Welcome to the Science of Sport, where we bring you the second, third, and fourth level of analysis you will not find anywhere else.

Be it doping in sport, hot topics like Caster Semenya or Oscar Pistorius, or the dehydration myth, we try to translate the science behind sports and sports performance.

Consider a donation if you like what you see here!

Did you know?
We published The Runner's Body in May 2009. With an average 4.4/5 stars on Amazon.com, it has been receiving positive reviews from runners and non-runners alike.

Available for the Kindle and also in the traditional paper back. It will make a great gift for the runners you know, and helps support our work here on The Science of Sport.

Sunday, November 23, 2008

No need to swallow?

Rinse your mouth out with sports drinks - it beats water...!

Once again, apologies for the absence - our posting frequency has fallen right off the charts, and we're down to an embarassing one per week! The end of the year tends to bring with it a log-jam of work, as all the outstanding work from the previous 10 months is suddenly squeezed into the final two months of the year! Combined with "post-fatigue", I'm tempted to say that like Olympic athletes, we're in the off-season period!

Also, very little seems to be happening in the world of sport, unless you count cricket and rugby, but there are already about a billion blogs and articles being written about those "small" sports, so I thought it best to leave them alone!

Instead, I came across a very interesting study this past week, which is actually a repeat of a study that was first published in 2004, in which simply rinsing the mouth with a glucose-containing drink (like Gatorade for those in the USA, Powerade/Energade for SA) improved performance during a time-trial lasting about one hour. You can see the abstract for the orginal 2004 study here, and the latest one here.

The 2004 study - rinse with a carbohydrate drink, not water

In the 2004 study, cyclists had to complete a given amount of work in as fast a time as possible. This is important to note, because as we've discussed previously in our series on fatigue, the MODE of performance trial can influence the conclusions drawn quite substantially. In both these studies, the cyclists were able to freely select their work-rate and slow down or speed up in response to the swishing around of the carbohydrate drink or the water.

The figure below shows the performance part of the research study, and summarises its results:

So, the rinsing with the malto-dextrin solution improvement performance by almost 3%, quite independently of the usual metabolic effects of glucose on performance. That is, our usual thinking would be that glucose is superior because it is rapidly metabolised, converted to energy, and delays the onset of fatigue. This study clearly challenges that, for a number of reasons.

First, the performance trial is quite short - the theory that glucose ingestion has a profound effect on performance is particularly true for long-duration exercise, where the depletion of liver or muscle glycogen may be possible. That said, there is some evidence that glucose ingestion improves performance even during much shorter exercise trials, which is difficult to explain because only a small part of your energy comes from that ingested glucose when you do shorter, high intensity training - that was one of the reasons this study was done in the first place.

Also, in this study, the performance of the cyclists was better on CHO from the very first quarter of the trial. So when performance is broken down, it turns out that you go faster within the first few seconds, and that clearly has nothing to do with metabolic effects.

Another important point is that the malto-dextrin solution was chosen because it is colourless and non-sweet, which means it's less likely to be obvious to the subjects when they're on it - this would otherwise introduce the possibility of a large placebo effect. It turns out that some subjects did pick up the difference: Four of the nine cyclists guessed they were on the CHO-trial, based on taste and viscosity. Of these four, three improved, one did worse. I guess the placebo effect can never be ruled out, but it does seem to have been controlled as much as possible in the study.

Increased central drive - feedback from the mouth, output to the muscles?

So what then is the explanation? Before that, a couple of cautionary words. Firstly, this study did NOT show that rinsing the mouth is better than swallowing the solution. It simply compared rinsing with CHO to rinsing with H2O. It is not, therefore, an excuse not to drink glucose-containing drinks during exercise! (the second study discusses this a bit more).

However, what it does do is help us understand how shorter exercise might be improved by carbohydrates. The authors of this study speculated that there are receptors in the mouth, which are stimulated by the CHO-drink (but not by water), and then trigger centres in the brain that then increase the central drive for exercise.

Those who've followed our site for a while now will recall our Fatigue Series, in which we described how it is the brain that regulates performance, by controlling muscle activation in response to various cues (like heat, oxygen availability etc.). This CHO-finding would fit in with that kind of thinking - the brain receives information from the mouth that glucose is available, and it allows an increase in muscle activation.

