Peter Attia MD212 - The Neuroscience of Obesity | Peter Attia, M.D. & Stephan Guyenet, Ph.D.
EVERY SPOKEN WORD
135 min read · 26,567 words- 0:00 – 0:08
Intro
- PAPeter Attia
[upbeat music] Hey, everyone. Welcome to "The Drive" podcast. I'm your host, Peter Attia.
- 0:08 – 7:17
Stephan’s neuroscience background and his focus on the nuances of obesity
- PAPeter Attia
Hey, Stephan. Thanks so much for making time. I've been looking forward to this for such a long time, uh, probably since I started a podcast, which has been now we're coming up on four years. Uh, always knew we'd, we'd have to sit down, uh, so I'm glad we're finally doing this, albeit not in, not in person, but that's more, uh, an artifact of my laziness. Um, so anyway, uh, you know, again, look, I think a lot of people listening to this, uh, will be familiar with, with you and your work, but I think a number of people won't be. So, um, let's tell people a little bit about kind of, uh, your, your path to, to where you are now, which is sort of being one of the most, uh, thoughtful people on the, on the nuances of obesity. Um, what, what, what did you study in college? Were you a neuroscience major?
- SGStephan Guyenet
Biochem.
- PAPeter Attia
Biochem.
- SGStephan Guyenet
Yeah.
- PAPeter Attia
Okay.
- SGStephan Guyenet
But I had, uh, neuroscience in mind when I was doing biochem. My idea at the time is that it would provide a foundation for going into neuroscience later, which I'm not sure that, uh, reasoning really works out so well, but, um, it worked out okay in the end.
- PAPeter Attia
And did you go straight from your undergrad to Mike Schwartz's lab, or did you do your PhD with Mike Schwartz or a postdoc?
- SGStephan Guyenet
Postdoc.
- PAPeter Attia
Okay.
- SGStephan Guyenet
Yeah. Ph-
- PAPeter Attia
So where did you do your PhD?
- SGStephan Guyenet
PhD was with Al LaSpada at the University of Washington studying neurodegenerative disease.
- PAPeter Attia
Interesting.
- SGStephan Guyenet
Yeah.
- PAPeter Attia
So tell me about that. Was that, uh, a detour that was always part of the plan, or at that point were you not yet fully interested in, uh, obesity?
- SGStephan Guyenet
Yeah, I think more the latter. So, um, I've always been fascinated by the brain, but I didn't know which area of neuroscience I wanted to get into for a long time, and, um, I became interested in neurodegenerative disease for a few different reasons. One, they're just absolutely horrible diseases, and two, my grandmother had Alzheimer's disease. Um, and the, um... But so in, in grad school, I was studying neurodegenerative disease, but I wasn't studying Alzheimer's disease. I was studying a class of neurodegenerative diseases called triplet repeat diseases that includes Huntington's disease, or polyglutamine repeat diseases would be another, uh, name for them. So Huntington's dis- disease is the most common. That's the most common heritable, um, heritable neurodegenerative disease. But-
- PAPeter Attia
And as you said, it's a, it's an absolutely awful disease with, I mean, almost 100% penetrance, correct? I mean, it's a deterministic gene?
- SGStephan Guyenet
Well, it's actually [chuckles] more complex and interesting than that. It's actually ... The, the genetics of it are real interesting because they are non-Mendelian because the length of the CAG repeat actually changes intergenerationally.
- PAPeter Attia
Hmm.
- SGStephan Guyenet
The weird thing about it is there's, there's these CAG repeats that code for polyglutamines, uh, polyglutamine stretches in the protein. They are unstable in replication, and so what you tend to see is an enlargement of these, uh, polyglutamine repeats from one generation to the next. And so, um, yeah, so it has this really weird, um, non-Mendelian pattern. I think, I think... Okay, you asked specifically about penetrance. I think the penetrance actually is pretty high. So in other words, if you have a polyglutamine repeat in the wrong protein of a certain length, yes, I think very high likelihood you're gonna, you're gonna develop the disease. Um, but like anything, it, it is not, you know, 100% fixed, but I don't think we really understand what makes it not fixed. Um, but I was, I was studying a, uh, one of the less common ones called SCA7, spinocerebellar ataxia type 7, and, uh, you know, it's an interesting disease. It's a neurodegenerative disease with some relevance to other more common neurodegenerative disease. But, um, I used to joke that there were probably more scientists studying it than people [chuckles] with the actual disorder. I don't think that's actually true, but, um, I think you get the, the point of the joke. Um, and I just kind of wanted to study something with greater impact, and I also... I al- I've always been interested in fitness and nutrition, um, kind of on a personal level, and so when I started learning about the neuroscience of obesity during my PhD work, I got really into it because that was a way to kind of satisfy my criteria for something that's impactful. It's hard to imagine much that's more impactful than that in the world we live in. Um, it's very common. It, um, is... It relates to my interest in fitness and health, and it has a strong, uh, relationship with neuroscience. And so once, once I figured that out, which I think to a lot of people is not obvious, the relationship with neuroscience, and I think that's a topic we'll get into. Um, but once I figured that out, I started realizing that not only was this really fascinating, but there was a ton of information in this space that was incredibly enlightening that was not making it to the public.
- PAPeter Attia
What year did you finish your PhD?
- SGStephan Guyenet
2009, I think.
- PAPeter Attia
Okay.
- SGStephan Guyenet
Yeah.
- PAPeter Attia
And did... And Mike Schwartz was also at the University of Washington, correct?
- SGStephan Guyenet
Correct. Yeah.
- PAPeter Attia
So at, at what point as you were wrapping up your PhD did you connect with Mike and, uh, or at least become familiar enough with his work that you thought, "You know, this is kind of my finishing school"?
- SGStephan Guyenet
Yeah, I'm not sure. I, I, um- I don't remember exactly what all the options were that I was considering, but I was particularly interested in obesity. And staying at the same institution after your PhD is atypical. Um, it's something that I wanted to do in part for personal reasons. Um, I don't-- I'm not really a big fan of the typical academic thing of it's almost like a military life. You know, you're moving around like five or more times before you, you finally settle down. Um, so that was part of it, but Mike was also a really good fit. And there are other labs that could have been a good fit in other places. Um, but Mike's lab was a really good fit, and he was willing, so I, uh, I did that.
- PAPeter Attia
And God, I think the, I feel like the first time you and I met would've been probably at, at a conference in 2012. Um-
- SGStephan Guyenet
Mm.
- 7:17 – 14:45
How obesity has changed for humans throughout history
- PAPeter Attia
So, so let's, let's tell people a little bit about the problem that you work on today. Um, I think everybody knows directionally that obesity is a significant issue, but, uh, you can probably quantify this for people a little bit better and help people understand maybe even over a few thousand years how things have changed in terms of the, uh... Let, let's just talk about the phenotype. We're gonna obviously talk about the environment and the, the triggers, but let's just talk phenotypically how have we as a species who, you know, we've been around what? Maybe 6 million years in our current rendition. But what, what's changed over the last thousands, uh, uh, thousand years in terms of our phenotype?
- SGStephan Guyenet
Yeah. So if we're talk- if we're comparing the, uh, body shape of people in modern affluent societies like the United States to what the typical human would've looked like 1,000 years ago, I think it's clear that we're much fatter today on average with a much higher, uh, percentage of obesity. And 1,000 years ago, there was obesity. I mean, we have evidence even from Egyptian mummies that, um, among the wealthy there was obesity. Not to say that it necessarily was super common, but I don't think it was that uncommon, um, among the wealthy, I think probably for similar reasons that we have obesity today. But, um, certainly the prevalence was much lower. And when we start to get into the more modern historical period where we start to actually get data on this, the first data that we can find on this in the United States, or at least that I have found that is somewhat informative, are from, um, Civil War veterans from 1890 and 1900. They did height and weight measurements on middle-aged Civil War veterans. So these people were, I think, almost exclusively white men. Um, however, if you compare to the same demographic, so middle-aged white men today, you see that there was almost no obesity back then, and today the obesity rate is something like four- probably 45% for that same demographic.
- PAPeter Attia
And, and just to be clear, we're defining obesity in the most traditional way, which is the use of body mass index, and we're defining it as a BMI of more than 30, correct?
- SGStephan Guyenet
Yeah. Correct. And so, yeah. So the advantage of BMI is it's really easy to measure, and you can calculate it from these really simple measures that go back a long time. Unfortunately, they didn't have DEXA machines-
- PAPeter Attia
Yeah. Yeah, exactly
- SGStephan Guyenet
... in 1890.
- PAPeter Attia
[chuckles] Yeah.
- SGStephan Guyenet
Um, which would've been, of course, a more informative way of looking at it. Um, but yeah, using measures that we can compare over long periods of time, like body mass index, um, I don't remember the exact numbers, but it was like a few percent. It was like low single digits of people actually classified as BMI over 30 at that time, 120, 130 years ago. And then if we, um, look toward more recent data, what we see is that between... So the first, the first really good data we have starts in the 1960s for the United States. That's when the, uh, NHES surveys started, which later became NHANES. And in those surveys, what you see is by the time they started measuring it, it had already gone up from that previous time in the, in the late 1800s, early 1900s.
- PAPeter Attia
So you're saying, Stephan, that there really wasn't a lot of longitudinal data from 1900 to 1960 that, that kind of con- you know, to, to check what that trend line was doing. It was sort of this big effort in 1900, and then another big effort didn't take place till about 1960?
- SGStephan Guyenet
Yeah. I mean, I, I don't know that I'd call it a big effort, but the biggest that I'm aware of in the late, in the, the one in the, the late 1800s, early 1900s. Um, and then the true, like true representative national sampling started in the 1960s and then got better through the '70s, and, and now we have this, this, um, NHANES survey methodology that is the best source of evidence that we have. And so it started, started getting good in the '60s.
- PAPeter Attia
And what were those levels there, Stephan, in the 1960s?
- SGStephan Guyenet
Yeah. I think the earliest measures we have are like something like, I don't remember the exact figure, but it's like 12%. of US adults had obesity at that time, something like that, in the earliest measures.
- PAPeter Attia
And do you have a sense of -- because I'm sure this is going to become more relevant today, is looking at what, what, what is the term that's used if BMI is over 35? Isn't there a extreme category of obesity?
- SGStephan Guyenet
Yeah.
- PAPeter Attia
Or morbid obesity, is that defined as 35 or 40 or something?
