Cramping in Ironman triathlons: not dehydration or electrolytes
Another interesting study in the British Journal of Sports Medicine, posted online in December while I was away. It’s a prospective study that looked at 209 Ironman triathletes before and after a race, and tried to detect any differences between the 43 who developed muscle cramps and the 166 who didn’t. The major finding is that there was no significant difference in the levels of dehydration or electrolyte loss between the two groups, challenging the prevailing electrolyte-depletion hypothesis of cramps. The cramping group lost 2.8% body mass compared to 3.1% in the non-crampers; crampers’ sodium levels dropped 0.1% (+/- 1.9%), while the non-crampers increased 0.4% (+/- 2.6%).
The study comes from Martin Schwellnus’s group at the University of Cape Town. I wrote about Schwellnus’s theory of “altered neuromuscular control” for cramps last July, after returning from a visit to his lab. At the time, he was analyzing some data that he said suggested that crampers tend to be those who set faster time goals and start faster relative to their fitness, and who have trained more in the final week before the race (thus leaving their muscles fatigued). That suggested that realistic goal setting and an appropriate taper would minimize your risk.
The new study only bears part of that out. The three factors that predicted cramping were (1) faster predicted race time, (2) faster actual race time, and (3) previous history of cramping. Training volumes and paces for the final week before the race were more or less identical in the two groups, which means that in this group of athletes, a bad taper wasn’t to blame.
The fact that faster predicted and actual finishing time was associated with cramping is sort of bad news, because it doesn’t offer any simple solutions. The two groups were matched in terms of training history and personal best performances, so on the surface you’d expect them to aim for similar times. But it turns that those who aimed higher were more susceptible to cramps — and achieved better performances. So it seems (albeit from a single, isolated study) that cramping is just one of the many unavoidable risks associated with getting as close as possible to your limits, especially if you happen to have a (genetically determined?) history of cramping. You’d probably find a similar pattern if you measured the likelihood of dropping out or having a very big positive split: those who aim highest are most likely to blow up or drop out, but are also most likely to produce faster performances.
UPDATE Feb. 3: There’s quite a lively debate on this topic going on over at Slowtwitch. I just posted to address a couple of questions; my response is below:
Interesting (and lively, as usual for this topic!) discussion. A couple of quick points/answers:
“My point is that the result of salt loss measured before/after, only reflect that, not the actual level before.”
The pre-race sodium levels were 139.8 (+/-1.8) in the non-crampers and 139.8 (+/-1.5) in the crampers – so there was no difference in pre-race sodium levels.
“crampers’ sodium levels dropped 0.1% (+/- 1.9%), while the non-crampers increased 0.4% (+/- 2.6%). Doesn’t this bit from the study at least suggest that sodium loss could be a factor???”
One way of trying to understand whether an apparently small difference between two groups of numbers is significant is to look at the “p-value,” which tells you the probability of getting this difference purely by chance. Most scientific studies consider a result significant when p<0.05, meaning there’s only a 5% chance the result could have occurred by chance. In the case of these this data set on % change in sodium levels, p=0.836. That means there’s an 83.6% chance that difference between -0.1% and +0.4% is just random statistical fluctuation, and a 16.4% chance that it represents a real physical difference.
“the user suggesting pickle juice – the reason why it works at replenishing electrolytes is because of the salt content, not really a special characteristic to pickles.”
One of several reasons that researchers don’t believe that electrolytes explain pickle juice’s effect on cramps is the time frame. In the initial pickle juice study, it took an average of 85 seconds for pickle juice to relieve cramps. Earlier studies have found that it takes at least 30 minutes for small volumes of pickle juice to even leave the stomach, let alone be distributed throughout the body.
“It is almost never my working muscles that cramp first, but my stomach one.”
Stomach cramps are a completely different phenomenon than muscle cramps, with a completely different set of causes. Given that electrolyte levels do affect how quickly fluid leaves the stomach, it wouldn’t be at all surprising if you noticed a correlation between electrolyte use and the incidence of stomach cramps.
One last general point (even though I’ve already rambled on far too long): this is a simple correlational study. It most certainly doesn’t disprove the possibility that electrolytes influence muscle cramps through some pathway that isn’t clear to us. But if they are a factor, the question is: why can’t anyone find any solid evidence to show it?