The Peter Attia Drive - #380 ‒ The seed oil debate： are they uniquely harmful relative to other dietary fats？

00:00:00.000 Hey, everyone. Welcome to the Drive podcast. I'm your host, Peter Atiyah. This podcast,

00:00:16.540 my website, and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:21.520 into something accessible for everyone. Our goal is to provide the best content in health and

00:00:26.720 wellness, and we've established a great team of analysts to make this happen. It is extremely

00:00:31.660 important to me to provide all of this content without relying on paid ads. To do this, our work

00:00:36.960 is made entirely possible by our members, and in return, we offer exclusive member-only content

00:00:42.700 and benefits above and beyond what is available for free. If you want to take your knowledge of

00:00:47.940 this space to the next level, it's our goal to ensure members get back much more than the price

00:00:53.200 of a subscription. If you want to learn more about the benefits of our premium membership,

00:00:58.000 head over to peteratiyahmd.com forward slash subscribe.

00:01:04.140 My guest this week is Lane Norton. Lane is a nutrition scientist and accomplished power athlete

00:01:10.300 and a very evidence-based thinker in the space of diet and metabolic health. And while Lane's been

00:01:16.680 on the podcast many times, today's episode is a little bit different from sort of the usual podcast

00:01:21.680 dynamic. Now, originally, this was supposed to be our first in a series that we've been toying with,

00:01:29.380 which is kind of a debate episode. I've talked about these a little bit on other podcasts,

00:01:33.580 but the long and short of it is I wanted to do a debate that was going to be incredibly rigorous.

00:01:39.340 In fairness, I've never really seen a podcast debate that was anything other than nonsense,

00:01:45.520 if I'm being blunt. And the reason is the format is basically impossible because people can make up

00:01:52.760 anything they want. They can cite anything they want out of context. And no human has the capacity

00:01:57.900 to know the entire body of literature up front. And therefore, it's very difficult to counter claims

00:02:04.720 regardless of whatever the topic of discussion is. I've just never found this appealing. And so

00:02:09.080 over the past year, we've been sort of noodling this idea of, well, what if we ran a debate like a

00:02:14.280 court case? And of course, in a court case, everybody has to sort of present their evidence

00:02:18.920 up front during a process called discovery. And that allows both sides to view the evidence of

00:02:25.820 the opposing arguments along with their own arguments in equal time. That way, there's no

00:02:30.820 ambiguity about what's being presented. We came up with an idea, which was we were going to start with

00:02:35.620 this first topic, which is to be the seed oil debate, which I'll clarify a little bit more in a

00:02:40.160 moment. And the idea was, we'll take two very thoughtful individuals who have opposing views.

00:02:45.580 And we did just that. So Lane was on the side of the argument, which was that seed oils are not

00:02:50.940 harmful. We had an individual who took the opposite point of view on that. The idea was that Lane and

00:02:57.280 this other individual would share all of the information that they plan to present with our team and with

00:03:03.580 each other. And that I would serve really here as a moderator of this discussion. And my research team

00:03:09.540 would serve as the fact checkers and that you as the public would basically serve as the jury to my

00:03:15.880 judge. However, the individual on the other side of this argument, for reasons that I'm not entirely

00:03:21.060 clear, decided that they just didn't want to do this. And we felt that it was still a worthwhile topic

00:03:26.080 to explore. And while the format of today's podcast is, in my view, not as potentially interesting as

00:03:33.680 it would have been had we done the original debate, we still think that it brought a lot of value. So

00:03:37.140 what I did was I kind of attempted to steel man the argument against seed oils and therefore allow

00:03:44.100 Lane to represent the pro-seed oil position for why seed oils may or may not be harmful when consumed

00:03:51.480 in isocaloric quantities relative to other fatty acids. That really is kind of the jugular question

00:03:57.320 here, which is, I don't think anybody would dispute that consuming excess anything is harmful. So

00:04:03.220 consuming excess calories in the form of seed oils versus monounsaturated fats versus polyunsaturated

00:04:07.780 fats versus carbohydrates, probably not many people are going to debate the downside of that. But what we

00:04:12.680 were really trying to understand is, is there something that under even isocaloric conditions would make

00:04:17.640 seed oils particularly metabolically harmful. So in this episode, we're going to discuss the role of

00:04:23.660 bias, evidence evaluation, and scientific interpretation as a lead into this, the historical

00:04:28.180 randomized control trials that shape the seed oil conversation, the mechanistic biology of LDL

00:04:34.560 oxidation and atherosclerosis, which is central to this argument, or at least one of the arguments,

00:04:39.260 the methods of oil processing and the chemistry behind modern seed oils, how evolutionary and ancestral

00:04:45.740 arguments fit into these nutritional debates, the relationship between seed oils, ultra processed

00:04:50.200 foods, and modern dietary patterns, the broader lifestyle factors that influence cardiometabolic

00:04:55.640 health, and then ultimately practical considerations around dietary fats, cooking oils, and real world

00:05:00.420 food choices. So in the end, I think we salvaged a lot of hard work that had gone into the preparation

00:05:07.040 for the debate over the course of a year. And ultimately, I think you're going to find a lot of value in

00:05:12.280 this. So without further delay, please enjoy my conversation with Lane Norton.

00:05:21.340 Lane, thank you for making another trip out to Austin.

00:05:24.020 Always. Pleasure to be here.

00:05:25.700 Okay. This is going to be a kind of a different episode. So I'm going to set this up and we're

00:05:29.960 going to give it a shot. So originally we were planning to do this as our inaugural debate series.

00:05:36.400 People have heard me talk a little bit about how I desired to do a debate series, which was to have

00:05:42.860 two people on who had opposing views on a topic. But I've been very vocal of my criticism of debates

00:05:49.840 on podcasts in that I think I could charitably call them useless, which is to say anybody can sort

00:05:56.020 of say anything. And in real time, it's almost impossible to verify what people are saying.

00:06:01.840 And it's not to say that people are necessarily lying. I think it's that people are maybe taking

00:06:07.020 liberal interpretations or not interpreting things the same way. And it would be much more

00:06:11.440 valuable if everybody could be looking at the same thing. So anyway, we had this whole idea where we

00:06:16.820 were going to have two people that were going to pre-submit all of their evidence to me and my

00:06:24.240 analysts, so the entire research team. And everybody was going to agree upfront what the papers were,

00:06:31.600 what the questions were that we were asking, what the data were. And during the process of the

00:06:36.940 debate, people could only reference things that were pre-submitted. In other words, we were going

00:06:41.400 to make it feel a lot like a courtroom. And I say that through the lens of we have a process in court

00:06:46.920 where you have discovery and the opposing lawyers have to submit everything. So this idea made a ton of

00:06:52.160 sense. My role was really to play judge, not jury. The public, the people listening to this would be the

00:06:57.520 jury. They would ultimately be the ones that would decide. And the first topic we're going to focus

00:07:02.000 on was the one we're going to talk about today, which is seed oils. How long did we spend on this?

00:07:05.960 About nine months, a year maybe?

00:07:07.180 Oh, I think it was over a year when we started talking about it.

00:07:08.960 So we identified you as the person who would speak to the argument that seed oils are not uniquely

00:07:15.600 harmful. We identified another individual who seemed incredibly qualified to speak to the other

00:07:20.840 side of this debate, which is to say that seed oils do pose a unique nutritional risk. And for

00:07:27.820 reasons I honestly don't even remember, that individual at some point just decided they didn't

00:07:31.600 want to do it. I think there was some concern that my personal view leaned more towards the side

00:07:38.900 that seed oils are probably not that harmful. I always am pretty vocal about my biases and I was

00:07:44.080 very vocal about stating, I don't really see something here, guys. But I was also clear to point out

00:07:48.620 that I'm simply the judge and not the jury. And ultimately the jury decides, and they're going

00:07:52.000 to also decide if I can be a fair judge. Nevertheless, the person decided not to do

00:07:55.520 this. And that left us asking the question, well, it is still a topic that people care about. And

00:08:01.120 therefore we view ourselves on this podcast as kind of the authority about going really deep on topics

00:08:06.820 that matter. So we thought we would do it anyway. However, we are going to do this a little

00:08:10.980 different than a normal podcast. Instead of just a regular interview, I am actually going to make my best

00:08:17.380 attempt to steel man the case for the other guest who is not here. Because again, in the process of

00:08:24.220 whatever we spent a year together, I did come to better understand the arguments for why a person

00:08:31.020 would think seed oils are uniquely harmful as a class of fatty acids. So with all of that said,

00:08:37.840 would you like to add anything before we jump into this?

00:08:40.960 I think that when speaking to bias, I think it's important to point out that everyone has bias.

00:08:46.640 Everyone has personal beliefs they developed. And that is just a human characteristic. There's no

00:08:52.760 way to get rid of that. I have my own personal beliefs. But what I will say is, I'm very upfront

00:08:59.080 about my biases. If we're on a topic where I have a different opinion than perhaps the consensus of the

00:09:05.480 literature or some other experts who I do consider to be good evidence-based experts, and I have a

00:09:11.080 different opinion, I will say, hey, look, this could be my bias showing here. Or I have a bias towards

00:09:16.360 this. I understand what this literature says. Here's why I think that maybe it doesn't capture it all

00:09:22.380 right now. And I think that that's about as best as you can do. And one of the things I told a friend the

00:09:28.700 other day was, people think that funding or money is by far the biggest driver of people essentially

00:09:37.400 like not sticking with the evidence. And I would say that in some cases, that's true. But I think

00:09:43.420 that personal beliefs are actually just as powerful, if not more powerful. I mean, look at how many

00:09:48.020 people spend hours online arguing over politics that get zero money from arguing about politics.

00:09:52.900 I just think that the current day and age with social media, with clickbait, that things are very

00:09:59.120 information siloed, and there can be a lot of talking past each other. And I hope today, what we can do

00:10:06.100 is present this evidence. And I will acknowledge where I think that there is something really there.

00:10:13.580 And then I will also explain why I think overall, my view is accurate and in line with the best data

00:10:19.900 available. And just to be clear, upfront, if anybody has a bias against seed oils, it probably should be

00:10:27.560 me. I came from a lab that was very much in line with the lower carb way of thinking of maybe saturated

00:10:37.720 fat isn't as bad as we thought. Maybe LDL doesn't matter. And I got into graduate school in 2004,

00:10:44.040 when that was a pretty popular idea that, okay, well, maybe it's not just saturated fat, LDL. Maybe

00:10:50.140 it's the particle size, the oxidation status, LDL to HDL ratio. And what I'll tell you is I said those

00:10:56.900 things for a long time and eventually changed my mind with the evidence. And just to point out one

00:11:02.800 more thing, my research was funded. I got money from the National Dairy Council, the National Cattlemen's

00:11:09.000 Beef Association, and the Egg Board. I'm painting with a broad brush, but I would say most of the

00:11:13.960 anti, the very, very rigorous anti-seed oil people tend to err on the side of either low-carb,

00:11:21.360 animal-based, or carnivore. And if anybody has a bias towards high-quality animal protein, it's me.

00:11:26.960 So I just want to start there and say that I think a lot of people, when the research conflicts with

00:11:33.620 whatever their viewpoint is, they immediately jump to funding source or think that there's something

00:11:38.160 nefarious going on. And what I will say is the scientific method is perfect. It is a perfect

00:11:44.480 method, but it is done by people who are not. And that is why it is so important to look at the

00:11:50.880 overall consensus of the evidence and looking at the different converging lines of evidence, which is

00:11:58.160 something you'll probably hear me talk about a lot today, because I do think there's a lot of converging

00:12:02.040 lines of evidence here. And that can give us a relatively strong or weak amount of confidence in how

00:12:07.920 accurate something is or a statement is. And so I just want to put all that out there, because when you're

00:12:13.520 looking through scientific research or you're scrolling social media, if you have a bias towards

00:12:18.740 something, you can always find a study or phrase something in a way that supports whatever you wish

00:12:25.280 to be true. And so that is why it's important. I think people like what you and I do, which is trying

00:12:31.880 to cut through that noise that other people who aren't equipped to read research simply can't do.

00:12:37.720 I'm going to tell you what I've heard as the four main arguments for why seed oils should be viewed as

00:12:45.740 potentially harmful. And we're going to kind of talk through these, not necessarily in this order,

00:12:49.760 but I'm going to kind of go through these. One comes down to the mortality literature on some

00:12:55.360 of the large RCTs. We're going to talk about two in particular. In other words, when we go back and

00:13:02.200 look at the literature, particularly in the era when people began to appreciate that saturated fat

00:13:09.020 raised, at the time it was total cholesterol, eventually when it got fractionated, it became

00:13:14.020 another subset of that called LDL cholesterol. We saw the association between LDL cholesterol and

00:13:19.760 ASCBD. The question became, hey, can we substitute something else for saturated fat, think margarine

00:13:26.900 versus butter, to lower cholesterol? So a couple of these studies were done and these studies,

00:13:33.120 while lowering cholesterol, did not lower mortality. We're going to talk about that. We're then going

00:13:37.840 to get into some really mechanistic stuff and talk about LDL through the lens of oxidation. You alluded

00:13:43.820 to this a little bit with your change in thinking around LDL particle size, but this goes kind of a step

00:13:48.160 further than just particle size. We get into the really granular biochemistry of what is happening

00:13:53.320 to an LDL particle that renders it pathologic versus maybe not so much. We're then going to talk

00:13:59.020 a little bit about not just the seed oil per se, but the industrialization of how a seed oil is refined.

00:14:06.660 In other words, is there something about the process of making a seed oil commercially that introduces

00:14:13.520 something that's harmful as a byproduct? And then finally, we're going to talk about this from maybe

00:14:18.220 an evolutionary or first principles perspective, which is, look, if we were having this discussion

00:14:22.920 a hundred years ago, there were no seed oils and people were a lot healthier. So maybe the

00:14:27.800 introduction of seed oils should be viewed as potentially problematic. So let's kind of start

00:14:31.800 with the Minnesota Heart Study. So I've talked about the Minnesota Heart Study before. This is a study

00:14:37.420 that took place in the 1960s. I believe it ran seven years-ish. It's notable because it was carried out

00:14:47.440 in an environment that would be very difficult to do today. It's very difficult to do long-term

00:14:52.480 nutritional studies with complete control over what your subjects eat. And yet that is essentially what

00:14:58.180 this study was able to do because it was carried out in institutionalized patients. So these were

00:15:03.240 mentally institutionalized patients, they were in patients in a hospital. And the experimental design

00:15:10.860 was quite elegant, which is the subjects were divided into two groups. One group was given a diet that was

00:15:20.100 higher in saturated fat. The other group was given a diet that was lower in saturated fat, but had an

00:15:26.440 isocaloric substitution of polyunsaturated fat. I believe it was mostly linoleic acid that made the

00:15:32.980 substitution. So in other words, you can think of this as substituting saturated fat like butter,

00:15:38.180 lard, meat for other sources. But the oils would, of course, then be the canolas, the safflower,

00:15:44.420 sunflowers, things like that. Yeah, I think in MCE, it was mostly corn oil and then margarine.

00:15:48.440 Yep. Now, here's what was really interesting, Lane. This study completed in, I believe, 1973.

00:15:56.400 So I think it ran from 66 to 73. I think that was the seven years of the study.

00:16:01.100 And the study found that indeed, the total cholesterol of the patients on the higher

00:16:08.820 polyunsaturated fat diet, again, this was isocaloric. It wasn't like we were playing

00:16:13.220 tricks with calories. No, no, no. This was, they were getting the same amounts of calories from

00:16:17.400 the macros even. It was just, we were substituting saturated fat for polyunsaturated fat. The patients

00:16:22.460 on the low saturated fat diet had a significant reduction in total cholesterol. At the time,

00:16:27.160 that was the only tool you had at their disposal. You didn't even have LDL cholesterol, let alone

00:16:31.760 APOB or particle size or any of that stuff. But you saw that total cholesterol went down.