- SGStephan Guyenet
Gosh. Yeah. I think the, the terminology has changed to, you know, try to avoid stigma. Uh, it has been extreme obesity or morbid obesity. I can't remember what the current, um, term is for it. But yeah, there's a category over 35 as well.
- PAPeter Attia
And, and presumably there's been a change in that as well. I'm, I wouldn't be surprised if, if-
- SGStephan Guyenet
Oh, yeah, yeah. Sorry. There's-
- PAPeter Attia
Yeah
- SGStephan Guyenet
... a class system, so class one, class two, class three. So I think that corresponds to 30, 35, and, and 40, if I'm not mistaken. Anyway, you were saying?
- PAPeter Attia
Yeah. So, so, so basically this progression was not just at the kinda median level, 'cause I'm sure the median BMI was also moving, the mean BMI was also moving, the fraction over 30, and presumably the fraction over whatever that highest threshold is, be it 35 or 40.
- SGStephan Guyenet
Yes. And actually, the most extreme changes happened in more severe obesity. So you see very, very little, very, very few people had BMIs over 35 in the earliest measures, and then now it's like, it's, it's something like 9 or 10% today, so, um, of, of adults. So there has been, uh, more movement at the extreme end than, than at the mean. Yes. And, um, and that's what happens when a distribution spreads out, which is what happened. If you look at the, the distribution of BMIs, it used to be a lot tighter, and it just got less tight. So there are still people who are lean. Um, there are still people in every BMI category just like there used to be, but since it has spread out, you get a disproportionate increase at extreme values.
- PAPeter Attia
What about underweight? What was, d- if you happen to know, what was the fraction of underweight in 1900 if we would define that to, say, BMI below, I don't know what underweight is. Is it below 18 or below 20? And then how has that changed over time?
- SGStephan Guyenet
Yeah. Typically, the cutoff is 18.5, and I don't know the answer to your question. I think it was higher than it is today, uh, but I'm not sure about that.
- 14:45 – 25:49
The association between obesity and adverse health outcomes, the “obesity paradox,” and confounders when relating BMI to longevity
- PAPeter Attia
When did people first make the connection that there is an association between obesity and adverse health outcomes? I mean, presumably back when you think back to 1,000 years ago or back with the Egyptian, you know, uh, aristocrats, obesity probably would've been a sign of affluence, and I don't think anybody would've, uh, looked down upon it too negatively. When we -- Back in medical school, I remember learning about gout and how gout emerged around this time, and it was really this disease of excess, right? Excess alcohol, excess sugar, excess protein would manifest itself in gout, and, uh, this was the ultimate rich man's disease, uh, basically a sign of affluence. So I can't imagine people were too upset to be obese back then. Um, w-when, I mean, we t-- and, and of course, today, we take it for granted, right? I mean, I think there's, despite some of the, the, the political pressure to, uh, understandably try to destigmatize obesity and somehow now suggest that it's completely un-, uh, completely healthy, I, I think the reality of it is it's, it's pretty unambiguous that obesity is indeed associated with poor health outcomes. When did that become clear?
- SGStephan Guyenet
Yeah. That's a good question. So I'm not real knowledgeable about the deep history of this, but I know I've heard that, uh, you know, there were physicians in ancient Greece and India who recognized that, uh, being, uh, very heavy was associated with health problems such as having sweet-tasting urine, for example, sign of type 2 diabetes or any kind of dia-diabetes. Um, and, but I think, yeah, and, and then there were these insurance life tables, um, earl- in the early 1900s that suggested that, uh, people who had obesity had, uh, shortened lifespans and greater risk of certain diseases. But it actually became pretty controversial with a series of studies that was published, um, Katherine Flegal was intimately involved in this work, suggesting that there was actually not the relationship people thought there was between body mass index and mortality. So these studies, um, which were, this was labeled as the obesity paradox because what they found is that there wasn't really much of an association between obesity and poor health outcomes. And often, if you look at the relationship between, um, BMI and mortality, if you just look at a graph of it, the, the nadir, the lowest point on that graph, was in the overweight range or sometimes even on the low end of the obese range, depending on what the study was. And they were finding this in meta-analyses of millions and millions of people. And so suddenly, people were saying, "Well, maybe higher body fatness is not bad. Maybe it's actually protective." Um, so there's been this, this big debate about it. Um, I think what has emerged in recent years, especially the last 10 or 15 years, and, and by the way, let me take a step back. The reason this is called a paradox is because there's all this evidence that excess body fat contributes to all kinds of diseases, type 2 diabetes, cardiovascular disease, certain cancers, and so how could it be protective, uh, for mortality when it's driving all these diseases that are the leading causes of mortality? So that's why it's called a, a paradox. Um, and I think the research that has come out since then is suggesting that it's probably not a paradox, as paradoxes often are not, and that it's a, an artifact of those observational data, that essentially, um, there are probably a few different things going on, but the biggest one is that people who are sick often lose weight. So if you... There are many different health conditions that can cause a person to lose weight. So type 2 diabetes, if you're, if it's not well controlled, you can be losing weight.
- PAPeter Attia
Certainly renal failure, COPD, things like that.
- SGStephan Guyenet
Yeah, those are great examples. Alzheimer’s disease, cognitive decline, um, those things can cause a person to lose weight. And so there are many health conditions that can cause a person to lose weight, and essentially the concern was that that makes leanness look worse than it really is, because you're getting all these people in the lean category that are lean because they're sick. They're not sick because they're lean. So some people have called that reverse causation. David Allison, uh, corrected me. It's technically, um, confounding.
- PAPeter Attia
That's right, yeah. You and David and I had a nice email exchange [laughs] about that because we, we did discuss this.
- SGStephan Guyenet
Yeah.
- PAPeter Attia
I'll, I'll share with you a very glib example. I had shoulder surgery less than three weeks ago. So in, I don't know, in the 18 days since I've had my shoulder operated on, [clears throat] I've lost nine pounds. Now, I don't know what my BMI was before versus after. I probably went from BMI of 26 to BMI of 24, so on paper that looks good. I would argue there is nothing about me today that is superior in health to where I was 18 days ago, right? If that, of those nine pounds I've lost, I'd be willing to bet seven of it is lean body mass. So it's, again, it's a silly example, but it illustrates that one can see an improvement in BMI with probably a deterioration in body composition and an increase in morbidity.
- SGStephan Guyenet
Absolutely. BMI is a crude measure, you know. No doubt about it. It's useful. I mean, yeah. You talked about this with, with David Allison in, in your episode, and I think you guys had a nice conversation. But CliffsNotes is, you know, it's useful for a population level studies. It can be useful for screening, but it's, it's a crude measure. Um, yeah, so, you know, how do you get around this issue of confounding or reverse causality? Um, there are some methods that have been developed for this. One of them that I particularly like was developed by Andrew Stokes and his colleagues, and that is the maximum attained weight method. So instead of using the exposure variable, ex- instead of looking for BMI right now and seeing that, how that correlates with mortality risk right now, you say, "What's the heaviest you've ever been? And how does that correlate with your health outcomes?" And what you see when you do that... So essentially that's saying like, you know, whatever health condition you've developed that might have caused you to, to lose weight, we're kinda screening that out. We're looking before that. And when you look at it that way, you get a sharpening of the association between BMI and mortality, and you find that the nadir, the lowest point, shifts to a lower BMI. So essentially whereas before it looked like maybe overweight was the best, now you see that in the lean range looks the best. And the, and there's a stronger difference, a larger difference between being lean and having obesity in terms of mortality. So it really sharpens things up. And the reason that happens is because when, when he looked into this, you're excluding a bunch of people who were formerly in the obese or overweight category and went into a lower category. And those people, if you look at their health outcomes, they're terrible. So those people who used to have obesity and now just are overweight or are lean, those people have massively elevated rates of chronic disease and of mortality. And so they're essentially, uh, bringing all this excess mortality into lower weight categories.
- PAPeter Attia
And why, why is that, Stephan? Because th- that, that almost sounds like the worst news you could ever hear, right?
- SGStephan Guyenet
[laughs]
- PAPeter Attia
It would suggest that if your BMI is 33, all hope is lost, that your fate was sealed the moment you hit that BMI. You know what? Just dig in and eat more Häagen-Dazs because y- you know, if you, if you bring that BMI down to 26, you don't assume the health of the 26. You bring the health of the 33, which th- that, that doesn't sound right either, right?
- SGStephan Guyenet
No, it's not right, and this is another limitation of the, the observational data because these people... So it's hard to lose weight, right? And so most of these people, it's hard to lose a lot of weight. It's hard to change or to go from-
- PAPeter Attia
And it's hard to keep it off.
- SGStephan Guyenet
Yeah. It's hard to go from 35 to 25 BMI. That's a lot of weight loss. Most people, frankly, are not able to achieve that and maintain it, and so through voluntary means. And so these people who are losing weight, it's predominantly unintentional weight loss. This is not people who start a diet and lifestyle plan and lose a bunch of weight. This is predominantly people who are losing weight unintentionally. And as a doctor, I'm sure you know this, when a person starts losing a bunch of weight for no reason, that's probably not a good sign, right? Even if they start off overweight and they're, and they're lean, maybe they even look good, that's probably gonna start ringing alarm bells. Um, so if you look at studies that have measured the impact of intentional weight loss on mortality, this has been done both for diet and lifestyle weight loss, randomized controlled trials, so you know, we're talking about good, good quality of evidence, and also for bariatric surgery, you see a reduction in all-cause mortality.
- PAPeter Attia
That's what I was gonna ask you. I was gonna say, how does bariatric surgery, and pretty soon I'm very n- be very curious, how does semaglutide affect this? It's too soon to say, but when we start to look at what I think is the most impressive weight loss drug out there, and then of course when you think about Roux-en-Y gastric bypass, which has been around for a while, it'll be interesting over time to see if that can flip that paradigm, huh?
