Thursday, May 14, 2020

Low fat vs low carb again

I guess everyone knows about this pre-print

A plant-based, low-fat diet decreases ad libitum energy intake compared to an animal-based, ketogenic diet: An inpatient randomized controlled trial

There is a wealth of data to enjoy and a lot to say from the Protons and insulin point of view but just a brief look gives us equal weight loss, equal fat loss and the sort of changes in fat free mass you would expect from likely shifts in glycogen and its associated water:

Clearly the extra 600kcal eaten under ketogenic conditions did nothing to blunt fat loss, much as we would expect from the low carb perspective. The extra calories did not evaporate, they were lost through increased energy expenditure, especially during sleep and while sedentary:

These people have uncoupled metabolism during the period of eating the ketogenic diet, they generate heat. As measured within the limits of indirect calorimetry. You could argue about a greater faecal, urinary or breath mediated loss of calories too but that's less important than a measured equivalent weight loss despite higher, extremely accurately measured caloric intake.

That's all pretty boring.

What is really, really interesting is the equivalent spontaneous weight loss under the period of high carbohydrate intake. Over the years I've looked at the carbosis vs ketosis for potential mechanisms and this study may go some way to clarifying what is going on. The very low fat eating certainly does not limit the penetration of either glucose nor insulin past the liver. Both spike systemically after every meal. But still there is spontaneous weight loss due to a suppressed appetite.

Under low fat eating less "waste" heat is generated, metabolism is coupled. Tightly coupled metabolism means people needed less calories. The subjects, under very low fat eating, lost weight without any biochemical markers of inadequate calories. Just as they did under ketogenic eating.

That's really interesting. With data, lots of it including important things like the effect of a typical meal on blood glucose, insulin and lactate. Plenty to work with. Needs a lot of thinking about.


BTW does this sound like a metabolic advantage to ketogenic eating? Rhetorical question, 24h energy expenditure combined with utterly accurate food intake measurements tells us something...


JR said...

This is quite well organized study at a first glance. Maybe best so far from Hall. Ad lipidum, no isocaloric, no doubt of the ca. 200kcal energy expenditure difference of diets, these are real lc vs. Lf diets.

But, generous snacks. But, biased title focusing on "ad lipidum energy intake". Better forget energy from inside stores and the difference in expenditure.

Looking forward to the final version, as well as further analysis.

raphi said...

Quoting you Peter from a few posts back, "NB It is difficult to emphasise how good this study is. Just ignore anything the authors have to say". Maybe you should have this as a default sign-off on all your posts!

Peter said...

Haha! I've read the methods and results...


ctviggen said...

I just wish for once he'd do a study for keto that is longer than two weeks. Was listening to a podcast with Nina Teicholz (The Big Fat Surprise), and she remarked that Kevin Hall never studies keto/low carb diets for more than two weeks...then I see this study. It's almost like he wants keto to "fail"...

I ate a very low fat diet for years. Kept fat calories under 10% per day. Was younger, exercised a ton, ate lots of pasta, brown rice, beans, rice cakes, hot cereals (oats, malto-meal, etc.). Did it because I thought fat was deadly (this was the 80s-90s).

I found over time that (1) I was always hungry; (2) there's lots of gas with this diet; (3) it caused depression, mood swings, and more.

For me, I'm mentally happier, more stable and less hungry on my "carnivore-ish" diet (some veggies sometimes) than I ever was on my Pritikin-style diet.

Frédéric Lacroix said...

Notice the high intake of Omega 6 fatty acids on the ABLC diet (16,5g/1000 kcal) compared to the PBLF diet (3,1g/1000kcal).

Notice the "ranch dressing" and the "creamy Ceasar salad dressing" (both usually soybean oil based in the US) in Menu 1 and "mayonnaise" in Menu 2 (usually Canola oil) or "spicy sauce" in Menu 3 (canola oil).

Of course, one really following a "well formulated" keto diet would steer clear of those.

A pity in an ortherwise very interesting study.

Frunobulax said...

@Peter I love it if results are not what we expect. That's when we're starting to learn something :)

@Frederic You hit the nail on the head I think :) I often argue that fat composition is as important as carb content. (As are antinutrients. I started to really lose weight only after eliminating lectin-rich food from my diet. My keto menus would look very different.) The snacks even had 28g n-6 PUFAs. And I know I would gain like crazy if I ate 5 meals/snacks a day, keto or not.

