Does this study on fasting show that fasting doesn't work?

(Richard Morris) #1

Extended Fasting Causes Muscle Loss

… well that is the headline glommed onto by some low carb bloggers to make the claim that their own proprietary dietary formulations are a replacement for extended fasting.

But is that what this study shows? And does the study tell us something new about metabolism? And for a post meal appetite cleanser, let’s look at those proprietary dietary formulations and see if the claims stack up mechanistically.

So let’s start with the study

A randomized controlled trial to isolate the effects of fasting and energy restriction on weight loss and metabolic health in lean adults

This study took 36 lean (BMI: 20.5-24:9, and body weight < 120 kg) weight stable (for the prior 6 months) individuals with no previous experience fasting or eating disorders, tracked their dietary intake for 4 weeks while they ate their habitual diet to see how many calories they ate to set their individual baseline caloric intake. They were then randomized into 3 groups and equally distributed by BMI.

  1. 75:75 eating 75% of their baseline calories every 24 hours
  2. 0:150 fasting every alternate day, and eating 150% of their baseline calories every other day
  3. 0:200 fasting every alternate day, and eating 200% of their baseline calories every other day

The 75:75 group was looking at change in both fat and lean compartments from a 25% caloric reduction.

The 0:150 group was looking at change in both fat and lean compartments from alternate day fasting with a 25% caloric reduction.

The 0:200 group was looking at change in both fat and lean compartments from alternate day fasting with no caloric reduction from baseline.

Jumping straight to the chart reposted by bloggers … the subjects on the 25% reduction with no fasting saw the greatest reduction in total weight with the least reduction in lean mass.

Actually that isn’t exactly how that figure was republished by the bloggers with the proprietary diet to replace fasting. That looked more like this …

Just for fun: see if you can see what information the blogger really don’t want you to be distracted by … just like Ancel Keys really didn’t want to distracted by an additional 15 countries when his case is made with just 7 countries. Reply below when you see it.

What were they eating when they were eating? The usual balanced diet of carbs:protein:fat:alcohol trhat you would expect from a study by a researcher employed by Nestlé. Worth pointing out that the 75:75 group significantly reduced protein intake, the 0:150 and 0:200 groups did not - let’s put a pin in that for now.

Energy expenditure changes from baseline were a drop on all calorie restricted diets but a non-significant increase in caloric expenditure (both resting and activity) above baseline on the non calorically restricted ADF group - let’s put a pin in that too.

And we can see where they got their energy from; everyone was metabolizing mostly glucose before and after the experiment, but the non fasting group decreased, and both fasting groups increased their use of lipids - let’s put a pin in that too.

Let’s focus for now on the quantitative weight changes in fat and fat free compartments comparing the CR (75:75) and CR+ADF (0:150) treatments. This is directly testing ADF while holding caloric reduction stable. I’ve also averaged the total energy expenditure and caloric intake for each group from the spreadsheet in the supplementary information.

We can predict how much maximum energy on average they can liberate from body fat every day using the formula given by Seymore Alpert of 31.5 kCal/day/lb of body fat

The difference between the 2 groups is that the 75:75 group have a consistent caloric reduction of 277 kCal/day that is less than the amount that can be extracted from body fat and so this group need to use very little lean tissue to cover the deficit.

But consider the 0:150 group who on the fasting day have to get all their energy from storage.

75:75 change 0:150 change
Body mass (kg) 72.1 ± 10.2 −1.91 72.3 ± 8.2 −1.60
Fat mass (kg) 18.3 ± 4.1 −1.75 15.9 ± 5.2 −0.74
non-Fat mass (kg) 53.0 ± 10.3 -0.03 55.4 ± 9.8 −0.75
Energy expenditure (kCal) ₁ 2525 2617
Daily energy intake (kCal) ₂ 2248 2052
Peak shortfall (kCal) ₃ 277 2617
Energy available from body fat (kCal) 1268 1101
Min Req’d energy from amino acids (kCal) ₄ 0 1515

₁ Daily energy expenditure at the end of the intervention
₂ Net daily energy intake during the intervention
₃ For the 75:75 group this is the difference between use and intake, in the 0:150 group this is a surplus on the feeding day and the entire expenditure budget on the fasting day.
₄ Number of calories required to meet the daily expenditure after body fat energy delivery has been saturated.

