Sunday, March 07, 2021

A chat with David Gornoski

Okay. Head briefly above water for a few minutes!

I had a chat with David Gornoski last week (or rather it was the week before, things are a little hectic here). You can find it at

and on Youtube here

David had emailed me after a number of people he'd already interviewed had mentioned the Protons/ROS hypothesis and Hyperlipid, so mostly I was trying to to get across the where the key concepts came from. I guess I rabbited on a lot about about the four main papers which shaped the idea. Then we wandered away on to more general things.

I think the microphone works!



Bob said...

I think it was a damn good presentation, perhaps the most "user friendly" explanation I've heard from Peter to date on the ROS / Protons concepts.

I know nothing about David Gornoski. From his website he seems more a philosopher than a science person, certainly compared to Paul Saladino or Raphi. But he appeared to keep up pretty well.

Yes, the microphone worked. Neither of you were especially loud, but I didn't have to strain to hear either. Thanks for that!

You seem to be increasingly comfortable with the podcast format. I'm glad you're doing this.

Lucas Tafur said...

Hi Peter,

Nice podcast, almost done with it.

What do you think of data like this? (wrt to your comment that you don't think there is a strong genetic component to obesity)

ctviggen said...

Lucas, I'd say that genetics is one component of susceptibility to become obese, such as when subjected to a high-PUFA diet. But if you're not subjected to a high-PUFA diet, maybe you don't get obese?

I personally think the protons theory is one aspect of obesity. I kept losing weight over time on low carb/keto, getting to less then "obese" by BMI and getting into size 34 inch (US) pants. Then tried the Croissant diet and immediately gained 20-25 pounds. Could not fit into any of my 34s, or a lot of my 36s. This is while trying to lower my PUFA as much as possible, though still eat chicken and pork at times. And I'd been trying to lower PUFA for years.

On the other hand, I did get some hunger reduction if I ate a TON of saturated fat with stearic acid. But if I ate dinner, and my kids were home all the time, so it was hard NOT to eat dinner, I basically ate a normal meal.

And, oddly, I gained weight all in my belly. Like a beer belly, only caused by croissants and other carbs/"starch" with saturated fat.

Anyway, I found I can easily eat 3+ "real" croissants (not with added stearic acid), with added butter, with zero issues, and be hungry not too much later.

For me, high saturated fat is not the sine qua non for obesity that I see people saying it is (dare I say that I can design a low-PUFA, high sugar diet upon which many, including me, can get fat?), but I do believe it's a factor.

That's the way I think of genetics: assuming nothing "weird" (eg, Lipedema), it's a factor, not a single element.

Puddleg said...

Historically, there were some people who became obese on the starch-plus-SFA menu that was the norm. However the rate of this was fairly low.
Replacing some SFA with PUFA has increased the % of the population vulnerable to carb-insulin effects. But the people who didn't need the extra PUFA still exist.

Passthecream said...

Peter, good Pod!

Puddleg I am one of those people. I was raised on a diet of butter and dripping plus all the carbs you could count and lots of homegrown fruit with no oil or marg in sight. Combined with innapropriate genetics wrt that diet, means that I have been overweight since I was a child.

CTViggen I think the croissant diet proves something different to the saturated vs polyunsaturated fat hypothesis, rather it provides good support for the insulin theory of obesity. There is so much sugar and flour in three croissants that it wouldn't make much difference if you added extra butter on top --- just measure your blood sugar afterwards. All those carbs make for a blood sugar maelstrom and end up driving the fat into storage. Then you get hungry again etc., as you discovered! And there is wheat which a complex source of other problems. Sorry about the ads on this link, it is the first one I could grab for supermarket type croissants:

100g is approx two croissants, so per item you have 5.5g sugar in 23g total carbs plus 4.5g of non-saturated fat and 6g of sat, before you spread butter on your delicious snack. With all due respect to Brad and his work with good healthy, fatty foods, can we put the croissant diet back in the cupboard now?

Peter said...

Thanks Bob, the computer mike was always fine until I relaxed, then I’d lean back in the chair.

Hi Lucas and George, the problem with a podcast is that they lack the ability to correct relatively subtle things which you mean’t to say but didn’t quite. I have no doubt that many aspects of physiology are under genetic control, obesity tendency included. What genetics cannot explain is the *rise* in obesity since the 1970s. You can’t go back to a podcast and clarify such a statement. I think my favourite Venus figurine is the one from Willendorf ( These figurines may be stylised but the artist knew exactly what they wanted to depict, and they weren’t fat-shaming. At 25,000 years ago the AHA was not involved.