The ultimate outcome of this is that the perceived exertion is regulated, and performance is a function of that regulation. In other words, athletes pace themselves according to how they feel (this is obvious, I'm sure), and the rinsing with CHO changes that perception, which allows them to cycle faster, and thus achieve the same perception at a higher power output (again, read the Fatigue Series for more discussion on this)

If you think about it, this is really the ONLY way to explain how power output and hence performance are improved when you swish a drink around in your mouth. It cannot be metabolic, only sensory. And sensation (or perception) is, as you may recall, the key component of this regulatory system where the brain monitors perceived exertion and performance. Think also of the last time you went for a run or ride in very hot conditions, and what the first drink you had afterwards felt like - it's instantly pleasurable, and that's because of feedback to the brain, a different kind of physiology from the classic thinking that something in the body has gone "wrong".

The second study: Repeating the finding

This one study could be dismissed as a spurious finding, but it has just been repeated, in a study that has yet to be published but is available online. It followed an almost identical method to the first, so we won't bore you with the details, but will point out that it differs in that this second study also looked at ingestion of CHO and H2O.

The figure below summarizes the design and main finding:

So, a repeat of the finding that rinsing with the CHO-solution improves performance compared to rinsing with water (this time by almost 4%). There was not, however, a difference between ingesting the same CHO-mixture and drinking it, and nor was there a difference between rinsing and ingesting.

A couple of interesting points about this study - they did not use a malto-dextrin solution, but rather a commercially available sports drink, which means the sweetness/taste issue could be a factor. Quite why drinking the solution did not have an effect is difficult to know - it would have to pass the mouth and presumably trigger the same sensations as if it was rinsed. That remains an unknown.

However, the key point, a repeat of study 1, is that the subjects rode at the same Perception of Effort (or RPE), for a higher power output in the CHO-rinsing trial. Once again, this would suggest that the brain picks up the "reduction" in effort as a result of the glucose in the mouth, and the cyclist rides harder, with the result that the perception increases to the desired level at a higher overall performance level. It supports the notion that the sensation and perception are the regulators of performance.

A third study in the "sandwich" - no effect in runners

Finally, I must make mention that a study in 2007, between these two, did the same thing in runners and found NO DIFFERENCE - that is, rinsing with malto-dextrin did not improve performance in a 45-minute running trial. Is it a running vs. cycling difference? Or something related to habitual diet, a placebo effect? I suspect it may be related to the intensity, and the "cluster" of signals that determine how the athlete paces themselves. Running and cycling differ with regards to the eccentric loading on the muscles and joints, and I suspect this might be a key difference, negating the effect of the signals from the mouth. Difficult to know.

Perhaps a discussion for another day!

We'll hopefully post sooner than next weekend!



Pete-0 said...

Some years ago (mid-80's) I did some undergraduate work in neuroscience, specifically in the area of taste and smell. At the time, there was a phenomenon being explored called the "pre-absorbtive glycogen response." If I'm recalling correctly (and it's a big if!), the mere act of contemplating sugar consumption was enough to initiate small but measurable changes in blood chemistry as the body prepared for sugar intake.

Unknown said...

The perceived effort being influenced by swishing with CHO makes sense, after all, the placebo effect with medication has long been known to be true. If you believe you will do better with CHO, then you will!

Anonymous said...

Thanks for this interesting topic! In connection with the post from last week (doping), there are some very interesting studies comparing the absorption of i.v. glucose to that of ingested glucose (remember drips are "banned" by the latest anti doping regulations). Its some time since I read these investigations, but as far as I recall, the ingested glucose was faster in triggering cellular uptake etc. compared to the i.v. glucose (!) The authors postulated some central neural mechanisms connected to the act of swallowing that induced the reaction to glucose. So this topic is not new and has been explored to some extend. Interesting topic and another example of the fact that the human body isnt as simple as it seems ;-)

runtilyoudrop said...

If I had to guess I would say that as a result of the sugar in the drink the stimulation of insulin production occurred, via some higher cortical process rather than purely as a result of a raised blood sugar, and this enhanced glucose uptake into cells thus improving performance. It may be that the higher stresses associated with running complicate this and the effect is somehow lost.

Giovanni Ciriani said...

Hello Drs.,
I confess that I've not read the first study. However, back-of-the-envelope calculations would indicate there is something fishy in the findings.

Based on pure physics considerations, if you multiply Watts by time, you obtain the energy expenditure: 928 kJ with H2O, and 926 kJ with CHO. Uhm .. this is weird, almost the same, actually a little less for CHO. Why?

This is strange because energy expenditure is caused by friction of the tire, and aerodynamic drag. You can expect friction of the tire to be the same in both cases. However, aerodynamic drag goes with the square of the speed. Therefore, since in the CHO case speed increased by 2.8%, drag increased by about 6%, and energy expenditure caused by drag would have increased by as much. Assuming that aerodynamic drag is 2/3 and tire friction is 1/3 of total, you'd come up with a 4% increased energy demand. So something is wrong here. Perhaps CHO makes you crouch in a better aerodynamic position? :-)
Giovanni dot Ciriani at Gmail dot com

Anonymous said...

