Adipose Flux: Energy & Linoleic Acid

What are the physiological limits of body fat transfer, and how much are we continually PUFA'ing ourselves?

pictured: calories, in flux

As you probably noticed if you’ve been reading this blog for a while, I think a lot about the physiological processes involved in obesity.

One of the most curious ones is the rate limit on energy transfer from our body fat - what I call adipose flux.

People love to talk about CICO and deficits, but somehow almost nobody seems to ever think about how this actually works.

Many people seem to assume this just magically happens somehow, but liberating fatty acids from body fat is a physiological process. Those processes often have rate limits, and they can break.

The rate limit of body fat: 32kcal/lb/day

The only study that I have even seen attempting to answer this was done by Alpert in 2003: A limit on the energy transfer rate from the human fat store in hypophagia.

He actually looked at the old Minnesota Starvation Experiment done by Ancel Keys during World War II. These trials were done on American conscientious objectors, to mimic the conditions in prisoner of war camps, and learn about how to deal with populations facing starvation. Lots of data came out of this, and people can still analyze it, like Alpert did here.

Alpert was interested in finding out how much energy could actually be delivered from body fat, and he found his answer: 290±25 kJ/kg/d. 290kJ are about 69.3kcal, so it’s about 70kcal/kg/d or 32kcal/lb/day fat mass.

This means it’s a linear effect: the more fat you have, the more energy it can give you. Makes sense intuitively, but it’s actually interesting how different the numbers work out:

• A 300lbs man with 150lbs of body fat should be able to generate 4,800kcal/day

• A 188lbs man with 38lbs of body fat should be able to generate 1,216kcal/day

• A 220lbs man with 68lbs of body fat should be able to generate 2,176kcal/day

I find these interesting because 300lbs was me at my most obese (50% body fat), and 188lbs would be the top end of “normal BMI weight” for me, representing about 20% body fat. And 220lbs was me recently, before beginning a water fast.

They also show why this number is probably pretty theoretical, and not necessarily applicable for us in daily life.

Getting obese should be impossible

If morbidly obese me at 300lbs was getting a massive 4,800kcal/day of energy dumped into his blood stream, how did I not wake up every morning, completely satiated & stuffed to the brim? That’s more than even the bizarre 4,600kcal/day TEE estimate I got recently, and 60% more than the more recent estimate of 2,900kcal/day. How was I able to eat anything at all? How was I hungry all the time?

Clearly, something else is wrong with obese people. If it truly worked only like this, nobody would ever manage to get obese.

Lean people would immediately starve

On the other hand, it’s probably also a bit misleading that a guy at 20% body fat can only get 1,200kcal/day out of his adipose tissue. Even given my 2,900kcal/day estimate, that doesn’t even fuel half of it! Would a normal-BMI person be unable to go half a day without beginning to starve, get hungry, and start losing lean mass?

Now there are people who get super hungry when not eating for half a day, but I wouldn’t describe those people as particularly healthy. They probably have whatever problem I had when morbidly obese, just without the obesity part.

Healthy, lean friends of mine can easily go a day or two without noticing much hunger, if any. Maybe they can’t quite fast for 5 days like I recently did, cause they have less adipose flux. But they certainly don’t start catabolizing their muscle for energy after 12 hours.

Buffers

I don’t really know the answer, but I suspect the answer is buffers. We have a pound of glycogen or so in the body, which would give us another 2,000kcal we can draw on before we get low. We also have ATP preloaded in our tissues that we can burn before needing to refill it. There are probably a bunch more similar buffers, so that we can easily go 1-2 days without much trouble at all, before being rate-limited by our adipose tissue in refilling all these buffers.

Is this why I could only fast 5 days?

I recently tried a water fast after years of not fasting. I had no idea how long I’d be able to go - my record water fast in the past was 7 days, but I was white-knuckling it back then, barely sleeping at the end, consumed by thoughts of food and not very functional.

This time, I just quit after day 5, starting to recognize the signs of bad sleep.

And in this context, the rate limit from Alpert kind of makes sense. If I can really generate around 2,100kcal/day from my adipose tissue, and I really have a TEE of about 2,900kcal/day, then I’m missing 800kcal/day. The first 2 days I might burn through some glycogen. But, eventually, this internal deficit will catch up with me and my body will begin down-regulating my metabolism.

Let’s assume an initial glycogen buffer of about 2,000kcal. That means I was able to go 2 days just covering the missing 800kcal/day from that. By noon on day 3, I was really missing energy - any deficit from now on would not able to get covered; my body would have to downshift and reduce some functions. After 2.5 days of this, by the evening of day 5, I started noticing these effects - inability to sleep, waking up in the middle of the night, unable to fall back asleep, restlessness.

With all the speculation about the buffer sizes, and the 32kcal/lb/day being pretty averaged and vague, I think it’s hard to make precise predictions here.

But it does explain a lot: why can fat people sometimes fast longer than thin people? Why do very lean people start losing lean mass pretty rapidly?

