My husband, Andrew Rettek, has a blog you should read. As he’s gotten into fitness, he’s started following exercise science, which is the (very new!) field of running small controlled experiments on diet and exercise on athletes who do exactly what you tell them to, under strict observation.

This is in contrast to fields like nutrition science, which study the effect of the “intervention” of telling people (usually non-athletes) what diets and exercise programs to follow. Exercise science tends to have much smaller sample sizes, but it can come to more unambiguous conclusions because it’s testing the intervention itself, not people’s ability or willingness to follow it. Contra the “nobody knows anything about diet or exercise” conventional wisdom, we do know some things! It’s just…not very many things. And only about college athletes.

One surprising thing Andrew noticed is that organ size, especially liver size, has a lot to do with overall metabolism.

Larger Athletes’ Higher RMR is Partly From Bigger Livers

Athletes have higher resting metabolic rates than non-athletes; their bodies use more energy, even when they’re not exercising. That means they can eat more without getting fat.

Some part of this effect is due to higher muscle mass, which “costs” energy to maintain.

But muscle isn’t the only tissue athletes grow; bigger athletes also have bigger livers.1

In fact, muscle is a lot less metabolically expensive than other organs. Muscle consumes only 13 kcal/kg/day, while liver, brain, heart, and kidney consume 200, 240, 440, and 400 kg/kg/day respectively. In fact, 60-70% of resting energy expenditure2 in adults comes from these four organs, even though they’re only 6% of body weight.

If you look at “small”, “medium”, and “large” athletes3 (male, mean weights 147, 170, and 200 lbs), obviously the larger athletes have more REE than the smaller ones. What’s surprising is that most of this is due to non-muscle fat-free mass differences — i.e. organs, mostly liver. The difference between “small” and “medium” athletes’ REE is 73% due to this “residual mass”. The difference between “medium” and “large” athletes’ REE is 56% due to “residual mass”.

Same deal for female athletes4 — 76% of the REE difference between “medium” and “small”, and 36% of the REE difference between “large” and “small”, is due to increases in “residual mass”.

If you compare sumo wrestlers to untrained college students, they of course have more of everything — more muscle, more fat, and heavier organs, especially liver — as well as more REE. However, if you compare untrained obese to normal-weight subjects, obese people don’t have bigger livers or kidneys. Big organs are a thing in big athletes, not just big-in-general people.5

Similarly, in a comparison of college athletes to college non-athletes, the athletes had 25% higher body weight, 40% more muscle mass, and also about 30% more liver and kidney mass. Within both groups, bigger people had bigger livers, but the slope of the correlation was much bigger for athletes than non-athletes.6

Intriguingly, even as liver fat tends to accumulate with age, liver mass and liver blood flow both decrease with age. 7

Causality? Looks Like Yes: More Protein = More Liver

The implications are intriguing.

Does exercise, high protein intake, or something else big athletes do, cause liver growth?

If you somehow managed to grow a bigger liver, would that cause a higher resting energy expenditure?

One longitudinal study of college football players, instructed to eat 500-1000 extra calories a day while engaging in weight training, interval training, and skills training, found that they gained an average of 21 lbs, which was 29% muscle, 46% fat, and 45% neither — including the liver, heart, and kidneys.8 So that looks like a causal effect: intentional weight gain combined with exercise results in liver growth. The authors speculate it’s due to the high protein intake of these athletes.

We also have a bit of animal evidence that dietary protein intake affects liver size.

When mice are fed diets with different amounts of protein but the same total calorie amount, the high-protein-fed mice (46% kcal from protein) don’t have significantly higher body mass than the low-protein mice (7% kcal from protein) but do have significantly higher (15-18%) liver, kidney, and stomach mass. 9

Similarly, lambs fed high-protein diets (125% of recommended daily value) have significantly higher kidney and liver weights than lambs fed normal-protein diets, but no significant differences in overall weight. (The high-protein lambs had 12% bigger livers).10

And mice fed low (5% kcal from protein), medium (15% protein), and high (55%) protein diets had, accordingly, small, medium, and large livers, while there was no difference in total weight or muscle mass. The high-protein-diet mice had 15% larger livers than the medium-protein-diet mice.11

Rats fed a typical 20% (by mass) protein chow or a high-protein 40% chow gained similar amounts of body weight, but the high-protein rats had 11% bigger livers.12

This all points to “eating a higher % of your calories from protein leads to increasing liver mass”. And, since we consistently see that livers (and other organs) have similar energy consumption per kg regardless of how big they are, this would suggest that a high-protein diet, maybe in conjunction with exercise, would result in higher REE due to liver growth.

What Else Grows Livers?

Ok, but if you’re a gym rat, you are already exercising and eating a bunch of protein. Is there anything else that’s reasonably safe and increases liver size?

Liver enlargement, or hepatomegaly, is usually a bad thing. You will get a bigger liver through inflammation (due to infection or toxin exposure), fat accumulation, or iron accumulation. But you don’t want this.13

For instance: inject LPS, the toxin produced by E. coli infection, and their livers quickly become 20% bigger, which appears to be due to both larger liver cells and increased numbers of liver cells. But this is, of course, a very bad idea (it’s essentially inducing sepsis.)14

Likewise, the inflammatory marker IL-6 will make a rat’s liver double in size…and also cause muscle wasting everywhere else. Not something you want to try at home.15

In general, livers are excellent at regenerating after injury, but they stop growing once they hit a fixed percent of body weight. Negative growth regulating signals kick in at a certain point and stop the liver from growing arbitrarily big. What happens if we artificially mess with those signals?

Well, tentatively that looks like a pretty bad idea.

Mice lacking the Yap and Taz genes that control liver size have larger livers…but they also have liver cancers, and worse regeneration from liver injury.16

Similarly, mutant mice lacking Hippo signaling have unusually large livers that don’t stop growing when they hit the usual “maximal size”…but they also get lots of liver tumors not seen in wild-type mice.17

Rats given a liver-growth-stimulating solution including insulin, glucagon, and thyroid hormone (T3) could more than double liver size without increasing body size…but 60% of the animals treated died within 8 days. Again, not a good idea to try.18

One avenue that looks a little less terrible is follistatin; a viral gene therapy in mice that induced follistatin overexpression in the liver resulted in 40% bigger livers, with no apparent signs of impaired liver function. The growth appeared to be due to increased cell division. However, the animals were only observed for 12 days, so we don’t know about long-term risks.19

In short, I’m not, so far, seeing examples even in animal studies where livers can be lastingly enlarged beyond the usual maximal size without producing cancer, cachexia, or other quite serious problems.

On the other hand, maybe someone will find a solution (or maybe there’s one buried deeper in the Google Scholar results than I care to look today.)

Meanwhile, I’m intrigued by the prospect that eating a high-protein diet could grow the liver. Compared to all this other stuff, protein consumption is very safe.

What we don’t have, but would be interesting, is a human interventional study that varies protein consumption and looks at its impact on liver size and RMR.