The role of carbohydrates in endurance sports

How ‘carbohydrate rinsing’ can boost sports performance

Everyone knows that carbohydrate drinks can enhance performance. But as Asker Jeukendrup explains, new research suggests that actually swallowing your favourite sports drink might not be necessary. Bizarre as it sounds, you can rinse your mouth, spit the drink out on the ground and still be faster!During the 1980s, carbohydrate CHO supplementation received a lot of attention and a large number of studies demonstrated that fatigue was delayed and performance was improved when exercise lasted two hours or longer. These studies are summarised in recent reviews(1,2). The reason for the improved performance may be the prevention of hypoglycaemia. However, probably more important is the maintenance of higher rates of carbohydrate oxidation. These high rates of carbohydrate oxidation allow higher work rates to be maintained.

The optimal carbohydrate dose is still open to debate, as dose-response studies have not given clear answers. Some studies have shown a dose response relationship and concluded that more carbohydrate is better. However, a larger number of studies have found no relationship (1, 2). A recent ACSM position statement recommended athletes take 30-60 grams of carbohydrate per hour (3). This is partly based on the finding that ingested carbohydrate cannot be oxidised at rates higher than 60 grams per hour(1,2). But these guidelines may be out of date already, especially for exercise lasting three hours or longer.
Recently it was demonstrated that when a combination of carbohydrates is ingested that use sugars with different intestinal transporters (ie glucose:fructose) this can result in very high  ingested carbohydrate oxidation rates – as much as high as 105 grams per hour (4). When this glucose: fructose drink (2:1) was compared with a glucose drink and performance was measured during a 3-hour cycling trial, it appeared that the glucose:fructose drink improved performance by 8% compared with glucose and 17% compared with a placebo (7).

These data suggest that higher exogenous oxidation rates may result in better performance, at least during very prolonged exercise. These effects are only seen when large amounts of carbohydrate are ingested (ie 90 grams per hour). The potential benefits of glucose/fructose drinks were discussed in Peak Performance issue 233 and an overview of up to date recommendations is given in table 1 below.

Table 1: Recommended carbohydrate intake during exercise in order to optimise performance
Exercise duration CHO needed CHO intake Comments
15-45 min Very small amounts of CHO
45 min-2h Small amounts of CHO Up to 30g/h Can be achieved with most forms of CHO
1.5-3h Moderate amounts of CHO Up to 60g/h Can be achieved with CHO that are rapidly oxidised (glucose, maltodextrins)
>2.5h Large amounts of CHO Up to 90g/h Can only be achieved by intakes of multiple transportable CHO
(glucose:fructose; maltodextrin:fructose)

Carbohydrate and short duration exercise

Twelve years ago, we discovered that carbohydrate feeding can also improve performance during shorter duration exercise of higher intensity (8). We studied cyclists who performed a 40km time-trial with or without carbohydrate and on average, they were 1 minute faster with carbohydrate. This was a large and unexpected effect for which we did not have an explanation at the time.

During exercise of one hour or less duration, hypoglycaemia does not develop and blood glucose concentrations may even rise. Also, it takes time before any ingested carbohydrate is absorbed, transported to and used by the muscles, so we calculated that only a small percentage of the carbohydrate ingested during these time trials was actually used as a fuel. This amount was thought to be too small to provide additional fuel and result in a beneficial effect.
In order to further study the potential role of carbohydrate as a fuel during this type of exercise, we asked cyclists to perform a 40km time-trial. On one occasion we infused them with a glucose solution, and on another occasion we infused saline (salty water)(9). The cyclists did not know what they were getting on each occasion.

We observed that when glucose was infused blood glucose concentrations were twice as high and glucose was taken up into the muscle at high rates. However, even though this glucose was going into the muscle and was probably utilised, there was no effect on performance. This tells us that providing fuel during this type of exercise is not that important and other factors determine performance. But if carbohydrate does not exert its effects through being used as additional fuel, how can we explain the performance benefits during a 40km time trial?

Mouth-brain connection

An alternative explanation could be that the carbohydrate somehow influences the brain. For example, there is evidence that taste influences mood and it may also influence the perception of effort. An interesting observation provides support for a central nervous system effect. When you are hypoglycaemic after a long ride or run without food and you are feeling weak and dizzy, all you have to do is bite into a chocolate  bar to feel better. Almost instantly the feelings of weakness and dizziness are reduced, and you feel better long before the carbohydrate has reached the blood circulation and the brain. This means that there must be connections from the mouth directly to the brain.

This may also explain why we found improved performance during a 40km time trial. In the following study, we asked cyclists to repeat the 40km time trial but only rinse their mouth with a carbohydrate solution without swallowing any of it(10). The carbohydrate used in this study was a non-sweet maltodextrin solution, containing carbohydrate, but tasteless. The rinsing protocol was standardised. Subjects would rinse their mouth for 5 seconds with the drink and then spit the drink out into a bowl. They were not allowed to swallow any of the drink.

The results were remarkable: performance was improved with the carbohydrate mouth rinse compared with placebo and the magnitude of the effect was the same as the effect we had seen in the early study with carbohydrate ingestion! They were about 1 minute faster, even though none of the carbohydrate had actually entered the body (no carbohydrate is absorbed in the mouth). Perhaps the carbohydrate in the mouth rinse had connected with a receptor in the mouth that subsequently sent a signal to the brain. This signal probably informed the brain that food was on its way and this reduced the perception of effort, making the exercise task easier. These results were reproduced in studies that followed, although not all studies have found these effects (see discussion below).

