Monday, April 27, 2020

The miracle of safflower oil (3)

TLDR: PUFA in a mixed diet are obesogenic. PUFA under hypoinsulinaemic conditions are not. I doubt they get a free pass long term.

There have been some interesting snippets on Twitter recently, triggered by Diet Doctor's discussion about vegetable oils here. Perhaps the most controversial quote is this one:

"Disclaimer: Vegetable oils are routinely recommended as “heart healthy.” There is high-quality evidence demonstrating that replacing saturated fat with vegetable oils reduces LDL cholesterol levels. But at this point, there is inconsistent evidence whether this translates into fewer heart events or lower rates of cardiovascular mortality".

This is absolutely incorrect for people with pre existing cardiovascular disease as it was found, in a randomised control trial using safflower oil, that increasing vegetable oil for bulk calories will increase all cause mortality (p = 0.05), cardiovascular disease mortality (p = 0.04) and coronary heart disease mortality (p = 0.04). Mortality is an utterly hard end point and particularly the all cause mortality is an end point which cannot be argued with.

Let's rephrase that: in the context of a mixed diet in people with established heart disease vegetable oil (from safflower seeds) is going to increase you risk of death, especially from cardiovascular disease.

The main issue is to ask whether this still applies under low carbohydrate eating conditions. Given the role of insulin in CVD this is far from certain. But context will be crucial here and who would like to be the guinea pig?

The interesting twitter conversation goes like this:

Dr Westman: "In 20 yrs of clinical research and practice using LCHF/keto, I’ve never even mentioned reducing omega 6s, and it works wonderfully. Just cutting carbs gets the job done!"

Tucker: "I disagree, but @drericwestman is an excellent physician who does great work. This is more about determining ultimate causation so we can address people who can't just go low-carb, which is most of the planet".

I think both people are correct. I came to LC because it works. Over decades I've read studies where it works pretty much invariably on a group basis and studies from the mainstream usually advise progressively increasing carbs if they want to knock low carb and secure future funding. You have to pay the mortgage.

I am perfectly willing to accept that consuming carbohydrate in a rapidly absorbable form will overwhelm the liver's ability to protect the systemic circulation from hyperglycaemia so will require systemic hyperinsulinaemia to control that systemic hyperglycaemia. In particular hyperinsulinaemia comes with its attendant problems (ie most of medicine) but obesity only occurs when hyperinsulaemia is marked enough to overcome insulin-induced insulin resistance. I have no doubt this can occur without PUFA but I think it is massively easier in the presence of PUFA, which delay normal insulin-induced insulin resistance in the immediate post prandial period.

The role of polyunsaturated fatty acids is to stop adipocytes developing insulin resistance by limiting ROS generation. Combining hyperinsulinaemia with hypersensitive somatic cells is a recipe for maximising lipid storage in adipocytes and simultaneous packing lipid in to muscles, pancreas and anywhere else you care to imagine that sprouts an insulin receptor (most brain cells excepted).

Eating a low carbohydrate diet side-steps the problem by reducing absolute levels of systemic insulin. Down a set of unrelated rabbit holes I'm looking at what might control hunger under LC eating and PUFA may have some influence on this, but it is clearly a small effect when compared to the same dose of PUFA combined with an insulogenic diet.

Ultimately at low levels of insulin it doesn't matter how well or badly adipocytes respond to/resist insulin. There is so little insulin about that FFAs and ketones are able supply the body's energy needs, given some excess fat (especially visceral fat) available to be utilised.

Back to long term speculation: Do PUFA matter for non-insulin reasons on a low carb diet? Recall that López-Domínguez et al used a low calorie semi-starvation model (which is a partial mimic of low carb eating) to look at longevity in rodents (post is here). It certainly matters under their study conditions but the effect is small enough that I doubt it would show in any way for someone at 40 years of age under a year or two's exposure to a high PUFA but low carbohydrate diet. For those of us in this for the long haul it's much easier not to be the test case and PUFA avoidance seems prudent to me.

And I am undoubtedly still a low carb eater.



Tom of the Missouri said...

Outstanding summary! I have been avoiding PUFA's and carbs for years but nice to see things boiling down to this debate. Would be nice to see a long term controlled trial of the all cause mortality endpoints of low carb only and low carb seed oil avoiders. Will never happen of course. In the meantime I will avoid both like the plague except for some occasional full fat ice cream and dark chocolate when I cheat.

Outlaw said...

If the virus is not reason enough to avoid PUFA...

Not sure what to make of this:

Although at the end he comes so close... is soybean oil (or its surrogates) really the elephant in the room??

Michael Eades said...

