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.

Wednesday, November 21, 2007

Muscle Cramps: Part II

The electrolyte depletion model of muscle cramps

In part one of this new series we tried to set the scene by providing some history in this area of muscle cramps. At times it might seem like we are a bit heavy on the historical side, but as we mentioned in one of our comments to Part I, understanding the historical record is crucial as often it helps us understand why we think what we do---and this affects one's interpretation of the science. In this post we will focus on the prevailing premise that dehydration and electrolyte disturbances cause muscle cramps.

The first important thing about this area of research is that Professor Martin Schwellnus is hands down the one researcher who has consistently moved this area forward. As a sports physician he has treated many a runner with cramps, and his curiosity and what he was seeing in the medical tents lead him to challenge this paradigm that dehydration and electrolyte problems cause cramps. What he found was that this model was based on not one shred of scientific data, and instead relied heavily on anecdotal evidence. Since 1997 he has published some of the only evidence available that has even attempted to determine what actually is causing the cramps and who is prone to this condition. The first paper he published in 1997 proposed a novel hypothesis for muscle cramps, but we will address that in Part III of this series.

The lab vs. the field

In our series on dehydration we discussed how the lab is not always translatable to the field, and vice versa, but that each has its own important role. Field studies are often cross sectional in nature, and although important we cannot assign direct cause and effect from them. However it is observations and findings from field studies that often lead to the very precise and mechanistic lab studies that are important in advancing our knowledge.

However one major obstacle in this area (cramping), is that no one has yet created a laboratory protocol in which we can reproduce muscle cramps in a controlled manner. Being able to do this is a crucial step in eventually identifying what causes them because it will allow us to make specific interventions to test what the effect is on cramps. So although we are still in the infancy of this area of research, the field studies are a very important starting point and have so far yielded important findings.

One study published in 1990 showed that there was no association between potassium levels and cramps. In that study cyclists rode for up to five hours. Some of the subjects did cramp, but their potassium levels were not uniformly high or low, thus showing no association between that variable and the cramps. However beyond that study (and one more that was presented at a conference but apparently not published) there is little real data out there to support or refute this hypothesis that dehydration or electrolyte disturbances cause cramps.

Study 1: Two Oceans Ultra Marathon

In a 2004 study published in the British Journal of Sports Medicine, Professor Schwellnus and his colleagues examined runners before and after the Two Oceans 56 km marathon in Cape Town. They measured quite a few variables, but since we are discussing changes in electrolytes and hydration, we will talk about those results. Remember that many people, both scientist and personal trainer alike, will profess that cramps are caused by dehydration and/or some disturbance in the electrolytes (sodium, potassium, magnesium, etc.) So the important finding from this 2004 study was that when the crampers were compared to the controls---who were matched for body mass and finishing time---the only differences were that the crampers had lower sodiums and higher magnesiums. The problem with this is that a lower sodium concentration suggests overhydration and not dehydration, and also if magnesium deficiency is meant to cause cramps then surely the crampers should have been lower here?

Crampers (N = 21)

Controls (N = 22)


139.8 ± 2.1

142.3 ± 2.1


4.9 ± 0.6

4.7 ± 0.5


0.73 ± 0.1

0.67 ± 0.1


280 ± 6

284 ± 10

The relevance of this study is that if dehydration and electrolyte disturbances really play such a large role in cramps (as they are proposed to), then the crampers should have much higher electrolyte concentrations since they would be losing fluid and causing the concentrations to rise. Yet instead we see something entirely different, first that the crampers had lower sodium concentrations, and second that the crampers were not really different compared to the controls.

What is also noteworthy from this study was that the crampers had an average loss of body weight of 2.9%, compared to 3.6% for the non-cramping controls. In otherwords, the people who DID NOT cramp lost more weight than the people who did. It goes further than this, because Schwellnus et al were able to measure the change in plasma volume as well - a more direct measure for what is happening to fluids. Here, they found that the crampers actually gained a small amount of 0.2% during the race. The non-cramping control subjects LOST 0.7%. So the sum effect of this data is that it suggests very strongly that cramping is not associated with dehydration, or with lower serum electrolyte levels, which is what we have had drilled into us for many years!