There has to be a rate limit on adipose flux, and it makes sense that it isn’t crazy low, or crazy high.

This could help explain the exponential decay shape that many people see in their weight as they lose fat. Here’s mine:

There’s gotta be something that messes you up, or you wouldn’t be able to get to 300lbs to begin with. Remember, somebody with 4,800kcal/day just from adipose flux should wake up satiated to the brim.

But let’s assume that you can get morbidly obese (cause there are many such cases, lol) by somehow blocking this. And then, one fine day, you begin a crazy cream diet and suddenly the blockage is removed.

And you lose 20lbs the first month. Then 10. Then another 10.

But over time, your fat loss starts leveling out. Finally, a -3lbs month is a good one.

It kind of makes sense, doesn’t it? At first, your adipose tissue is dumping immense amounts of fat into your blood. With the blockage suddenly removed, this gives you strong satiety, drives up your metabolic rate, and enables you to lose fat very, very quickly.

But at some point you have way less fat on you than you used to, I’m down about 65% of my original fat mass. Therefore, my available adipose flux is also down 65% just by Alpert’s linear model.

Naturally, I would start eating more just intuitively, and I wouldn’t lose any further fat nearly as quickly.

This rate-limit actually explains why people see this exponential decay shape: it doesn’t matter what your “deficit” is if you can’t supply it. If you’re short 1,000kcal of energy, but your body can only get 500kcal from body fat… well, then you’re only losing half of that weight from body fat, by definition.

That’s why I think forcing deficits is dumb. If your body can’t get at the fat, you’re not gonna lose fat. You might lose weight, but it’s probably going to be lean mass.

If your body can easily get to that body fat, then you’ll naturally be slightly more satiated, eat less, and drive your metabolism up a bit. So you won’t have to force a deficit to lose the weight.

You do have to remove the mystical “blockage” though.

How many tablespoons of soybean oil are you getting from your body fat every day?

I use the informal term “PUFA’ing myself” frequently. This refers to the idea that we store the fats we eat in our adipose tissue, and they are liberated all day long, every day, just as new fat is deposited, in a slow exchange.

This is a core concept in Modern PUFA Theory: you can’t just stop eating seed oils today and expect to be completely free of their effects in a week. It took years to deposit lots of PUFAs in your adipose tissue, and it’ll take basically just as long to get rid of them again.

This is why seed oils are so pernicious, and difficult to study: you’d have to deprive a control group for 4-8 years to completely wash out the PUFA from their body fat, and THEN you could start a trial.. but it might take another 4-8 years.

Introducing: The Adipose & Linoleic Acid Flux Calculator

Ha, tricked you! This whole post is just an explainer for the newest calculator I made.

After a discussion on Twitter, I was curious: we talk about PUFA’ing ourselves, but how much of that are we actually doing? What’s sort of a “critical mass” when it starts having deleterious effects?

This calculator tries to answer that.

Thanks to Eli for design help, the colors & spacing were way more eye cancer inducing before he gave me some pointers.

You can see it shows both expected adipose flux (at 32kcal/lb/day) per Alpert, and lets you configure a linoleic acid percentage. It’ll then tell you how many kcals of linoleic acid you are dumping into your own blood stream, and roughly how many tablespoons of soybean oil that is. (Soybean oil is 51% linoleic acid, so it’s roughly 2x by weight.)

As you can see, a person with 25% linoleic acid of 68lbs fat mass would be dumping 8tbsp of soybean oil into his blood stream, which is around 10x the “safe, ancestral” amount of ~2g/1,000kcal TEE.

Unfortunately I can’t tell you exactly how to determine your own linoleic acid percentage. We can’t really do adipose tissue biopsies. Trust me, I’ve tried getting one. Nobody even knows where you would get that done.

But we know from the LA Veterans Study that adipose fatty acid composition will eventually approximate dietary fatty acid composition, with some small differences for fats generated by the body itself (lipogenesis).

These veterans were fed a high-linoleic acid diet, and their adipose linoleic acid % rose from 11% at the beginning of the study to 32% after 5 years.

From: https://www.jlr.org/article/S0022-2275(20)39591-2/pdf

We therefore know that 32% isn’t a particularly crazy number, and 11% was common in elderly men in the early 1960s. See, even back then, people were pretty PUFA’d! These were WWII veterans, maybe even WWI, and they would’ve been born around 1900-1920. And already, they had 4-5x the ancestral 2% level in their adipose tissues.

Since then, seed oil consumption has skyrocketed. It’s not unrealistic to assume that a modern American eating the Standard American Diet has 32% or more adipose linoleic acid.

Linoleic acid range: 2-50%?

I decided to limit the percentage slider to 50% on this. Soybean oil has 51% linoleic acid. So even if the entirety of your fat intake is from soybean oil, you could only just barely crack the 50%. I suppose you could switch to corn oil (58%) but at some point you really have to try.

