2009) 105:69–80. Colours of the bars indicate where results were statistically significantly different (or not) from each other



There are also studies of hundreds of marathon competitors where body weight change (and therefore sweat losses not replaced) were compared with finishing times. It’s been found that there is an association between the amount of fluid lost (and not replaced) with finishing time in a marathon – the more “dehydrated” athletes are, the better the finish time! However that’s not to say that dehydration improves performance, because this sort of study cannot prove if one factor caused the other. What it does suggest is one of two possible scenarios. Firstly the faster runners may tolerate dehydration better than slower runners. Secondly and more likely is that this level of dehydration neither helps nor hinders performance during a marathon. The slower runners probably drink more simply because they are much more focused on simply getting to the finish line in one piece, and they drink more because they think it will assist them with this.



So what does it all mean?



The results from the University of Cape Town study suggest that whilst dehydration does reduce performance in an 80km cycling time trial, it only does so when fluid is not made available at all during exercise. It also suggests that an athlete’s natural instinct to drink provides enough fluid to prevent a reduction in performance, even if 2% or more dehydration occurs (at least for exercise up to two hours). However because no studies have ever compared ad libitum drinking to aggressive fluid replacement in longer performance trials, it remains unknown whether or not athletes need to drink  more than they instinctively would in order to maximise performance for events such as a cycling road race, ultramarathon or ironman triathlon.



Hyponatraemia – too much fluid can be life-threatening



There is also a very real threat from overly aggressive hydration strategies in endurance sports. Consuming more fluid than is lost in sweat results in a dilution of the electrolyte levels in the blood. The main electrolyte in the blood is sodium, and a low level of sodium (known as hyponatraemia) can have catastrophic consequences. There have been numerous cases of athletes who have become severely ill or have even died from severe hyponatraemia. Those at highest risk are those who run slower and those that set out to replace all (or more) of their sweat fluid losses. It’s also more common in cooler weather when sweat losses are lower. Research shows that adding sodium to sports drinks makes very little difference to the risk of hyponatraemia – the only thing that can definitely prevent hyponatraemia is not over-drinking fluid.



Summary & Final Thoughts



For years sports scientists and dietitians have recommended athletes drink enough fluid to prevent a loss of body weight from sweat of more than 2% during exercise. However this recommendation is based on studies using performance tests that don’t resemble real world sporting events, and in most cases didn’t provide any fluid at all to the “dehydrated” group in the trial. Only one study to date has been published comparing no fluid, ad libitum and aggressive fluid consumption, during a time trial performance test of around two hours duration. They found that ad libitum fluid consumption was sufficient to maximise performance, and additional fluid did not improve performance further. Whether ad libitum fluid consumption is sufficient in longer events remains unknown. Athletes should be cautious about overly-aggressive fluid replacement strategies (especially in cooler weather), because excessive fluid consumption can lead to potentially fatal hyponatraemia.



I have one final thought on the topic of how much to drink, and that’s to do with the practicalities. The human stomach empties at a maximum rate of just over 1 litre per hour during exercise. Even if sweat rates are higher than this there’s not much you can do about it. On the other side of the coin remember that sports drinks during endurance exercise are not just there to provide fluid, they also contain carbohydrate. Given that aggressive carbohydrate consumption in longer events has been shown to enhance performance (for information on this refer to my previous post on the topic), there’s a minimum amount of fluid that an endurance athlete would want to consume to get their carbs, particularly if they don’t tolerate solid foods well. Therefore on a bike most athletes would probably want to be drinking at least 500mL an hour to help maximise the carbs (unless sweat losses are less than 500mL an hour), but less than 1000mL an hour to prevent gastrointestinal discomfort. Runners may find 500mL an hour difficult to comfortably drink, let alone 1000mL an hour. In that case the upper limit to fluid consumption may be determined by tolerance.

6 comments:

  1. Hi guys,
    This post is very useful and very interesting to read.
    Really, this post is providing nice information, Keep it up!!!!
    calvincaline

    ReplyDelete
  2. Thanks Calvin, it's much appreciated. Good science is all about taking new learnings & knowledge, then going back to test it against your previous assumptions. This is true no matter whether you're building spaceships, curing cancer or just trying to maximise sports performance. Hopefully these posts will help to explain to athletes about how this process applies in the field of sports nutrition.

    ReplyDelete
  3. "unless sweat losses are less than 500mL an hour"

    How does one determine sweat losses?

    Is it sufficiently accurate to weigh before/after and discount any drinking and urination?

    ReplyDelete
    Replies
    1. Hi there,

      Yes weighing before and after exercise is about the only practical way we have of measuring sweat losses. It's been suggested that weight loss in ultra-endurance exercise (say >6hrs) would overestimate fluid losses, but unfortunately there's no better way of working it out. Also bear in mind that in longer events the intensity/pace is usually lower, which will act to reduce the sweat rate compared to a shorter but higher intensity session.

      Delete
    2. Well done for all your hard work in providing this high quality blog.Thanks for great information you write it very clean.I am very lucky to get this tips from you.

      Delete
  4. This comment has been removed by the author.

    ReplyDelete