I do think that the issues with a low fat diet are more of a long-term nature, and 14 days is an order of magnitude too short. (Discarding the consequences of high insulin which are even more long term.) From my own past, I always did well for a few weeks on low fat, but suddenly the food wouldn't satisfy me anymore and I would start to eat a lot more, ravenously in some cases, or go off diet. And I think this is because a low-fat diet lacks certain nutrients (micro or macro), like essential amino acids and vitamins. After a few weeks we have to eat more, because we slowly become deficient and we stay hungry until we've got enough of everything. But this will take a few weeks to kick in. (And maybe this is why people in metS have issues with leptin regulation. People stay hungry because our body knows that something is missing. Just getting enough macros to produce Acetyl-CoA isn't enough.)

Add to that: Those people are fairly young (29), only moderately overweight, and we know that insulin sensitive people lose weight well on a low fat diet (

I admire the effort that went into the study, though. If more studies were like this, we wouldn't be in the mess we're in now.

Side note: We need more data on keto weight loss, because there seem to be different weight loss phenotypes out there. Some people lose weight easily (and consistently) with 3 meals a day, others like me are stable unless we eat OMAD. Some people lose weight with a high protein ratio, others have to go at least 70% fat to lose weight. (And I'm talking about people that have fat to lose.) Some hit long plateaus, some don't.

cavenewt said...

As pointed out in an earlier comment, a two week keto diet isn't a keto diet. And that was a very good catch about the O-6, Frédéric.

Another thing that bothers me about studies like this, in addition to the short term and the wrong fats, is the crossover bit. Two weeks on keto your body is just starting to change and rejigger for fat burning. Then all of a sudden you switch to the low fat arm? It just doesn't seem to be a good idea. Poor confused mitos.

Hall's keto is keto in name only. I don't think there's any doubt he's setting it up for an intentional fail. I just hope Peter can get some useful data out of the study.

Alex said...

1. so many details about the food. wow. good.
2. it is a VEGAN PBLF vs ABHF. % of animal foods in the PB group is zero. Omega 6 are in AB 16g vs PB 3g.
3. round 3850 kcal/day? wow. that is huge, no?

LeenaS said...

The exessive calories are probably a must for Hall,
since he needed to push his own line in favor of lowfat.
Most LCHF folks do not eat between the meals :)

karl said...

Interesting.. Speaking of appetite control - my continued experiment with the suet I rendered (stearic acid) is interesting. Sometimes after eating - I actually feel warm - and - I keep forgetting to eat. Is it just me? Has anyone else here tried this?

Also - in my continued digging into lead(Pb) (lead - as in Pb is hard to search for as it "leads" to homophones) -- I ran into this paper from 1977:

Two of the bits of info - fats increase absorption and vit-D reduces absorption. So could it be that the observational diet studies were looking at the effect of less absorbed lead rather than the fat itself?

Peter said...

Hi all,

Valid points. I suspect most will get addressed in the posts coming up based around the data in this post. I have to say thet the PUFA was the first thing I looked at, pretty much as expected...


Alex said...
This comment has been removed by the author.
JR said...

About fats: "all fats include all fat types. There is no exception". Learn from Zoe Harcombe. So, there must be more pufas within "animals" vs. "plants" group. How much?

So, from fig 2. "foods consumed", "animals" ate 2057 kcal/d of fats daily vs. "plants" ate 218 kcal. Coefficient >9, because the total energy intake was lower by "plants". Study design suggested ca. 7.

If that were all butter (let's not include proteins and water for simplicity), we get
animals: 283 g of butter (727 kcal/100g), which includes sat-mufa-pufa 53-19-3 g/100g
so the pufa content is 8,5g. Omega3/6 is 0,36/1,04 (g/100g) ca. 1:3.
Plants: 30 g of butter; 30*3/100 = 0,9g of pufas. 9.5 coefficient to "animals" pufas.

From "foods available" table2 we see that "animals" got little possibilities for omega-3, the ratio was much better within "plants" group. Pufa coefficient less than 6 from "plants to animals" (and this is only as available, theoretically).

Even if the "plants" would consume their little fat as butter, there would have been 9.5 times greater pufa consumption within "animals". In reality, the fats differ inbetween -but it did not show yet I guess. However, no stacking of the deck as it seems (my butter example should favor "animals").

JR said...

PS. to fats above.
if the "plants" ate olive oil as their staple fat (884 kcal/100g) i.e. 25g, pufa share being 10,9 g/100g i.e. less than 3g of pufa. There would still be 3 times more pufas in "animals"... but soyoil would reach butter's pufas.