Hi Guys,
The other 2004 Carter paper with glucose vs saline infusion also supports a central effect of oral glucose ingestion.
Three questions about this topic. Firstly, when the Lode is set on linear mode, does this mean that to lower the power output one must slow their cadence down from their preset preferred rpm? If so, that seems like a limitation.
Secondly, in the swallow trials of the Pottier study, did the subjects first rinse then swallow or just quickly swallow the solutions? If the latter was the case could that be the difference?
Finally, in your diagram it shows that the CHO trials were maltodextrin, but in your post you said they were commerically available sports drinks. I haven't read the study, can you clarify?
Thanks, Matt

Anonymous said...

I wonder if it is the 5s of CHO rinsing, or the residual CHO left in the mouth (after spitting) that is important. What happens if you rinse and spit with CHO, and immediately follow with rinse and spit with water (to reduce/eliminate leftover CHO in the mouth?

Ross Tucker and Jonathan Dugas said...

Hi everyone

Thanks for the feedback. I'll try to reply one by one...

Peter Holstein

Interesting comment, thank you. We know that there are quite a few of these "feedforward/anticipatory" responses in the body. For example, if you tell someone they're about to exercise, their heart rate, breathing rate, stroke volume etc. all increase, mediated by the sympathetic nervous system. I suspect that your recollection is spot on and is a very plausible part of this performance improvement. It does again emphasize that the brain is sensitive to some change (the presence of sugar in the mouth) and kicks in systemic responses that are similar to those occuring if the actual change had occurred.

It must be pointed out that in both studies, subjects were "blinded" (as far as possible) to the nature of the drink they were rinsing with, so the "conscious" component was minimized as far as possible. However, I have no doubt that you're right, and some changes will have occurred, mediated by the brain.

Thanks for the insight!


As always, thanks for the comments (a regular!). You're right, the placebo effect is potentially large. I think the researchers did manage to limit this as much as possible, and only 4 of the 9 subjects "knew", and only 3 did better when they knew. The other 5 didn't know, yet still improved. So there is an argument to be made that it's not 100% placebo-mediated. I think (speculatively) that the sub-conscious is well aware, and mediates a big part of the change in performance.


Thanks for that - the issue of IV glucose infusion has been studied - the highest rates of glucose oxidation are achieved from IV infusion DURING exercise, much higher than ingested glucose. This is likely due to faster uptake/absorption. I'm not sure how it works at rest. But certainly, the act of swallowing is a part of a mechanism that helps the ingested glucose be used more effectively. I must point out though, that in both these studies, there was no swallowing, just spitting, which means the presence of glucose, not the act of swallowing it, is sufficient to cause whatever performance benefits were found.


I disagree - the role of insulin DURING exercise is very minor, and the reality here is that there is an IMMEDIATE change in performance, within the first few moments of the trial, and that's simply too rapid to be explained by some metabolic effect at the site of the muscle. It has to be related to the nervous system function,and the immediate response triggered by the afferent feedback from the mouth. I cannot see how a metabolic effect would explain this, particularly given that glucose is not likely to be limiting in this trial, lasting only 60 minutes.


The reason the energy expenditure is the same is because it's designed to be. If you read the post, you'll see that by definition, the cyclists had to complete a given amount of work, and the difference is the rate at which the work was achieved. So the fact that you can calculate the total energy to the same is no co-incidence - it was intended to be that way!

So in both trials, cyclists had to complete the pre-determined amount of work, but in the CHO trial, this was done at higher power output and thus in a shorter time period. So I don't see anything fishy with the result at all...


Yes, you're right, the pedalling rate issue is a limitation on the Lode bikes - we've moved away from using them for that reason. It's not so much a limitation though, because the same applies to both conditions, and so in the H2O trials, the ability to maintain cadence is impaired compared to CHO-trials. That's still an interesting finding, if a little "constrained" by the equipment.

Second, the "rinsing" method was the same as in Carter - 5 seconds of "swishing" the solution around the mouth, followed by spitting it out. I'm not sure what you mean by "the difference"? Is that referring to the running study where no effect was found? I'll have to have a look at that, because you may be right...

And then third, you're onto us - the diagram is wrong. Sorry - I copied the original and thought I'd edited it and changed the necessary details, but I forgot to change that block referring to maltodextrin. So to be clear, the Carter study used maltodextrin, the Pottier one used Gatorade, the "commercially available drink".

Thanks for the spot!


Good question. And I don't know. I suspect the "residual" glucose is important, in that it triggers the afferent feedback that the authors and us have speculated is responsible for the performance benefit. But honestly, I don't know, that's something for someone to test.