00:16:36.080 Interestingly, and much to the surprise of the investigators, mortality did not.

00:16:41.120 This was such a surprise to the experimenters that they chose not to publish the results of

00:16:46.220 this study for another, I believe, 13 years. So the Minnesota Heart Experiment or the Coronary

00:16:52.080 Minnesota, I forget, there were several different names for it, but this study was not published

00:16:55.020 until the 80s. But as someone trying to make the case that there's got to be something wrong with

00:16:59.860 polyunsaturated fats, how could we otherwise explain that there was no improvement in mortality

00:17:05.320 despite the fact that total cholesterol went down? And total cholesterol went down quite a bit.

00:17:09.620 We don't have the data to say if it was atherogenic particles that went down, but given the magnitude

00:17:14.920 by which it went down, you have to assume at least the fraction of LDL cholesterol went down with

00:17:19.240 it, even if HDL cholesterol went down with it as well. So let's maybe start with that study

00:17:23.900 and say, doesn't that at least suggest that there could be something nefarious about

00:17:27.900 the substitution of lard to those polyunsaturated fats?

00:17:31.740 Yeah. And this is probably the single most popular study that gets cited by people who are

00:17:36.760 making the case that polyunsaturated fats are actually bad for you compared to saturated fat.

00:17:41.380 I want to be really clear. When I make any criticisms of this study, I think it was a very well-designed

00:17:46.220 study for the tools they had at the time and what they knew at the time. And so it's always easy to

00:17:51.240 Monday morning quarterback these things. But I think it was a very well-designed study. The big

00:17:56.380 takeaway was for every, I can't remember exactly the number, I think it was a 30 milligram per

00:18:02.760 deciliter decrease in cholesterol, total cholesterol, that there was like around a 20% increase in

00:18:09.360 mortality.

00:18:10.660 22%.

00:18:11.240 Yeah. So the biggest thing that I'm going to say right off the bat that really confounds all these

00:18:17.500 outcomes is the inclusion of trans fats. So the polyunsaturated fat group, they were getting

00:18:23.440 quite a bit of their polyunsaturated fats from margarine. Margarine at the time was around 25 to

00:18:31.140 40% trans fats. And we know that trans fats are absolutely atherogenic.

00:18:38.460 Can we pause for a minute? I maybe should have done this earlier. We should probably tell people

00:18:43.160 the difference between a saturated fat, a monounsaturated fat, a polyunsaturated fat,

00:18:47.060 and a trans fat as a subset of polyunsaturated fats. Let's go one-on-one just to speed this up.

00:18:52.900 A saturated fat is a hydrocarbon where every bond is saturated. That means there are no double bonds.

00:19:01.800 So every carbon is attached to another carbon, but it has two hydrogens on it. They can swiggle around

00:19:08.540 and that gives it unique properties. So one of the properties of a saturated fat is it is more likely

00:19:15.900 to be solid at room temperature. There are a whole bunch of reasons that are going to come up later

00:19:20.560 when we talk about why saturated fat plays a role in cardiovascular disease through its impact on LDL

00:19:28.260 receptors and cholesterol synthesis. But that's what a saturated fat is. So then tell us what a monounsaturated

00:19:34.020 fat is. So a monounsaturated fat means there is one double bond in the fatty acid chain.

00:19:40.380 Important to point out when it comes to these double bonds and why this probably makes a difference

00:19:44.820 is it changes the fluidity of these membranes because fatty acids and lipoproteins in particular,

00:19:52.860 they're not just like individual fats. They're arranged into what are called micelles,

00:19:58.140 which basically the polar head of the fatty acid is on the outside and then the inside you have all

00:20:03.340 the fatty acid tails. And the reason for that, of course, is that if it wasn't that way, they could

00:20:08.200 never travel around our body because to travel through the medium of our blood, you have to be

00:20:14.620 able to be repelling water on the inside, hydrophobic, while being attracted to water, hydrophilic on the

00:20:20.740 outside. Correct. So you have all these tails kind of pointing towards the center. And then if you think

00:20:26.400 about even on cells, the phospholipid bilayer, the tails are pointing towards each other. And for natural

00:20:32.480 unsaturated fats, like monounsaturated fats or polyunsaturated fats, most of those double bonds

00:20:38.340 are what are called cis double bonds. And I don't want to get too far into the biochemistry of it,

00:20:42.680 but essentially, if you have a cis double bond, it puts a kink in the fatty acid tail. If you have a

00:20:47.900 trans double bond, it doesn't change it. It still essentially looks like a saturated fat in terms

00:20:52.680 of structure besides the double bond. And it's actually easy to picture that. I almost wish I brought

00:20:58.060 a chalkboard to draw on. But anybody who's taken a biochemistry class will remember this. But a

00:21:02.800 cis double bond forces the two carbons up or down, but they are on the same side of the double bond.

00:21:09.460 So that's the real kink. Whereas a trans double bond, you have a carbon here, a double bond, and

00:21:15.840 then the other carbon is here. That stays much straighter than if you force the kink up or down.

00:21:21.360 And as we're going to talk about later when it comes to like LDL aggregation and whatnot,

00:21:26.420 the actual membrane fluidity is actually very important when it comes to the progression of

00:21:32.680 cardiovascular disease. So the membranes and the fatty acid composition of these lipoproteins

00:21:39.200 actually becomes very important. And it's very important to keep that in mind. But when we're

00:21:44.380 talking about unsaturated fat, it's important to point out that trans fats are very unique

00:21:49.820 in terms of the research literature very clearly showing an atherogenic effect.

00:21:55.940 Yeah. And the atherogenic effect of trans fats was so significant that they have effectively been

00:22:02.160 banned by the FDA. It's also important to understand how they came about. When the belief

00:22:06.960 and realization around saturated fats, which have been, I think, probably over demonized historically,

00:22:13.820 when that took place, food makers looked for a substitute. And you brought up an interesting

00:22:19.440 point a moment ago, which is if you have a trans fat, you can have something that is not saturated,

00:22:27.040 but behaves as saturated. So if you think of one of our favorite saturated products, it's butter.

00:22:32.700 Of course, as maybe we will talk about later, there's no food that is purely one thing. So it's not like

00:22:38.720 butter is just saturated fat. In fact, it's probably made up of mono and polyunsaturated fats,

00:22:44.600 if my memory serves correctly, as is even the fattiest ribeye. And actually has some natural

00:22:49.100 trans fats in it as well. Yeah. Low, low amounts. But what makes butter appealing is that it is solid

00:22:54.020 at room temperature. So in an effort to create something that looked, felt, tasted, and behaved

00:23:02.340 like butter, and we were going to deprive you of saturated fat, we had to put in something that at least

00:23:08.080 behaved like saturated fat. So when that margarine that was made up of high amounts, 25% is really

00:23:14.080 high, 25% of that is trans fats, initially thought that was a win because you got the benefit of

00:23:21.120 solid at room temperature. It was only after a few years we realized, actually, this was creating

00:23:26.600 far more heart disease than we were seeing even with saturated fat.

00:23:30.580 And part of that is likely because, and again, this we'll talk about in a little bit,

00:23:34.320 it makes the membrane very rigid because those fatty acid tails can get packed in tighter with

00:23:40.640 saturated fat and trans fat. Whereas when you have those kinks with mono and polyunsaturated fats,

00:23:46.180 it essentially creates space in the membrane. And that is actually very critical in terms of

00:23:52.920 particle recognition by the LDL receptor and also aggregation. But we'll get into that a little bit

00:23:58.640 later. So not only do you have trans fats being able to be packed into lipoproteins in a similar

00:24:05.340 way as saturated fat, but now they have a double bond that can be oxidized as well. So you're getting

00:24:10.020 kind of the worst of both worlds with trans fats.

00:24:13.540 So do we know how much trans fats was consumed by the low saturated fat group in the Minnesota

00:24:20.200 coronary experiment?

00:24:21.600 So as far as I understand, we don't have the specific numbers. We just know that it was likely

00:24:26.120 a significant portion of the polyunsaturated fat based on they essentially got a lot of their

00:24:31.560 polyunsaturated fats from either corn oil or margarine. Now, as far as I know, we don't know

00:24:37.520 the exact composition of each.

00:24:40.200 Was Chris Ramsden at the NIH, who I believe did a re-evaluation of this, was he able ever to identify

00:24:46.760 the raw data on that?

00:24:48.640 I'm not sure to be quite honest with you. I didn't see it anywhere. Maybe if people smarter than me are

00:24:53.140 watching or are familiar with it, I would love to know if they did. I do know there's some other

00:24:58.080 studies where they kind of looked at dietary adherence by looking at LDL in the blood or looking

00:25:06.480 at like linoleic acid incorporation into lipoproteins. But as far as this specific study, I don't think

00:25:13.060 they did. We have to remember, as you said, this study started in 1966. So this was very shortly

00:25:18.780 after it was identified and accepted that saturated fat raised cholesterol and that that seemed to have

00:25:26.080 a pretty strong association with heart disease. So when it came to doing this study, they had no reason

00:25:31.800 to suspect that these trans fats were going to be uniquely deleterious. And the other thing that's I think

00:25:39.440 important to point out, two things are important to point out with the MCE. The second is that during the

00:25:45.240 time that this study was going on, I think laws changed where people who were in psychiatric wards

00:25:50.920 could just check themselves out. And so many of the people in this study were not continuous. In fact,

00:25:57.180 it kind of threw a wrench in the researcher's protocol because I think they had originally planned that

00:26:01.840 these people were going to be continuously housed in these psychiatric wards for the duration of the

00:26:08.020 study. So now you've got another confounder where, okay, they're going in for a period of time and now

00:26:13.840 they're coming out and we don't know what they're consuming while they're out. Now, what I would say

00:26:18.540 if I was going to make a counter argument to that is, well, this is why randomization is important

00:26:23.340 because the likelihood is, okay, if they're changing what they're eating while they're out,

00:26:27.440 it should affect both. Yeah, it's probably equally distributed. So I think that's a far less

00:26:31.940 significant criticism compared to the inclusion of trans fats. And then I think the other thing that's

00:26:38.320 important to point out is it's really hard to do very long RCTs in humans. And this is something

00:26:45.340 uniquely difficult when you're trying to do cardiovascular disease research with hard endpoints

00:26:51.420 because cardiovascular disease is not something that develops in a couple of years.

00:26:56.020 It develops over the course of decades. And so Peter, for example, what I like to compare it to is

00:27:01.400 investing. If you and I start investing, same time, and I get into a mutual fund that on average over

00:27:09.040 the course of let's say 40 years gets me a 9% return and you invest in something that gets 8.5%.

00:27:15.160 If we look a couple of years out, there's really not going to be that much difference. I mean,

00:27:20.100 statistically in terms of significance, probably won't be a significant difference. But if we look 40

00:27:25.460 years out, there's going to be a major difference. It's important to understand that we get into the

00:27:30.500 mechanisms of LDL cholesterol. This is a total lifetime exposure risk. And so when you have

00:27:37.420 people coming in who we don't know what their baseline LDL was, because now if we were doing

00:27:43.620 this experiment, what would probably happen is you would randomize the groups based on their baseline LDL

00:27:49.200 or based on some other marker, maybe calcium score, whatever it is, so that you can have some degree of

00:27:56.080 confidence that you don't have differences at baseline. But they didn't know any of this stuff

00:28:00.840 back then. They didn't have those tools available to them. And so I think it's also important to

00:28:05.120 point out when you're looking at these studies where I think the average follow-up time in this

00:28:08.800 study was about one to two years. That's a pretty short period of time to actually see any real

00:28:15.060 differences in progression of cardiovascular disease and to try and actually find hard endpoints.

00:28:20.980 And as we'll talk about here with some of these other studies, the overall number of deaths are

00:28:25.860 very, very low, even to the point in one trial, like the corn oil trial.

00:28:30.460 Let's talk about that trial, because I was going to say, to get around your point about duration,

00:28:35.340 the Sydney heart study, which is the one you're referring to, attempts to solve this. So it was a

00:28:40.460 much smaller study than the Minnesota coronary experiment, which had nearly 9,000 subjects. This one had

00:28:45.740 a little under 500 subjects, but they were selected to be very high risk. So each of these men had just

00:28:53.300 suffered an MI. So you took a group of men who had just suffered an MI, but they were in the fortunate

00:28:59.980 group at the time who didn't die. And remember, back at the time of the Sydney heart study, you were

00:29:06.060 most likely to die from a heart attack. So you're already pre-selected to be pretty lucky. You haven't

00:29:12.060 died, but your risk is now very, very high. The baseline characteristics of this group were as

00:29:17.920 follows. Their baseline saturated fat intake was 16% of total calories. Their PUFA intake,

00:29:25.080 that's polyunsaturated fat intake, 6%. So the instruction set for the intervention group,

00:29:32.280 they were randomized. The low saturated fat group was instructed to increase PUFA to 15%

00:29:37.400 and reduce saturated fat to 10%. Now that's not draconian, but of course, these people,

00:29:44.580 they were not housed. The manner in which they were doing this was basically through

00:29:49.600 safflower oil and safflower margarine. And let's talk about what happened. So in this study,

00:29:57.420 there was a higher mortality in the control group, meaning the group that stayed on the saturated fat,

00:30:03.660 and it was 32% versus 20% at three years, which again, it's good in a bad way. It's good that you

00:30:10.020 can see a high mortality in three years because you've started with such a sick group. So again,

00:30:15.060 this would at least suggest that that group that lowered saturated fat, raised polyunsaturated fat,

00:30:20.520 they had a higher mortality risk in a short period of time. It could suggest that there's something

00:30:25.680 wrong with the safflower oil. Yeah. And important to point out, a lot of anti-seed oil people

00:30:30.780 will specifically talk about linoleic acid and safflower oil is very high in linoleic acid.

00:30:37.140 So the strengths of this study, it was longer, like you talked about, and these were people who

00:30:43.780 already had cardiovascular disease. So the likelihood that you could see more hard endpoints

00:30:49.740 during the duration of the study was higher. Now that also comes with the opposite side of the coin,

00:30:55.240 which is they already had a cardiovascular disease event. They've already had a lifetime

00:30:59.200 of accumulation of plaque. So how much difference can you really make on this truck that's already

00:31:07.220 rolling down the hill? Again, the main criticism of the study, which I think is quite frankly,

00:31:12.460 the biggest confounder with these trials is the inclusion of trans fats. A large portion of what

00:31:18.480 they consumed was safflower-based margarine, which at the time, again, was 25 to 40% trans fats.

00:31:27.240 It's really difficult to pick out, is this a polyunsaturated fat problem or is this specifically

00:31:35.260 a trans fat problem? And I would say, and I will argue that the human randomized control trials

00:31:43.780 that were not confounded by trans fats were actually probably better designed studies and better powered.

00:31:51.600 But I acknowledge that, okay, we're seeing some of these trials where higher cholesterol in the blood

00:31:59.320 is actually associated with lower mortality. One other thing it's important to point out is really

00:32:03.960 sick people, sometimes you can have what's called reverse causality, especially with cholesterol.