The follow-up study from Iron Man - further evidence against serum electrolytes

The next year they published a study in Medicine and Science in Sports and Exercise, and instead of runners it was Ironman triathletes. According to what most of us hear day in and day out, it is these ultra-distance athletes who are exercising for 10+ hours at a time that must be most susceptible to dehydration and electrolyte deficiencies. After all, they are sweating for hours on end, and the numbers tell us that with so many liters of sweat lost then they must also be losing grams and grams of "essential electrolytes" such as sodium. Below you will see the basic data on these athletes, and the important finding here is that we see the crampers and controls were the same age and were similar in mass, had similar pre to post cahnges in mass, and also finished the Ironman in similar times:

Crampers (N = 11)

Controls (N = 9)

Age (years)

33.5 ± 8.8

35.4 ± 8.1

Pre-race mass (kg)

79.1 ± 5.9

77.7 ± 6.4

Post race mass (kg)

76.3 ± 5.6

74.6 ± 6.5

Body mass loss (%)

3.4 ± 1.3

3.9 ± 2.0

Total race time (min)

660.8 ± 77.9

685.7 ± 48.5

So the two groups were essentially the same in that the crampers did not spend longer in the course or lose more weight (a crude measure of dehydration). Yet again the crampers and the controls looked remarkably similar on paper---except as in the 2004 study the crampers again had a statistically significant lower sodium concentration, and, we will repeat this, that suggests they were more hydrated compared to the controls. . .yet they were cramping. Here are the data from the electrolytes in the two groups:

Crampers (N = 11)

Controls (N = 9)


140 ± 2

143 ± 3


4.4 ± 0.06

4.2 ± 0.5


0.9 ± 0.2

0.8 ± 0.1

Recall that what is most often put forward as the cause of cramps is either dehydration or some electrolyte disturbance, but the data from these two studies do not support that hypothesis. Although these are field studies and we cannot assign a cause and effect relationship, this available evidence suggests that these (normal) levels of dehydration do not appear to cause cramps. If these levels of dehydration did cause cramps and were largely responsible for cramps, then what we should see is a very high incidence of cramps in all of the race finishers with the same physiological characteristics as these subjects----or in other words, the vast majority of the race finishers.

Rejecting the old models

In science when the available evidence does not support the hypothesis, we must change the model. Based on this available evidence we see clearly that dehydration and electrolyte levels are not associated with muscle cramping during or after exercise, and therefore we must adopt a different model to explain what is causing them. We cannot just ignore the data we have shown here and keep on telling people that it is dehydration and electrolytes when new evidence suggests otherwise.

So in Part III of this short series we will lay out the newest hypothesis that tries to explain the "why" and the "how" of muscle cramps. It is novel and, as you might have guessed already, has nothing to do with electrolytes and dehydration! So come back and join us for Part III of this series, and then join us for the comments and debate!

See also:
Part I: Theories and fallacies of muscle cramps

Brouns F et al., "Ammonia accumulation during highly intensive long-lasting cycling: individual observations." International Journal of Sports Medicine. 1990 May;11 Suppl 2:S78-84.

Schwellnus MP et al., "Aetiology of skeletal muscle 'cramps' during exercise: a novel hypothesis." Journal of Sports Sciences. 1997 Jun;15(3):277-85.

Schwellnus MP et al., "Serum electrolyte concentrations and hydration status are not associated with exercise associated muscle cramping (EAMC) in distance runners." British Journal of Sports Medicine. 2004 Aug;38(4):488-92.

Schwellnus MP. "Muscle cramping in the marathon : aetiology and risk factors." Sports Medicine. 2007;37(4-5):364-7

Sulzer NU et al., "Serum electrolytes in Ironman triathletes with exercise-associated muscle cramping." Medicine and Science in Sports and Exercise. 2005 Jul;37(7):1081-5.


Anonymous said...

Looking forward to the final answers on cramping! Seriously though, I am a marathoner and calf cramping has been my big problem. I never cramp in training, even when it is far hotter than any marathon I've run. I only have the cramping problem when racing. I can think of two obvious differences between racing and training for me. One is that I run harder overall in marathons than in training. (Of course I do tempo and intervals and marathon pace runs, but I don't crank out 26 miles at marathon pace in a training run). Two is that my legs take more of a pounding in a marathon because they usually are on concrete, and I try to avoid long training runs on concrete.

Anyway, looking forward to tomorrow's post.

Ross Tucker and Jonathan Dugas said...

Hi Anonymous, and thanks for joining us here at The Science of Sport.

You have correctly identified an important factor in cramping---muscle fatigue. Without spilling the beans too much for Part III, I can say that muscle fatigue is normally associated with the cramping muscle(s), hence the fact that they occur in the muscles used most actively for that activity. So the quads, hamstrings, and calves for running (or cycling).