- SGStephan Guyenet
Well, we have data for type 2 diabetes, for people with type 2 diabetes for semaglutide, and it does reduce all-cause mortality in meta-analyses of randomized controlled trials. So it will be interesting to see whether that extends to people without type 2 diabetes. Obviously, people with diabetes, um, are probably gonna benefit the most from that kind of drug class in terms of, you know, the physiology, uh, of it. But I think there are promising signals that at the very least it's probably not gonna kill you. [chuckles]
- 25:49 – 33:52
The sharp increase in obesity across demographics
- PAPeter Attia
We go from the '60s to the '70s to the '80s. When does it really, from an epidemiologic standpoint, just completely take off in terms of... You know, 'cause even if I remember what you said a moment ago, back in the '60s, we might have been at 12% obese. And I mean, today it's hard to imagine such a low figure given 40, 40 to 45% today. Where, where-
- SGStephan Guyenet
Yeah, 43 is the latest
- PAPeter Attia
That, that's, that's simply unbelievable. I, I, I... Not that I don't believe it, it's just hard to fathom that nearly half of the US adult population, um, could have a BMI above 30. When, when did it really hit, hit its stride?
- SGStephan Guyenet
Yeah. So this is an interesting question. So probably a lot of people listening will have seen graphs of NHANES data where it kind of spikes around 1980. But the interesting thing about that is that point around 1980 is actually the average date of that survey, but it was actually a multi-year survey.
- PAPeter Attia
Mm.
- SGStephan Guyenet
It was a survey that started, I believe, in '76 and ended in, like, '84, something like that. I, I don't remember the exact years, but it's, it was a range. That's how the NHANES used to be. Today, their, the range is much narrower, but at the time it was a broader range. And so what we're seeing is, you know, we can turn that into a point and put it on a graph, but really it represents a range of years. So we don't know exactly when it happened, but it's, it... If you look at the average value for that, it's, I think 1978 is the average value for that survey, the average year. Um, but we don't know exactly where in there it started to turn up, and we don't really know how sharp it was because we don't have that resolution. It looks real sharp on the graph. Again, when you-
- PAPeter Attia
Mm-hmm
- SGStephan Guyenet
... when you make it a single point, it looks sharp. But we don't really know exactly how sharp it is. Um, and so anyway, I'm, I'm kind of putting a lot of nuance into this, but to, to answer your question-
- PAPeter Attia
Well, it is actually an important point because I know that many people, I've probably been guilty of this myself 10 years ago, would also look at changes in macronutrient composition that occurred around the same time and say, "Boy, it's hard to uncouple those." But in reality, if they occurred over a, a different timescale, it might make that correlation a little less robust.
- SGStephan Guyenet
I mean, I would say that we have more, um, we have a sharper focus on the changes in dietary intake because we have annual data from the Economic Research Service, USDA Economic Research Service. Um, the data on BMI are less sharp, so I would say that there's... Those two, those two changes are absolutely compatible and, you know, just on a logic level I'm sure they're related. Um, you see big changes happening around the same time in the diet. So yeah, anyway, to, to give a kind of simpler, hopefully more satisfying answer, sometime around between the late '70s and the early '80s, we see an uptick, an apparent uptick in the obesity rate. So the rate starts to increase, goes up and up and up, and then there's a couple of places where it slightly goes down for a year or two, and then it keeps going up. Um, yeah, it's, it looked like it was gonna plateau or maybe start going back down, but really in recent years it's really just been skyrocketing. So, um, yeah, so essentially from somewhere in the late '70s to early '80s to now, we went from something like 15% of obesity to 43% of US adults. And I wanna point out something else too that I think is relevant. One way I like to think about this is the lifetime risk. So that's just the population prevalence.
- PAPeter Attia
Right.
- SGStephan Guyenet
That includes people who are 20 years old.
- PAPeter Attia
That's a snapshot. Yeah.
- SGStephan Guyenet
Yeah.
- PAPeter Attia
That includes people who may be growing into it, and-
- SGStephan Guyenet
Exactly.
- PAPeter Attia
Yeah.
- SGStephan Guyenet
If you look at the lifetime risk, I don't know what the exact figure is, but I think it's well over 50%. So I think more than half of US adults will be classified as actually having obesity at some point in their life if the current context is maintained. And you see the same thing for type 2 diabetes. The, the, the prevalence is like... The prevalence of all diabetes is something like 10 or 12%, um, but if you look at the lifetime risk, it's like double that or maybe even more. I can't remember what the exact figure is, but it's like mind-blowing the number of people who will at some point in their life develop type 2 diabetes in the US.
- PAPeter Attia
Y- yeah. I mean, if, if you take a 50-year, [clears throat] uh, look at the change in type 2 diabetes prevalence, so the snapshot, I believe it's a fivefold difference in risk over 50 years, the past 50 years. So if you go back 50 years ago and look today, just the prevalence delta is 5X. So-
- SGStephan Guyenet
I could believe that
- PAPeter Attia
... so you have to then look at what does that rate of change tell you about lifetime prevalence? What does it tell you about a person who is only five years old today or a person who's 15 years old today? Yeah, it's... I, I would complete... I mean, I haven't done the math, but I would completely agree that their lifetime risk is probably at least one in five of having type 2 diabetes.
- SGStephan Guyenet
Yeah, absolutely, and this, this has been quantified. I tweeted out a paper about this a while back, and unfortunately I don't remember the top-line figure, but there are actual data on this.
- PAPeter Attia
So [clears throat] what about the rest of the world, Stephan? Where, where are they in relation to us? I mean, we, we, we probably led the way, uh, along with a, a handful of other developed nations, but o- o- one of the things that seems to have changed is this is no longer a condition of affluence, correct?
- SGStephan Guyenet
Well, it, it, I-
- PAPeter Attia
At the individual level. Uh, let me-
- SGStephan Guyenet
I think that-
- PAPeter Attia
Let me say that again. Yeah
- SGStephan Guyenet
... yeah, I mean, I think it really-- sorry to, to be pedantic, but it really hinges on how we define the term affluent because if you go to the very poorest people in the world, there still is not a lot of obesity in those places. Places that are really challenged with food security, where the diet is very limited, like subsistence farmers in sub-Saharan Africa, you're still gonna see that there's a low prevalence of obesity in that pl- in those places.
- PAPeter Attia
Yeah, sorry, I think what I meant was at the individual level, so if you look at the United States from the bottom 10% of the population economically to the top 10% of the population economically, is there a difference in obesity there?
- SGStephan Guyenet
I think bottom 10 to top 10 you would see a difference. If you're looking in tertiles, so just bottom third to top ter- top third, there's very little, there's very little difference actually. Um, w- if you start slicing and dicing it by, uh, sex and race, then you will start to see larger differences emerge. For example, uh, women who are in the top tertile are leaner than women in the bottom tertile, but there's no difference for men in terms of income. Um, so you can start to find patterns when you slice and dice. Um, my overall feeling, though, is that there is no demographic in the United States that has not gotten a lot fatter over the last few decades, even though I think if you look at certain demographics, particularly with regard to education, you're gonna see gradients emerge. But, um, even among highly educated people, you're gonna see a higher prevalence today than there was 50 years
- 33:52 – 45:32
The hypothalamus and its role in obesity
- SGStephan Guyenet
ago.
- PAPeter Attia
So let's now bring this up to your work. What did you, what did you do when you got to, to Mike Schwartz's lab? How did you begin your re-education around neuroscience as it applies to everything that has to do with obesity, which let's just talk about this. There's an input side and an output side. Um, we, I think we should talk a little bit about... I mean, Rudy was on the podcast, God, it's probably been two, three years ago, so maybe it's worth a refresher on the neurobiology of how the different parts of, for example, the hypothalamus can regulate, uh, energy expenditure, can regulate appetite, and what we can learn under very, very controlled experimental settings with animals. Um, and then how we can start to think about how that applies to, to humans with a, you know, uh, a primitive brain in a modern world, so to speak.
- SGStephan Guyenet
Yeah. So I wanna take a little step back before getting into that and say, you know, answer the question of what does the brain have to do-
- PAPeter Attia
Mm
- SGStephan Guyenet
... with obesity at all? Just to make sure we're bringing everybody along, 'cause I think it's not obvious to everyone what the brain might have to do with obesity, but I think if you start to think about it, it becomes pretty obvious that the answer is just about everything. And the reason is that, you know, the most obvious reason is that the brain is the organ that generates behavior. And so if you think that behavior of any kind relates to body fatness, what we eat, how much we eat, how we use your bo- how we use our bodies, how we sleep, whether we're stressed or not, if we think any of that relates to body fatness, then we think the, the brain is, is playing a role. I think most people would agree that food intake quantity and quality is, is pretty important there. Um, and then the second reason is that the brain actually contains a regulatory system for body fat, for body fatness, I should say, body fat mass, and it's the only known system in the body that does that, and it's located primarily in a part of the brain called the hypothalamus, which is like a, I forget who described it this way. It's maybe it was Herman Pontzer. It's like a wad of bubblegum on the bottom of your brain near where your optic nerves cross. So it's this little tiny part of the brain that specializes in homeostasis, maintaining the stability of body systems
- PAPeter Attia
I always describe it as a walnut, but maybe a little bit-
- SGStephan Guyenet
A walnut, okay.
- PAPeter Attia
Yeah, yeah.
- SGStephan Guyenet
Yeah. So just to give you an example, it, it's the part of the brain that regulates body temperature, and there's a thermostat in there, and effectively a thermostat and thermometers. There are thermometers that measure your core temperature. There are thermometers on your skin that measure future threats to your core temperature. Like, if you jump into a cold lake, your core temperature doesn't instantly drop, but your brain knows that it will drop because of the temperature ch- sensors in your skin, and so it can respond adaptively. And then that system in the hypothalamus engages a behavioral, a, a suite of behavioral and physiological responses to maintain temperature homeostasis. So you'll... On the physiology side, you get vasoconstriction, you get, um, [lip smack] non-shivering thermogenesis through brown fat. You get shivering. And then through the behavioral side, you wanna get out of the cold water. You wanna put a sweater on. You wanna, you know, drink some hot tea. Uh, you want to adopt a heat conserving posture. And through this coordinated physiology and behavior, you get incredible regulation of temperature, of core temperature, I should say. And, uh, you know, like, plus or minus one degree Fahrenheit when the exterior temperature could be varying by 50 degrees. And so it's, it's an incredible regulatory system, and the body fat regulatory system, unfortunately, is not so precise. Um, but I, I give that as an analogy just to give you a sense of what the hypothalamus specializes in. There's also a regulatory system for body fatness, and a nice name for that is the lipostat, so lipo, fat, stat, the same. And really, we've known about it since 1840, or we've known there was something going on since 1840 when the Viennese physician Bernard Mohr published a case study about a woman who had extreme obesity, that rapid onset extreme obesity. He did an autopsy after her death, and she had a tumor in her hypothalamus. And to this day, hypothalamic obesity is a thing that we have to deal with with people with tumors or other damage to the hypothalamus. It often causes extreme obesity.
