If you are one of the thousands of people taking on this weekend's London Marathon, or any of the other spring marathons, do you know how much fluid you need to drink on the run? A litre per hour? Or 400ml? There's a simple indicator to help you find out. It's called thirst. Despite years of advice to the contrary, many experts, including the International Marathon Medical Directors Association (IMMDA), now recommend exercisers drink only when they are thirsty – not as a strategy to prevent dehydration.
If you haven't been following the debate about fluid intake over the past couple of decades, this may strike you as blindingly obvious - but when I ran my first London Marathon, back in 1996, the American College of Sports Medicine (ACSM), a leading authority on sport science, had just issued a statement saying: "During exercise, athletes should start drinking early and at regular intervals in an attempt to consume fluids at a rate sufficient to replace all the water lost through sweating (body weight loss), or consume the maximal amount that can be tolerated." Thirst, the statement said, is not the best indicator of dehydration or the body's fluid needs.
The notion that it's no use relying on thirst to tell you when and whether to drink has become commonly accepted among sportspeople. For example, research last September from Loyola University in the United States found that 36.5% of runners drink according to a preset schedule or to maintain a certain body weight, while 8.9% drink as much as possible. But according to Dr Ross Tucker, an exercise physiologist and consultant scientist to the Sports Science Institute of South Africa, such practices are wholly misguided.
"The body's thirst mechanism is absolutely exquisite," he says. "It is controlled primarily by the sodium level in the blood, and the moment this rises, as a result of losing water, we become thirsty." Even during exercise, when we're more likely to be tuned into our pounding heart and burning lungs than a dry mouth? "It's likely that during exercise, thirst is adjusted so that we don't need to drink all the time (which isn't efficient) but provided we have access to fluid when we do become thirsty, we stay safe," says Tucker.
While the ACSM guidelines were updated in 2007, and no longer recommend either specific volumes of fluid per hour or guzzling down as much as you can, they do still define 'excessive dehydration' as greater than 2% body weight loss from water deficit. And yet according to Ruth McKean, a dietician at the Scottish Institute of Sport, studies show far greater levels of dehydration among many elite distance runners – who frequently cross the finish line 6-7% dehydrated.
"What we're learning is that we can't give blanket advice about fluid intake that will be appropriate for everyone," says McKean. 'Some people can tolerate greater levels of dehydration than others and the oft-quoted 2% dehydration being the threshold for problems isn't borne out in evidence."
Tucker agrees. "In races, athletes who lose the most fluid are often the winners – which makes sense because they're running the fastest, and hence have the highest sweat rate. And when you are running along at 3 min/km, you don't have the time to drink all that much either. A race winner will typically lose three to four kilograms in a marathon, and they hardly drink at all - you'll see they sip water or sports drink every few kilometres, but that's it. There's none of this 'one litre per hour' doctrine that others are following."
The idea that there is a 'one size fits all' volume of fluid that everyone needs is now firmly in the past among the sport science fraternity. Both the ACSM and IMMDA's current guidelines support the use of customised fluid replacement programmes – that is, working out your own individual fluid needs through a 'sweat test' (see below). The theory is that by weighing yourself naked before and after exercise, you can work out how much fluid you lose over a given timescale, which can inform your decision of how much to drink in future sessions. But it's not watertight.
"The problem is that the test is so environment-specific," says Tucker. "On a given day, you might measure that your sweat loss is 1200ml/hour, but on another it may be 800ml/hour. Unless you know exactly what your sweat rate is for a given exercise intensity, temperature and humidity, you're guessing. So doing tests is unnecessary and potentially unsafe because of the risk of getting it wrong."
McKean agrees that sweat tests can oversimplify things. "Use the results as a rough idea rather than an exact guideline, and if you're serious about monitoring your fluid loss, perform the test regularly, under different conditions, different temperatures and paces to build up a picture," she advises. "And use this knowledge in conjunction with thirst, not instead of."