At least you realize that it is perhaps some difference in the nature of the activity, and did not waste piles of cash on over the counter "cures" for muscle cramps.

As a side note, Part III will not go up on Thursday, but probably Friday or on the weekend. So you might have to hang in there for it, but we promise it will be worth it!

Kind Regards,

Meg & Dave said...

This is interesting...

It seems like you are having more and more fun with your posts, really well written. The suspense is kind of fun!

I'm not sure you've really defined "cramping" I associate cramping with three distinct things.

One being sprinting past the anaerobic limit where muscles spasm while cooling down, one where the spasms happen at about mile 90 of a hilly century ride, and one from kneeling too long with my legs squished. All three of these result in hamstring, quad, calf, shin, or foot cramps.

Are these all the same outcome? They feel pretty similar but result from very different things.

Just a curiosity that came up while reading.

Looking forward to the next post!


Ross Tucker and Jonathan Dugas said...

Hi energetich20,

Your comment helps illustrate an important point, which is a similar one as Anonymous made earlier.

In Part II of this series we have tried to illustrate that electrolyte concentrations between crampers and non-crampers are similar, which suggests that any changes in electrolytes are not related to cramps.

The other important evidence is that cramps occur in muscles the we use most often during that activity. Therefore local muscle fatigue seems to be an underlying factor in cramps.

In the first two scenarios you describe, local fatigue would normally be associated with them.

Perhaps to explain the third one, we have to elaborate a bit more on what a cramp is. As we mentioned in Part I, it is an involuntary contraction, and we should add that the muscle does no relax. It is stimulated to contract (shorten) and does not relax, which means it keeps getting shorter and shorter, which causes the painful feeling.

Therefore in this third scenario you describe, the fact that you are placing the muscle in a shortened state (kneeling for a long time) is what pre-disposes that muscle to cramp.

In addition, this is why stretching a muscle is the best way to alleviate a cramp. Recall that during the cramp the muscle is activated and is being stimulated to contract (shorten), and therefore by stretching the muscle to counteract that shortening and turn off the stimulation (thus alleviating the cramp).

So to sum it up, local fatigue seems to associated with the cramps, especially in the first two situations you describe, and furthermore placing a muscle in a shortened state for prolonged periods of time will also predispose it to a cramp.

Hope that helps. And do come back and join us for Part III where we describe the novel hypothesis of cramps!

Kind Regards,

Anonymous said...

Very good article. I really enjoyed reading it. A few quick points though. I am not saying that I don’t agree with your conclusions, I’m just confused about a few things. I don’t understand how a loss of sodium will always indicate over-hydration and higher electrolytes will always indicate an electrolyte imbalance. You say this near the end of the segment about runners. They had lower sodium (indicating over-hydration) and you said that if electrolyte disturbances play a large role you’d expect higher concentrations of electrolytes. I think this only applies if the subject does not drink any water. If a subject looses sodium through sweating and then drinks some water (not enough to replace all the water he lost) he can have too few electrolytes (an imbalance) and be dehydrated at the same time. Maybe I’m wrong, but it just seems like when you comment on the decreased sodium you only mention how it means that they are over hydrated, and any possible electrolyte imbalance must have higher electrolyte concentrations. Lastly, you say how Mg cannot contribute to cramping since it’s higher and one would assume that it should be lower, but who says that higher Mg levels cannot contribute to cramping? You don’t even humor the idea?

For the triathlons you again mention how they had lower sodium levels which indicates over-hydration, yet they lost around 3.5-4 lbs, so how can they be over-hydrated? Again you seem to overlook the possibility of lower sodium not indicating hydration level, but indicating a loss of electrolytes. Not once do you mention lower electrolytes without mentioning over-hydration, you never comment on how lower electrolytes per se could contribute. You don’t seem to look at the evidence with a skeptical eye, and only report information that goes along with this story you are trying to tell. Maybe that’s what one needs to do when writing an article for the public vs. one for a peer reviewed study, I don’t know. But I do want to emphasize that I don’t disagree with your conclusions, I just have a different route of getting there. Thanks again for a great read.

Ross Tucker and Jonathan Dugas said...

Hi Ryan, and thanks for participating in this discussion on cramps. I hope I can address the points you have made.