- PAPeter Attia
And what's the nature of the obesity? How much of it is due to, um, hyperphagia, uh, y- excess eating? How much of it is due to loss of, uh, activity or even just a shutdown of metabolic rate?
- SGStephan Guyenet
Yeah, that's a great question. So if you look at probably the closest experimental analog of, uh, hypothalamic obesity that, uh, sort of human-like hypothalamic obesity would be, um, VMH lesion. So this is something that's been done since, I think, the '20s. They go in with a very precise instrument called a stereotaxic instrument in animals, and they lesion this part of the hypothalamus called the ventromedial hypothalamus. So basically, they're, they're replicating, they're trying to replicate the damage of hypothalamic obesity. And as soon as the anesthesia wears off, these animals are cramming food into their faces. And if there's no food in their cage, they'll eat bedding. They'll eat the bedding. They will just, like, put anything they can get ahold of into their bodies, and they will continue binging until they have rapidly gained a large amount of weight, and then it will start to plateau off. Um, but they have extreme hyperphagia. And if you restrict their calorie... The first experiments that were done on this showed that if you restrict them to a normal level of calorie intake, so that of a non-lesioned animal, it prevents the fat gain, suggesting that the... Or s- I should rephrase that. It prevents the weight gain, so they were just weighing them at the time, suggesting that it's primarily a phenotype of hyperphagia. However, later experiments, uh, that were more precise found that it, it doesn't completely eliminate the weight gain. It only eliminates about 80% of it. And so there is a component from energy, coming from energy expenditure. So, uh, primarily hyperphagia, but there's also an energy expenditure component that's smaller.
- PAPeter Attia
Yeah, uh, it's funny. I just remember in medical school, one of the neurobiology professors saying, "If you, if you were asked to give up a piece of your brain, like, the last one you wanna give up is a piece of your hypothalamus." If you, if you had to give up, like, one cubic centimeter of brain tissue, you just fight like hell to make sure you preserve your entire hypothalamus.
- SGStephan Guyenet
[laughs] I agree with that. I would say the brainstem and the hypothalamus, probably not places you wanna lose.
- PAPeter Attia
Let's go a little bit further in that. Um, we've... You, you referred to the lipostat. So say a little bit more about what, what we've learned about this, this lipostat and what circulating factors might play a role in governing this and what's been learned through some of the, uh, work that was done using parabiosis.
- SGStephan Guyenet
Yeah. So I'm gonna give you the simple version. There is a more complex version that is emerging, um, but I'll start with the, the simple version, which is that, um, it is a negative feedback system similar to your home thermostat, similar to the thermostat in your brain in that it, uh, there is Um, the hypothalamus measures levels of a circulating hormone called leptin that circulates in proportion to your body fat mass. So the same way your home thermostat measures the temperature in your home, your hypothalamus is measuring the lev-level of leptin in your circulation. Still some controversy about how exactly or where that measuring happens, but the signal gets to the hypothalamus. And, um, it uses that to determine whether you essentially have the amount of fat that your hypothalamus wants you to have. So the same way that your thermostat has a set point and, um, your internal thermostat in your body has a set point, your hypothalamus has a certain idea of how much fat it wants you to have on your body, and y- if you deviate from that, it starts to engage a coordinated series of physiological and behavioral responses to restore the previous level of body fat. And this system works better at protecting against fat loss than it does against fat gain. And certainly over long periods of time, we see that the average person in a country like the US tends to gain fat. The lipostat is not stopping them, or at least it's, it's not preventing them. It might be resisting, but it's not stopping the process of weight gain. Um, but we see that it actually is quite vigorous at defending against weight loss, and this is part of the cruelty of, or the unfairness of how obesity works is that your set point or your defended level of body weight, there's controversy about what, what to call that, but whatever it is, it goes up. And so a person with obesity, their body defends against weight loss as if they were starving, just like a lean person losing weight, their body and brain would defend against weight loss. And it's, it's literally a starvation response. It's the same behavioral and physiological process that ramps up your hunger, that makes you, uh, more focused on food cues, greater cravings. It, uh, down regulates your energy expenditure and, uh, you know, it does everything it can to try to bring the fat back. And so that is, I think, a key reason and possibly the primary reason why weight loss is so difficult and so temporary. If there was no regulation happening, weight loss would probably be pretty easy, and weight maintenance certainly would be very easy. But it's not. You see that people tend to regain back to their former level unless they're being really well supported in that weight loss, and even then, there's usually some amount of weight gain. Weight regain,
- 45:32 – 53:24
The role of the hormone leptin in obesity
- SGStephan Guyenet
I should say.
- PAPeter Attia
So let's go back to leptin for a moment. So leptin is a hormone. Is it made in the adipocyte?
- SGStephan Guyenet
It is, yeah. It's made in the adipocyte and secreted in proportion to body fat mass.
- PAPeter Attia
So the more adipose tissue you have, the more leptin you have. Is it generally pretty static? Does it change with meals or exercise or anything like that?
- SGStephan Guyenet
Yes, it does. So the-- So over the long run, the amount of leptin in the bloodstream is strongly correlated with fat mass. However, it's also strongly impacted by short-term energy balance. So if you, if you, uh, let's say, cut your calories by 25% for a couple of days, you're gonna see a drop in leptin that is disproportionate to your amount of fat mass.
- PAPeter Attia
And so where is the leptin receptor? Is there a leptin receptor in the periphery or are they central?
- SGStephan Guyenet
There are leptin receptors in many parts of the body, but the ones that are relevant for body weight regulation are in the brain.
- PAPeter Attia
Uh, are they in the hypothalamus?
- SGStephan Guyenet
There are-- There is a high concentration of leptin receptors in the hypothalamus, yes. Um, there are leptin receptors in other parts of the brain too. Um, however, it's not 100% clear what, where the important ones are for body weight regulation. Probably somewhere in the hypothalamus, but there are a lot of different papers where they take mouse models and they knock the leptin receptor out of different cells in the brain. And, um, if you knock them out of GABAergic cells, you get a really big... Basically, you recapitulate animals that don't have the leptin receptor at all in terms of their body weight. Um, GABAergic, that's a, a major, one of the two big neurotransmitters. That's the main inhibitory neurotransmitter in the brain. And then you can knock it out of certain subtypes, cell subtypes in the hypothalamus and get big effects. There was a paper suggesting you can just knock it out of AgRP neurons and get, um, and recapitulate the obesity phenotype of animals that have no leptin receptor. However, that, um, is controversial. Not every paper has shown that. So I'm not sure what people who are on the cutting edge of this field would say about where that evidence is. Um, but certainly there are cell populations, and probably the most important ones are in the hypothalamus. There are cells in the brain, probably mostly in the hypothalamus, that are receiving that signal and conveying it to the key cell types in-- that are kind of at the center of this, uh, lipostat, which are AgRP neurons I refer them in my book as NPY neurons just for storyline simplicity, but more commonly they're called AgRP neurons and POMC neurons. So the AgRP, those are the, the hunger neurons. The POMC are the satiety neurons. Or you could think of them as hunger/body fat increasing neurons. The POMC are the, the, uh, the opposite essentially.
- PAPeter Attia
And so what is leptin resistance, and how does it manifest? Why does it manifest? Um, and how frequent is it?
- SGStephan Guyenet
Leptin resistance, so when leptin was discovered first in 1994 by a team led by Jeff Friedman and, and Rudy Leibel, they found that this was the gene that was missing in an obese mouse model called the ob/ob mouse. This animal was extremely obese as a result of lacking of a defect in the production of, of this protein. One single amino acid substitution that destroyed this protein and caused, uh... Or sorry, one, one single base pair that destroyed this protein and caused a loss of function and this massive obesity. And when this was discovered, and also discovered that humans have leptin, then it was like this scientific bonanza. It was like, "Well, maybe we've discovered the cause of obesity. Maybe people with obesity don't have enough leptin, and so their brains, you know, think they don't have enough body fat when really they do." They... The failure to perceive the body fat, that's what causes obesity in these ob/ob mice. And so they started measuring leptin levels in people with obesity, and it turns out they were actually-
- PAPeter Attia
They were probably high
- SGStephan Guyenet
... elevated.
- PAPeter Attia
Yeah.
- SGStephan Guyenet
Yeah, because it's correlated, we now know it's correlated with fat mass. And so what, you know, what's the deal? If this is a hormone that regulates body fatness, why is it that people with obesity have so much of it and it's not suppressing their excess body fat mass? And the concept that has been, um, invoked to explain this is leptin resistance. So in the same way that people can develop insulin resistance where it takes more insulin to do the physiological jobs in the body, people with obesity, um, require more leptin for the hypothalamus to be satisfied. So another way to say that, they require more leptin to avert the starvation response that the brain has that, uh, where alarm bells start going off because it thinks you don't have enough body fat. So they require more, more leptin to achieve that state. And so we call that leptin resistance, but we don't really know how it works yet. So, you know, that's just a general term for requiring more leptin to avert the starvation response, but we don't know whether that is something where there's fewer leptin receptors on a certain kind of cell, whether there's a downstream signaling impairment in intracellular signaling cascades, whether there is a change in cell-to-cell communication. Maybe the cell that's receiving the leptin is getting the message just fine, but there's some kind of downstream change in neural processing where the signal gets, gets clouded or modified. We don't know the answer to that question.
- PAPeter Attia
Yeah, it seem, it seems like the only thing we know is it's not too low an amount of circulating leptin, as evidenced by two things. One, the high circulating [chuckles] levels of leptin, and the fact, I think more importantly, that when you give exogenous leptin, it doesn't improve the condition, suggesting that that's not the defect.
- SGStephan Guyenet
Yes. I mean, you can give high levels of leptin and it will cause weight loss, but it doesn't do much, and if you look at leptin signaling, there's, there were some early studies done in, in animal models suggesting that if you're just looking broad... If you mash up the hypothalamus and you look at what's going on in it broadly, on average, you find that the amount of leptin response, the, the intracellular signaling cascade that's activated by leptin is not really impaired in animals with obesity. It's like, it's like they're getting the same leptin signal from a much higher level of leptin. So that, that's kind of like some evidence that we have.