While the original aim of such tests was to prevent dehydration, McKean is more concerned about people overdrinking as a result of well-intentioned hydration strategies.
If drinking too much on the run simply meant spending too many precious minutes in the portable toilets, it wouldn't be such a big issue. But cases of exercise-associated hyponatraemia (a potentially fatal condition in which blood levels of sodium are dangerously diluted as a result of too much water) have risen significantly over the past two decades (including a number of deaths in marathons and other endurance events), and the 'drink, drink and drink some more' message has undoubtedly been a contributing factor. "People don't die of thirst in races, but there have been many cases of hyponatremia and the symptoms can be quite similar," says McKean (see below).
And if you thought that drinking an isotonic sports drink instead of water rendered you immune from the dangers of hyponatraemia, think again. "The sodium content of a sports drink helps attenuate the drop in blood sodium levels, but not by much," warns Tucker. "People who drink too much could probably get away with drinking 1.4 litre/hour of sports drink, but only 1.2 litre/hour of water. But it's a moot point - the root cause is still overdrinking, and that's what should be avoided."
The other reason Tucker is set against basing your fluid needs on anything other than thirst is that it's simply unnecessary. "Our bodies are adapted to lose fluid. They had to be when our survival depended on running to hunt, when we would often be unable to drink. Humans are delayed drinkers. We tolerate fluid loss really well, and we can drink later."
Recent research on the drinking habits of elite endurance runner supports this. A study published in the Journal of the International Society of Sports Nutrition last year found that among elite Ethiopian runners, no fluids were consumed before or during training, with only modest amounts being drunk afterwards. Compare that to the average western recreational runner who won't set off for a 5km trot without a bottle.
An earlier study on Kenyan runners found the same thing - athletes did not consume fluid before or during training, with significant amounts of body weight being lost through sweat. Theoretically, this should spell disaster – body weight loss means that the runners were dehydrated. But when the researchers measured 'total body water' and pre-training body weight on a daily basis, they discovered that these were well maintained – in other words, over the course of the day the runners spontaneously replaced lost fluid and rehydrated by drinking 'ad libitum' or to thirst, rather than on a schedule.
But what if, like me, you've spent your running years surgically attached to your water bottle? Could your thirst mechanism be a little rusty? According to McKean, thirst can be a less effective indicator in older people, who may need to be more vigilant – but she stands by the 'drink to thirst' message, adding only that we "still need to exercise common sense."
Tucker says that studies seem to show that drinking to thirst does result in some people replacing more fluid than others. "But in no case have I ever seen anyone voluntarily run themselves into dehydration danger by losing more than say, 7 or 8% of their mass. Some lose 1%, others lose 4%, some 6%. All of them are OK."
The sweat test
• Weigh yourself (in kg) naked before a training run of a measured time (30 or 60 minutes).
• Do not drink any fluid during the run.
• Towelling off excess sweat first, weigh yourself again when you get back (don't go to the toilet between each weigh-in).
• The amount of weight you have lost represents water loss – each gram lost equates to 1ml fluid – not a shred of it is fat! If you ran for 30 minutes, multiply the figure by two to get your hourly fluid loss. If you ran for 60 minutes, the figure you have represents your hourly fluid loss.
• Bear in mind that factors such as the temperature, humidity and your running pace will influence fluid loss. So the figure you come up with may need to be adjusted up or down on future occasions.
Example: before a half-hour run, you weigh 63kg. After the run, you weigh 62.6kg. That means you lost 0.4kg (400g), equalling 400ml fluid. So in an hour you would have lost 800ml.
Symptoms of dehydration
Disorientation
Fatigue
Nausea
Headache
Muscle cramps
Symptoms of hyponatraemia
Disorientation
Fatigue
Nausea/vomiting
Headache
Dizziness
Seizures
Coma
Swollen hands and feet
Breathing difficulties
Muscle weakness
Serum sodium levels <135 mmol/L (determined by blood test)