1. Sodium
Although we do lose some sodium in the sweat during exercise, the fact that we are sweating and losing volume as well leads to the concentrations of the electrolytes rising. Sweating will increase the concentration of the electrolytes, and in fact a very small rise (1-3%) will stimulate us to drink. So during exercise we sweat. . .the concentrations rise. . this stimulates us to drink. . .the concentrations fall a bit. . .and so it goes on and and on until we finish.

A study published in 1992 in marathoners by Robert Cade showed that the group ingesting water to thirst maintained their sodium concentrations just fine.

So sodium losses do not really feature during exercise, in spite of what so many mags and websites and products say. This is because 1)sweat does not contain much sodium when compared to body fluids, and 2) it is the changing (decreasing) volume of the extra-cellular fluid that has a much more profound effect on these concentrations, and it causes them to rise, which again stimulates us to drink so that they remain in a comfortable range.

Furthermore, much of the info out there says that cramps are caused by dehydration, and we just do not see dehydration and low sodium concentrations. This is because losing fluid volume will (as mentioned) cause the concentrations to rise, not fall.

So our point was that if the crampers had lower sodium concentrations then that suggests that they are a bit less dehydrated than the controls.

2. Magnesium
The assumption by the product manufacturers is that lower magnesium concentrations cause cramps, and this is why you must take their supplement---because it gives you extra magnesium.

If you do a quick google search for "magnesium cramps" you will receive a plethora of pages that advise taking magnesium to prevent cramps.

So that is where the assumption comes from that low magnesium plays a role. It is not our assumption, but one proposed by many others.

3. Weight losses
The biggest problem with prolonged exercise and using weight losses to estimate fluid losses is that during the 225 km of an Ironman we burn off a substantial amount of carbohydrate and fat---probably in excess of 750 g, which is perhaps 30% of their weight losses.

On this note, if one were to finish the Ironman with 0% losses, they would be overhydrated since they have burned off 500-750 g of fuel. Therefore losing only 0.5-1.0 kg represents a net change of 0%, and as we have discussed on our dehydration series, losing a little bit of weight is the normal response as the body is not trying to regulate the weight. It is trying to regulate the concentration of the blood.

There are two "take-home" messages from the Ironman data. First, the two groups were not different from each other. Second, I would not say that losing 3-4% of your pre-race weight represents any significant amount of dehydration.

Clearly we need to do a post that attempts to explain sodium balance during exercise, because electrolyte losses are just not part of the equation. Do we lose some sodium via the swat? Yes, we do. Does that result in a lower sodium concentration? No, it does not, because we are losing volume, which causes an increase in concentration.

Just to finish this comment, this is a really contentious issue as we are seeing with so many comments and questions. The main reason for this, I believe, is that so much misinformation exists around this topic. There is a true dearth in the scientific literature, and all that leaves is what product manufacturers tell us----it is electrolyte losses and dehydration that cause cramps.

Do we know exactly what causes cramps during exercise? No, not yet, but the only available evidence suggests that it does not have anything to do with dehydration or electrolyte losses.

Thanks for the critical questions, and please keep them coming until Ross and I can explain this clearly enough. Asking these questions is the only way we will really get to the bottom of it!

Kind Regards,

bk said...

Interesting article. Im stil a little confused about the whole sodium loss during sweating issue.

Clearly sweat tastes salty, and once it has dried it leaves a whitish streak behind, both clear indications that there are fairly large amounts of salt being lost thru sweating.

Secondly, if cramping is not caused by electrolyte loss, what are the main problems caused as a result of things like low sodium, chromium, magnesium etc.



Ross Tucker and Jonathan Dugas said...


Thanks for the questions, I hope I can provide some clarity.

First question: Why does sweat taste salty? Well, the answer is that it does have sodium and chloride in it - we've described how the concentration of sodium in your sweat ranges between about 20mM and 100mM, depending on your level of acclimatisation (the fitter you are, the lower it is).

Now, this is SUBSTANTIALLY lower than your plasma - so in otherwords, your sweat has a much lower concentration of sodium than the fluid. This is why when you sweat, you don't actually cause an OVERALL LOSS of sodium - in fact, your sodium concentration will go up - you are losing MORE FLUID than SODIUM.

However, the fact that you still have sodium in your sweat gives it the salty taste. NOW, here's the thing - by the time that sweat reaches the surface of the skin, the fluid part will evaporate - that's the point of sweating, after all. What it leaves behind is sodium, in a far more concentrated form. That's why it tastes so salty - by the time you 'taste' it, the concentration is far higher because the water is GONE! However, the amount is still quite low.