00:32:10.340 And let me explain. Once you get to a certain age, wasting diseases become a problem. High cholesterol

00:32:16.640 or low cholesterol, more specifically, low cholesterol can actually be an indicator of kind

00:32:21.600 of just overall poor health. That people with really low levels of cholesterol, they're more prone

00:32:27.680 to wasting. It may be more of a downstream effect than it is an upstream effect. And so this is where

00:32:34.180 it's really hard to pick out these sorts of things because it does get confounded by the reverse

00:32:40.340 causality, especially in people who are really sick. But even though the Sydney heart health study

00:32:45.160 was, as you said, a longer study, it was lower number of people. And even though they'd had a

00:32:51.880 cardiovascular event, the overall number of deaths was still pretty low, I believe. I think it was

00:32:56.540 somewhere around, I think it was under a hundred. Yeah. Total deaths were pretty low. 37 in the treatment

00:33:03.620 group, 28 in the control. Right. And that's a situation where when you're comparing 28 versus

00:33:09.260 37, just a few deaths, this is where you can have a sampling error, just a few deaths would have swung

00:33:15.360 the significance. And if you look at the confidence interval, the confidence interval nearly crossed

00:33:20.600 the one. And when a confidence interval crosses the one. Now, was that in the original analysis? I know

00:33:26.340 when the Ramsden re-analysis, it was 1.03 to 2.8. Was that what was originally published?

00:33:34.860 I actually don't recall if that was what was originally published.

00:33:36.860 Or maybe that was the original one. No, I'm sorry. That was the original published, I believe.

00:33:41.980 Yeah. That confidence interval is relatively wide considering the risk it's showing. And so,

00:33:49.260 because of course, the studies that support my contention, there's limitations on those as well.

00:33:54.580 But I think the most powerful thing, if I was going to pick it out, it's really the inclusion

00:33:59.060 of the trans fats, which is through no fault of the researchers of their own, because when these

00:34:03.400 studies were done, they just didn't know that trans fats had that effect.

00:34:07.840 Okay. So again, I'm playing the role of, I believe seed oils are bad. So the biggest contention we would

00:34:14.840 have if we go through the RCTs is because these RCTs were done at a time when trans fats were

00:34:23.780 the substitution fat du jour, I don't know how we're going to reconcile that. Basically,

00:34:29.320 it comes down to, do I believe that trans fats are less problematic than you believe? Because if

00:34:34.960 I believe that trans fats are actually not harmful, then that would take away your argument. You

00:34:41.200 wouldn't have an argument. Is there any other argument you've got besides trans fats on this?

00:34:46.060 I think the relatively short duration is another thing, because if we talk about some of the other

00:34:50.360 studies where it wasn't confounded by trans fats, some of those were longer, let me put it a different

00:34:54.780 way. Let's just say we take the trans fats out of it. Let's do that. And this was in the Ramsden

00:35:00.020 reanalysis as well. If we include all the human randomized control trials, looking at substituting

00:35:06.000 polyunsaturated fats for saturated fat, the overall effect is null. There was no effect one way or the

00:35:13.020 other. When you say that, you're referring to the largest Cochrane analyses. There are two, correct?

00:35:19.040 Yeah. Ramsden also did an analysis, I believe, where some of the trials that showed a benefit

00:35:24.260 to PUFA were confounded by the fact that there was omega-3s included in them. Now, I would argue that

00:35:28.520 if you look at the literature on omega-3s, it's actually not super strong that they decrease

00:35:33.540 hard cardiovascular disease endpoints. But nevertheless, it's a confounding variable.

00:35:37.760 So he took those out and then showed, okay, when we take out omega-3s, we see an increased risk.

00:35:44.240 But of course, that still includes the trans fats confounders. So if we just include both and lump

00:35:51.680 them all in together, the net overall effect is that one way or another, reducing saturated fat,

00:35:57.740 raising PUFA, it was equal risk. And so what I would say, if the statement is seed oils are uniquely

00:36:05.160 deleterious to human health, even if we take the trans fats out of it, and if you're going to allow

00:36:12.000 those to be included, you have to have, I call it, logical symmetry, which is if I'm going to allow

00:36:18.000 this confounding variable to support my point, I also have to allow your confounding variables to

00:36:22.420 support your point. Otherwise, we just got to find the ones that take both of them out. And so in that

00:36:27.160 case, the net effect is still no harm. What's the other confounding variable, the EPA and DHA?

00:36:33.380 Correct. If you want to start talking about some of these other studies, I will touch on the one other

00:36:37.540 study they cite is the Rose Corn Oil Trial, which I, quite frankly, I don't know what to make of it

00:36:42.320 because it was only like 70 people in the entire thing. Yeah, it was 26 people in a control group,

00:36:50.640 26 people on an olive oil group, and 28 people in a corn oil group. So again, this is a two-year

00:36:57.700 study. This was in people with significant cardiovascular disease. So the people in the

00:37:03.000 control group, business as usual, three of those people died. Five people died in the olive oil

00:37:10.020 group, but eight people died in the corn oil group. Sorry, that was total deaths for what it's

00:37:16.280 worth. Cardiac deaths, one in the control group, three in the olive oil group, six in the corn oil

00:37:23.080 group. Now again, the confidence intervals on this were very large because the sample size was very

00:37:29.320 small. This one is a bit challenging because the participants were given their oils. You could

00:37:35.080 think of this as kind of an early version of the PREDIMED study, where it was a free-living study,

00:37:41.360 but the participants were given olive oil when they were in the olive oil group. There were three

00:37:46.140 groups. There was low fat, and then there was high MUFA, high monounsaturated fat through olive oil,

00:37:52.100 and you were given a bottle of olive oil every week, or through nuts, and you were given the nuts

00:37:56.060 every week. So here, it was a free-living study, but you were given your corn oil, or you were given

00:38:00.600 your olive oil. Other than that, we don't have dietary recall. Now, the benefit, not confounded

00:38:05.820 by trans fats. There was no confounding with trans fats here, so that would be a positive. But I think

00:38:11.160 the first thing I would say is, I think you said this, the cardiovascular disease deaths were one,

00:38:16.440 three, and six. Yes. I don't know what to make of that. Those are such small numbers that you're so

00:38:22.540 prone to sampling errors, where if you'd had 10,000 participants, that's very likely going to get lost

00:38:29.200 in the wash. And let's just take the olive oil group. Even people who are anti-seed oil typically

00:38:35.720 acknowledge that olive oil is heart-healthy. Well, you had three times the deaths from cardiovascular

00:38:40.000 disease in the olive oil group compared to the control group. And so it doesn't fit with the

00:38:46.500 evidence we have. And it's so small. And like you said, the confidence intervals, I mean, if we want to

00:38:51.740 talk about a confidence interval being wide of 1.03 to 2.8 as being wide, the confidence interval in

00:38:58.240 this, I believe, was like 0.6 to 37. I mean- So not statistically significant either.

00:39:05.580 No, no. It was absolutely massive. And here's where my PhD advisor used to say,

00:39:11.040 if you torture the data enough, it will confess what you want it to show. And so if you just say,

00:39:15.700 well, there was six times the number of deaths in the polyunsaturated group, you're technically

00:39:20.560 correct. But you're leaving out a really big portion of the data. And again, when we plop all

00:39:25.800 these studies into the meta-analysis, don't take out any confounding variables. What do we see?

00:39:31.840 We see a null effect. Now, if we take out the trans fats, there was a meta-analysis, I think,

00:39:37.340 in 2017 where they put together all the trials. There were human randomized control trials looking at

00:39:43.440 substitution of polyunsaturated fats for saturated fats not confounded by trans fat. This was a very clear

00:39:49.940 benefit to substitution. I think it was around 20% reduced mortality risk.

00:39:57.180 No, it was even more. I mean, I have the luxury of cheating because I'm looking at the actual

00:40:01.740 figure. So the rose corn oil trial has the largest hazard ratio of any trial ever done.

00:40:08.440 Of this, I don't know, of any trial, of these trials, it was 4.64. What that means is there was

00:40:16.220 a 364% increase in risk if you took corn oil, but it did not reach anywhere near statistical

00:40:26.940 significance to your point because the hazard ratio confidence interval, the 95% confidence interval

00:40:32.400 was 0.58 to 37.15. So if we limit ourselves to this body of literature and the 1, 2, 3, 4,

00:40:44.600 5 studies, we do not achieve statistical significance, though we do have a trend towards risk.

00:40:52.700 That trend is 1.13, meaning a 13% relative risk increase.

00:40:58.640 If we include the trans fat, the trials.

00:41:00.240 Yes. So this is a bit of a messy analysis to your point because we're including the MCE,

00:41:05.900 we're including Sydney, but we're also including rose corn oil. We're just trying to get as many

00:41:10.860 bodies as possible through the engine of the meta-analysis and we don't reach significance.

00:41:16.340 In fact, the only study that reaches statistical significance is the Sydney heart study. That had

00:41:24.120 a 74% increase with a confidence interval of 1.04 to 2.91. So it got there, but I guess your argument

00:41:32.680 is going to be yes, but they were at 25 to 50% trans fats. So this is a tough one to score, Lane,

00:41:39.880 because on the one hand, it looks pretty bad for polyunsaturated fats here. Every time you eat

00:41:45.740 them, it just seems to move you in the wrong direction, except for the linoleic VA study.

00:41:50.560 Well, there's a few different human randomized control trials. So the VA study,

00:41:55.160 so there's several strengths to this study.

00:41:57.420 Tell us what the study was about.

00:41:58.700 So it was in veterans' homes. The food intake was controlled. They provided it to the participants.

00:42:04.260 And it was, I think, around 850 participants. And the average follow-up, I believe, was just under

00:42:09.680 nine years. So that's actually a pretty long study for this. Not confounded by trans fat,

00:42:15.740 and they did show, again, I don't have the raw numbers in front of me, but I think it was around

00:42:20.940 a 20 or 30% reduction in risk. It was, though it did not reach statistical significance. It didn't

00:42:26.980 have a wide confidence interval. It was actually quite narrow, but it crossed unity. So it had a

00:42:32.080 18% reduction in overall risk, but it was 0.56 to 1.21 was your confidence intervals. But this is one

00:42:41.580 that's on the other side of what you said a moment ago. They included omega-3s, if we believe that.

00:42:46.460 So the question is, did the omega-3s help, or was the study underpowered? Is that why it didn't

00:42:52.200 reach statistical significance? Or does it just not matter?

00:42:56.080 I mean, this is why we do meta-analyses. Because when it comes to single studies,

00:43:00.660 sometimes, especially when things are messy like this with hard endpoints, even if 800 people sounds

00:43:05.780 like a lot, when you're trying to get hard endpoints like mortality and cardiovascular disease events.

00:43:11.220 It's pretty small.

00:43:11.800 It's pretty small. There are some other trials out there. The Oslo Heart Health Study,

00:43:16.600 that was only 400 people, but that showed, I think, like a 47% risk reduction. Again,

00:43:22.220 confounded by omega-3s. Then there was probably one of the strongest studies, I think, is the Finnish

00:43:28.540 hospital study. And the reason is, not confounded by trans fats, not confounded by omega-3s. People

00:43:37.240 did each diet for six years, so it was a crossover design. So I guess it's important for me to cover

00:43:42.760 what a crossover design is briefly. So a crossover design is when you take people and you have them

00:43:48.100 do both treatments. Now, there are downsides. The downside is, okay, what about a crossover effect

00:43:55.960 where, okay, maybe you have, let's say, a cardiovascular event on one of the diets, but you did the other

00:44:02.820 diet before that. So is it from the diet you're on now, or is it from the previous diet? The way researchers

00:44:09.140 attempt to get around that is by essentially making the order in which they do both diets split, so that

00:44:17.160 there's 50% of people did one diet first, and then the next diet, and then they reversed it. Usually, you do that

00:44:23.620 by randomizing each individual person. Now, here's the weakness of the study. They didn't

00:44:29.000 randomize individually. They had two hospitals. One hospital started with one diet, one hospital

00:44:34.780 did the other diet, and then they switched them after six years.

00:44:37.900 Completely makes sense. Understandable.

00:44:40.040 Logistically, it makes sense.

00:44:41.080 Yeah.

00:44:41.340 Yeah. Otherwise, you're trying to like-

00:44:43.660 Figure out if Bob in bed A is on this diet versus Sally in bed B.

00:44:47.720 Exactly. I understand why they did it, for sure. I guess if you were going to try and make the argument

00:44:52.540 that this introduced a lot of bias, the argument you would make is that perhaps there was inherent

00:44:57.660 characteristics about one hospital versus the other that were more prone to people getting

00:45:02.480 cardiovascular disease. I would say that risk is probably pretty low, but it's an argument that

00:45:08.400 can be made. The benefits of this is it was 1,200 people, but since they crossed them over,

00:45:13.380 effectively, it was like 2,400 people. One of the benefits to a crossover experiment is

00:45:18.240 you can up your power.

00:45:20.560 I hope somebody listening is sharper than I am in my statistics because it's been so long,

00:45:25.500 and I used to know the answer to this. I think it's actually even more impressive than 1,200 to

00:45:31.620 2,400. I think it's even more, not to overuse the word power, but it's even more powerful

00:45:37.360 statistically when you do a crossover because every person gets to be their own control.

00:45:41.880 Exactly. That is probably the largest benefit, which is, as you said, it's the way you do the

00:45:48.220 t-test and the statistical comparisons. Each person is their own control because when you're

00:45:54.140 doing a parallel group design, which is where you just have one group on one diet, one group on the

00:45:59.300 other diet, you are assuming that the randomization process randomly distributes the baseline

00:46:05.060 characteristics that may be different between groups to where they're equally distributed amongst

00:46:09.060 the groups. But you don't know that for sure. But when you cross over, now you're guaranteed,

00:46:14.280 I don't want to say guaranteed, but you are very confident that inherent baseline characteristics are

00:46:19.880 now no longer a confounding variable. And it's likely that the people's overall lifestyles are

00:46:26.520 going to be similar throughout the course of the experiment. There's less risk of bias from

00:46:30.940 inherent lifestyle differences as well. Somebody having more stress versus less stress,

00:46:35.080 those sorts of things. And in this study, again, not confounded by trans fats,

00:46:40.300 not confounded by omega-3s, and pretty well-powered for a long duration, six years on each diet,

00:46:47.320 12 years overall. And they saw a pretty significant reduction in the risk of cardiovascular disease

00:46:53.420 events and mortality.

00:46:55.380 Yeah, 41% reduction in the treatment group that was, and just to give folks what the actual intervention was,

00:47:05.080 so the high saturated fat or control arm was 18% saturated fat, 4% polyunsaturated fat,

00:47:13.440 three to four, they gave some wiggle room. The treatment group was 14%, so they increased to 14%

00:47:19.320 polyunsaturated and cut saturated fat in half to 9%. These were kind of similar targets to the MCE study.

00:47:26.540 I want to say they measured linoleic acid in tissue too. I believe they did that.

00:47:31.080 I don't know.

00:47:31.820 If I'm wrong, I'll eat crow, but I think that they actually looked at like tissue and blood

00:47:36.220 biomarkers of linoleic acid and showed an increase in linoleic acid content to verify the

00:47:40.760 dietary treatment worked.

00:47:43.880 What would be your best explanation for this? Because again, the finished study,

00:47:47.960 I mean, 41% relative risk reduction in cardiovascular disease with that low saturated fat,

00:47:55.320 high polyunsaturated fat, and the confidence intervals are really tight.

00:47:59.580 I think, again, this is why this is a situation where, okay, I'm acknowledging

00:48:03.780 what seems to be a negative effect of polyunsaturated fats in some of these studies,

00:48:09.460 but I'm explaining why I think that the studies that show a reduction in risk are on balance better.

00:48:17.980 Again, not making a criticism of the researchers because, as we talked about, I think they did the

00:48:23.520 best they could at the time with the information they had. So this isn't an indictment on the

00:48:28.520 researchers whatsoever. But when you look at these studies that showed a reduction in risk,

00:48:33.480 they're longer, they're typically better controlled, and they don't have the inclusion

00:48:38.400 of trans fats. I think those things are the most powerful movers.

00:48:42.580 So we think that that is the best explanation for why the Finnish heart study came up with a

00:48:47.260 different answer.