Frunobulax said...

Micronutrients and hunger:

Very underrated. If your food is not dense in micronutrients, you'll eat more and become obese. And this is why you should never do a study that is only 14 days long.

Frunobulax said...

Thinking more about this... We know so much that obese people are usually deficient in micronutrients ( There are speculations why deficiencies could lead to obesity (

However, has -- to your knowledge -- anyone considered if appetite/hunger change if we are deficient in minerals or vitamins? You know, if we lack salt then we'll crave salty food. If we have too much, we get thirsty and the kidneys do their thing. Hunters-and-gatherers had no RDAs, yet they were not deficient. What could have regulated their vitamins and minerals, except appetite/cravings?

The simplest explaination (occams razor :) appears to be that appetite/hunger is moderated by deficiencies. We'll not be satiated (long) if we lack nutrients. That may be a reason why obese people are deficient (obesity comes because they lack nutrients and therefore eat until they are full and can eat no more). This may be the reason why malnourished kids are sometimes fat (they eat a lot energy wise but not enough micronutrients). And we fall off the cliff on a diet after a few weeks, because diets are deficient in micronutrients ( and the appetite/hunger is harder and harder to suppress.

Obviously there are other mechanisms (insulin influences leptin production, and all that) that may cause overeating. But still, has anyone tried to formulate a diet that is lacking in a specific micronutrient, and checked if supplementation of this micronutrient changes the amount of food consumed? Sounds like a worthy experiment to me.

cavenewt said...

@Frunobulax said "...appetite/hunger is moderated by deficiencies. We'll not be satiated (long) if we lack nutrients. That may be a reason why obese people are deficient (obesity comes because they lack nutrients and therefore eat until they are full and can eat no more)."

Don't forget to consider the effect of the microbiome on appetite and hunger. Cravings can also be caused by manipulative microbes wanting us to eat something.

"Role of the vagus nerve in the development and treatment of diet‐induced obesity"

"The vagus nerve, food intake and obesity"

"Autoantibodies against appetite-regulating peptide hormones and neuropeptides: putative modulation by gut microflora."

(The most memorable scene in one of my favorite childhood books, My Side Of The Mountain by Jean Craighead George, involved a kid living by himself in the wilderness. One day he kills some animal or other and finds himself looking at the liver with incredible craving—he had an iron deficiency. This concept fascinated me ever since, and I was gratified when science recently discovered a physiological basis for it. Jean Craighead George comes from a family of scientists; the Craigheads are the famous grizzly bear researchers from the 60s.)

Frunobulax said...

@cavenewt Thanks for the links, I'll have a look at them. Absolutely, the microbiome can create cravings too. I'm not saying that micronutrient deficiency is the only mechanism there, or even the dominant one. But it may make a difference, and the more pronounced the deficiency, the stronger.

Alan2102 said...

"Clearly the extra 600kcal eaten under ketogenic conditions did nothing to blunt fat loss"

1. The table in question gave data on weight, not body fat.

2. The table in question only went to 14 days on the diet -- not nearly long enough to see anything truly meaningful. I thought everyone knew that a great deal of water weight is lost in the early days/weeks of keto or carb-restriction. Further, even if the table had given us data up to, say, 14 WEEKS instead of a totally-insignificant 14 days, that is still not long enough to say much. All diets work in that time frame... and all diets fail when followed up couple or five years later (with of course outlier/exceptional individuals).

Peter said...

Hi Alan, Fig 3 C gives the fat loss. I agree, two weeks is nothing. But you get a clear cut metabolic advantage to ketogenic eating under this time frame and I doubt the insulin response to low fat will go away over a longer time frame. What will happen to hunger on low fat would need in excess of 22 weeks to pick up (from other studies) but the 4% PUFA content is physiological, so hunger may never kick in until a physiological bodyweight is reached, when appetite would be expected to increase to stop excessive weight loss...


bill said...


Many years ago I came across a study where Goats
were fed some feed ad lib for a while to get a baseline,
then one nutrient was taken out of the feed and
the goats were allowed to eat ad lib. They wouldn't
stop eating, presumably in an effort to obtain the
missing nutrient. They got fat. After the nutrient
was put back into the feed, they stopped gorging on
the feed and went back to the baseline weight.

Wish I could find that study again.

Puddleg said...

Is it a metabolic advantage to need more food?
Maybe if it gets you more micronutrition...