The problem here is that the subjects are too lean to fully fuel a bad day of hunting … AKA a fasting day.

So how much body fat would someone need to not lose lean tissue in a fast?

Well assuming their energy expenditure was 2617 kCal, the amount of body fat required to fully fuel that at a minimum would be 83 lbs or 27 kgs (instead of the mean 15.9 kgs of body fat in the 0:150 cohort)

Just 11kgs more body fat and you might see most of the weight lost from fat mass, and almost none from lean tissue.

So let’s go back to some of those pins.

Protein <> Muscles

Let’s clear one thing up to start out, when you see energy production from lean mass that is not from muscles. That is now how this works, that is not how any of this works. Muscles are made from protein (and fat but shhhh let’s not get distracted now). We don’t make energy (or even glucose) from protein. If we did then during a fast the liver would eat itself to make glucose.

What we do is use amino acids from a buffer called our labile pool of amino acids. The labile pool comprises roughly 1% of your total lean mass and consists of amino acids dissolved in all the aqueous compartments in your body (Circulation, Cells, interstitial spaces. Those compartments are filled by proteases working in liposomes in your cells recycling proteins (including other enzymes) into their constituent building blocks. Your cells first recycle proteins in excess, then those that are optional (that’s what we call a metabolic slowdown), and then finally those that are essential (we call that process starvation). That entire process is sometimes called autophagy.

So what happens in an ADF when you don’t eat protein on the fasting day is you draw down from that buffer, and on the over feed day you replenish it. If you don’t have enough body fat then you might draw down on it faster than it can be fully replenished which explains why fasting these lean people resulted in significant loss of lean tissue.

But the people who are on the consistent caloric reduction are drawing down on that pool every day - do that for long enough and eventually you run into a metabolic slowdown, and then starvation.

Just eat moar protein

So you might say well when I diet I’ll just eat more protein and make sure that I replenish that buffer. The problem is the way you use amino-acids for energy which is to strip off the amine moiety by making ammonia as a toxic byproduct. Normally we safe that by turning it into urea and peeing it out. But there is a rate limit of how much urea we can make. We can also see what the maximum safe limit of protein oxidation is as that was determined by Rudman et al to be in a linear relation to ¾ body weight (as a proxy for lean mass). It’s roughly 3.21g of protein for every kilo of lean mass.

Above that and you start accumulating ammonia because you can’t make it safe faster than you make more new ammonia.

Let’s say that we put the 0:150 group of people with not enough body fat to safely fast onto a protein sparing modified fast on the fasting days. Can we do that?

Let’s do the math. We need to replace the 1515 kCal from protein used to feed our energy shortfall on the fasting day. So to calculate that we divide 1515 by the Atwater approximation of 4 g/kCal for the energy density of protein = 378g of protein.

OK so how many kilograms of lean mass do I need to have for 378g of protein being oxidized to be below my rate limit to not accumulate ammonia? Well we divide 378g by 3.21 g/kg = 117kg of lean mass.

Arnold when he was competing was reportedly 108kgs with 6% body fat = 101.5 kgs of lean mass. But if you can put on an additional 15.5 kgs more than the Austrian Oak then you can likely safely use a protein fast to replace a regular fast.