00:48:48.040 Yes. There was a meta-analysis in 2017, and I can't remember the name of the lead author,

00:48:54.600 but they basically did a meta-analysis of the studies not confounded by trans fats. Now,

00:49:00.000 they included some of the ones that had fish oil in them, or omega-3s rather, but they showed overall,

00:49:05.420 again, I want to say it was around a 21, it was 29% reduction in risk. So again, if you look at

00:49:12.160 the Ramsden analysis, in their best case for polyunsaturated fats being bad, it's a smaller

00:49:18.280 increase in risk, and the confidence intervals are very wide.

00:49:21.780 So let me play devil's advocate for a moment. And again, this is sort of maybe a silly argument,

00:49:26.520 but what if this says more about saturated fats harm than polyunsaturated fats benefit?

00:49:34.540 Well, and this is a situation where when you're doing nutritional studies-

00:49:38.320 You have to substitute something.

00:49:39.560 Right, right, because-

00:49:40.760 If it's going to stay isocaloric.

00:49:42.000 One of the things when we were talking about doing this debate with the other individual,

00:49:45.660 I was very clear in saying, I'm not saying that there's no deleterious effect to seed oils

00:49:50.640 whatsoever, because added oils in the diet are a major source of calories, and excess calories

00:49:57.640 are not innocuous. So we have to compare apples to apples, which is when you are substituting in a

00:50:03.460 one-to-one ratio. What is more beneficial? The reality is, I think you could take any food or

00:50:10.780 any nutrient, and you can find both positive and negative pathways that it activates. The question

00:50:17.920 is not whether or not a nutrient or a particular food activates positive or negative pathways,

00:50:23.260 and I say that very broadly, because as you and I both know, there's not really such a thing as a

00:50:28.440 positive or negative pathway. But just in general, there could be positive and negative outcomes.

00:50:32.800 What matters is, on balance, what is the net effect? And I explain this, I'll use another

00:50:38.800 financial example. So we're talking about these pathways, we're talking about mechanisms,

00:50:44.440 which are important. If an outcome exists, there's a mechanism or mechanisms to explain it.

00:50:48.940 But those are like single stocks. An outcome, like a cardiovascular disease event, that is like

00:50:55.140 a mutual fund. And so what I mean by that is, I could take a mutual fund that's doing really well,

00:51:01.000 say, up 20% year over year. But I could find a few stocks in it. I'm like, oh, you don't want to

00:51:05.520 invest in this. Look at these, they're down like 50%. But what matters more, those individual stocks

00:51:10.200 that are down or the overall mutual fund is killing it? You care about what the overall mutual fund is

00:51:15.600 doing. And just an example of this, smoking decreases the risk of Parkinson's disease. This is

00:51:21.460 a very consistent effect in the research literature. But on balance, I don't think anybody's going to say

00:51:25.960 smoking is good for you. We should just point out to listeners, it appears to be the nicotine

00:51:29.900 that is causing that benefit. I don't want anybody with Parkinson's disease or at risk for Parkinson's

00:51:35.240 picking up cigarettes. If you're going to do anything, just choose some nicotine, choose

00:51:38.960 non-tobacco nicotine, please. But perhaps a better comparison would be, we know, I believe aspirin is

00:51:44.700 an anticoagulant overall, but it also activates some pro-coagulant pathways. But the overall effect is

00:51:50.780 anticoagulation. So if I just wanted to pick out and say, well, look, it activates these pathways,

00:51:56.540 not that it doesn't matter, but on balance, on balance, it is a net positive for anticoagulation.

00:52:03.780 And so when it comes to looking at polyunsaturated fats versus saturated fats, and we will get into

00:52:08.900 the mechanisms of these. Yes, there are mechanisms that exist that would suggest that polyunsaturated

00:52:15.080 fats can have a negative effect. But on balance, there are more mechanisms that exist

00:52:20.600 that show saturated fat to be a negative. To circle back to your original question,

00:52:26.180 is this a effect of polyunsaturated fats being beneficial or saturated fats being a negative?

00:52:32.420 I think it's almost impossible to disconnect those two questions because when it comes to

00:52:37.560 nutrition research, to do it accurately, you're looking at substitutions. So if you're going to

00:52:41.900 take saturated fat out, you've got to put something in. And so if we look at studies where you substitute

00:52:47.560 carbohydrate for saturated fat, not really much change. I would say that probably depends on the

00:52:53.460 type of carbohydrate very much. If you were doing like fiber dense sources of carbohydrate,

00:52:57.260 I'm pretty sure you'd see a reduction in risk. Monounsaturated fats, there appears to be a

00:53:02.260 reduction in risk of cardiovascular disease. It appears to be not quite as powerful, at least in

00:53:07.440 the cohort studies. But again, since we have to substitute something, let's make it very basic.

00:53:15.700 Even if there was no net, if polyunsaturated fats didn't do anything cardioprotective, it's still

00:53:22.240 cardioprotective in that when you put them in in place of saturated fat, you have improvements in

00:53:27.820 outcomes. I would argue that there are mechanisms in place that explain why polyunsaturated fats

00:53:32.640 are cardioprotective. My point I'm making is I think it's difficult to disconnect those two

00:53:37.800 questions. Do we know if the Cochran analyses on this question have blended and included the studies

00:53:46.060 across those that contain trans fats and those that don't? The one from 2017? Yeah. When they look at

00:53:52.000 all the PUFA versus SFA studies? No. So they specifically excluded ones that were

00:53:57.420 confounded by trans fats. Ramsden had one where it included basically all of them,

00:54:02.640 net effect being null, nothing. And when he pulled out the ones that were confounded by omega-3s,

00:54:08.920 they showed a negative effect of polyunsaturated fats. With trans fats still in? But with trans

00:54:13.760 fats still in. Okay. And Cochran had no trans fats. No trans fats. But did have omega-3s. But did

00:54:19.520 have omega-3s. And had a slight benefit to PUFA. Yeah. Well, pretty decent. I think, what did you say

00:54:24.560 it was? It was 20, I want to say it was like 23%, or no, 31%. Yeah, 29 or 31. Okay. Yeah. Yeah.

00:54:30.360 Somewhere around there. I apologize for not remembering the raw numbers. There's a lot

00:54:34.340 floating around in there. Has anybody done the analysis of excluding omega-3, excluding trans

00:54:40.440 fats? Well, I mean, you're basically down to- I know you're down to very few studies. I think

00:54:44.220 you're down to like two studies. I just want to ask a question. I apologize to interrupt.

00:54:49.540 That's good. When you say omega-3s, are you talking linoleic acid or are you talking EPA DHA?

00:54:57.640 I believe they're specifically talking about the EPA DHA. But I know like alpha lino-

00:55:02.960 Alpha linoleic acid. Yeah. Yeah. Alpha linoleic acid is omega-3 as well.

00:55:06.820 Alpha linoleic acid doesn't- Although no, that was part of it. That was part of it as well.

00:55:10.880 Because it doesn't convert very efficiently to EPA and DHA. So does that mean that these

00:55:16.400 studies that contain omega-3s are going out of their way to supplement with EPA and DHA,

00:55:22.300 which of course you can really only get in high quantities from marine sources?

00:55:25.260 It wasn't clear in the research literature I read. It was just basically like, yeah,

00:55:30.520 there was some omega-3s in the diet. There's basically, I think, two studies that were not

00:55:35.740 confounded by trans fats, not confounded by omega-3s. And that was the Finnish study that

00:55:39.900 we talked about. And then STARS, which STARS was not looking at hard cardiovascular disease endpoints.

00:55:46.860 This was looking at plaque progression. So I believe this was a one-year study and they looked

00:55:52.000 at either people doing low saturated fat, higher polyunsaturated fat versus higher saturated

00:55:56.660 fat, lower polyunsaturated fat. And they looked at the progression of plaque. So not hard outcome,

00:56:04.340 but I would say that the strength of this is since you are looking at the progression of what we know

00:56:10.880 causes these myocardial events, even if you don't have a myocardial event, you can get a really good

00:56:17.680 idea of what's going on. And so, for example, and I'm just going to make a hypothetical here,

00:56:22.940 you could have somebody in one of these other studies where they got right up to the point

00:56:27.020 where they got like a 99% blockage, but no myocardial event in the time where they were

00:56:31.620 measuring it. And they just came up as that's a risk reduction. But in reality-

00:56:36.400 They had disease progression.

00:56:37.520 Right. So the benefit of this is you still get a harder endpoint than just biomarkers,

00:56:42.860 but you can get a little bit more granular than just looking at, I guess, the way to describe

00:56:47.380 mortality and cardiovascular disease events as big blunt instruments.

00:56:51.820 So let's talk a little bit more about these mechanistic things now, because again,

00:56:55.480 most of these trials were done with a brute force tool of like the hardest outcome is all-cause

00:57:02.200 mortality. But again, that's a high bar. When you do clinical trials, you trade off between outcome

00:57:08.200 and barrier to outcome, for lack of a better term, right? So if you want to use all-cause mortality,

00:57:13.840 that is the ultimate measuring stick, but it's a high bar to clear. And then you can go disease-specific

00:57:19.040 mortality. Okay, we're now going to look at coronary death. Then you can go one step lower,

00:57:23.240 and an example would be MACE, major adverse cardiac events, where we're going to use heart attack

00:57:27.700 and stroke and cardiac death. And then you can go one step below that and say, well, we're just going

00:57:32.300 to look at a change in cholesterol, or we're going to look at a change in LDL cholesterol.

00:57:35.780 And you get more and more insight into disease process, or I guess below MACE,

00:57:40.400 you would go disease progression would actually be your next thing.

00:57:43.060 Right. That would be what stars did.

00:57:43.560 Yeah, yeah. Let's go from the macro to the micro.

00:57:46.820 Before we do that, you may disagree, but I think now would be a good time to talk about

00:57:50.600 the Mendelian randomization studies, unless you want to do progression first, or unless you want

00:57:55.320 to do the mechanism first.

00:57:56.840 Okay. Do you want to talk about Mendelian randomization through the lens of LDL?

00:58:00.460 Well, yes. Okay. So first off, I think it's important to point out why I'm bringing this

00:58:06.040 up now. Because when it comes to these studies, I think that these are the most powerful. Because

00:58:11.200 as you mentioned, mortality is a difficult outcome to get. It's a high bar to clear.

00:58:16.200 The benefit to Mendelian randomization is that essentially you have a lifelong randomized

00:58:21.000 control trial. Now, Mendelian randomization takes advantage of the fact that at birth, genetic

00:58:27.000 variants are randomly assigned. So in a research study, if you and I are in a research study,

00:58:31.960 Peter, the researchers are just going to, through whatever randomization process they

00:58:35.580 have, okay, Peter, you're doing this. Lane, you're doing this. Completely random. Mother

00:58:40.080 nature actually does that by design, which we now know identified, I think around a dozen

00:58:45.860 variants that will essentially change your LDL cholesterol levels. They're variants that deal

00:58:53.220 with LDL clearance, LDL production, how much cholesterol you absorb. There's all different

00:58:58.440 kinds of variants. Now, the benefit to this is you can look at someone's lifelong LDL cholesterol

00:59:04.680 exposure and attempt to see what is the risk of mortality and cardiovascular disease. That

00:59:12.080 is very powerful. And these are studies where some of these have hundreds of thousands of people

00:59:16.800 in them. And again, the randomization of this process by nature is so important because now we

00:59:25.480 can, if we see differences between groups, we can assume it's due to that assignment of that genetic

00:59:32.160 variant versus some kind of confounding variable. Now, there's a caveat here, which is for a Mendelian

00:59:40.140 randomization to be useful, a couple of things have to be true. What you said has to be true.

00:59:44.840 There has to be genetic assignment to the variable of interest. That's true for many things. It's true

00:59:51.360 for height. It's true for body composition. It's true for many psychiatric disorders. I mean,

00:59:57.420 there's lots of things for which genetics play a significant role. And LDL cholesterol turns out to be

01:00:04.060 one of them. But this is the other important thing that people often forget when talking about MR,

01:00:08.860 which is that those same genes cannot have a direct impact on another variable that affects

01:00:17.820 the outcome of interest. Because if they do, your randomization just got wonky.

01:00:23.520 For sure. And that's why they do tests for pleiotropy, which we talk about. And when I talk

01:00:28.880 about the results, I'm going to explain why it's very unlikely that these differences were due to

01:00:32.880 pleiotropy. But it does require us to make one assumption. And this may seem ticky-tacky,

01:00:38.220 but I'm trying to be logically consistent here. Which is, this is not a lifelong test of saturated

01:00:44.260 fat versus polyunsaturated fat. This is a lifelong test of what we expect to happen to LDL cholesterol

01:00:50.280 by substituting polyunsaturated fats for saturated fats.

01:00:53.720 Stated another way, this is a test of LDL causality.

01:00:58.060 Correct.

01:00:58.980 It's a leap to then make the statement about what does nutrition do to this. I'm going to give

01:01:04.520 you another example of this that is off the beaten path, but related to LDL, but unrelated to our

01:01:09.480 topic. And I believe I included this in my book, although it may have got left on the editing floor.

01:01:15.400 I know I wrote about it. I can't remember if it made the final cut. And it came down to the question

01:01:19.480 of, so this is me taking off my debate hat and just doing an aside on MR because it's interesting.

01:01:25.280 A handful of studies showed that people with lower cholesterol were more prone to cancer.

01:01:30.880 And so the concern became, well, gosh, we shouldn't be lowering people's cholesterol

01:01:34.420 in an effort to prevent heart disease if it increases the risk of cancer. And of course,

01:01:39.540 on average, that didn't seem to be the case, but there was always one study that might've

01:01:43.160 suggested that. And you never knew if there was a confounder because cancer, you weren't

01:01:46.620 randomizing to find that. So this is where MR became potentially valuable, provided you believe

01:01:52.040 that the genes that regulate cholesterol, synthesis, absorption, and LDL receptor expression don't

01:01:59.260 also regulate a cancer process. And if you believe that, then the Mendelian randomization was quite

01:02:05.380 clear that there was no relationship, neither good nor bad, between LDL, cholesterol, and cancer.

01:02:13.600 So that meant that LDL had no causal role one way or the other on cancer, whereas, as you'll probably

01:02:20.280 talk about it as a relationship towards ASCVD.

01:02:23.420 So that's a great explanation. And I think it took me a while to wrap my head around Mendelian

01:02:28.380 randomization when I first started reading about it. But these really were the studies that made me

01:02:32.320 change my mind on LDL. I want to point out the strengths to this. Lifelong exposure, which based

01:02:38.720 on the lipid hypothesis, we would expect to see a linear effect of lifelong exposure to LDL on the risk

01:02:45.480 of cardiovascular disease and mortality. And you can get a large number of subjects for their lifetime.

01:02:52.940 And it's randomized. So it's not a cohort study. Those are the benefits. The one downside is, again,

01:02:58.420 we're having to make a leap of, okay, we're not directly measuring people eating saturated fat versus

01:03:02.920 polyunsaturated fat. But I would say that this leap is pretty small because it is very well established

01:03:08.880 that increasing saturated fat intake increases LDL cholesterol and increasing PUFAs decrease LDL

01:03:16.140 cholesterol. And raising PUFAs and lowering saturated fat significantly decrease LDL cholesterol. I think

01:03:22.460 in most studies, you get like around a 15% change in LDL cholesterol from substituting in polyunsaturated

01:03:28.860 fats for saturated fat. Now, I will tell you, there's a paper from, I think it's Forenz in 2012.

01:03:35.520 It was the first big MR study with cholesterol. And if you look at the figure where they take

01:03:42.140 all the genetic variants and you look at how the line goes almost straight through it. I mean,

01:03:48.860 we're talking about an R of probably above 0.9, which in studies like this, that is an incredible

01:03:55.420 association for every one millimole reduction in LDL. Which is about 37 milligrams per deciliter.