On the other hand, 2% is often assumed to be the ancestral linoleic acid number. This is what has been measured in a few hunter gatherer type populations, somewhat untouched by the Evil Western Diet with its Ultra Processed Foods.

And, going again by the idea that our adipose fatty acids will eventually approximate those in our diet, it would also likely happen to end up around that number: grass-fed beef has about 2% linoleic acid.

Again you could eat lots of nuts to increase your linoleic acid percentage, but nuts are regional & seasonal. No nut butters from Whole Foods for Grok the caveman.

And, of course, we cannot generate linoleic acid endogenously: it is considered an “essential” fat, meaning we need to consume it to get it; we cannot create it via lipogenesis. So 100% of the linoleic acid we store in our adipose tissue would’ve had to come from our diet.

Mapping OmegaQuant results to adipose LA%

This is the holy grail of Modern PUFA Theory, I think, and the reason why I created the Omega Tracker and maintain almost 100 results in there now.

Unfortunately, the OmegaQuant doesn’t measure adipose fatty acids. It measures “whole blood” - which includes red blood cells and their phospholipids, triglycerides & free fatty acids from your adipose tissue.. maybe more.

But if we can roughly map the relatively easy to acquire (Amazon.com) & inexpensive ($100) OmegaQuant Complete test to your adipose tissue LA%, we can then tell if our depletion protocol is working, and how close we are to reaching the goal of 2% linoleic acid in adipose.

Here are some key assumptions I’m making:

• The people who’ve been strictly avoiding for 5-10 years and have “bottomed out” at just over 10%, or in some cases around 5-6% in their OmegaQuant, are probably at or close to the 2% LA in adipose tissue - that’s about as low as you can go.

• Likely nobody is over 50%, as that would require eating exclusively corn oil or grape seed oil as your dietary fat.

Now let’s take a look at the 99 data points from OmegaQuant tests we have so far:

The highest OmegaQuant LA% we see are typically between 20-25%, with a single outlier at 28%.

The lowest are sort of split into 2 groups, some hovering around 10%, and some way below, at 5-6%. I think these actually represent roughly the same adipose LA% and I’ll write about why I think that in the future.

• Let’s consider anyone <11% OQ LA as having arrived at around 2% adipose LA.

• Let’s consider the 20-25% OQ LA area as the highest you can reasonably go, eating a Standard American Diet, absent literal corn oil chugging.

• This puts most of the data points we see somewhere in the middle, at 15-20%. I myself have hovered between 15-21%, and I’m 2 years into this. Most people who do these tests are like me: they’ve cut out PUFAs pretty strictly, but only for a few years at best.

• We only have a single person testing over 4 years, a handful at 3, some at 2. Most of us simply haven’t known about this for long enough yet.

• Over time, I expect that we’ll see some of these lines slowly drop toward the 10%. We can already see some very successful trends, people who’ve gone from very high to moderate, or from moderate to very good.

OmegaQuant LA% → Adipose LA%?

Can we reverse-calculate a reasonable estimate for the Adipose & Linoleic Acid Flux calculator?

It might be a bit tricky. I think it’s a fair assumption that somebody with an OQ LA% of <11% is at around 2% adipose.

But is somebody with a 25% OQ LA at 50% adipose? Probably not. A literature review done by Stephan Guyenet, PUFA avoider par excellence, estimates that average U.S. linoleic acid intake & adipose storage in 2008 was around 21.5%. (Thanks to Tucker Goodrich for pointing me to that study.) And maybe in the 15 years since 2008, the number has grown by another 5% or so? I wouldn’t be surprised.

From: https://www.sciencedirect.com/science/article/pii/S216183132300114X?via%3Dihub

We could therefore assume that the “average-high” people on our OmegaQuant List, say from 20-25% OQ, also have 20-25% LA in their adipose tissue. But as adipose LA comes down over the years, it won’t come down as much or as quickly in the OQ tests, where there’s a much higher floor than the roughly 2% we should have in our food and adipose tissue.

We could attempt a very rough translation like so:

OmegaQuant LA      Adipose LA      Judgement

25%                25%             Very bad (SAD)
20%                20%             Bad
15%                10%             Not great, but normal for the 60s
11-15%             3-10%           Getting there
<11%               2%              Very good

Now this is pretty vague and rough. And, unfortunately, the OQ doesn’t represent 100% adipose flux as of the test day even if fasted. It contains a large percentage of red blood cell phospholipids, which average the last 3-4 months of fatty acids in the blood. That, plus the fact that the phospholipids have a certain floor in how low LA they can become, means we don’t have a perfect signal, and it can be pretty difficult to discern our progress.

Let’s take another look at all the OmegaQuant LA numbers in our database, but only show those who’ve taken more than 1 test so we can judge progress:

This looks pretty promising! A good number of repeat testers have managed to bring their OQ LA% down significantly, even if it often took 3+ years. Some have a pretty linear & progressive down trend. Others haven’t made progress or their number has even gone up (including yours truly, the whacky red line).

I’m hoping that we’ll see more and more people slowly trend down over time as this trend progresses.