The TL;DR is you need fat to fast - fat on your plate, or fat from a Krispy Kreme that you ate a decade ago

Too much food
All Animal August---Triple A
(Mark Rhodes) #2

This explains very succinctly why I could fast 5 days every six weeks my first three years of keto and add lean tissue from 149 pounds LBM of lean to my top lean weight of 181 LBM pounds. according to Dexascans, which some consider to be the Gold Standard of measuring body tissues. Currently I am about 174 LBM at 230 pounds and 20% BF. Indeed, I wrote a blog post about how when I ate less calories averaged out I gained lean but when I limited calories I often dropped weight. A prominent Ketoverse people called me an “outlier” at some Keto festival I was at. I fasted this way not so much as a method of weight loss track but to repair some osteoporosis and did so successfully raising my T Score from -T1.9 in 2013 to +T 1.7 as of this year.

Clearly if I am not misrepresenting my data my N=1 confirms what that I have promoted , not just preserved, protein sparring. Without adding protein to my fasts. Besides reversing my bone density I have also improved my heart health, dropping my Agatston Score by 8.2% from 386 to 355 and improving my Ejection fraction from 45% to 60%. I do not think a heart that is losing lean tissue does that. I just don’t. My next scan is September 2 as preliminary work/data for 90 days of carnivore to help address my wife Jamie’s autoimmune issues. I am doing labs as well.

I am not opposed to a “fast” that adds fat. Currently I do that. At 230 pounds currently when I fast I only have 45 pounds of fat to work with and that equates to 1395 Kcal ( 45 x 31 = 1395) from body fat,. My average calorie load is 2800, 70% fat, 28% protein and the rest carbohydrates. If I don’t eat that much protein I stay hungry as I just cannot seem to leverage more fat to satiety, only this much protein per day works. AND with that I have a BHB between .4 to 1.0 mmol per Keto-Mojo.

I am not opposed to a fast that adds Protein in principle. In one aspect this is similar to Longo’s Fasting Mimicking Diet. If you think you need to eat because you’ll wither away otherwise, well go ahead but the science doesn’t bear this out. Of course you could be limiting other health benefits of not eating at all including a deeper level of cellular repair. That’s just a personal belief about cell repair. I did test HGH levels at 96 hours about 6 times and each time it was below baseline so I have no empirical evidence to highlight that. I do know that my insulin at 96 hours fasted is about 1.0 uIU/ml, normally 6.0. I DO leverage that insulin spike into hopefully muscle growth. I break a fast after a heavy lift at 100 plus hours and eat a Keto-Chow followed by a Ribeye to use the insulin spike as a means to pack my skeletal muscle with the goodies it wants. Seems to work. That and the glycogen combining with watwer at a 1:3 ratio I gain 7 pounds the day after I eat. My muscles swell up. Nice.

However if you want to accuse a reputable doctor whose clinic has used ADF fasting as an effective means to control Type Two Diabetes and other scourges of metabolic disease and have your admins call him a charlatan? I take issue with that. I find it a disservice to health in general and it appears to be a bit of carnival barker…don’t look over there, look here, hire us. I am glad you have looked at this BoolSheet.

That is MY OPINION only. I do not want to imply Richard thinks or feels that way.

(Polly) #3

The thing that leapt out at me was that all their subjects were lean to start with. Surely a study of lean folk tells us nothing about how overweight or obese folk react to a set of dietary or fasting circumstances.

(Polly) #4

Oh and just to add, the lean subjects are probably metabolically well rather than metabolically deranged. So many folk who come to keto, and IF are in the metabolically deranged camp.

(Richard Morris) #5


I read through the supplementary literature and it included oral glucose insulin assays and HOMA scores and there are 3 people who are lean’ish but certainly on the path to type 2 diabetes. Probably a decade away from a diagnosis.

Most of the rest have HOMA scores under 0.5 so twice as healthy as a reference 30 year old man with no metabolic syndrome.