I've been on the same stage with Dr. Westman when he said he's never mentioned reducing PUFA to his patients, and they've all done fine. In fact, I agreed with him, as I did the same thing for years when I was in active practice. It's difficult to find a lot of PUFA in a low-carb diet. Most of the foods loaded with linoleic acid and other PUFAs are basically processed snack foods that are high in carbohydrate, and they're eliminated in a low-carb diet. I guess you could slather a salad with PUFA, but most people use olive oil, which does have some linoleic acid, but not a huge amount. There is a little more in pork and chicken than in beef, but, again, not a huge amount. As long as you go on a whole-food, low-carb diet, it's difficult to get a lot of linoleic acid, so maybe part of the reason a low-carb diet works as well as it does is that it restricts linoleic acid along with carbs. And you typically get more saturated fat, which, in my view at least, is protective. So you get the best of all worlds: lower insulin, negligible PUFA, and plenty of saturated fat.

karl said...

@Michael Eades
Not just a little in chicken - it is so high I stopped eating it. I remember a paper that showed it was the biggest source in the US diet - can't remember where I saw it. Chicken skin is really high in LA 7.7% by weight according to - Chicken fat clocks in at 19.5%... Lard(pork fat) is about 10% - Beef fat - is around 2% (One of their stomachs has a lactobacillus that converts LA into CLA) .

Olive oil still has about 10% - concentrated seed oils are not human foods. My hunch is the PUFA in salad dressing makes people gain weight.

The amount of corn fed to chicken and pigs appears to effect the amount in the fat.

There are lots of indication that it bio-accumulates - and the amount in human fat keeps increasing:

I now consider beef a health food.

CIAB(Crap In A Bag) - is basically carbs cooked in LA.. first thing that went when people were panic buying.

@Outlaw -
What happens to people with ARDS if they get a feeding tube? That goop they pump in has lots of PUFA..

I rendered some suet and by adding it to my diet - appears to ramp down my appetite - time for someone to write a new diet book.. I'm thinking the Stearic acid effect is the real thing.

Frunobulax said...

I always wondered if safflower/soy/corn/canola oil contains other stuff that is bad, except n-6 PUFAs. All source plants are high in lectins. Not sure how they extract the oil, if they cold press the stuff the oil may contain lectins, if they extract it chemically it may contain other bad stuff. But I googled it a while ago, couldn't find a thing.

My basic assumption is that we should do well on whatever our hunter-and-gatherer ancestors ate. (@Michael Eades I'm a big fan of yours, sir!) And they certainly ate no grains and refined carbs. But as for carbs, I'm not a true believer: The longer I'm in the keto community, the more doubts I have that keto is really the optimal diet for us. If we're metabolically healthy (and that's a big if), why should a vLC diet be better than a "natural" diet that contains 15-20 percent of calories from unrefined carbs (which is what our ancestors ate, as per Or perhaps even some more carbs (-> "croissant diet") if we make sure that we stick to healthy fats?

Sure, we see dramatic health improvements with keto. I consider Keto as antidote to a high-carb diet, and many people (especially diabetics) will have to stick to keto for the rest of their lives. But this is true for any nutrient that has a U-shaped optimal curve: If we ate too much of it, we need to avoid it for a while. If we ate not enough, we have to supplement. And "a while" could be many years. In the case of carbs, we may have caused permanent (beta cells are dead) or semi-permanent (insulin resistance) damage. In this case we must stick to the antidote (keto) for a long time, possibly forever. But if there is no damage, do we really have any evidence that strict keto is better than eating some carbs (possibly seasonal)? Look up Robb Wolfs talks on metabolic flexibility, I think he makes some good points.

Personally I do eat some carbs (a bit of rice, or some fried potatos, once or twice a week), but I strictly avoid n-6 PUFAs.

Peter said...

Thanks Tom,

Hi Outlaw, haven’t listened to it all but PUFA seem to be very important to met syndrome and obesity. Sugar appears able to give met syndrome without obesity (skinny fat?)…

Mike, yes. Early in Atkins induction I recall frequently sharing a whole roast chicken with my wife. Often ready roast from a supermarket, in the carpark. Weight loss was fine… I recall a decade or so ago there was a tendency for people to ask me what type of fat might be best on a low carb diet and I didn’t have the data in those days. I have the data now.

Karl, yep, beef certainly is a health food!

Fruno, also yes, the Protons approach to PUFA does not need carb restriction but I think there is enough information on carbs, glycolysis, insulin and signalling pathways that I prefer not to go there.


Frunobulax said...