01:04:02.940 It's 39.4 milligrams per deciliter. Yeah. There was a 50 to 55% risk reduction in cardiovascular

01:04:09.880 disease. Again, the pleiotropy argument is pretty much null and void because it didn't matter what

01:04:15.600 kind of variant it was. If it increased LDL clearance, if it decreased production, no matter

01:04:22.340 how it affected the metabolism of LDL cholesterol, it had the same dose effect. The only exception to that

01:04:30.520 was some of the CETP variants, which when it came to drug trials as well, CETP variants,

01:04:36.540 basically, I believe they raise HDL and they lower LDL. But, and here's the big point,

01:04:41.460 the lowering of LDL with those variants and those drugs, it lowers the cholesterol mass,

01:04:47.860 but there's a discordant lowering of ApoB or LDL particle number. So every single LDL particle

01:04:55.520 has an ApoB protein on it. And when we get into the mechanisms, this is going to be very central to

01:05:00.540 the lipid hypothesis. If you lower cholesterol mass, but your ApoB doesn't drop that much,

01:05:06.020 you basically have just made each particle smaller. And what we see is in that particular

01:05:12.060 subset of variants, there is a small risk reduction, but it's basically explained by the

01:05:17.460 small decrease in ApoB. You don't get the predicted risk reduction that you would get

01:05:22.460 with reducing LDL cholesterol. But amongst the variants that decrease LDL and correspondingly

01:05:28.960 decrease ApoB, it is a very, very consistent effect. Again, regardless of the genetic variant,

01:05:36.080 it has the same risk reduction. And then when we look at the statin trials, regardless of the way

01:05:41.400 that LDL cholesterol gets lowered, and there's different statins that work different ways,

01:05:45.940 it is the same dose response, which is about 22%.

01:05:48.460 Yeah. So I was going to ask you, why do you think that when we look at the totality of the

01:05:54.920 Mendelian randomization, and Tom Dayspring has a figure that lays every single MR study,

01:06:02.940 every single RCT study, and every single observational epidemiology study on the same

01:06:09.680 graph with three lines, basically the linear regression through each. And what do you think

01:06:16.000 is the best explanation for the following observation, which is when you do all of the MR

01:06:19.940 studies, and just again, so people, especially because most people are listening to us, not

01:06:24.220 watching us, so I don't want to use my hands too much, but on the x-axis, you have LDL concentration,

01:06:29.840 and on the y-axis, you have mortality or cardiovascular mortality. And as you pointed out,

01:06:35.400 these lines are just going straight down, meaning as LDL cholesterol is going down, either genetically

01:06:42.500 or through drugs, cardiovascular mortality is going down. But why is it that in the MR study,

01:06:49.680 which in theory would be the most pure study, every millimole or every roughly 40 milligrams

01:06:55.200 per deciliter reduction in LDL cholesterol is giving you 50 to 55% reduction. But when you do the same

01:07:02.160 thing with a statin, you still get a benefit, and that benefit appears non-ending, but it's only 22%.

01:07:08.500 Doesn't that tell us that statins are bad? Well, again, as you mentioned, there's still

01:07:12.820 a risk reduction. Yeah. Let me state it another way. Does that suggest that while statins are net

01:07:18.480 positive, they're doing something bad? So this is because think about when people are getting on

01:07:25.360 statins. So I say this as somebody, my bias should be that I hope that LDL cholesterol is not bad because

01:07:31.800 my whole mother's side of the family runs high LDL, even with low saturated fat diet, even with high

01:07:37.820 dietary fiber intake, I run about 150. I started taking a statin when I was 40 years old. But my

01:07:44.020 endothelium, my blood vessels have already had 40 years of LDL exposure at that level. And so even if

01:07:52.960 I get on a statin at age 40, every single day to the day I die, there is still some LDL cholesterol

01:07:59.780 that has penetrated that endothelium and has contributed to some degree of atherosclerosis.

01:08:06.980 I mean, I had a coronary calcium score done. It was pretty low. Overall, my net risk was very low for

01:08:12.240 my age because I have good insulin sensitivity, all those sorts of things. Important to point out as

01:08:16.260 well, when we're talking about LDL, we're not saying everything else doesn't matter. This is an

01:08:20.260 independent risk factor is what I'm saying. Insulin sensitivity, metabolic health, all those things

01:08:24.600 still matter. These are not mutually exclusive. But when you have a genetic variant that lowers LDL

01:08:31.080 from the day you are born, your entire circulatory system is exposed to less LDL cholesterol. And when

01:08:38.400 it comes to how much gets into the intima, it is concentration dependent. And it's basically dependent

01:08:46.200 on the number of particles in your bloodstream with ApoB that are under 70 nanometers. Because

01:08:54.900 any lipoprotein under 70 nanometers in diameter with an ApoB can penetrate the endothelium and get

01:09:02.500 retained there. So when it comes to these Mendelian studies, the reason you see such a powerful effect,

01:09:08.260 it's just a time effect. Because most people, when are they getting, I don't know what the average age

01:09:11.600 is somebody's prescribed a statin, but I'm imagining it's probably around my age. Yeah.

01:09:14.800 Yeah. So you're really just looking at the difference in like that investment analogy we

01:09:20.000 used earlier, 40 years, you'll see a big difference. But if you start investing from the day you were

01:09:24.620 born in 70 years, you're going to see a massive difference due to compounding interest. And while

01:09:30.680 it may be kind of a rudimentary comparison, these problems compound over time. And so what you're doing

01:09:36.520 with somebody who has high LDL, put them on a statin, you're obviously like pumping the brakes,

01:09:42.420 but that truck still started rolling down the hill. Whereas with people who have lifelong low LDL,

01:09:47.920 it never really got started.

01:09:49.200 Yeah.

01:09:49.960 I want to point out one more thing. Important to point out, there was a regression. They looked at each

01:09:54.020 one millimole reduction. So regardless of the variant, same reduction. And in the drug trials,

01:10:01.760 the statin trials, regardless of how the statins worked. Also with surgeries, like I think it was

01:10:09.200 a little bypass or things that reduce LDL through absorption.

01:10:13.600 No, I think different drugs that are independent of the mechanism by which LDL is lowered.

01:10:19.140 Yeah.

01:10:19.540 You'll see the same benefit.

01:10:20.800 And then also with dietary reduction of it in terms of the risk when they look at some of these cohort

01:10:25.900 studies. If you want to argue it's pleiotropy or you want to argue it's possibly something

01:10:31.460 else, it's a really hard argument to make when it is a very consistent effect and the dose is also

01:10:39.080 very consistent. When we talk about converging lines of evidence.

01:10:43.240 There's an argument to be made, which is not necessarily the argument we're having today.

01:10:46.460 I won't even attempt to steel man it. I'll just state it. The argument is that if you look at all

01:10:51.220 the literature of statins and you see reduction in mortality, that doesn't mean that it's because

01:10:57.380 it's lowering LDL. It could be because it's doing something else. It's lowering inflammation

01:11:01.700 or it's doing something else. And you're arguing that that's a tough argument to make in light of

01:11:07.460 all of the MR coupled with all the clinical trial data.

01:11:10.980 If that was the case, you would see a difference in the dose response. You would see inconsistencies

01:11:17.000 in the trials with similar designs. I'll give a comparison that's kind of out of left field,

01:11:21.880 but maybe it'll make the point. And that is, for example, I don't believe that unprocessed red meat

01:11:28.240 specifically is inherently carcinogenic. And the reason is, even though you see it come up as

01:11:36.140 carcinogenic in some of these cohort studies, the effect isn't always consistent. And when they

01:11:41.700 control for overall diet quality, where people are eating enough fruits and vegetables, because again,

01:11:45.900 if people eat more of one thing, they tend to eat less of another. When you control for some of the

01:11:50.160 overall diet quality variables, you don't really see a consistent associated of red meat with cancer.

01:11:56.180 Now, I could be wrong, but I'm just not convinced by it. But when it comes to dietary fiber's effect

01:12:01.960 on cardiovascular disease and cancer, there's a dose response and it is very consistent in the

01:12:08.700 research literature. In fact, I'm not really aware of hardly any study looking at dietary fiber

01:12:14.860 and reducing the risk of cardiovascular disease, cancer, and mortality that doesn't show a benefit.

01:12:20.160 If there's a forest plot of all the studies out there, everything is going to be the protection

01:12:24.880 side. When you have that consistency, even though you could argue, well, it could be other things,

01:12:30.140 it could be other things. I guess if you want to make the whataboutism argument, it's hard for us

01:12:35.380 to ever actually say something causes something else. I mean, maybe it's not the smoking that causes

01:12:41.740 lung cancer because we can't really do randomized control trials on smoking because it wouldn't be

01:12:48.680 ethical, but we feel very strongly because of the dose effect and because of the consistency of the

01:12:53.280 results. So I get that argument that can be made, but I guess I would say, well, then what do you

01:13:01.840 think it's doing? What would explain this very consistent effect? It typically just ends up being

01:13:07.120 an argument of, well, you don't know for sure. And it reminds me of in graduate school, I was giving a

01:13:12.480 talk on leucine content of dietary protein sources, and we did a dose response experiment

01:13:17.800 with different dietary protein sources that varied in the leucine content, and pretty much showed almost

01:13:23.440 a perfect association between the amount of leucine in those protein sources, the ability of those

01:13:29.100 protein sources to increase plasma leucine, and the effect on protein synthesis. And I had people

01:13:34.140 that were other scientists in the audience say, well, but you can't say that. These other protein

01:13:37.880 sources, they have different other amino acids in them. It's not just leucine. There's other things

01:13:41.860 that are changing. And I said, okay, do you have anything else that would explain this? It was kind

01:13:47.840 of like they didn't really have much to say. And so yes, I grant that it's possible. It's just highly

01:13:53.700 improbable based on the data. Okay. So what about the idea though, that if you're eating a diet that's

01:14:00.260 high in polyunsaturated fats or seed oils to be specific, we acknowledge now you're getting a lot of

01:14:05.940 linoleic acid. Well, you now have LDL particles. Maybe you have not only fewer of them, but they

01:14:14.640 have more linoleic acid in them. And linoleic acid can easily be converted to arachidonic acid,

01:14:22.480 which is inflammatory. And we know that the single most important part of atherosclerosis is indeed the

01:14:29.320 oxidative inflammatory process. In fact, people don't die because their coronary arteries just

01:14:35.840 slowly get occluded. They die because the body in an effort to repair and respond to the oxidative

01:14:44.680 damage in the artery walls creates an immune response. So inflammation here is the game.

01:14:52.240 So are you not concerned with the fact that a diet that is high in linoleic acid, which is the

01:14:58.340 precursor to arachidonic acid, is going to lead to more inflammation, more oxidative LDL, and therefore

01:15:07.500 ultimately more atherosclerosis, even if you see lower LDL cholesterol? So this is going to be the

01:15:14.420 really fun part of this talk because I learned so much about this through researching this stuff.

01:15:20.600 Let's just take the broad 10,000-foot view first, and then we'll zoom in on the mechanisms and talk

01:15:27.160 about the lipid hypothesis because it's important to understand what it is and how this disease

01:15:32.900 progresses so that then we can unpack all these side quests. And I'll take what you said a step

01:15:38.780 further. Precursor to arachidonic acid, which is a precursor to prostaglandin production, which are

01:15:43.760 inflammatory. Also, and you did briefly mention this, polyunsaturated fats, much more prone to

01:15:49.300 oxidation than saturated fat. And we do know oxidized LDL is more atherogenic on a per-particle

01:15:56.100 basis. And people who have MIs, people who have cardiovascular disease, they have higher levels of

01:16:02.900 oxidative LDL. So really, when we nail it down, I believe that's one of their big core arguments is

01:16:09.100 when you substitute polyunsaturated fats for saturated fats, you are creating an unstable

01:16:15.480 lipoprotein, more prone to oxidation, and that is what is going to cause this disease to really

01:16:21.920 progress. Let's unpack this a little bit. From a 10,000-foot view, if that were true, what we would

01:16:28.460 expect to see is people who eat more linoleic acid have higher rates of heart disease. And what we see

01:16:35.460 is the opposite. I think there was a study that came out, a cohort study looking at like,

01:16:39.800 I think it was over a million people from various different countries showing that people who had

01:16:46.180 higher levels of linoleic acid intake had lower levels of cardiovascular disease.

01:16:52.240 What was that a substitute for? This was a free living study, right?

01:16:55.340 This cohort. So they're just looking at how much do linoleic acid do people eat?

01:16:59.660 And what was the primary source of the linoleic acid?

01:17:02.020 I'm not sure. I think they just looked at overall dietary linoleic acid

01:17:05.100 If I really dug back into the paper, they might list some of the primary sources.

01:17:09.500 But of course, these people are probably substituting linoleic acid because they're

01:17:14.320 healthier people to begin with.

01:17:16.060 Healthy user bias. What I would say when it comes to healthy user bias, what you typically see is like

01:17:22.800 no effect of something that should be bad or it'll be inconsistent in the research literature.

01:17:28.620 But it's hard to argue converging lines of evidence. If your position is that CEDOLs are uniquely

01:17:34.560 deleterious, it's hard to argue converging lines of evidence when one of the major things you really

01:17:40.300 should see is if people eat more linoleic acid, the effect should be so powerful. If they're the

01:17:44.800 primary driver of cardiovascular disease, which is what some of these people claim, that effect should

01:17:48.800 be powerful enough that even if they were doing other healthy behaviors, that you should still see

01:17:55.700 something and certainly not a protective effect. To take it one step further, because dietary recall studies

01:18:01.200 are problematic. Anybody who's ever done some of these knows. I mean, I don't even remember what I

01:18:05.820 ate yesterday. But one of the great things about the fatty acids you eat, the essential fatty acids

01:18:11.200 you eat, is if you eat more of them, it shows in your tissues. If you take an adipose tissue sample,

01:18:16.600 if you take a blood sample, you will see more of that essential fatty acid incorporated into your

01:18:23.360 lipid, your phospholipid bilayer of your cells.

01:18:26.500 And indeed, in studies where they look at tissue amounts of linoleic acid, they see the same thing,

01:18:33.600 a reduction in risk of cardiovascular disease. So this is from a 10,000-foot view. To me,

01:18:38.800 that's a pretty damning thing right off the bat. And what I would say is, if you're going to argue

01:18:43.320 that polyunsaturated fats are bad for you, you're going to argue that saturated fat is really bad for

01:18:47.840 you. Because if, again, I think that's where logical consistency is important. If the data existed

01:18:54.180 showing people who ate more saturated fat had lower rates of cardiovascular disease,

01:18:59.360 if you even thought about talking about saturated fat being bad, the people in the anti-seed oil camp

01:19:05.220 or carnivore camp would lose their minds. And so it's kind of like this picking and choosing of

01:19:10.160 what data I want to talk about that fits. So we don't see that. We also don't see that when people

01:19:17.040 eat more linoleic acid, they don't produce more economic acid. That conversion apparently is already

01:19:22.720 kind of at saturation. Just eating more linoleic acid doesn't have a feed-forward effect.

01:19:28.720 You're not actually getting more arachidonic acid production. So I think the prostaglandin pathway

01:19:33.980 is not something we really need to be too concerned with. Because again, we'd expect to see increasing

01:19:39.240 amounts of arachidonic acid. Now this is where I think the oxidized LDL is the argument that I

01:19:45.220 struggled with the most before I really dug into this. So I'm going to talk about the lipid hypothesis,

01:19:49.480 and then I'm going to also talk about why it's so important to understand where LDL gets oxidized.