Pass and ctviggen, Sucrose/fructose are interesting (as are very, very high PUFA diets). There is a lot of thinking to do about both.


Gyan said...

There used to be a hypothesis or model, based upon insulin, which claimed that one is safe eating low-carb or high-carb but there is a special danger zone associated with intermediate ratios of fat/carb irrespective of the fat quality. I think Stan (heretic) wrote something about it long back.

Which makes it strange considering that generally the starch-eating populations tried to incorporate some saturated fat with their starch.

Lucas Tafur said...


Sure, I understand how it is. Don’t sweat it; I think after hearing this podcast I’m on board with your protons idea, made everything more clear. Will binge on your proton series this week. I agree with your comment wrt to increase in obesity.


A key aspect of genetics and its influence on obesity is that a genotype needs the proper environment for expressing the phenotype; that is, genes and environment are inextricably linked.

What has probably happened during these last decades is that the environment is allowing these genetic differences to be expressed in the population. Before seed oil mass consumption, there were subjects genetically predisposed to be obese that didn’t develop obesity because the environment was not conductive to it, only in extreme cases it was manifested. The same genetic diversity is present nowadays but because it is conductive to obesity we can now see the genetic diversity being expressed. I also believe that in many cases the environment is superseding genetics; lifelong accumulation of “factors” makes people that are naturally lean start gaining weight during adulthood. Linoleic acid accumulation and turnover might be key.

The protons/ROS theory explains how the system works, under normal circumstances, at the core level. It is not “an aspect” of obesity; if true, different “aspects” are upstream of it.

In theory, the “croissant diet” (or equivalent) should work. But many times it doesn’t. It should work in the sense that you should be able to feel full and satiated and spontaneously eat less. That is the core of the hypothesis.

Now, why can it not work? I think a very likely culprit is linoleic acid. Unless you have been living under a rock or exposed since childhood to informed parents, and particularly if you’re obese, have plenty of linoleic acid stored in your cells and adipose tissue. Just like lipophilic toxicants, once you start mobilizing your fat stores, you’ll end up increasing plasma/bioavailabile LA. According to ROS theory, this should screw your satiety signals. We also have upstream mechanisms for LA increasing appetite (insulin sensitivity and endocannabinoid signalling). So even if you’re not eating LA *now*, you’re still using a lot. Hence, you’ll overeat easily on the croissant diet.

The solution for this, IMO, is to try to replace/deplete as much as possible your adipose tissue LA stores. This can be effectively done by fasting + high saturated fat ketogenic diet.

But there might be times in which you’re screwed no matter what, due to polymorphisms in nutrient-satiety signaling neurons (where most of the genetic differences in common obesity appear to be). Whether you can use normal physiology under this scenario is not clear to me.

Captain Sunset said...


Obesity and Pregnancy-to-term theory re Virgin Statues:

Peter said...

Cool Lucas.

Hi Captain Sunset. I have no doubt they represented an idealised desirable phenotype. I'm not so sure about the glaciers but then I've not read the paper itself!


Eric said...
This comment has been removed by the author.
Eric said...

nyt's notorious Jane E. Brody having a go at saturated fats again

This is her source:

It is interesting how unshaken Sacks is in his belief that LDL causes CVD. I was also struck by his assumption that sat fats raise LDL. Isn't it rather that by not eating as much PUFA you are not lowering LDL rather than actively raising?

cavenewt said...

Yeah, the Sacks article relies completely on the dogmatic assumption that saturated fat raises LDL which causes heart disease. He even says "There is no randomized clinical trial that determined the effect of coconut oil on cardiovascular events such as myocardial infarction, heart failure, or stroke," though to be fair, he goes on to acknowledge the limitations of doing such studies with all kinds of nutritional questions—however, without similarly acknowledging the limitations of observational studies.

I've got an elderly friend with many health issues who has become coconut curious and asking me for information. Now I have to start thinking how to talk to him when he sees this latest Brody execration.

Bob said...


Saturated fats in a low-carb diet certainly seem to keep my LDL-C quite high. Same with Paul Saladino. Same with Peter, I believe. Dave Feldman has commented extensively on "hyper-responders" whose cholesterol enters the stratosphere when they go high sat fat.

My votes for Worst Two Thoughts in the Sack's paper:

"Cholesterol-rich LDL is a major cause of atherosclerosis because it delivers its cholesterol load to the arterial wall and causes obstruction and inflammation."

Liver to heart: "Special Delivery from Me. PS. I hate you and want to destroy you."