(BuckRimfire) #6

I find this figure really perplexing. What is surprising is that the number is low enough that it seems like I should not be able to fast successfully. I weigh 157# and two years ago (when I also weighed 157#) I had a DEXA scan that said I have 13% bodyfat, so I should have about 20-21 pounds of fat tissue. (From looking in the mirror, I actually think I’m somewhat leaner now, but let’s skip that.) That suggests I can access only about 650 Kcal/day of energy from fat. So that, plus sometimes ~200 Kcal from coconut oil and HWC in my tea in the morning (I usually but not always have this when fasting, rarely have it when not fasting), would be only 850 Kcal per day of available energy. I would expect that to mean that I would feel terrible, but instead I feel fine and as energetic as on a feeding day. Before breakfast the next day (about 36 hours since my last full meal), I actually feel less hungry than I did the previous afternoon.

Last year, I did a 44 hour fast, with five mile runs at 18 and 40 hours of fasting, then a short weight workout before eating. (I’m not a big runner, but I dabbled in it last spring and summer.) I felt fine, and the 18 hour runs was one of my fastest, 8:15 per mile, while the 40 hour run was right on my average fed pace. (I’ll admit I probably pushed harder than usual to keep up my pace on those runs, particularly the 18 hour time-point.)

A couple of weeks ago, we did our first >48 hour fast, from dinner Wednesday to breakfast on Saturday. On Thursday afternoon I did a fairly fast 2-mile kayak paddle (36 minutes at 140 to 150 bpm heart rate; I’m 56 years old so that is probably Zone 4 for me) then a 4.5 mile walk. On Friday I did a moderately hard, by my standards, strength training workout: this plus some banded leg abduction and adduction that I forgot to log. Again, no problems with energy.

After three years on low-carb, I’d believe that I’m capable of partial glycogen repletion, at least, via gluconeogenesis (didn’t Jeff Volek demonstrate that in one of his studies?) but I’d still be surprised if I was using only 650 Kcal from my other stores…

(Michael - When reality fails to meet expectations, the problem is not reality.) #7

The numbers don’t mean you will necessarily ‘feel terrible’ rather than not ‘as energetic as on a feeding day’, only the max amount of body fat you can utilize. Being fat adapted on keto for several years means that you are metabolically flexible and change fuel sources seamlessly. The rest of the energy comes from another source: mostly lean mass, but as you note also probably some glycogen. The glycogen is muscle mass dependent specifically, plus a little more in your liver.

Another example: myself. I weigh 145 pounds and have 14-15% body fat - so about 22 pounds. Close to your numbers. I’m a bit older, though, so aren’t into long fasts or your intensity of exercise. But I’m no couch potato. I do what I call an overnight IF 3 or 4 times per week. That’s simply timing the last meal of a day to be about 12 hours prior to the first meal of the next. I feel very little hunger on those mornings. If my mid-day meal on a work day (I work full-time at Walmart) happens not to come until early to mid-afternoon - up 6-7 hours after my morning meal - I still don’t feel very hungry. Just a passing and gentle reminder that it’s time to eat. My energy remains the same regardless.

This is what metabolic flexibility feels like. If you and I were to fast for several days or longer, then we would quickly see first the fat mass and the lean mass disappear and overall energy expenditure drop as our bodies attempted to compensate for the lack of incoming fuel. If you don’t believe me, try a 14 day water fast.

Too much food
(Bob M) #8

I would think that theory about how much energy you get from fat reserves is highly suspect for many people, particularly those with high insulin resistance.

I’ve always thought that when you transition to keto from SAD, it can be quite hard to access your fat due to high “insulin resistance” (let’s ignore that I hate this term, as it’s basically meaningless without a definition). Now, you also have a lot of fat mass, so maybe you can access enough of your fat to have a fast.

But, for me, after fasting for the last 5.5+ years and gaining muscle, I have a harder time fasting now than I did when I started. I remember fasting 4.5 days and exercising twice during that period. Now, I do not exercise at all if I do a 4.5 day fast.

If I fast 36 hours, I will exercise at about the 32 hour mark. That seems to be OK.

On the other hand, I know people like Peter Attia who is really “thin”, yet also can fast 7 days (while exercising what I consider to be an extreme amount).