- 53:24 – 1:05:48
The genetic component of obesity
- PAPeter Attia
I looked at some twin concordant and discordant studies, identical twin, and I was surprised to see, maybe I shouldn't have been surprised, but I was surprised at how heritable obesity was. It was about 0.7. Um, and that's, I mean, when you see heritability of 0.7, that tells you something is very, very genetically predetermined. So even though 100 years ago virtually none of us were obese, and today let's just call it your lifetime incidence of obesity is 50%, tells you... And our genes haven't changed in 100 years, so clearly our susceptibility for obesity has been with us for a great period of time, and it is highly, highly preserved. It's just that in the last, whatever, 40, 50 years, we now have matched or mirrored our genes to an environment that is allowing that trait to flourish. What do we know about the genes that, that regulate obesity? Or fatness. Let's just talk about it through that lens, I suppose.
- SGStephan Guyenet
Yeah. The meta-analysis of twin studies that I like to, um, cite these days suggests an average heritability of 75%.
- PAPeter Attia
Wow.
- SGStephan Guyenet
So-
- PAPeter Attia
That's even stronger. Yeah
- SGStephan Guyenet
... it's, it's massive. And, and there's some debate about that.
- PAPeter Attia
Yeah.
- SGStephan Guyenet
You know, I think
- PAPeter Attia
But, but directionally this is, this is a really big deal
- SGStephan Guyenet
It's very heritable, yeah
- PAPeter Attia
Yeah
- SGStephan Guyenet
And a lot of things are very heritable. I think that's one thing we're learning. Um, but yeah, so you have this, this very high heritability of body mass index. So, um, variation between individuals and body mass index, about 75% of those differences between people is explained by their genetics. That's what that implies. And, um, if we look at other methods that have tried to figure out what are the genes that underlie this, what are the genetic differences, these are the genome-wide association studies that I think are particularly informative in this regard. They simply ask the question, if we look at the entire genome and we look at these representative genetic markers where different people have different, uh, genetic code called SNPs, single nucleotide polymorphisms, where in the genome, what markers correlate with differences in body mass index? Fortunately, body mass index is real easy to measure, so you can get really big sample sizes in these studies, which you need to get statistically significant results, because you're looking at, I think, like millions ... I don't remember exactly how many, but you're looking at a lot of genomic markers, so you need tremendous statistical power to detect anything with, with a high level of confidence. So you have these studies, I think the latest is, like, 800,000 people.
- PAPeter Attia
Hm.
- SGStephan Guyenet
Um, the leader of the pack is, is the height genome-wide association study. I think they have, like, 3 million. And now they've saturated the heritability. They've gotten all the information they can with that sample size out of what the common genet- genetic variants are that correlate with differences in height. So I think with body mass index, we may, in the near future, we may saturate it as well. We may know what are all the common genetic differences that correlate with differences in, in body mass index. So, um, essentially, so far these studies have identified I think, like, 900 variants that differ. So what this suggests is that differences in body mass index between individuals are very complex, that genetically very complex. They're determined by a lot of different genes with very small effect sizes. So you get this sorting of all these different genes, and whatever combination you get, lucky or unlucky, determines whether you're ... to a large degree determines whether you are susceptible or not susceptible to obesity in a, in a fattening environment, is, is the way I would put it. And, um, so they have various ways of looking at what these genes are doing because, you know, that's one way you can use these genome-wide association studies that's particularly informative is you could say what's the underlying biology that makes some people fatter and some people slimmer. And I wanna talk a little bit about why this is such a important approach. One is that you're looking peop- so you're looking at people in their regular everyday context. This is not an artificial lab scenario. You're just looking at people living their lives and experiencing higher or lower weight, and you're saying what genes correlate with that, so it's very naturalistic. Second, it's very replicable. These studies are highly replicable. In other words, if you do three studies of this nature, you're gonna tend to get similar results. So the methodology, it's one of the most rigorous, I would say, in the biological sciences that we have. And the third-
- PAPeter Attia
And I think that ... Um, sorry, I'll let you finish because I was gonna add something to that, but go ahead
- SGStephan Guyenet
Um, and the third one is that it's unusually objective as well. It has a higher level of built-in objectivity, resistance to bias compared to other types of investigation because it's not hypothesis driven. You're just looking across the whole genome and seeing what pops up. You're not saying, "I'm gonna focus on the connection between X biological process and Y outcome." You're just saying, "I'm interested in Y outcome, what correlates with it, and let's see what biology pops up. Could be anything. We're just gonna see." And so that really gives you a chance to, I think, check your thinking on what the underlying biology is in various traits and, and diseases.
- PAPeter Attia
Yeah, and I think, I think part of that comes from the strength of what ultimately makes genetic analyses like Mendelian randomization so powerful is the genes are randomly distributed. So y- that's what cleans out some of those biases, is when you are looking at a million people for, for whom the genes are randomly spread across them, and you take an unbiased view of the sample, and then you get those results over and over and over again, I think it becomes very powerful. And, and look, y- you know, if people are listening to us saying, "God, what are these guys talking about?" I mean, I think it's just important to understand the big picture here. The big picture here is 1,000 years agos, 1,000 years ago, to all intents and purposes, none of us were obese. But that still means directionally 50% of us at least had the genes that would allow us to become obese in an obesogenic environment. That's really all that ... That's, that's really what we're explaining here, is that there are a highly, highly heritable set of genes that will allow a subset of the population ... And actually, one of the things I'm just curious about your thoughts are teleologically why is it f- 50%? Why isn't it 100%? Right? Like why is it that more people don't have that? Was this just a fluke of evolution? Was this... You know, you would almost think that evolution would've wanted everybody to have those genes.
- SGStephan Guyenet
Yeah, absolutely. So I'm going to finish up the thought we were on previously just to wrap that up, but I would like to get back to that. So we have this incredible methodology for looking at... You know, and just to step, since you want to step back a little bit, I just want to also add like what we're talking about is why some people can effortlessly stay thin and other people have to really struggle to maintain their weight and maybe are not able to. So that's kind of like the everyday-
- PAPeter Attia
Yep
- SGStephan Guyenet
... thing that we're trying to explain here that I think many people recognize intuitively that that is a thing, that different people have different propensities for becoming obese or developing obesity or not. And yeah, so we can look at the underlying biology, and that's been done, and there are a couple different ways you can do it. One is you can say what are the genes that seem to be associated with these genomic differences and where are those expressed? What tissues are those expressed in? And there was a paper where they looked at I think 43 different traits of all kinds, diseases, personality traits, other stuff, and they asked what does the tissue enrichment look like? And if you look at body mass index, it looks like psychiatric diseases and educational attainment. So all of those are heavily enriched for brain-related genes to a similar degree. So conditions that we know are related to the brain like educational attainment, how many years of education you've attained, whether you are susceptible to schizophrenia, depression, Tourette's, like all these brain-related conditions, obviously the brain shows up in genome-wide association studies for those conditions, and you put those next to body mass index and you couldn't tell them apart. Like that's how heavily enriched for brain-related biology body mass index is.
- PAPeter Attia
And those diseases, by the way, that you just mentioned are some of the most heritable diseases we see in medicine. I mean, when you look at autism, when you look at schizophrenia, these have heritability indexes of 0.6 to 0.7. They're highly genetic conditions. So there's two things going on, right? Which is you have these parallel things that are highly, highly genetic and then they're disproportionately concentrated in the brain.
- SGStephan Guyenet
That's right. And so we know it's... And I don't want to say that it's literally 100% about the brain. I think that's unlikely to be true, but it's certainly the primary signal that emerges across the literature. And so I think that really validates this idea of the brain being important for body fatness. And if we look a little bit deeper at what is going on, for a lot of it we don't really know. We don't really know exactly how the brain is doing this, what it is about these genes, but we can see that it correlates with certain types of ways of interacting with food. So people that have obesity-promoting genes tend to have greater eating drive. They tend to have lower satiety. But this is an area that hasn't really been very well explored yet, so there's a lot we don't know. However, if you look at the monogenic obesity syndromes, so where there's one mutation that causes severe obesity, those really revolve around the leptin brain signaling axis. So those mutations tend to be in leptin, the leptin receptor, melanocortins, melanocortin receptor that are downstream of leptin in the brain. And those types of signals also show up in the genome-wide association studies, but they're not dominant. A lot of this stuff is really general. It's like-
- PAPeter Attia
Yeah
- SGStephan Guyenet
... stuff that affects general neuronal development and neurotransmitters that are involved in a lot of stuff. So I think there's a long way to go before we really understand exactly how those genes are affecting the brain in a way that impacts body fatness. But I do think we can say that differences in body fatness between individuals are primarily determined by differences in how the brain is constructed and how it operates.
- 1:05:48 – 1:13:32
Understanding the tendency of humans to store fat through an evolutionary lens
- PAPeter Attia
So Stephan, let's now go back and try to put all of this in the context we were just ready to get to a moment ago, which is it's 250,000 years ago. For all intents and purposes, we're the same creatures we are now, obviously minus the environment that we live in. But food and energy are one of our top priorities, right? Like most, I'm guessing, I'm not an anthropologist, but it would have to seem to me that security from other tribes and animals and the environment, right, weather, acquisition of energy and reproduction were kind of the only things that would've mattered, right? There probably wasn't a lot of other stuff that mattered at the time. And acquisition of energy was essential in that it could kill you very quickly if you failed to do that. So acquiring energy, storing energy was like kind of the struggle that defined us. Probably in the short term much more so than, uh, reproduction, which obviously is a huge other contributor here. So w- we evolved over millions of years, and everything you said about leptin now starts to make sense in that environment, right? Leptin is a signal that says there's not enough energy, and that's what should really trigger the response. So in that sense, it's not surprising that leptin isn't doing the opposite. It's not surprising that high leptin doesn't make you wanna stop eating. It's... Who cares? N- nature wouldn't have cared about that. But it certainly would care if leptin gets too low. That should be a screaming signal to go and eat, resist that, resist that sign. What do we know about the genetics, the efficiency with which we store energy? I mean, we haven't really talked about that, but this, this, this ability that we have to get fat is kind of a remarkable thing, right? I mean, we don't really store carbohydrates. We can't really store protein, and we don't wanna be breaking down muscle to get amino acids. So we do really have to rely on this ability to store fatty acids and excess carbohydrates as fatty acids in a relatively inert structure of white adipose tissue, yeah?