Anonymous said...

Any research on what effect using a sweet tasting but sugar/carb free solution has on performance compared to water and carb solutions?

Anonymous said...

During more-extended exertion (say 90 minutes or more), where carbohydrate availability can be limiting, it would be interesting to see whether swishing and swallowing produced better performance than swallowing alone (or rinsing by itself). In the former case, the athlete would have both the feedback from the mouth and higher carb availability.

Anonymous said...

So, a repeat of the finding that rinsing with the CHO-solution improves performance compared to rinsing with water (this time by almost 4%). There was not, however, a difference between ingesting the same CHO-mixture and drinking it, and nor was there a difference between rinsing and ingesting.

Not undrstanding the above..ingesting & drinking..? Is the difference (there is a difference 6.25-6.36) statistically to small?

Ross Tucker and Jonathan Dugas said...

Hi again

Thanks for the comments. Just a couple of short responses:


That would be an interesting study. Introduces a potentially large placebo effect, which is what study 1 here tried hard to avoid. So that's a confounder. The third condition would be a sweetened carbohydrate drink to correct for this, so that there are three conditions - water, sweetened water and sweetened carbs. Maybe someone will do it!


I suspect you're right, there'd be a temporary effect, but over time, as soon as the potential impact of hypoglycemia becomes important (after maybe 90 minutes, depending on pre-exercise diet), then performance would begin to decline. Interesting that you MIGHT be able to "fool" the athlete into starting too fast in the CHO-rinsing condition and potentially cause earlier "failure" in such long trials. That's a study also worth doing. I wonder if the overall longer trial, and resultant more "conservative pacing" would prevent you from finding the same effect though. The length of these performance trials is, as you can imagine, always an important consideration. Too short and you lose the effect, drowned out by very high power outputs. Too long, and the pacing is too conservative to have a physiological effect. Tricky balance...


Yes, pretty much, if I follow your question. The difference between ingestion of water and carbohydrates is not significant, either too few subjects, or the pattern of change was not consistent. As for ingesting vs. drinking, same thing. Rinsing is the difference. I'm not sure where I might have mixed them up, if at all?

THanks again

Anonymous said...

Very Interesting study. Reminds me of training with a friend for a marathon a few years ago. As the runs started to go over an hour and a half, her knees would start to ache. She finally started taking a sports gel when the aching began, and it seemed to subside immediately. Now there is no way anything could have gotten through her system that fast, but she swore by them. The brain is wacky!

Giovanni Ciriani said...

Lorne, I dont' think this study applies to gels.
Giovanni Ciriani

Ross Tucker and Jonathan Dugas said...

Hi GMC, Lorne

I think Lorne's example is a good one - it shows the power of the brain and perception during exercise. Gel or not, the amazing thing is that a systemic (or metabolic) effect is exerted simply through the action of taking a supplement, which flies in the face of traditional physiology that tells us that this cannot happen.

So while you're right, GMC, this study doesn't apply to gels, I can see the link and the example is an interesting one. The brain sure is whacky, whether it's in response to gels, placebos, or CHO-drinks (swallowed or rinsed!)


Anonymous said...

Sorry if I did not make my confusion clear (!), it is your terminology you say 'Not however a difference between ingesting CHO & drinking it, nor a difference between rinsing & ingesting.'
I got confused with ingesting vs drinking vs rinsing.

Ross Tucker and Jonathan Dugas said...

Hi Britspin

Thanks for clarifying - your confusion was my fault. I left a word out. That sentence should read "there was not a difference between ingesting the same CHO-mixture and drinking WATER", given the finding of the study that performance was not different in the two IN-trials (on the right of that graph). The next sentence is right - no difference between ingesting/drinking and rinsing.

So it's not a confusion of the word, but rather my incorrect use of "it" when I meant to say water.

Hope that helps. Thanks for pointing it out. Sorry for the error.


Anonymous said...

Perfect..now it makes sense..thought I was losing it for a while there!

Unknown said...

Coming to the party a little late... I've been told in the past that people do actually absorb simple carbohydrates directly into the bloodstream from the mouth (and thus if you're in the grip of a serious bonk that you ought to hold your gel/whatever in the mouth for a little while). If that's true, and 5 seconds of googling suggests it might be, it might have some bearing on the swishing.

Anonymous said...

There has been a study comparing sweet CHO solutions, non-sweet CHO solutions, and water on cycling performance. CHO ingestion improved constant-load cycling time to exhaustion independent of sweetness. Reference is:
Carter, Jeukendrup, Jones, Can J Appl Physiol. 2005 Aug;30(4):379-91