01:19:55.480 So the lipid hypothesis basically states that really any non-HDL cholesterol that is under 70

01:20:02.800 nanometers in diameter, which VLDLs can fall into that, LDL obviously falls into that, IDLs can fall

01:20:09.640 into that, depending on the IDL. But basically any lipoprotein that contains an ApoB molecule,

01:20:17.440 or an ApoB protein, that the lipoprotein is under 70 nanometers in diameter, can penetrate the

01:20:23.120 endothelium. Now just penetrating the endothelium and getting into the intima, and this is concentration

01:20:27.840 dependent, and this has been very well established in the mechanistic literature, but just penetrating

01:20:32.580 the endothelium is not enough to cause progression of cardiovascular disease because those molecules

01:20:39.360 can come back out into the bloodstream. What can happen is that ApoB molecule, or that ApoB

01:20:45.820 protein, can be enzymatically modified into a proteoglycan. Basically enzymes inside the intima

01:20:55.580 can act on that ApoB. That enzymatic modification causes that LDL or VLDL or whatever particle it is

01:21:04.240 to be retained inside the intima. Once retained, that causes an oxidation. Oxidation can increase

01:21:11.760 on those lipoproteins. That can attract macrophages, as you pointed out, the immune response. Those

01:21:19.860 macrophages begin to engulf some of these. These LDL particles can start to clump together, or VLDL

01:21:27.080 particles or whatever particle. They start to clump together in a process called aggregation.

01:21:31.840 Macrophages infiltrate inflammation, trying to, again, repair this. But those macrophages,

01:21:38.580 then engulfing them, this produces foam cells. And over time, you also get more, I think the smooth

01:21:46.680 muscle starts to thicken as well. And this is eventually what's leading to kind of this closing

01:21:51.780 of the blood vessel. Now, oxidation is part of this process. One of the core components that you

01:21:58.700 mentioned of this anti-seed oil hypothesis is that, okay, if you have lipoproteins that have more

01:22:03.440 polyunsaturated fats, they are more prone to oxidation. That is true. But we need to understand

01:22:08.640 where oxidation is occurring. I think anti-seed oil people would have you believe that the

01:22:13.760 oxidation occurs in the plasma, and that those oxidized LDLs in the plasma, those penetrate the

01:22:19.720 endothelium, and that causes the progression of cardiovascular disease. Less than 1% of LDL

01:22:25.380 is oxidized in the plasma. Because LDL is mostly cleared pretty quickly from the plasma. It's like

01:22:32.300 an hour or two. It's on the time course of hours. Once ApoB is enzymatically modified, that LDL,

01:22:38.580 that can be retained for weeks. And in your plasma, you have antioxidants, vitamin E, vitamin C,

01:22:46.400 beta carotene. Those stabilize those polyunsaturated fats, and you don't really have that much

01:22:52.640 oxidation occurring in the plasma. But in the microenvironment of once it penetrates the

01:22:58.440 endothelium, gets inside the intima, in that microenvironment, it's thought that those antioxidants

01:23:04.620 are not as available. And so the oxidation can begin to occur there.

01:23:10.380 So what's the proof? So you're saying that because plasma ox LDL concentration is such a small fraction

01:23:18.520 of total LDL concentration, say 1%, that means that we're not getting a lot. But it could be a lot

01:23:25.080 if that 1% disproportionately aggregates inside the subendothelial space. I mean, you don't need a lot

01:23:31.600 of LDL particles to cross the endothelial barrier, right?

01:23:35.100 I'm glad you brought up aggregation because that's important here. So remember when I said there can be

01:23:39.740 positive and negative aspects that you activate for any nutrient you're talking about. So aggregation,

01:23:46.800 once you have LDL inside the intima and you have this oxidation occurring, you have these things

01:23:52.500 occurring, aggregation is how these cells clump together. Lipoproteins that are enriched with

01:23:58.320 polyunsaturated fats per particle, they are more prone to oxidation. Yes, inside the intima. But keep

01:24:05.520 in mind, what gets into the intima is concentration dependent. Polyunsaturated fats overall lower the

01:24:11.680 amount of LDL getting into the intima. So you have less getting in, less being accessible for

01:24:18.800 oxidation since it occurs there mostly, not the plasma. But a bigger point is there's other aspects

01:24:25.480 of these lipoproteins that make aggregation happen. So when we talked about polyunsaturated fats

01:24:33.400 increase membrane fluidity, one, that actually helps with LDL receptor recognition. It helps LDL get

01:24:39.560 cleared so you have less in the bloodstream. But two, the ApoB molecule itself is less prone to

01:24:46.940 enzymatic modification on LDLs that are enriched with polyunsaturated fats compared to saturated fat.

01:24:53.320 Further, LDL molecules that are enriched with saturated fat, their membranes are stiffer and more

01:25:00.300 rigid because of the packing that we talked about. Whereas those enriched with polyunsaturated fats

01:25:05.940 are less rigid, they're more fluid. And that has a big impact on aggregation. There's an enzyme called

01:25:14.060 sphingomyelinase, which when it acts on a saturated fat enriched LDL molecule inside the intima,

01:25:21.580 it rapidly produces ceramides. And those ceramides actually, for lack of a better term, can collapse

01:25:28.420 these particles and cause them to clump together much more readily. So all that to say, oxidation is

01:25:36.220 part of the process of aggregation, but how much those aggregate is a more important factor than

01:25:42.680 oxidation of PUFAs. Because the oxidation is bad because it increases the aggregation of these

01:25:49.560 molecules. So the overall effect is, okay, polyunsaturated fats decrease the number of particles

01:25:57.040 that are getting into the intima. They also overall decrease them being retained there,

01:26:03.260 the ApoB is less prone to modification, and they are less prone to aggregation if they are retained

01:26:08.680 there compared to lipoproteins that are enriched with saturated fat. So think of it this way. I really

01:26:15.220 spent a lot of time trying to come up with an analogy for this because I realized a lot of people who are

01:26:20.120 listening to this, their heads are probably spinning because mine was spinning when I was reading about

01:26:23.380 this at first. Think about the LDL cholesterol in your bloodstream, or let's just say ApoB containing

01:26:28.300 particles in your bloodstream, being a bonfire. And there's a whole forest around it. Now the forest

01:26:33.200 around it is your blood vessels. And if you start a forest fire, that's cardiovascular disease.

01:26:39.080 Now, bonfires, they give off sparks. Let's say each spark is an LDL particle. You don't want the forest

01:26:45.720 to catch on fire. If you have polyunsaturated fat enriched fatty acids, maybe each individual

01:26:52.700 particle is a little bit more flammable, right? Or a little bit more prone to oxidation. But when you

01:26:59.780 eat high polyunsaturated fats versus saturated fat, your bonfire shrinks quite a bit. The amount of LDL

01:27:06.580 in your bloodstream shrinks quite a bit. You give off way less sparks, way less sparks hit the forest. And oh,

01:27:12.420 by the way, if I was being actually accurate, some of those sparks are much more likely to bounce back

01:27:17.160 into the fire compared to staying in the forest where they can start a fire. Also, these particles

01:27:23.280 tend to, even if they get into the forest, these sparks, they tend to not clump up and be prone to

01:27:30.700 causing a forest fire. They tend to scatter. That's polyunsaturated fat enriched lipoproteins. So even

01:27:37.260 though on an individual level, the sparks may be a little bit more flammable, the bonfire for saturated

01:27:44.240 fat is way bigger. It casts off way more sparks. And those sparks are more likely to clump together

01:27:51.760 and start a fire compared to the fire from polyunsaturated fats. That's about as good of

01:27:57.240 an analogy as I could come up with.

01:27:59.060 And you're rejecting my idea that even though only a small fraction of LDLs are being oxidized in the

01:28:08.560 periphery, that those ones don't disproportionately concentrate in their ability to either make their

01:28:15.940 way into the endothelial or subendothelial space and get retained. I feel like we're potentially

01:28:21.040 overlooking that as a potential driver, right? Because LDLs can traffic in and out of the subendothelial

01:28:25.760 space. So the question then becomes, what are the factors that would increase retention,

01:28:32.300 adhesion, oxidation, and then the cascading effects? And do we not believe that an oxidized LDL versus a

01:28:41.800 non-oxidized LDL would be more atherogenic?

01:28:45.680 On a per-particle basis, yes. An oxidized LDL is more atherogenic.

01:28:50.900 In the periphery.

01:28:51.720 So you're saying like in the bloodstream.

01:28:53.380 Yes. I know it is inside, but I want to make sure we would agree that potentially it would

01:28:58.340 also be more atherogenic outside because it has a greater probability of becoming retained

01:29:04.080 and remaining oxidized and inciting the inflammatory response.

01:29:08.500 It's just such a small amount that gets oxidized.

01:29:11.400 Has anyone looked at a study where, you know, asked the question if you are on a high polyunsaturated

01:29:18.180 or seed oil diet versus a high saturated fat diet and you normalize for total LDL, which

01:29:23.540 obviously will be quite different, do you have an equal percentage of oxidized LDL in

01:29:28.800 the periphery?

01:29:29.640 Say it again one more time.

01:29:30.960 Let's just assume you put somebody on a high saturated fat diet, somebody on a high seed

01:29:34.660 oil diet. And let's assume that the PUFA person, the seed oil person's got an LDL cholesterol

01:29:39.300 of a hundred milligrams per deciliter and the high saturated fat diet person is going

01:29:44.180 to be at 200 milligrams per deciliter. And let's just assume that that's concordant with

01:29:47.540 APO-B. So same number of particles, 2X the particles, 2X the concentration, I mean.

01:29:52.240 Do we expect there to be the same ratio or delta or fraction that's oxidized? Or do we think

01:30:00.120 that the person on the high seed oil is going to have disproportionately more ox LDL in the periphery

01:30:06.180 where you can measure it? And therefore is likely, even though their gradient is less

01:30:11.480 favorable, they might get more particles in there.

01:30:14.700 There was one randomized control trial. I think they fed soybean oil and saw oxidized LDL in

01:30:19.900 the periphery go up. But again, you're talking about like increasing from like, let's say,

01:30:24.860 and I'm making the numbers up, but it's on the order of, okay, normally maybe 0.5% is oxidized

01:30:30.060 and now it's 0.7, 0.8. That is such a small number compared to once a particle gets inside

01:30:37.400 the intima, the rate of oxidation can, I think the estimates I saw were anywhere from 30 to 80%

01:30:42.520 of those particles. And in fact, I actually had a long conversation with Tom Dayspring about this,

01:30:47.600 trying to understand it. Because if oxidized LDL in the periphery was really driving cardiovascular

01:30:53.900 disease, why have the studies where they give a bunch of antioxidants not decreased cardiovascular

01:30:59.160 disease? Because when we do that in animals, so they've actually done these studies, I think it's

01:31:04.100 in rabbits, where they'll put oxidized LDL into their bloodstream and they will see atherogenesis

01:31:11.220 progress. But when they do it with antioxidants, that doesn't happen. And that is because where most

01:31:17.360 of that oxidation is occurring is inside the intima. So your biggest lever to actually reduce

01:31:24.120 your overall amount of oxidized LDL is just to prevent as much getting into the intima and retained

01:31:30.920 as possible. And then when we look at people who have higher levels of oxidized LDL, it's typically

01:31:37.500 a downstream effect of how much LDL got into their intima in the first place. Because even after it gets

01:31:44.560 into the intima, retained for a while, oxidized, some of those oxidized molecules can still come

01:31:48.680 back out into the bloodstream. And so, yes, we do see people with greater amounts of cardiovascular

01:31:53.360 disease. Do we know if the oxidized LDL that we measure in the periphery was oxidized in the

01:32:00.560 periphery or is escaped LDL that was oxidized in the endothelial space? I don't know the exact answer

01:32:07.600 to that question because that would be difficult. You'd have to do some sort of metabolic tracer study.

01:32:12.120 Maybe you'd have to track it for a really long time, at least weeks to months to see if that

01:32:17.960 happens. I don't know for sure. I will attempt to answer the question as best I can. In people who

01:32:23.960 undergo myocardial infarctions, so where you have this kind of rupturing, they do see short-term

01:32:29.420 oxidized LDL go way up, coming out of the, presumably that's because it's coming out of the

01:32:34.920 intima since there's that rupture. But have they ever done a study to like kind of link it together

01:32:40.680 with like a stable isotope? I'm not sure about that. But again, I think the important point

01:32:46.560 is that the less ApoB that gets retained inside the intima, the less chance there is for overall

01:32:54.220 oxidation to occur. And really, aggregation is the endpoint that's much more important.

01:33:00.900 Oxidation is only bad, quote-unquote, because it's more prone to aggregate. But we know on balance

01:33:06.960 that saturated fat-enriched particles are more likely to aggregate than polyunsaturated-enriched

01:33:14.940 particles due to the differences in membrane fluidity, as well as the ability for ApoB to be

01:33:20.440 modified, and because of the sphingomyelin content and ceramide content of the saturated fat-enriched

01:33:26.840 molecules. So again, I would say, let's say I grant that those polyunsaturated fats per particle

01:33:32.820 more prone to oxidation. You're still having to weigh it against the other things that progress

01:33:37.840 cardiovascular disease. And on balance, you're still better off with the polyunsaturated fats

01:33:42.940 because they do lower the amount that gets into the intima. They lower the amount that gets retained

01:33:47.780 because it's less likely that ApoB will get enzymatically modified. And then those saturated

01:33:53.640 enriched particles, because they're rigid, because they produce ceramides, they're more likely to clump

01:33:57.500 together and cause that fatty streak and that lesion. Okay. So let's consider something else

01:34:02.960 though, which is for me to get a bottle of corn oil or any of the other seed oils on your table,

01:34:11.120 I have to do a lot of industrial processing. I have to heat these things up. I have to refine these oils.

01:34:18.380 I have to use industrial-grade solvents to extract them. It seems very likely that both of those

01:34:27.400 processes can contribute to the negative impact of them, independent of what we might see if we were

01:34:34.800 talking about something pure. In other words, everything we've talked about so far is assuming

01:34:39.020 a pure form of linoleic acid. But what if I'm now saying, yeah, but I'm going to heat, reheat, cool,

01:34:47.580 bastardize this molecule. And oh, by the way, I'm not going to be able to get all the hexane off this

01:34:52.220 molecule. And I needed to use hexane to extract it. We don't like to talk about it,

01:34:56.920 but food processing is big industrial chemistry. And what I would say is the actual processing

01:35:02.460 of the seed oils removes oxidants and removes some impurities that are maybe negative. There

01:35:10.540 are some things that do increase. We'll talk about that. But let's start with the hexane itself. So

01:35:16.060 to get the oils out of these seeds, you need to either do mechanical or chemical extraction.

01:35:21.400 Now, I think most people would say, well, I'd rather have the mechanical extraction, right? Because

01:35:25.320 less chemicals, but it is much more costly. The yield is lower and economics is a thing.

01:35:32.020 Is that an opportunity? Can you go into a grocery store and choose

01:35:35.240 to have safflower oil that was mechanically extracted versus chemically extracted?

01:35:40.720 I actually have no clue, but I would imagine there are probably places that do sell it.

01:35:43.960 Pay more for it?

01:35:44.780 For sure. For sure. So let's talk about why hexane is used. So they take these seeds,

01:35:48.460 they wash them with hexane. Why hexane? Well, hexane is a nonpolar solvent. And when you're

01:35:55.320 dealing with oil, polar solvents are much more popular because most things or most things that

01:36:00.940 we try to get are polar. Most things like to interact with water. It makes sense based on

01:36:04.540 our biology and our biochemistry. Oils are different. Oils you have to do very unique things to.