"Replacing coconut oil with nontropical unsaturated vegetable oils, especially those rich in polyunsaturated fat, will have a health benefit."

Tucker Goodrich just rolled his eyes right out of his head. He should, at a minimum, send Sacks a copy of Nina Teicholz's Big Fat Surprise. She spent a couple of chilling chapters on polyunsaturated fats, hydrogenated and not. Tucker's own work is probably too technical for Sacks.

cavenewt said...

"nontropical unsaturated vegetable oils"

Gee, I wonder what industry contributed that phrase. Let's hear it for Big Ag subsidies!

Eric said...

I stopped following Dave a while ago. I thought it was low carb & fat, not necessarily sat fat, that did it?

How high is stratospheric? I was always into butter, cheese, the odd steak and olive oil. For my last work check-up (I travel a lot), I had been doing IF for some time, lowish carb, but it was not like I was supplementing with extra coconut oil or stuff, my LDL was a little higher than usual but not beyond 200. The doctor was ok because my HDL was high, too. I didn't argue.

Passthecream said...

So what are _tropical_ unsaturated vegetable oils and why can't they also be miraculous?

Bob said...


I wonder if you remember the TV commercials from the mid 1970's extolling the virtues of corn oil. The two I recall are the indigenous "Indian" woman pushing Mazola corn oil with the "goodness of maize" and no cholesterol. And John Houseman hawking Puritan vegetable oil in his best stern schoolmaster voice:


I remember marveling at the greatness of American industry, that it could produce oils without that icky cholesterol. I was but a wee lad of 21.

Passthecream said...

"relies completely on the dogmatic assumption that saturated fat raises LDL which causes heart disease"

I've proven to my own satisfaction at least, that chronic carb exposure raise LDL and low carb eating drops it right away. One of Dave Feldman's hacks, to eat extra fat in the two or three days before fasting to get a blood draw for lipids drops ldl way down and probably for interesting reasons: Recently consumed fat is transported by chylomicrons rather than regular LDL type lipoproteins and chylomicrons have a short half-life but LDL a longer half-life. Chylomicrons suppress LDL production. Over the time of the short fast after high fat consumption the lipoprotein dynamics are such that chylomicrons dissipate and LDL stays low.

Feldman is probably mixing unsat, mono, and sat fat in that chylomicron hack but I think that fat type is a minor consideration and perhaps unsaturated fats only drop LDL levels for the same short term reasons. Chylomicrons are saturation agnostic afaict.

cavenewt said...

Bob—"I wonder if you remember the TV commercials from the mid 1970's extolling the virtues of corn oil." I left home in 1973 at 19 and lived without a TV for most of the next two decades, so while those commercials sound vaguely familiar, I can't say for sure. Mostly I remember speecy spicy meatballs and "Hey Mikey! He likes it!" It has not proved beneficial the way a little bit of nutrition "knowledge" appears, and then industry monetizes the hell out of it, at which point it becomes embedded in the culture and is really hard to weed out later.

Pass—"One of Dave Feldman's hacks, to eat extra fat in the two or three days before fasting to get a blood draw for lipids drops ldl way down..." I can't help wondering, if you're going to hack it, then what's the point of the test at all? My annual test I only look at the HDL and the trigs and the HBA1c. And a lot of places don't even give you the last one.

Passthecream said...

I agree, the tests are worthless but one annoying thing about them here is the conclusion " dyslipidemia" which is automatically generated by the result data and rubber stamped by the supervising lab doctor and which persists in medical records if you let yourself be seduced into having one. I'm sure it's the same in other places and the only good reason for trying the hack is for insurance or job/medical purposes.

I'm way past that but I pointed the possibility out to a friend who was being heavily statinated. It opened their eyes after they started putting extra butter on their breakfast toast for a few days before the draw and got a result which pleased their cardiologist immensely.

It's an easy way of demonstrating how unbelievably stupid this whole obsession with HDL/LDL/HH PUFA is.

Passthecream said...

An old talk but a goody:

Them bones, them stones ...
We ARE the planet of the apes.

kellyt said...

Peter, thank you.

Also, I sent the American Heart Association the link to this podcast..

"AHA Recommendation
For good health, the majority of the fats that you eat should be monounsaturated or polyunsaturated. Eat foods containing monounsaturated fats and/or polyunsaturated fats instead of foods that contain saturated fats and/or trans fats."

If I could add one thing to the top of their website:
How to royally F@#$ Up your insulin signal and drive obesity, toxicity, and disease!

Peter said...

You are welcome kelly, I hope the AHA enjoy. I vaguely recall being quite insulting about them!