- SGStephan Guyenet
Yeah, absolutely. Uh, you know, I think Herman Pontzer would be a great person to talk to about this. He-
- PAPeter Attia
Yes, I, I definitely lo- I... His book is on my list to read, and I definitely plan to have him on to, to get into this.
- SGStephan Guyenet
Yeah, he has some good thoughts. Jon Speakman has some good thoughts on this as well.
- PAPeter Attia
Another person I should probably have on the podcast.
- SGStephan Guyenet
Yeah, so I think there are good reasons to have a certain amount of body fat, and I think, you know, some of them, you know, the basic idea is pretty obvious. You want to have a way to cover your energy needs between eating opportunities. And, um, you know, we have other energy reserves. We have glycogen, but they're just far more limited. The thing that's awesome about fat is, first of all, it's a very concentrated source of energy, right? Like dietary fat is nine calories per gram, carbohydrate is four, protein is four.
- PAPeter Attia
Yeah, it's anhydrous. There's no water. It's-
- SGStephan Guyenet
Exactly
- PAPeter Attia
... like literally just pure energy.
- SGStephan Guyenet
That was the second thing I was gonna say is that it's hydrophobic. And so you can store it without having to hydrate it like you do with glycogen. Glycogen, the weight of glycogen I think is mostly water.
- PAPeter Attia
Three, three or four to one water.
- SGStephan Guyenet
Okay, there we go. And then the weight of adipose tissue, even if you include all the interstitial stuff and all that, I think it's like 85, 90% pure fat. So the energy density is just off the charts depending on, you know, uh, relative to any other storage method that the body has. And so it makes sense that that's kind of our long-term energy buffer.
- PAPeter Attia
By the way, just for people who think about EVs and stuff, like there's no battery that can come close to the energy density of our fat, just to put that in perspective, right? Like, or any hydrocarbon for that matter.
- SGStephan Guyenet
Yeah, so, um, and so I think the importance of that is obvious to have a way to cover times when you don't have as much energy coming in as you would like. And in the evolutionary context, you know, the thing that comes to mind from our modern perspective is whether they find food or not, but there's also the question of illness, and I think that's a really important one. So if we look at, um, the primary causes of mortality in children under five in low income settings, what we see is that it's strongly related to their weight for height, which is kind of a different way of measuring BMI. And, um, and it's also strongly related to disease pressure, especially diseases like diarrhea that interfere with nutrition. And so if you look at the correlation between, uh, weight for height and mortality, there's a massive correlation. So kids who have malnutrition, moderate or severe malnutrition, that's what we call being underweight to a certain degree, they have massively increased mortality 'cause basically if you don't have those energy stores, you can't defend yourself against infections. And so it's not just about energy. You know, there are other nutrients that are important, vitamin A and, and some other things, but energy is, is huge. And so there's this... Because it's such a huge s- uh, source of mortality, especially in kids, there's this massive selectiv- selective pressure to maintain a certain amount of, uh, energy storage in the body. So that would be, you know, just an example of a selective pressure that would, um, select for a certain amount of body fat. And it is interesting in this regard that humans have a lot more fat than our closest, uh, primate relatives. So chimps are like mid-single digits fat, and they don't develop obesity. They cannot physiologically develop human-like obesity i- is my understanding. Um, so we're kind of special physiologically in-
- PAPeter Attia
Hmm
- SGStephan Guyenet
... our capacity for fat storage.
- PAPeter Attia
Has anyone ever looked at different, uh, ancestral populations for w- And this might be just irrelevant to do because we don't have enough data, where there were different amounts of food scarcity and seeing if there's an inverse relationship between the food scarcity that that population emanated from, whether it's this part of Africa versus that part of Europe, and how that translates into the, um, genetic predisposition to obesity in their modern ancest- in their, uh, modern kin today?
- SGStephan Guyenet
Yeah. I don't know the answer to that
- PAPeter Attia
But you understand the question I'm asking? Like-
- SGStephan Guyenet
Yeah, yeah
- PAPeter Attia
... yeah
- SGStephan Guyenet
It's basically d- does the history with starvation select for more obesity types-
- PAPeter Attia
For even greater obesity today. Exactly
- SGStephan Guyenet
Yeah. I know that John Speakman has argued against this idea. He has pointed out that apparently people with obesity do not survive famines better than lean people, um, which is kind of counterintuitive. I'm not sure why that would be. I know that's a point that he's made. Um, yeah, so anyway, that's about all I know about it. I, I don't really know the answer to that question.
- 1:13:32 – 1:26:46
The hedonic aspect of food, and how the brain reacts to modern, highly-rewarding foods
- PAPeter Attia
Okay. So let's start to talk about [clears throat] the hedonic aspect of food. So we have five tastes, right? We can taste sweet, sour, bitter, salty, and umami. Those are the things, the five things we taste. Um, what's the best way to describe umami to somebody?
- SGStephan Guyenet
It's a meaty flavor that is present in cooked meat, bone broth, soy sauce.
- PAPeter Attia
Mm-hmm. Yeah, yeah. Okay. So I think people kinda get it. It's distinct from salt, I think is an important point here. Um-
- SGStephan Guyenet
Yes
- PAPeter Attia
... although they often go together, of course
- SGStephan Guyenet
Yes. Mm-hmm.
- PAPeter Attia
S- And I know different parts of our tongues have different, you know, uh, we, we sense, for example, sweet on the very front of our tongue, I recall, but that's, [chuckles] that's about the extent of my knowledge and recollection of where, where our tongue resides in that. But what do those... What does the taste that we experience today, you and I, if we go out to a restaurant, how does that compare to the range of taste that our ancestors experienced?
- SGStephan Guyenet
Yeah. I think, I think this is, this brings up a topic that, um, I like to, I like to talk about because I think most people, i- if you look at what hunter-gatherers actually eat, let's say we're, we're looking at contemporary and historical hunter-gatherers where data have been collected and using that as a proxy for the types of food that our ancestors would've eaten, it is radically different than, than what we eat today in many ways, but one of them is the hedonic properties of it. Like, if you look at what the Hadza eat, you know, if they go out and kill, uh, an antelope, they just, like, cut off pieces of meat and throw it in the fire or put it next to the fire in coals. They don't have sauces. They're not putting salt on it, and then they just, like, cut off the charred parts and eat, and sometimes it's, like, half raw on the inside. They eat rotten meat, meat that we would consider literally rotten, they will eat.
- PAPeter Attia
I, I was just thinking about that the other day, by the way, because I eat so much wild game that I've killed, but I realize, like, I'm still a baby because, like, A, I cook it. I don't eat it raw. Um, but more importantly, to your point, I season it, right? Like, I use salt. I use pepper. I put lemon on it, and I'm sure it would taste fine without those things. It would certainly be edible. I just don't do it, right?
- SGStephan Guyenet
Yeah, and that's right. And, you know, I, I'm sure, like... Yeah. I think if a person, a typical person were to try to eat at a Hadza camp for, like, a week, I think it would be really challenging for them. Um, you know, like, even that meat, even, even meat, as you said, cooked unseason- unseasoned, it's not... It doesn't taste bad. It wouldn't taste as good. But now imagine, like, the outside is charred. The inside is half raw. You're brushing sand off of it.
- PAPeter Attia
[chuckles]
- SGStephan Guyenet
Like, [chuckles] that's, that's kind of the context. And then, you know, that's the meat. That's, like, one of the most palatable things they eat. And then, you know, we have, like, probably the most palatable thing or, yeah, probably the most palatable thing and certainly the thing that they really like is honey. But they're not putting it on toast with butter on it. You know?
- PAPeter Attia
[chuckles]
- SGStephan Guyenet
They're literally drinking it-
- PAPeter Attia
With a little jam
- SGStephan Guyenet
... straight. They're just taking the honeycomb, eating it, and, like, drinking the honey. So it's, it's a bit of a different scenario even there. And then we have, like, they eat a lot of baobab. That is a very fibrous fruit that has sweetness to it, but it also has some off flavors, and it's not a very sweet fruit. It's not like an apple. It's, uh, got a lot of fiber. And then there's the, uh, the tubers, which is another major article of diet. And to be fair, this is their least preferred type of food.
- PAPeter Attia
Oh, it is? Okay.
- SGStephan Guyenet
[chuckles] Yeah. It's, it's... They would rather eat things other than tubers. Um, but they roast these things in the fire. They dig them up. They're these, like, long, stringy things that look like long sweet potatoes. There, there are multiple species, but, um, that's one of the common ones. And they're so fibrous that they actually have to spit out a wad of fiber after they're done chewing it.
- PAPeter Attia
So it's like a sugar cane or something where you sorta-
- SGStephan Guyenet
Yeah
- PAPeter Attia
... suck out the nutrient, but you're spitting out the, the pure-
- SGStephan Guyenet
Yeah
- PAPeter Attia
... uh, un- insoluble fiber.
- SGStephan Guyenet
Yeah. And so they're not sitting there sautéing onions on the stove. They're not putting sauces. They're not spicing. Like, they're just taking food out of nature and cooking it and eating it. And so I think it's, it's just a radically different type of diet than we're accustomed to. And I think in this regard, um, it's interesting to consider how reward s- circuits adapt to that. Basically, our brains are set up to not be satisfied with ordinary stuff once we have gotten good stuff, is a simple way to put it. And, uh, Michael Krashes did a really interesting neuroscience study on this in mice. Mice, normally they eat these unrefined food pellets. That would be like the default diet. But they much prefer these calorie-dense, refined, high-fat pellets. And if you give that to them, they will very much preferentially eat that over the healthier, unrefined pellets. And what happens is they actually neurobiologically devalue, if you look at the circuits, activity of their reward c- circuits, they devalue, once they've been exposed to the preferred food, they devalue the less preferred food. So it no longer satisfies them, no longer motivates them in the same way that it did before they were exposed to the highly preferred food. And so, you know, to bring this back to our context, if you have somebody like you or I who's been raised in a context where we have, you know, tasty, calorie-dense, easy-to-eat food and that's how we were raised, then going back to eat food more like how our ancestors would eat is, is really difficult. Like
- PAPeter Attia
Like, now twice you've mentioned the calorie density, so let's now talk about that because that's kind of different from taste, right? So the, the, the taste thing is interesting to me and, and I... This is something you and I actually remember speaking about probably one of the first times we met, this is, like I said, 10 years ago, which was I actually think table sugar is disgusting. Like, I truly do. Like, if you put a bowl of that white crap in front of me and said, "Dip your finger in and eat it," I could maybe do it once, but that's about it, right? If you said, "Just mix it into water and drink it," it's gross. It tastes fine in coffee and tea, but just by itself it's really disgusting. And similarly, if you just have me eat lard, it's really disgusting. Like, I, despite the fact that I was on a ketogenic diet for three years, I never developed a taste for putting coffee and butter and things like that. But I freaking love ice cream.