I haven’t quite figured out why some people fast so easily and others don’t. “Insulin resistance” provides part of the answer, but am I still “insulin resistant” after almost 8 years low carb/keto with many, many 36 hour, 3.5, and 4.5 day fasts under my belt?

(Bob M) #9

Should say I have to look at the original post in more detail. Will have to do that later.

(Michael - When reality fails to meet expectations, the problem is not reality.) #10

Studies on Starvation begins at the bottom of page 5.

Fuel Metabolism In Starvation
Too much food
(Joey) #11

I confess, the original post was longer than I have time for right now … but that won’t stop me from commenting :wink:

As a rather lean guy (5’9" hovering just below 140lbs) who typically eats TMAD - quite comfortably in terms of lack of hunger - I don’t fare so well when I reach around 24 hrs without eating anything.

My hunger is not great until about the 1 full day mark … then apparently I’ve got no Krispy Kreme storage left from 10 yrs ago (in truth, I was never much of a donut guy even pre-keto anyhow).

In short, being 2+ yrs fat-adapted on low-carb and feeling wonderfully energetic every morning/noon/night, I need to get my fuel from dietary fat within a day or so else I start to peter out. I can function, but I feel hungry :roll_eyes:


(Michael - When reality fails to meet expectations, the problem is not reality.) #12

Based on my experience, after fat adaptation I’d guess that serious hunger starts when your energy requirement exceeds your ability to access stored fat fast enough. Your metabolism does not want to eat lean mass since that leads rather quickly to bad outcomes. So the signal to eat fuel increases. But even so, it goes away eventually if you don’t. Read Scott’s Diary. Folks don’t starve to death feeling hungry, but very tired, weak and often befuddled.

Too much food
(Michael) #13

I have given this a lot of thought, and I (no surprise) do not have a definitive answer, but I “feel” that you do not get nearly the benefit of lowered insulin resistance from fasting until your insulin drops hard. This can vary greatly on the person, as you saw from my last fast, that took around 4 full days for me, whereas I have seen other insulin/time graphs of fasters who had it fall precipitously at 3 days or 2 days if they were metabolically healthy and of course, longer than 3 days if they are not. If you were very resistant, it would not surprise me if many of your fasts “just” got to the stage of actually dropping your insulin significantly before you broke your fast. Is that optimal? /shrug, not sure.
This is part of the reason I feel it is important to really note that time and then go for at least a few hours/days within that lowered range. I expect my next fast will be +2/3 days longer than it takes for my insulin to drop substantially (as I plan on an insulin lowering fast before another autophagy fast in 6 months or something). That might be 5-8 days total. Not sure when I will pull that trigger, but it will probably be not for a while, and totally spontaneous when it happens. Right now, I am just loving 3 steaks a day, so I am going to really enjoy eating for a few weeks fist. Only problem is I am eating those 3 steaks and NOT putting fat on my body, I need more fat, to fast, faster and harder!

(Richard Morris) #14

Yes the data that formed that hypothesis was from young fit mostly Mennonite men, conscious objectors of WWII being starved on 1400 kCal/day for 6 months.

Who knows what it would be in females? in the obese? in hyperinsulinaemics? Insulin doesn’t only blockade fatty acids in fat cells it also inhibits their transfer into mitochondria to be oxidized - so it surely must reduce the factor.

The problem is you couldn’t redo the study today. So it is all we have for now.

(Richard Morris) #15

“Oates’ last thoughts were of his Mother, but immediately before he took pride in thinking that his regiment would be pleased with the bold way in which he met his death. It was blowing a blizzard. He said, ‘I am just going outside and may be some time.’ He went out into the blizzard and we have not seen him since.”

Yikes. I once thought it might be a fine thing indeed to do a tour in the Antarctic.

“We want more food yet and especially more fat.”


(Richard Morris) #16

Just doing some back of the envelope math, your lean mass is 1-13%*71.3kg = 62kgs. You labile pool of amino acids available to use for making new protein or oxidizing to make energy is around 620g. Multiplying by the Atwater approximation gives you a buffer of roughly 2480 kCal that you can dip into that your ancestors could use to through a bad day of hunting.