@Peter I don't think we have enough data on "low carb (20% calories from unrefined carbs) vs. keto for metabolically healthy people", or do you know any studies? We would need to find subjects that eat no n-6 PUFAs, keto or low carb for a long time and are Kraft I. Can't see how we could find a study population of metabolically healthy people.

The usual study suspects in any nutrition study ate lots of refined carbs for decades and have a high PUFA content in their lipid cells, meaning even if they lose weight they will burn more PUFAs than your usual hunter and gatherer. In this setting, any amount of carbs beyond keto may be damaging.

But evolution is such a beautiful thing. From all I've seen we're perfectly adapted to hunter-and-gatherer living conditions, with numerous fallbacks and high tolerance. It would be extremely surprising to me if we would react adversely to carbs (as they ate). The one uncertainty is that we can't be sure if our estimates of the carb content in their diet is correct.

ctviggen said...

Thanks, Peter, for being a voice of reason in all of this. For me, reducing carbs caused instant weight loss. That was way before I started intermittent and long term fasting and way before I knew what PUFAs were.

Unfortunately, a low carb diet to some extent will also reduce PUFAs. No fried food, for instance (from a restaurant). By how much? That's a good question.

Based on Fire in a Bottle's theory here:

I've postulated a "theory of everything". ;-) We vilified saturated fat, meaning no butter (I did not eat butter for at least a DECADE) and no milk fat. Everything low fat, at least for me. That ramps up hunger. But when I did restart eating some fat, what's recommended? Chicken and some pork. Both of which can be, and likely are, "high" in PUFAs. So, there was a double whammy of avoiding saturated fat while at the same time encouraging foods that were high in PUFAs to be eaten. Add high carb to that, and it's not a great recipe.

But I think this can affect low carb. I see people eating chicken, bacon, almonds, avocados, olive oil/avocado oil mayo, etc. None of these alone might be bad, but you add them all up, and I estimated a quite high PUFA diet using one of the calorie counters. I think I used one of your posts to figure our whether it was enough to cause fat storage, and I thought it was.

Could this be the cause of the "stalls" you hear about? Maybe.

What about the benefits of a high-beef, high beef fat diet? Could that be not only a lack of PUFAs but an increase in sat fat? It might.

And I hear about hunter-gatherer populations who ate "high" carb. But how does this affect a 55 year old man with a 30+ year damaged metabolic system, from both high carb and high PUFA (at some point, anyway)?

My CGM does not lie, and when I eat carbs, it flies through the roof. I have tended lately to separate things into what causes me to overeat and what does not. If I eat pasta, rice, pizza, even low fat varieties, I want more. I could easily eat 5+ slices of bread with no problem, and be hungry a few minutes later. In fact, even in my very low fat days (daze?), I was always hungry. 3/4 pound dried pasta in one sitting? Check. Hungry 10 minutes later? Check. (I haven't tried high saturated fat and high starch enough to know its effects, and no longer have a CGM to help with that testing.)

So, while I think the hunter-gatherer populations can be illustrative, I'm not sure how they apply to me.

What do I tell my kids?

1) limit carbs
2) limit PUFAs
3) If you have to eat carbs, eat them last if you can, and/or eat them with a lot of saturated fat and low PUFA if possible. (Eg, you go out with friends, and they get bagels; try to only eat 1/2 a bagel, drowned in butter and cream cheese.)

tuul said...

This keto trial used a low-SFA (10%) high-PUFA (14%) diet, which resulted in higher hunger scores than the carb group and no weight loss advantage.

It's a pity, I respect them for promoting low carb to Australia but they weren't able to get over the SF dogma and might've bungled their own results because of it.

As you winked at in your post, it appears that the study design tolerated an increase in carbs after the first period too. And the carb group was told to go low GI.

Peter said...

Bear in mind that to a dietician a calorie is a calorie and and linoleic acid is absolutely a Good Thing and in no way obesogenic. Their ingnorance is astounding. They key to understanding their bent-ness is the application of 30% dietary restriction. Who does this on a LC diet plan? Second is to increase the carbs with time, as you say. But I would like to see the people running the study go for 30% caloric restriction for two years!


Peter said...

As a follow on thought it's worth noting that 70g/d of carbs was Lutz's preferred carb level, but he was a saturophile without a doubt.


JR said...

Yes indeed, "energy-matched" from the study aka isocaloric is the keyword. Use it and the race is lost at starting line. Cico is not real life.
It was dr. Phinney who showed that using your own inside energy reduces need for eating - In the beginning. At the final line, you eat what you consume i.e. Eat More than in the beginning. These dynamics does not include steady and fixed caloric defect - as it is always with these isocaloric comparisons.
You should rely on your hunger- focusing on educating it.

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