- SGStephan Guyenet
[chuckles]
- PAPeter Attia
Right? Like w- I think ice cream is about one of the most beautiful tastes in the world, everything about it. So... And it really isn't that much more than sugar and fat, right? I mean, yes, there's some flavors to it, and if you make it a coffee ice cream I like it even more. But, so, so I guess my question is, can you walk me through what my brain is doing when it's tasting sugar, when it's tasting butter, neither of which by itself I find remotely enjoyable, but then when I'm tasting ice cream? And, 'cause the ice cream's not that much more caloric than the butter, and the taste of those two things... You know what I mean? Like, there, there's something I'm trying to understand here, which is taste-
- SGStephan Guyenet
Yeah
- PAPeter Attia
... and energy density, and how are those figuring out? Because I believe they are instantly rewarding. So I'll give you one other analogy here. I remember when my daughter, who's now 13, turned six months old, and my wife and I were really fastidious about not feeding her any junk, right? Like, you know, fortunately my wife was able to breastfeed, so she didn't have all that formula. And, you know, we just, we were like, you know, your typical idiot first parents, right? Like, we, we spent way too much time thinking about what she was eating, I'm sure. But on her six-month birthday we got her ice cream.
- SGStephan Guyenet
[chuckles]
- 1:26:46 – 1:35:16
How we are hard-wired to think about food
- PAPeter Attia
Let's go back to our ancestors for a moment. Um, what apparatus was at their disposal subconsciously or consciously to help them understand and prioritize calorically dense food? Because I, I gotta believe that the three things that mattered most, correct me if I'm wrong, would be total calories, protein, and sodium. I mean, it can't be an accident that sodium is the only mineral we can taste, right?
- SGStephan Guyenet
Yeah. Correct. So, um, or I should say, by correct, I mean that is how I think about it as well. Um, and to break down the energy piece that would come for, for an animal with a digestive tract like a human, that would come primarily from carbohydrate and fat. So we have carbohydrate, fat, protein, uh, salt, and then sometimes I add glutamate umami to that list as well. And essentially-
- PAPeter Attia
And this is subconscious, Stephan? Like were we... Or is, is this again just part of that stuff that was now so wired into us that-
- SGStephan Guyenet
Yes
- PAPeter Attia
... you know, we didn't wanna eat grass. Like, we knew that even though you could sort of get the gastric distension from eating a lot of grass like a cow could, it was doing nothing for us. Both it didn't taste good, a- and so in the short term it wasn't pleasing, and in the long term obviously it didn't satiate us.
- SGStephan Guyenet
Yeah, that's right. And I think, you know, there's a couple angles on this. One, obviously humans have culture, so we figured out, you know, what foods are good over long periods of time. But a key aspect of this is dopamine mediated reinforcement, and essentially our bodies are set up to respond to certain types of nutrients like the ones you mentioned, and create a motivation and learning response that prioritizes and sets the motivational level on the seeking of those types of foods. And so presumably these are the kinds of nutrients that, um, our ancestors would have needed to prioritize to maximize their reproductive success, the currency of natural selection. So essentially we have these motivational systems that were selected to seek certain types of nutrients in the environment. And if you look at the, the modeling that's been done on foraging behavior in a wide variety of animals and in humans, you see that it revolves around maximizing the energy return rate of foraging. This doesn't describe every species, but it does describe many species.
- PAPeter Attia
It's amazing to watch it in big cats, for example, where they'll be chasing an antelope and it's literally almost like they have a sensor inside that says, "I'm gonna stop chasing now because my energy cost is not going to be met by my consumption over this period of time."
- SGStephan Guyenet
Absolutely. And these animals, you know, they don't know how to do math. They don't know that they're actually implementing a mathematical equation in their head.
- PAPeter Attia
Yeah.
- SGStephan Guyenet
But they are. It's just wired into their brains the same way it's wired into us. You can predict hunter-gatherer foraging behavior to a surprisingly, um, to a surprising degree just by knowing the calorie return rate of different foraging options. So our brains are very much wired, not just our brains but our bodies are very much wired around energy acquisition in terms of how our motivation and learning is set up on a non-conscious level. And, um, this is very much hard-wired. So we have dedicated sensors in the digestive tract. This, this is all pretty recent research since 2018. Um, they discovered these cells that they named neuropod cells in the small intestine primarily that have, um, receptors for specific nutrients that are directly, these cells are directly hooked up to vagal neurons. So when they detect glucose or amino acids, fatty acids, so that would be carbohydrate, fat, protein, they get the concentration and they start sending signals up your vagus nerve up to your brainstem, and from there it gets distributed to many parts of your brain. But, um, particularly relevant part is the parts of your brain that, uh, have to do with dopamine release onto your reward centers. And if the food that you're eating contains a high concentration of these valuable nutrients, particularly in combination with one another, you're gonna get a higher level of dopamine release. The more dopamine release you get, the more of a motivation you will develop toward that food.
- PAPeter Attia
Why is it there, there are, and maybe this is just me, I don't know what the literature would say, so th-this, this could be incorrect, but in me, [clears throat] like a ribeye is not something I seem to be able to eat into excess, and I feel like I should. Like, shouldn't I be wired to eat ribeye until I can't stand? Like, shouldn't I be wired to eat ribeye until the point of vomiting given how high it is in sodium, fat, and protein, and total calories? Like, the only thing it's missing is sugar and, and fiber and, you know, carbohydrates and things like that. But, but it's easier for me to overeat baked potatoes than it is to overeat a ribeye, um, and I'm not sure I understand why.
- SGStephan Guyenet
So let me, let me just clarify. With, with the potato, is that with or without toppings?
- PAPeter Attia
With. Let's say with. So yeah, yeah.
- SGStephan Guyenet
Okay.
- PAPeter Attia
So now let's, let's put on-
- SGStephan Guyenet
Okay
- PAPeter Attia
... let's put on butter, uh, sour cream, and salt. So, so I'm clearly making it much more than the carb, of course.
- SGStephan Guyenet
Got it.
- PAPeter Attia
Um, and it has to be crispy skin too, like if I'm gonna do it right.
- SGStephan Guyenet
[laughs]
- PAPeter Attia
You know, I can't be like some lame-ass buffet baked potato. It's gotta be my style. Um, but-
- SGStephan Guyenet
Yeah, that's an in-
- PAPeter Attia
But I don't know why, I, I, I don't know why, like, a fatty piece of meat is not something I have an amazing... Is that, is my experience typical, or would most people be able to just eat ribeyes until they puke?
- SGStephan Guyenet
Oh, that's a good question. I, I really don't know. I will say that when you look at, um, the foods that people cite as the most, um, typically associated with strong cravings and loss of control over eating behavior, meat does not usually come up high on that list. And so-
- PAPeter Attia
Which seems like it should, right?
- SGStephan Guyenet
Yeah, I mean, I can only speculate about that. Um, I'm not really sure why... I can understand where you're coming from. I don't know whether that's a kind of generalizable phenomenon. Um, I can only speculate about why that might be. So there are a couple things that come to mind for me. The first is that meat is about 75% water, so the calorie density of is actually, it's not low, but it's not especially high unless you're eating a really fatty piece of meat. So that's one thing. A-if we're comparing it to something like a brownie or, uh, something like pizza, which is more calorie dense than the steak.
- PAPeter Attia
Got it. Yeah.
- SGStephan Guyenet
Um, the second thing is it doesn't have any carbohydrates, so it doesn't have that fat carbohydrate combination that is most closely associated with foods that people lose control around. The third thing I would cite is the high protein level. So even though we have this strong protein specific appetite that's been demonstrated in many different species, protein doesn't work the same as carbohydrate and fat. And I think, I think we, we recognize that that's the case. Protein seems to... It's something that our bodies really wanna get enough of, but don't wanna get too much of. So there's really a, not only there's a drive to acquire it, but there's a drive to keep it within a certain range and not eat too much. And we see that, you know, if people go on high protein diets, their overall calorie intake will
- 1:35:16 – 1:50:06
A review of the “Carnivore diet”
- SGStephan Guyenet
drop.
- PAPeter Attia
Yeah, I mean, we're... I wanted to talk about the carnivore diet with you a little bit because I know you guys did a review on a book. Um, it's not a diet I've spent any time really thinking about, so I've basically spoken to, I don't know, a dozen people who have gone on it, and without exception, they all lose weight, um, which I think for some of them is their motivation for doing it. Um, and it must simply be that they just get tired of eating, right?
- SGStephan Guyenet
[laughs]
- PAPeter Attia
They just don't, they just can't take in the number of calories if they're doing it in that format.
- SGStephan Guyenet
I think there's probably... So I would say that, you know, we don't have any good data on, um, the impact of carnivore diet on, on weight, so, you know, there's no randomized controlled trials. But, you know, we have these anecdotes of people saying they lose a lot of weight. I certainly don't dispute that. Um, but I think if you came to me with this diet on paper and you asked me, "Would this cause weight?" I would say absolutely, because it has multiple properties that I would expect to make it a particularly effective weight loss diet. One, this is something we could talk about more if you want, but it has zero carbohydrate, so I think that anywhere, if, if you're on the extreme of the fat to carbohydrate, uh, ratio, in either direction, that's more slimming than being in the middle. So the most fattening diets are rich in both carbohydrate and fat. So there's zero carbohydrate. You're on the extreme... Or I shouldn't say zero. Very, very little.
- PAPeter Attia
Right.
- SGStephan Guyenet
So you're on the ex-
- PAPeter Attia
Outside of the glycogen in the meat.
- SGStephan Guyenet
Yeah.
- PAPeter Attia
That's about it. Yeah.
- SGStephan Guyenet
There's a little glycogen. So you're on the very extreme end of, uh, the macronutrient distribution. It's high in protein. That's also known to contribute to weight loss. You're eliminating almost every type of food, so your variety, the variety of your diet goes very low. I mean, you can prepare your meat in different ways. You can eat chicken or fish or beef or whatever, but your variety is greatly, greatly reduced. So that's, I think, part of it. Um, and you're cutting out all of these highly processed calorie-dense foods that are the foods that I think, you know, we could debate about the why, but I think everyone agrees that those are foods that drive excess intake and elevated body fatness. So I think all these things together, it's just even on paper, it's a diet that I would very much expect to cause weight loss more to greater degree than your average diet.