01:36:09.440 Hexane is a nonpolar solvent. So it will mix with these oils and it has a relatively low boiling

01:36:15.800 point. So you can evaporate it off. These seeds get washed with this hexane. It extracts the crude

01:36:23.060 oil. So now you've got the oil mixed with hexane. Well, now they bubble steam vapor through the oil

01:36:30.180 and that evaporates off the hexane. Now I will tell you that the steam and the temperature is

01:36:36.160 pretty low. In order to really start getting oxidation of seed oils, it depends on the oil

01:36:42.920 specifically. But most of them, you've got to be well over 200 degrees Celsius and you've got to

01:36:47.800 do it for hours. If we're talking about in like a large vat, I think I read like soybean oil, if you

01:36:53.520 heat it at like 240 degrees Celsius for like three hours, you will start to get a percent of the oil

01:37:01.280 being oxidized. But even after like five hours, it's still pretty small percentage points of

01:37:06.920 oxidation. And this process of removing the hexane is on the order of minutes or an hour,

01:37:12.380 90 minutes. Like it's a pretty short period of time. And hexane's boiling point is, I believe it's

01:37:17.820 69 degrees Celsius. So you only got to heat it up to a point a little bit above that to start getting

01:37:22.880 it off. Now, okay, can you get all of it off? Well, as anybody who's had basic chemistry, you know

01:37:28.440 that no compound you synthesize is 100% pure. I mean, you can get 99.999%, but you always have

01:37:35.800 residual atoms in there. You always have residual molecules in there. So the question becomes,

01:37:40.360 all right, how much hexane is in the end product and how much is required to cause harm? The hexane

01:37:46.160 in the end product, most of them are well under one part per million. In fact, a lot of them have

01:37:52.180 non-detectable levels of hexane, which means there's probably some in there, but the instruments

01:37:56.640 we have to measure it simply aren't sensitive enough to pick that out. So the amount of hexane in

01:38:03.300 these oils, I believe the research paper I read was anywhere from 0.05 to 0.5 parts per million for

01:38:11.640 most of these oils. Hexane specifically, the danger with hexane is not from ingestion. It's

01:38:18.060 actually from inhalation. So people who have had toxicity from hexane, it's from inhaling it.

01:38:24.480 When you actually look at how much hexane you'd have to get, I tried to look up hexane poisoning

01:38:29.380 cases where somebody died. It doesn't exist. There's a case where a guy drank, like literally

01:38:35.340 drank straight hexane and basically got a tummy ache. They've done rodent studies where they were

01:38:40.780 able to get toxicity and death, but basically just to get mild liver and neurotoxicity, it was 5,000

01:38:49.980 milligrams per kilogram of body weight. Now, when we do human equivalent dosage, that dosage becomes

01:38:57.100 smaller. But let me just put it in perspective as a bottom line. I did the calculation on this.

01:39:02.460 What you would need to consume from hexane to even have mild side effects, what you would need to

01:39:09.360 consume is 11,340 kilograms of oil at one time. Okay, but that's to die. No, that was for mild side

01:39:18.120 effects. Okay, but how do we know that the accumulation of hexane or some other industrial

01:39:24.020 solvent couldn't be leading to a chronic process? We've just talked about how all the diseases we

01:39:29.980 care about, whether it be neurodegenerative diseases or cancer or cardiovascular disease,

01:39:36.280 these things don't happen overnight. They don't happen in weeks, months, even years. Many times

01:39:40.720 they happen in decades. And so if we're talking about a lifetime exposure to these things, how do

01:39:44.800 we know that that's not increasing our risk? When we talk about lifetime exposure from something

01:39:49.180 like LDL, that's a relatively high concentration in our bloodstream. And it's always present. You

01:39:55.160 always have a baseline level of LDL. You don't really have baseline levels of hexane in your

01:39:58.920 bloodstream, I don't think, at least not to any appreciable level. And there is a process through

01:40:04.720 your body where your body converts this to something innocuous and gets rid of it. So really,

01:40:09.520 when it comes to things that don't, what we call bioaccumulate, the question is, if we have some of

01:40:15.100 this, is it in an amount that can be cleared quickly enough to where there's not negative

01:40:19.120 outcomes? And what I would say is, okay, the example I gave was the amount of oil you'd need

01:40:23.880 to consume to possibly get mild side effects. If anybody wants to, okay, say, let's just say your

01:40:30.880 body couldn't process this out. Who's drinking 11,000 kilograms of oil in their lifetime? I think

01:40:36.920 probably almost no one. So I just don't see the possibility for hexane having a negative outcome

01:40:44.180 for people, especially when you consider that it's a very, very low concentration. It doesn't

01:40:49.760 bioaccumulate and your body has a way to process it out. And the amounts that you get are incredibly

01:40:55.060 small from these seed oils. Okay. Now let's consider the fact that about a hundred years ago,

01:41:01.880 less than 3% of total food availability was made up of linoleic acid. Today, that number is,

01:41:09.560 I mean, it's probably closer to 10%. What I read, because I actually read a book that was

01:41:15.660 anti-seed oil because I was trying to understand the arguments. The figure they quoted was a 75X

01:41:21.440 increase in linoleic acid over the last like 150 years or so. Okay. That's even more than what I've

01:41:27.420 got. So that's an enormous increase. Huge. To make it even more stark, this means that we did not

01:41:33.820 evolve in an environment where people were consuming seed oils in much quantity at all.

01:41:39.000 And yet today, people are probably getting 10 to 15% of their total calories from these things,

01:41:45.500 right? Some people do. Yeah, for sure. And tissue levels are up more than a hundred percent.

01:41:51.380 Yep. Isn't there just sort of a first principles argument to be made here that says,

01:41:56.180 how would that be a good thing? How would that be anything but negative?

01:42:01.200 Yeah. And this kind of gets back to the, we're starting to tie into the naturalistic argument.

01:42:06.480 So what I will say is, again, you have to be logically symmetrical with how you approach these

01:42:10.380 things. And even people who think they eat natural these days, the human diet now is not in any way,

01:42:16.980 shape, or form what it used to be. So people will point out, well, look at how these vegetables and

01:42:21.480 fruits and plants have been modified. Yeah, but we've done the exact same thing to our animals.

01:42:25.360 If you think having a fatty ribeye is a ancestral diet, it's not. Those cows are much different

01:42:34.020 than they used to be. And it's not wild game. These are very different things. Take that out

01:42:40.120 of it for a moment though. Try to really zoom out and think about what is the purpose of biology?

01:42:45.900 The purpose of biology is to pass on your genetic material. So when it comes to survival and longevity,

01:42:52.720 and I'm sure you're very well familiar with this, there's a reason things start to kind of go

01:42:56.660 downhill after like age 40 or whatever. It's because you're past breeding age. Evolution's

01:43:01.380 done with you. You've done what you can do. You've passed on your genetic material. Hopefully you can

01:43:05.080 stay around a little bit longer to raise the next generation, but you've done your job.

01:43:08.920 The idea of longevity, living a very long life, that's not really something that's essential to a

01:43:16.820 species surviving or even thriving. Some of the most prevalent species on the planet don't live

01:43:23.660 very long. What matters is that they get to pass on their genetic material, that the favorable traits

01:43:28.540 are passed on in the genetic material. When it comes to cardiovascular disease, yes, rates of

01:43:34.720 cardiovascular disease have gone up because you actually have the chance to get cardiovascular disease

01:43:38.980 now because you're not killed by a virus or you're not killed by a warring tribe or you're not killed

01:43:44.840 by bacteria. Even if we go back into the 1860s, this book I read, one of the things that said was

01:43:52.320 cardiovascular disease is a 20th century disease. Didn't exist before that. No, it existed. People

01:43:57.700 just fell over dead and nobody knew why. And for the most part, people didn't have much chance to get

01:44:04.620 cardiovascular disease. I think you, forgive me if I'm wrong, but you stated basically if you live long

01:44:10.260 enough, everyone at some point will get some form of cardiovascular disease. It's just a time and

01:44:15.800 exposure issue as we talked about. This massive increase in cardiovascular disease, which, oh, by the

01:44:20.800 way, I point out that everyone lives longer now. I don't know if it dipped recently, but I think we're

01:44:24.720 still right around like the longest age lived on average. The point being, even in the 1860s,

01:44:30.980 if you live past age 10, I think your average life expectancy was still something like 55 to 60

01:44:38.160 years old. So most people didn't have the chance to get cardiovascular disease or they died from

01:44:43.640 something else. And again, I mean, this is back when we used to bleed people to try and treat them, get rid of

01:44:49.400 the toxic blood. We didn't have the tools to understand what we were looking at. This is one of those arguments

01:44:54.480 that people, when they say, well, everybody's sicker now, we're more unhealthy now. That's true. That is true.

01:45:00.460 But part of that is we just have had the chance to get unhealthy. And evolutionarily, I will also say

01:45:07.720 we're taking one step out like we do with the MR studies. But LDL cholesterol, high LDL cholesterol

01:45:13.020 is not ancestral. If we look at the best estimate we have of what our ancestors did, which is the

01:45:19.240 Hadza, which are basically a tribe that are essentially untouched by humanity. We have studied

01:45:24.780 them quite a bit because this is our best guess as to what ancestral was. The foods they eat,

01:45:32.040 it's about as untouched as you can get. If you look at the LDL of the Hadza, who have very low rates of

01:45:38.480 almost non-existent rates of cardiovascular disease, they're 50 to 70 on average. I think they even had

01:45:45.460 one Hadza who was like under 30. I think they only found one Hadza individual who is over 100 LDL.

01:45:53.640 I would argue, again, we're not talking about linoleic acid, obviously, but we do know that

01:45:57.700 linoleic acid lowers LDL. I would argue that what the anti-seed oil people would suggest should be an

01:46:06.120 ancestral diet is not supported by the evidence either. And so regardless, there's a reason naturalism

01:46:12.760 has its own fallacy associated with it. Because you can find a lot of things in nature that are

01:46:17.620 horrific toxins. You can find a lot of synthetic things that are quite good for you. And so I think

01:46:24.440 we tend to romanticize the past on every single level, right? Like we romanticize past relationships,

01:46:31.240 we romanticize like 50 years ago, things were so much better, there was less crime. And then when you

01:46:35.420 actually go pull up the FBI statistics, it's not even close. Crime is way lower now. But we romanticize

01:46:41.080 the past. And we romanticize how things used to be. And what I would say is that I don't think what

01:46:47.180 we think might be natural is necessarily a good barometer for what is conducive to living the

01:46:53.940 longest, healthiest life. I think you're kind of getting the order reversed. Mankind evolved in this

01:47:00.840 environment where I think one of the reasons we were able to thrive is we were some of the most

01:47:05.120 adaptable creatures out there. Obviously, smarter as well. But being so adaptable to our environment

01:47:11.180 helped us greatly. Because we used to think, well, the strongest survive. And really, we know it's

01:47:16.600 actually the animals that are most adaptive survive. So when we look at all the data together,

01:47:23.060 the question really shouldn't be, did we evolve eating seed oils? Or did we evolve eating this?

01:47:28.940 The question should be, based on the best evidence we have,

01:47:32.140 what is the overall net effect of these things? And I mean, again, like we were talking about the

01:47:37.560 processing earlier, I'll just run through a few more things. They say, you know, sodium hydroxide

01:47:41.600 gets used. There's very little of that in the end product. Activated clay gets used. Basically,

01:47:48.680 if we look at these things, the amounts that you would need to consume of this oil, and these are all

01:47:54.500 theoretical negative effects, assuming that, for example, pure sodium hydroxide stays in the end

01:47:58.620 component, which we know, it basically turns into soap and water. When you have a

01:48:01.980 chemical reaction, but even if the amount you put in stayed in there till the end, you still need

01:48:08.100 to eat two to 700 kilograms at one time of the oil. When we look at the end product of the oil

01:48:17.160 manufacturing process, it actually decreases the amount of oxidated. So there's a measure that can

01:48:22.540 use like a peroxide status to look at how much is in there. It goes down by a factor of about five to

01:48:27.820 tenfold. And then another measurement of like the aldehyde amount in the crude oil versus the actual

01:48:34.080 refined oil is much lower. Now, you do have, I think, polymerized trisoglycerides increase a little

01:48:41.220 bit. You do have some trans fats formation during this process. It's about 0.5% of the oil, but they're

01:48:47.080 all so low that it's very far below the threshold of what would cause negative effects. Now, there's no

01:48:52.740 safe amounts of trans fats, but again, we're taking this on balance. If we have this refined oil with a

01:48:58.740 very small amount of trans fat, but we know it lowers LDL so much, and then we have all the other

01:49:04.600 mechanistic data, what we call the converging lines of evidence, the mechanism is clearly elucidated.

01:49:10.760 The cohort trials agree with it. And then the studies that are not confounded by massive amounts of

01:49:16.160 trans fats agree. The Mendelian randomization trials agree. I guess the one other point I would

01:49:22.660 make is linoleic acid and polyunsaturated fats. When we trade them out in a one-to-one ratio for

01:49:28.800 saturated fat, they either have neutral or positive effects on inflammation, liver fat, insulin

01:49:36.160 sensitivity, and overall metabolic health. And that's been very clearly shown in numerous studies.

01:49:41.860 So again, if we boil down to it, regardless of what we think was an ancestral diet, which we don't

01:49:47.420 even know if that's necessarily the healthiest diet on balance, if you're going to make the argument

01:49:52.840 that polyunsaturated fats are bad, you have to be okay with the argument that saturated fat is really,

01:49:57.340 really bad. Okay. So how do we land this plane for folks? If you're out there and you're trying to

01:50:02.640 make sense of social media, you know, it's sort of funny. I was out at a restaurant recently and the

01:50:08.700 menu made a point to say there were no seed oils used in the preparation of the food.

01:50:14.480 So this is clearly a part of the popular zeitgeist. I'm pretty sure that the chef at that restaurant

01:50:21.420 isn't familiar with a single argument that has been made here today. So we're talking about an

01:50:28.360 argument that has sort of transcended a scientific discussion. So seed oils are culturally persona

01:50:36.520 non grata. The question is, is that warranted? You have, I think, fairly convincingly argued,

01:50:44.320 no. So for the individual listening to us, is there a precautionary principle? If someone says,

01:50:51.020 you know what, Lane, I've heard everything you've said. I can't poke holes in it, but why should I go

01:50:55.640 out and eat seed oils? What would you say to that person? I would say, okay, if you don't want to

01:51:00.960 consume seed oils, fine. But find something to displace the saturated fat in your diet with.

01:51:06.700 So leaner cuts of proteins, of meats, lower saturated fat sources of protein. And I guess

01:51:14.960 while monounsaturated fats don't seem to have the same effect on LDL cholesterol as polyunsaturated

01:51:21.360 fats, they do lower it when exchanged for saturated fats. They do appear to be cardioprotective to a

01:51:26.620 certain extent. It doesn't appear to be as cardioprotective as polyunsaturated fats. But if you are concerned

01:51:32.200 and you're not going to listen to logic that we've laid out here for three hours, then okay, try and

01:51:38.020 find some monounsaturated fats. So like olive oil, avocado oil, there's other sources of oils that you

01:51:44.820 could use that are still relatively cardioprotective or beneficial. And I didn't point this out when

01:51:52.380 talked about the processing. It should be pointed out that this is unique in that when these oils

01:51:56.820 are in a large volume, the rate of oxidation is low, even with heating. By the way, all the heating

01:52:01.980 in the processing of these oils is done under a vacuum, which means there's no oxygen, which means

01:52:07.880 virtually no chance for oxidation, even when heated. In restaurants, however, when you are frying

01:52:14.200 something, especially if you are frying in a thin layer of oil, the research shows going from like a,

01:52:20.300 I want to say it's like a one centimeter to like five centimeters of oil, huge difference in how

01:52:26.740 quickly oxidized and negative products will start to form. And if you are having oil that you are

01:52:34.540 frying, refrying in over and over and over, yes, certainly with a thin layer within 20, 30 minutes,

01:52:41.160 you can start to have significant amounts of these negative products accumulating. And then if you have

01:52:46.060 it in a vat and it's being heated all day, yeah, you're probably gonna have a significant amount of

01:52:50.020 oxidized, trans fats.

