Monday, April 7, 2014

Glycogen and Water Weight

One regularly hears that when one goes on a low-carb diet, one loses glycogen, and the weight loss is due to the fact that glycogen is stored with water, so the weight one loses is due to the water lost with the glycogen.
"Glycogen losses or gains are reported (2) to be associated with an additional three to four parts water, so that as much as 5 kg weight change might not be associated with any fat loss."
Here's that footnote:
2. Olsson K. Saltin B. Variation in total body water with muscle glycogen changes in man. Acta Physiol Scand 1970:80: 1 1-8.
From here:
"Glycogen storage: illusions of easy weight loss, excessive weight regain, and distortions in estimates of body composition."
Interesting. How did they get to that number?
"The total body water increased 2.2 1 which is assumed to be caused by the glycogen storage in the muscles and the liver. The amount of glycogen stored was calculated to be at least 500 g, which means that 3—4 g of water is bound with each gram of glycogen."
"Variation in Total Body Water with Muscle Glycogen Changes in Man"
Emphasis mine. Wait, they assumed? Has anyone checked this?
"Muscle water content expressed as mumol H2O lost/g wet tissue weight or made relative to protein content showed no consistent relationship to the glycogen content. These data, therefore, do not support the commonly accepted muscle glycogen-to-water ratio of 1.0:2.7 (g:g). Further work is necessary to quantify the exact amount of water that is actually associated with the glycogen complex."
"Muscle glycogen storage and its relationship with water."
And here:
"The extent to which muscle glycogen concentrations can be increased during exposure to maximal insulin concentrations and abundant glucose was investigated in the isolated perfused rat hindquarter preparation.... Total muscle water concentration decreased during glycogen loading of the muscles."
"Mechanisms limiting glycogen storage in muscle during prolonged insulin stimulation"
I can't find any evidence that anyone's tried to confirm this assumption in man.

An assumption does not equal knowledge. It's a guess, however educated. And a guess is not "science".

I do think, however, that body weight can go up with glycogen increase. But the water's not in the muscles, it's in the gut, buffering the sugar that is generally used to replenish muscle glycogen:
"What he discovered was that if he drank sugary water, his body reacted to it by flooding his stomach with water. 
"...Two lights went on over my head on hearing this. The first is that this explains diabetics’ frequent thirst and urination: they’re thirsty because diabetics consume a lot of sugar, which requires their bodies to dump water into their guts. They need to replenish this water in their systems, and perhaps drinking the water helps in diluting the sugar in their guts."
And going on a low-carb diet allows the body to lose the water that's in the gut (doing you no good from a running performance perspective), not the muscles.

At least that's n=1, not n=SWAG.

[Scientific Wild-Ass Guess]

P.S. OK, it's not just me:

"...Several attempts have been made to estimate the amount of water stored in muscle in association with glycogen, but this is not easily quantified. Early data suggested a value of about 3 g of water for each gram of glycogen (Olsson & Saltin, 1970). However, a subsequent study by Sherman et al. (1982), in which the muscle glycogen content of rats was manipulated by exercise and diet, suggested that there was no consistent association between the amount of glycogen stored in a muscle and the muscle water content. Richter, Hansen, and Hansen (1988) found a decrease in muscle water concentration after muscle glycogen loading. Nygren, Karlsson, Norman, and Kaijser (2001) have also suggested that glycogen loading may alter the disposition of water molecules within the muscle. These apparently conflicting data may result from different amounts of water storage with different structural forms of glycogen and changes in the proportions of these different forms as the total amount of glycogen changes.
"Notwithstanding the uncertainties, there is good evidence of gross changes in body mass as a consequence of diet and exercise manipulations designed to induce alterations in glycogen storage in humans. Any major change in the amount of glycogen stored in muscle will result in a change in body mass, with the major part of the mass change being a consequence of the storage of the associated water (Olsson & Saltin, 1970)...."

The studies they cite are the studies linked to above. They're not particularly logical with that last statement, however, as there's still no evidence for the proposition that glycogen is stored with water.

But that's how scientific myths stay alive!

P.P.S.  The whole point of the glycogen molecule is to store glucose without water...

"Glycogen is also a suitable storage substance due to its insolubility in water, which means it does not affect the osmotic pressure of a cell."


  1. Water weight and glycogen stores seem highly related on a personal level. There was a time when I tried 45 hour fasts and I'd watch my weight drop near 10 lbs over that period. Sure, maybe 1-2 lbs (tops) of that weight loss was bulk, but given that my max caloric burn over that time would be about 4K calories, you just can't get to the 10 lb figure without a lot of _something_ else being lost.

    This also explains why after a carb bender my weight goes way up (and I look bloated).

    It's more anecdotal observation, but bodybuilders seem well aware of the correlation between carb refeeds and water retention. Hard to imagine there's not _something_ at play with carbohydrates and water.

  2. I've lost 10 lbs on a Maffetone-method run that is designed *not* to deplete glycogen.

    It turns out that there's a good bit of evidence in the literature that ingesting glucose does cause the body to dump fluid into the gut to buffer it.

    Maybe, if I've got some time, I'll post it. :)

    I'd also be interested if there's any evidence to the contrary. So far the only research that was supposedly to the contrary was a link to the first study above.

    Whoops... ;)

  3. Justin, answer to your question, it appears that the body also buffers glucose in the blood:

    "Infusion of hypertonic glucose during dialysis results in a greater increase in relative blood volume (RBV) than equal volumes of other solutions. As mannitol has the same osmolarity, molecule mass and charge, the greater increase in RBV following hypertonic glucose appears to be a specific effect, possibly related to a decline in vascular tone."

    Which suggests that a low-carb diet could simply allow the body to lose water that it's using to buffer glucose. That wouldn't mean that it's also losing glycogen, since glycogen is stored without water.