Brain imaging

In follow-up studies conducted at the University of Birmingham, brain scans using a technique known as fMRI were used to see if there was increased activity in certain areas of the brain with a carbohydrate mouth rinse that was not present with a placebo mouth rinse (11). Indeed the study revealed that with a carbohydrate mouth rinse, certain areas of the brain such as the reward centre and areas involved in motor control were activated. The areas investigated included the insula/frontal operculum, orbitofrontal cortex and striatum (see figure 1).

During strenuous exercise many incoming signals arising from muscle, joints, lungs, skin and core temperature receptors are sent to the brain. Over time, these signals will be perceived as unpleasant and consciously or unconsciously this will lead to an inhibition of motor output. This is often called ‘central fatigue’.

Athletes tend to regulate their physical activity to keep their levels of discomfort within acceptable limits. It is not clear exactly which pathways are involved in this inhibitory activity but it seems plausible that the signals arising from the carbohydrate receptors in the mouth are counteracting some of these negative signals. Perhaps the sensors are telling the brain that: ‘you have nothing to worry about, because energy is on its way!’ The exact nature of the communication is unknown but the studies clearly show that there is a huge amount of communication between mouth and brain, even before any carbohydrate is delivered.

Sweetness or carbohydrate

Another question that arises is whether it’s the carbohydrate that has this effect or the sweetness or taste of the drink? Interestingly, the brain centres that responded to a carbohydrate mouth rinse did not respond to sweetness. When a drink with artificial sweetener was used to rinse the mouth no activation of these areas occurred. However, when a maltodextrin solution, which is a carbohydrate that is not sweet and has virtually no taste, was used, these areas of the brain were equally activated to glucose.

Together, these findings suggest that there are receptors in the mouth that detect carbohydrate and that these receptors are separate from the receptor that detects sweetness. However, such ‘carbohydrate receptors’ have not yet been identified in humans. These findings are also in agreement with some of the performance studies we conducted. When we compared a sweet and a non-sweet carbohydrate and asked cyclists to perform another 40km time trial we observed similar performance improvements (12).

During what type of exercise does mouth rinsing work?

Carbohydrate ingestion seems to work when the exercise is longer than 30 minutes. Exercise shorter than that does not seem to benefit from carbohydrate intake (13). Recently a study showed that mouth rinsing with a CHO solution increased total distance covered during a self-selected 30-minute run in comparison with mouth rinsing with a colour and taste-matched placebo (14). Similar results were obtained during a 60-minute self-paced run (15).
In another study, the influence of ingestion and mouth rinsing with a carbohydrate solution on the performance during a high-intensity time trial (~1h) was investigated in trained subjects (16). Subjects either rinsed around the mouth or ingested a 6% carbohydrate solution or placebo before and throughout a time trial. In the mouth rinse conditions, time to complete the test was shorter with the carbohydrate mouth rinse (61.7 minutes) than with placebo (64.1 minutes). Interestingly in this study, when drinks were swallowed and not rinsed, there was no difference between the placebo and carbohydrate drinks.
Also, in another study at the University of Birmingham, a 1.9% improvement in time-trial performance was observed with a carbohydrate mouth rinse compared with placebo. In two other studies no effect was observed when subjects ingested a breakfast before the time trial(17) or during running (18). So overall, the effect of a carbohydrate mouth rinse is convincing and seems to be significant for exercise lasting 30-60 minutes. It is not clear whether shorter exercise can benefit and it is unlikely that the mouth rinse effect can override some of the other factors that cause fatigue during more prolonged exercise.

In practice

So what does all this mean in terms of practical advice? Well, it appears that it’s not necessary to take on board large amounts of carbohydrate during exercise lasting approximately 30 minutes to one hour. Simply rinsing your mouth with carbohydrate may be sufficient. I have often seen athletes with lollypops and little sweets in their mouth before and during competition. Maybe that’s a practical solution?
It must also be said that in most conditions, the performance effects with the mouth rinse were similar to ingesting the drink, so there does not seem to be a disadvantage in taking the drink (although occasionally athletes may complain of gastro-intestinal distress when taking on board too much fluid). Of course when the exercise is more prolonged (two hours or more), carbohydrate becomes a very important fuel and it is essential to take it on board.
We will see what the future holds but it’s not difficult to imagine some messy feed zones with athletes taking sports drinks and then spitting them out on other athlete’s shoes. A final word of caution however; if you use this practice in Singapore you may be fined $500!

Asker Jeukendrup is professor of exercise metabolism at the University of Birmingham, has published over 150 research papers and books on exercise metabolism and nutrition and is also consultant to many elite athletes


1. Nutrition 20, 669-677, 2004
2. European Journal of Sport Science 8, 77-86, 2008
3. Med Sci Sports Exerc 39, 377-390, 2007
4. Br J Nutr 93, 485-492, 2005
5. JAMA 82: 1778-1779, 1924
6.  Am J Physiol 108: 203-209, 1934
7. Med Sci Sports Exerc 40, 275-281, 2008
8. Int J Sports Med 18(2): 125-9, 1997
9. Med Sci Sports Exerc 36(9): 1543-50, 2004
10. Med Sci Sports Exerc 36(12): 2107-11, 2004
11. J Physiol 587, 1779-1794, 2009
12. Can J Appl Physiol 30(4): 379-91, 2005
13. Eur J Appl Physiol 104, 831-837, 2008
14. Int J Sport Nutr Exerc Metab 18, 585-600, 2008
15. Med Sci Sports Exerc 41 (5), p 21 (abstract), 2009
16. Scand J Med Sci Sports, 2008
17. Int J Sport Nutr Exerc Metab 19, 400-409, 2009
18. J Sports Sci 25, 1385-1392, 2007