01:52:52.520 So would we be better off when it comes to heating oil using lard? In other words,

01:52:57.580 if I'm gonna have French fries, should I at least have my French fries made in lard as opposed to

01:53:02.220 polyunsaturated fat and seed oil?

01:53:04.020 So here's what I'd say. Both are bad.

01:53:06.320 Let's just say, I understand that French fries are hypercaloric, and let's just put that aside.

01:53:11.600 I'm gonna have French fries sometimes. So when I do, do I want McDonald's going back to

01:53:17.080 lard, or do I want them sticking with whatever seed oil they're using?

01:53:22.080 That's kind of a hard question to answer, right? Because you have competing mechanisms at play here.

01:53:27.560 And if we don't have a human RCT looking at frying with one way versus frying with another way,

01:53:33.160 and I'm not aware of any, but maybe there will be some, maybe a young potential scientist listening

01:53:37.960 to this would want to do this. But looking at, okay, what happens with LDL and then the components

01:53:42.820 of LDL?

01:53:43.820 But you're answering this purely through an LDL lens.

01:53:47.040 Right, right, right.

01:53:47.680 Yeah. Is there any other reason to care? It just feels to me intuitively that at least when you

01:53:53.860 heat up the saturated fat, you're less likely to introduce more ROS and other things. And by the

01:54:00.080 way, if I can control my LDL through other means pharmacologically, do I really care about my

01:54:04.420 saturated fat consumption?

01:54:05.820 We'll touch on that here in a second. So yeah, the saturated fat's less prone to oxidation.

01:54:09.820 Again, when we're looking at balance, what's going to negatively affect cardiovascular disease the

01:54:14.140 most? I don't know. What I would say is, if you're going to have french fries, just have the french

01:54:19.800 fries. If you want to have it fried in lard, okay, fine, whatever. You can decide what you want to do.

01:54:24.920 You're basically saying, don't treat my fries in lard as health food.

01:54:27.840 No. That's actually a really important point you bring up. You have to understand, people think

01:54:32.580 food companies care about which foods you buy. They just want you to buy. And so the pivot to,

01:54:38.460 oh, we're going to have tallow or lard or whatever. Food companies don't care. It's okay. Well,

01:54:43.740 we'll just make those then. That's fine. Oh, you don't like red dye 40? Yeah, we'll take that out.

01:54:48.320 And then we'll market about how healthy our cereal is now. We're marketing how healthy our french fries

01:54:53.120 are. The danger becomes, this really only becomes a problem if you're consuming french fries pretty

01:54:58.500 regularly. And then we have to ask the question, all right, which is worse out of these two really bad

01:55:03.660 options. But when you're marketing as some kind of victory that, okay, we're using beef

01:55:09.400 towel or using lard or whatever it is, as opposed to seed oils. If you're not having this sort of

01:55:14.800 communication, people, what they're going to interpret that as is, oh, these are actually

01:55:20.440 healthier now. I can eat more of them. I think that's one thing I've realized as being so in tune

01:55:25.860 with the public and reading comments on social media over years and years and years. I realized how if

01:55:30.500 I'm not extremely careful with how I word things, how misinterpreted it can get. I think as

01:55:37.520 communicators, like in a format like this, this is great. And when you say like, how do people navigate

01:55:43.120 on social media? That's where it's really tough because it's not this, right? It's 30 seconds.

01:55:48.240 How can I hook somebody in? Five reasons why seed oils are toxic. That's going to get a lot of attention.

01:55:54.020 And they're going to list things that there is an element of truth to every single thing that they

01:56:00.380 say. But they are leaving out all of the context that we just put multiple hours into covering.

01:56:07.100 I mean, I hope this podcast gets listened to by hundreds of millions of people, but the likelihood

01:56:10.540 is pretty unlikely. What's more likely is somebody puts up a TikTok and it goes viral and 10 million

01:56:15.680 people see it. I think it's very difficult to communicate this stuff with the public. To them,

01:56:22.780 there are so many mixed messages. Peter, I hear this all the time where people say,

01:56:27.980 you know, I don't trust scientific research because one study says this and one study says

01:56:31.540 that and they all contradict each other. And what I say to people is I say, did you actually read the

01:56:36.660 study or are you just looking at the social media hot takes? Because my guess is you're probably

01:56:41.500 looking at the hot takes. Because what we just did going into those studies, when they seemingly have

01:56:46.700 a weird outcome, I can tell you almost any time, 99% of the time, when I've seen a headline or a

01:56:52.620 social media hot take on a study that I go, that doesn't make sense. And then I go and read the

01:56:57.740 actual study. 99 times out of 100, I walk out going, oh, okay, I see why they found what they

01:57:02.720 found. Either the way the control group was designed or the difference in levels between

01:57:08.260 groups or whatever. My PhD advisor used to say, if I wanted to design a study to show no effect

01:57:13.540 or the study to show an effect, easiest thing in the world. And so again, this is why we look at

01:57:18.940 converging lines of evidence. We look at what does all the evidence state and what do the most

01:57:24.160 high quality, most rigorously controlled studies find? Yes, there are elements of truth to the

01:57:31.040 criticisms of seed oils. But on balance, when we look at these hard outcomes, when we look at what

01:57:37.260 we are very sure we know to be true, I think one of the things to point out in science, you can never

01:57:42.180 prove anything, right? We can only disprove things. But we can have relative degrees of confidence in

01:57:47.720 various data. I would say I have a relatively high degree of confidence that ApoB-containing

01:57:54.360 lipoproteins are atherogenic. Based on everything I've read, the converging lines of data,

01:58:00.920 especially the Mendelian randomization studies, especially the statin trials,

01:58:04.940 I feel relatively confident about it. Could I change my mind? Sure, but it would take a lot of

01:58:09.180 data over a long period of time. Now, you asked one question that I want to circle back to.

01:58:14.240 Why care about nutrition if you can just control this with statins?

01:58:16.600 No, that was in the context, I think, of why care about the effect on LDL if you can

01:58:22.120 pharmacologically regulate that anyway? And therefore, should we be focused on other negative

01:58:27.440 health benefits in the case of the frying? That was really my question.

01:58:30.620 Oh, I see what you're saying. So let's take care of the LDL piece, and then now oxidation or

01:58:35.460 aldehydes or whatnot become more important.

01:58:37.440 I was asking that specifically in the context of the frying oils.

01:58:40.200 That makes sense, yeah.

01:58:41.080 And cooking.

01:58:41.480 Yeah, I don't have a great answer for that, because I think one of the other frustrations

01:58:45.620 with the general public is when we put out limitations of studies, you and I know, we're

01:58:51.680 not necessarily saying, hey, these researchers are idiots. They did it wrong. They should

01:58:55.540 have done it this way. Every study has limitations. Every single study that's ever been done in

01:58:59.920 the history of mankind. There is no unifying study that explains the entire universe.

01:59:04.100 So pointing out limitations is not necessarily saying that a study is bad. It's just pointing

01:59:11.060 out, okay, we got to be careful how much interpretation we give to this, how broadly we

01:59:16.240 interpret it. And yes, there are studies that are more well-designed, well-conducted, that

01:59:21.560 have more statistical power, that have better measurements. And scientists try to account

01:59:27.080 for that when they look at, okay, how much weight am I going to give to something? But again,

01:59:31.940 at the end of the day, if I have to give a recommendation for people on this stuff, I would

01:59:37.160 say, when it comes to seed oils, if you don't want to consume them, okay. I would just say,

01:59:41.740 try to limit your saturated fat, eat enough fiber. But outside of that, there's so many

01:59:47.180 bigger levers that you can pull for your health than just worrying about seed oils. Put up a

01:59:52.560 thing a while back, I said, the average calorie consumption in the United States is 3,500 calories

01:59:59.020 per day. And the average physical activity is less than 20 minutes per day. You're spending

02:00:03.520 all this time worrying about what your fries get fried in. Not you specifically, but just

02:00:07.700 people in general. We're stepping over $100 bills, picking up pennies. Again, I'm not saying

02:00:13.300 don't worry about the little stuff, but you got to keep it in context of what really is driving

02:00:18.900 so much disease in the developed countries. A lot of it really is an energy toxicity issue.

02:00:25.940 Just to put a final bow on this, if you're looking at the macro trend of declining health

02:00:32.120 in the last 50 years, you're going to argue that caloric imbalance and activity levels are

02:00:39.600 contributing how much more than increasing seed oils? A little bit more, a lot more, medium

02:00:46.700 more?

02:00:47.120 I would say a lot more. Here's why I come to that. So there's no direct comparisons. I'm

02:00:52.380 going to say it right now. I'm going out on a limb. These are tenuous assertions by me,

02:00:57.740 my opinion. You look at the hazard ratios for mortality, talking about obesity. You're

02:01:02.860 not talking about, we're talking about like class three, class four obesity, even with

02:01:07.540 healthy blood markers, healthy obesity. You're not talking about 20, 30%. You're talking about

02:01:13.460 80 to 200%. You're talking about massive increases in the risk of mortality. When you're talking

02:01:19.260 about things like exercise, I mean, you talk about this all the time, your strength, your

02:01:25.460 lean mass, your activity levels, enormous predictors of how long you will live to the

02:01:32.880 order of hundreds of percentage points of magnitude when it comes to VO2 max, grip strength, overall

02:01:39.080 strength. I don't think there's anything unique about grip strength. It's just a proxy for overall

02:01:42.700 strength. And so again, that's not to say, hey, if you can make adjustments that make a

02:01:48.400 benefit overall, I don't want to be a whataboutism person either, but just make sure that your time

02:01:54.780 effort is being spent in your mind space. I got that from you. I liked what you said about that.

02:01:59.540 Your mind space is being spent on the things that will move the lever the biggest. And okay,

02:02:05.140 so you're controlling your caloric intake. You're exercising regularly. You don't want to eat seed

02:02:09.500 oils. Cool. Don't eat them. But I would also say, try to limit your saturated fat as well.

02:02:14.900 And try to make sure your LDL is under control. Get your APOB measured. I mean,

02:02:19.780 these are things that are modifiable. So if you can modify them, why not? And the last thing I will

02:02:24.100 say, because people do this too, I am not saying, and I don't think that you would say that APOB and

02:02:30.120 LDL are the only things that drive cardiovascular disease. Blood pressure, cardiovascular fitness,

02:02:38.140 insulin sensitivity, inflammation, these things all matter. They all matter. We are just saying,

02:02:43.340 and I'll give an example. You could have high LDL, but every other factor is low and your overall

02:02:48.900 risk for cardiovascular disease might be low. You can have high LDL and you might live to 90,

02:02:54.940 100 years old. That can happen. But there's also people who smoke for long periods of time and live

02:03:00.540 to be old. That doesn't mean that you should smoke or that it's not negative because statistics are just

02:03:07.500 probabilities. These are just probabilities. These are not hard. This is definitely going to happen.

02:03:12.760 So of course, you can always find an individual to show a difference. But all we are saying,

02:03:18.840 or all I am saying, is that everything else being equal, having your LDL lower is better,

02:03:24.960 all things being equal to having it higher.

02:03:27.780 All right. Well, Lane, I think that kind of brings this discussion to an end. I don't think this was

02:03:32.680 necessarily as interesting as it would have been in its original format, where we could have done it

02:03:38.560 as a genuine two-person point of view. I tend to agree with the point of view you've put forth.

02:03:44.500 My own nomenclature on this is that we're majoring in the minor and minoring in the major. I just don't

02:03:49.320 think this matters all that much, frankly. And my lack of enthusiasm around this topic is probably

02:03:54.700 palpable. I also think I'd make one final point to what you said, which is there is another confounder

02:04:01.220 with seed oils, which is that they tend to show up in low quality foods. And therefore, if you make

02:04:07.240 the decision to restrict your seed oils, you are probably doing a net benefit to yourself because

02:04:12.680 you are simply going to eat less Oreos, less potato chips, less junky salad dressings, and crappy

02:04:22.340 sauces and things like that. So the substitution effect will probably work in your favor, but you

02:04:29.460 don't have to be maniacal. If, for example, when you're making a salad, you prefer the taste of

02:04:35.820 safflower oil or canola oil over olive oil, it doesn't seem like you're killing yourself by doing

02:04:41.360 it based on the data. And you probably don't need to go to restaurants that are adamant that they

02:04:47.540 exclude seed oils because if it's a good restaurant, whether it uses seed oils or not, it's probably

02:04:52.220 using good ingredients otherwise, and you're probably going to be just fine. Yeah. I mean, you make a

02:04:57.660 great point about, it's probably more about what comes along for the ride. Another comparator real

02:05:02.100 quick, like sugar intake. People cut out sugar and they say, well, I felt better. You cut out a bunch

02:05:07.720 of junk food, but then you have people who get maniacal with it and start cutting out fruit

02:05:12.020 because fruit has sugar. Biochemically, it's basically the same thing. Once it gets in your

02:05:16.440 body, now there's a bridge too far. And I would argue the same thing that I think most people who are

02:05:22.140 experiencing negative effects from seed oils just have an overall probably low quality diet.

02:05:27.500 This isn't a problem for people who are going, you know, I think I'm going to cook with some canola

02:05:32.000 oil today and maybe use a seed oil-based solid dressing here and there. I think that that's

02:05:39.240 probably not what's happening. And what is happening is people are eating a lot of potato chips,

02:05:47.140 french fries, whatnot. And then the kind of anti-establishment people come out and they

02:05:51.920 say, look at how much our seed oil intake has increased. And we did what the government told

02:05:55.480 us to. Yeah. So my only wish that come from this podcast, in addition to public education,

02:05:59.940 is if you are a restaurateur and you're listening to this, please take the no seed oils used off your

02:06:05.580 menu. It insults me and it insults anybody who's been patient enough to listen to this episode.

02:06:12.380 So with that, thank you. Yeah. Thanks for having me. This was fun and a really cool

02:06:16.420 educational experience. I loved it. So thank you for having me and let me rap about this stuff for

02:06:22.340 a couple hours. Thank you for listening to this week's episode of The Drive. Head over to

02:06:27.920 peteratiyamd.com forward slash show notes. If you want to dig deeper into this episode,

02:06:34.740 you can also find me on YouTube, Instagram, and Twitter, all with the handle peteratiyamd.

02:06:40.340 You can also leave us review on Apple podcasts or whatever podcast player you use.

02:06:45.580 This podcast is for general informational purposes only and does not constitute the practice of

02:06:50.860 medicine, nursing, or other professional healthcare services, including the giving

02:06:54.700 of medical advice. No doctor patient relationship is formed. The use of this information and the

02:07:01.020 materials linked to this podcast is at the user's own risk. The content on this podcast is not intended

02:07:07.040 to be a substitute for professional medical advice, diagnosis, or treatment. Users should not disregard or

02:07:12.980 delay in obtaining medical advice from any medical condition they have, and they should seek the

02:07:17.640 assistance of their healthcare professionals for any such conditions. Finally, I take all conflicts

02:07:23.280 of interest very seriously. For all of my disclosures and the companies I invest in or advise, please

02:07:29.180 visit peteratiyamd.com forward slash about where I keep an up-to-date and active list of all disclosures.

The Peter Attia Drive - January 19, 2026

#380 ‒ The seed oil debate： are they uniquely harmful relative to other dietary fats？｜ Layne Norton, Ph.D.

Episode Stats

Length

Words per Minute

Word Count

Sentence Count

Misogynist Sentences

Hate Speech Sentences

Summary

Transcript