You can see this if you ever sit in a steam bath for a long time - you'll produce a lot of sweat and if you lick your skin, you'll taste almost nothing. Because in the steam bath, you cannot evaporate the sweat, so what you taste is the sweat in "original" form - then you see just how low it is.

Now, second question.

If you lose electrolytes as a result of diarrhoea or some other clinical complication, then a number of different complications can occur - the worst of these is that you can fall into a coma as a result of a very low sodium concentration, which then pulls water INTO THE CELLS - your brain swells up, and you can die as a result, if untreated.

The point we are making is that this does NOT happen during exercise as a result of sweating or electrolyte loss. So if your question is referring specifically to electrolyte loss during exercise, the answer is that NOTHING HAPPENS. There is no risk of electrolyte depletion during exercise. Rather, there is probably more risk of the electrolyte levels rising too high - I have a colleague who believes that high sodium explains much of the nausea, headaches and discomfort felt by runners at the end of the Comrades Ultramarathon.

But really, there is no risk. I realise that this may sound difficult to believe, but it's only because we've been indoctrinated for so long by the companies who've convinced us to drink and drink and drink, and that's it's dangerous not to. In that regard, they've done an excellent marketing job!


bk said...

Hi Ross, thanks for the response.

You're right, its like convincing the masses for the first time that the world isnt flat!

So are you now suggesting that people involved in longer endurance events like marathons and ultras should in fact just drink plain water? That there really is no use for gatorade and such?

thanks again.

Ross Tucker and Jonathan Dugas said...

Hi bk

Thanks for the feedback.

Regarding your question, not at all. In fact, I'd recommend the sports drinks to anyone, because they contain glucose and that's a non-negotiable for endurance sports. In order of importance, the sports drinks provide carbohydrates, then fluid, then electrolytes.

In fact, there are some sports drinks companies (here in SA at least) who are gradually changing their market positioning to emphasize the energy side of the drink, where it previously was fluid, then electrolyte. I met with the brand manager of a local sports drink the other day, and he was telling me how the market has traditionally been seen as "sports drinks" on one side, and "energy drinks" on the other, and only now are they themselves beginning to realise that there should not be that difference...!

As far as the electrolytes go, they help with taste/palatability, and so that's not insignificant, because a lot of people don't like the sports drink well at the best of times - the salt helps in that regard.

However, for exercise up to an hour, you don't NEED anything more than water, even as far as energy goes. Of course, people may WANT to drink more, either because they like the taste/effect, or because they just hate drinking water.

Once you go longer than about 60 minutes (there's no distinct cut-off), then you start to enter the area where you need energy. Your choices are to eat something - jelly babies, chocolates etc., and drink water. or you can drink a sports drink. Some people like to drink Coke, but for others, it's just too sweet. So that's where the sports drinks fit in.

I certainly wouldn't attempt a marathon on nothing but water. But then you don't need to overboard on either the fluid or energy, either. Drink fluid to thirst, and just make sure there is some energy intake as well.

We'll try to tackle this in a bit more detail in our Part IV post of this series.


Ross Tucker and Jonathan Dugas said...

In response to bk's comment. . .

There are two large obstacles we have to clear when we are trying to get fluid/fuel into te body.

1. Emptying from the stomach
2. Absorption from the gut

So the one thing I would add to Ross's answer is that the inclusion of sodium to the drink will enhance the absorption from gut.

Drinks up to 6% carbohydrate empty from the stomach at the same rate as water (which empties fastest). Then, once it is safely in the gut, the presence of sodium and carbohydrate enhance the absorption of fluid. As the solutes are absorbed, they create and "osmotic drag" which pulls the fluid behind them.

Therefore adding sodium enhances the taste a bit, especially in the presence of something sweet like carbohydrate, and it will also enhance the absorption from the gut.

But beyond that there is no real need to ingest sodium with your food and fluid.

The very important thing we must never forget about nutrition during exercise, though, is that we can be armed with all the scientific evidence in the world, but if you find something that works for you, then you must stick with that.

For some people that means ingesting water and eating energy bars/gels/etc. For others it means only fluid and no solid food. But you must use the science as a guide and find what works best for you on the day.

It is entirely possible to complete a marathon (or longer) on only water + solids----the important thing is to realize that for exercise of that duration you need energy and also fluid, and therefore you must take carbohydrate at some point in addition to the fluid.

Thanks for joining in this discussion, and please keep the comments/questions coming!

Kind Regards,