Your athletic results are fascinating because you know your regular fed results and can compare like for like changing just one factor - length of fast.

(BuckRimfire) #17

Whoa. A bit disturbing to think that I might be burning hundreds of grams of protein to get through a short fast!

Sadly, I don’t really have much data. I’m too lazy to take careful notes (although I might be able to extract the data from my Garmin GPS watch’s files) and my running career was very short: started some time in March 2020 due to pandemic boredom and quit in August 2020 when I realized that there was no way the Seattle Marathon would be held in November 2020. Then I tore up my knee cartilage doing yoga in December (Half Moon Pose bad!), so may need a partial knee replacement before I do any more running.

I ramped up pretty steadily from 3 to 4 to 5 mile runs along the same route (then extended to 7-9 mile runs along varying routes), so I had a set of about a half-dozen 5 mile runs to compare. My 5-mile pace was usually 8:30 to 8:45 per mile, worst was 9:00, which was fed but I think I’d eaten a bunch of peanuts before, so I am now suspicious of peanuts. The 40 hour fasted run was close to 8:45 pace, IIRC.

As I said, because I was afraid I’d be slow I deliberately pushed the pace on the uphill blocks for the fastest fasted run, rather than trying to keep a more nearly constant effort as was my usual habit, so it’s a bit screwy.

Thanks for the reply!

(Richard Morris) #18

Knees are why I cycle.

Not precisely protein. If I am right about how much energy you could get from fat, and assuming glycogen isn’t sufficient to make up the arrears, then you used amino acids from your labile pool. The labile pool is a buffer for amino acids which by weight comprises roughly 1% of your total lean mass. Those amino acids are potentially going to be used to make proteins only if you don’t need them for energy.

That buffer is filled by your cells recycling proteins excess to requirements, or digestion of dietary protein, and emptied by your cells making new proteins (and other molecules based on amino acids), or using them for energy.

Sure if you use a lot for energy persistently and don’t refill them you may inhibit the production of new proteins and because we’re continually tearing down and rebuilding proteins you could be losing muscles for a fast at that point.

Same - knees are why I cycle.


best answer!!!

in general to all--------- life. natural life of what it was meant to be.

eat, thrive, live and live another day to eat and live and if no kill, we ‘can make’ it like all predators do cause we don’t eat weeds or grass for our survival and predators like lions and ALL of them go longer cause THEY must which is why I say all humans are ketogenic burn vs. glucose burn…glucose burn is for 'secondary back up for survival…to survive life…fat burn ketone life is what life is all about YET humans went glucose nasty fake food sugar burn in full and we all are losing health point blank now.

and remember our lives ARE NOT and WILL never natural.

We work in cubicles. We fret over money for survival to pay bills and survive on a dime. We have noise pollution as an assault to us, we ‘sleep’ on demand thru work issues to pay for it all, family-past experinces that hit and LIFE of today for humans is not and won’t be natural again for each of us…in that big term of ‘real life natural flow of life’ and what it takes.

to SG–If you are hungry SomeGuy then you ain’t eating enough, simple as that if one is chatting carnivore plan :sunny: but I also do not no your life in any way that could be effecting why you feel hungry and why your eating on carnivore is not what it should be to make ya not say that…if your fat content means more to you on vitality then darn eat the fat to help you at all times :slight_smile: no need to peter out truly if you know you, ya know LOL

but you know you, you see it and correct it as you need, I did the same, I found me on what works with my time on plan.

(Ohio ) #20

“Fasting doesn’t work”

Comfortable, safe dry fasting is what motivates me to extended fast. Keto allows me to extended fast. OMAD makes exercise easier.

Fasting works. Look at Terry Crews. 53 years old with baby skin. He’s been doing it before science had much to say about fasting.

Eating is an inconvenience. Fasting always works when it allows you to avoid public restrooms.