- PAPeter Attia
And while we're on the topic, um, what are... I mean, tell me a little bit about your review of this, because I know you've put some time into this. Uh, and for folks that are wondering, you know, I, I don't really plan to do a, do a podcast on the carnivore diet. I, um, I, I... Doesn't seem to make a lot of sense to me, although I don't dispute that there are people who I think have had very, um, successful outcomes on it with respect to dealing with some of their physical ailments. Um, so maybe it has a role in sort of overcoming some acute illness. Um, and maybe I'm just biased towards thinking plants are valuable, but I, I guess I, I... It seems to me that one of the core tenets of the, of the diet is that plants are sort of low, low-grade toxic, right? I mean, isn't that sort of part of the thesis?
- SGStephan Guyenet
Yeah. The thesis is basically everything is toxic except grass-fed animal foods. [chuckles] Um, so, like, even tap water is considered, uh, not optimal. But, um, yeah. So the, the book spends a lot of time, like, going through the litany of all the potentially harmful compounds in, in plant foods. And actually, you know what? I, I sympathize with some of this. I think there is a bias toward thinking if it's in a plant, it's healthy, and I don't think that's true. I think the book is right that that's not necessarily true, and I think there are some plant compounds that, at least for some people, are not so good. And, you know, there are well-characterized examples of this, like if you eat a lot of spinach, you can get kidney stones from all the oxalate. Um, there are studies suggesting that the glucosinolates in, um, cabbage family plants might contribute to type 2 diabetes. Uh, you know, there's, like, kidney beans. If you don't cook them enough, th- they can be really toxic because of the lectins. So it's not like there aren't examples of this. It's, it's definitely true that to some degree, I think it just gets taken far beyond where the evidence is. And, you know, the way to think about how healthy a food is is not to say does it contain toxins. It's to say what's the cost-benefit analysis on this food, and most importantly, what are the empirical outcomes that we can see when it, when it's directly... when its impacts on health are directly studied. So, you know, like, this is something that I've kind of focused on in my evaluation of, uh, some of the ideas that are put forth in the public sphere, is that a lot of people with, um... a lot of people who are coming out with, uh, let's say, unusual ideas in this sphere, they really... they take a mechanism, and they run with it. Like, you know, X toxin is really bad, like lectins, for example. Gundry, "Lectins can do XYZ. Lectins are in plants, therefore we shouldn't eat these types of plants." And that's really, like, a bottom-up approach, like extrapolating empirical effects on health from mechanism, when really, I think in a complex field like nutrition, it's better to start with the empirical evidence like, "Oh, we have this study that suggests that there's actually an effect on health. Let's see if we can understand the mechanism."
- PAPeter Attia
What are some of the, um, biochemical changes that occur in people on a, a carnivore diet? I mean, the obvious one must be the dyslipidemia, right?
- SGStephan Guyenet
Yes. So, you know, there is, there is a shift, uh, toward a ketogenic metabolism because of the fact that it's very low carbohydrate. Um, that would be an obvious shift that occurs. Um, there... You know, I w- I don't know if I'd use the term dyslipidemia, but one of the terms that... One of the potential downsides I focused on in the review that is, um, downplayed by many carnivore diet advocates, including Paul Saladino, is, um, the change in LDL cholesterol and LDL particle count. So... And again, we don't have great evidence here, and this is kind of the crux of our review of the book on Red Pen Reviews, is simply that there are a lot of claims made that are not supported by any kind of, uh, convincing evidence. But, um, we have some evidence. So there's this survey study that was done on something like 2,000 carnivore dieters. Um, I think David Ludwig was involved in that. And they just reached out to people in social media groups, like their Facebook groups and things, and they administered this survey. And one of the questions was: What was your various blood lipid values before and after this diet? And you see that there are changes in positive and negative directions. Triglycerides go down as, as you would expect. Um, I don't remember what HDL did, but it probably went up. Yeah. And then there was a large increase in LDL cholesterol, and so that's a concern, you know, as far as I'm concerned, and I think you would agree. And Paul Saladino himself, you know, I'm, I'm not trying to pick on him. I don't wanna make it personal, but he's, uh, you know, been public about some of this stuff, so I, I think it's, it's fair to point out, to, to just repeat what he himself has said. But his LDL cholesterol is 533 mgs per deciliter, and his LDL particle count is also absolutely through the roof. And, um, so it... Not everyone responds like that. Like, if you look at the, the survey data, I think there was a mean increase of, like, 30 mgs per deciliter in LDL, 30 or 40, something like that. Um, so I think it depends on the individual. I think some people... Like, Shawn Baker's lipids are fine. He's another carnivore diet guy. Last time I saw, his lipids looked just fine. So I think it depends on the individual. Um, but some people do experience a large increase in LDL cholesterol, and, I mean, you know more about this than I do, but that certainly raises red flags for me in terms of cardiovascular risk over the long run.
- PAPeter Attia
Yeah. The thing I've never understood is, um, [clears throat] and this is, uh, probably true of not just carnivore but ketogenic or anything that, um, that does produce that hyperbeta lipoproteinemia. Um, uh, it, it's, it, it almost seems to be worn by some as a badge of honor as opposed to saying, "Well, maybe this diet is doing a lot of really good things for me. It's improving my insulin sensitivity. I feel better. I have fewer energy swings, but this one thing isn't so good." But here's the thing. Of all the things that could go wrong, that's about the most treatable one out there. It's very easy to treat elevated ApoB. [chuckles] Uh, and so m- that, that's the part... I, and this is what we do clinically, right? This is how we treat patients, right? We have patients who only get better on very, very carbohydrate-restricted diets. But then, you know, if they develop that, that pattern, that elevated LDL pattern, we have a choice to make, which is we abandon the diet or we treat the elevated ApoB. Um, and that's not a failure. That's simply using modern medicine to help us achieve the best of both worlds. So [lip smacks] um, I think I've always struggled to understand why a person will go on that diet, have, you know, an ApoB or LDL go from the 50th percentile to north of the 99th percentile, and instead of being curious about what the implication is, sort of dig their heels in and say, "Clearly, this is a good thing, and LDL does not cause heart disease."
- SGStephan Guyenet
I absolutely agree. I mean, it's, it's a dietary i- ideology. It's an ideology that has emerged to, uh, defend a certain type of dietary pattern. This, these kind of ideas emerged from the low-carb community essentially to defend against the idea that there might be some downsides to certain types of low-carb diets, and they've been taken to an extreme, I think, in the carnivore community because that's a particularly, uh, potent stimulus for increasing LDL. So people don't wanna believe that there's a downside to the thing they're doing, and, you know, like, I kinda get it. Like, people go on this diet. They lose weight. They feel better. Some people s- you know, they say their skin cleared up or XYZ condition improved. Like, there's all these tangible things they can see that are getting better. They don't wanna believe that there's an intangible thing that's actually putting them at severe risk 'cause, I mean, as you know, like, I say severe risk. I just mean cardiovascular disease is a big risk generally. Like, as you know, this is the number one killer. Cardiovascular disease is a huge, big deal. Even if it doesn't kill you, it, it can do really bad things to you physically and cognitively. So, like, it's not a risk you wanna be ignoring, so I, I absolutely agree. I think it's an irrational part of dietary tribal ideology that is holding people back from experiencing their beth- best health, and it's so treatable. I said that in the review, too. I was like-
- PAPeter Attia
Mm
- SGStephan Guyenet
... "You don't even have to stop the diet."
- PAPeter Attia
Right.
- SGStephan Guyenet
"You can just get it treated, or you could modify the diet."
- PAPeter Attia
Yeah, and, and, and by the way, it's the s- you, you see the same sort of equally, uh, stubborn ideology at the exact opposite end of the spectrum, where you use, uh, you know, we see these patients that'll go on these incredibly restrictive plant-based diets, and it's usually some combination of micronutrient deficiency and/or protein deficiency that's going to be the death of them. But there's no deviating from it. There's no, like, "I'm gonna supplement with protein shakes. I'm gonna take B vitamins. I'm going to do these other..." And, and again, it's the same sort of thing. It's like somehow if I acknowledge the fact that I need to supplement with these other things to work around a diet that I otherwise like or that is congruent with a belief system I have around the treatment of animals, which I can respect that, right? If that's, if that's your belief system, then by all means be true to it. But, um, but yes, I find it, um, s- somewhat self-destructive.
- SGStephan Guyenet
Yeah, absolutely. I think it's very analogous to what we see in certain corners of the vegan diet community. They want it to be best for everything. It's gotta be best for the environment and for ethics and for health. And, like, it's just very hard to swallow that there might be some downsides. And for the carnivore diet, too, you see, like- They try to justify the environmental aspects of it too. Like, um, in "The Carnivore Code," he talks a lot about regenerative agriculture, um, which, you know, is, is a concept that I think is, uh, interesting, and I think I support it, but, um, it's a certain, I would say, spin on it that makes it seem particularly favorable, and also, like, there's an underlying assumption that the average carnivore is gonna eat nothing but regenerative agriculture beef, um, which I think is not the case.
- 1:50:06 – 1:58:41
The energy balance model, carbohydrate-insulin model, and unifying the theories around adiposity
- PAPeter Attia
Let's talk about one more big topic, Stephan, which you've, you've, you've written a lot about. God, you're, I, I, I don't mean... I don't know if you're sick of it or you're, you're, you're still enjoying it, but this idea of kind of energy balance, carbohydrates and insulin, and unifying theories around adiposity. Um, now r- very recently, I heard you and, um, Kevin Hall on a podcast. I don't remember what the podcast was called, but it was a wonderful discussion because I know you well, I know Kevin well, um, and anytime I can listen to a podcast with people who I know a lot about and I know most of what they have to say, but yet I still pick something up in the discussion, it's fantastic, and that was, that was an example. So [clears throat] in it, you guys did a pretty good job, I thought, of really explaining the history of these models, um, which I think you acknowledged are probably not perfectly named, so there's a little bit of historical baggage that goes into the nomenclature. And I think anybody who's able to sort of pay attention long enough will realize that there's a lot in common with these models, but there are some fundamental differences that are important to understand. So maybe we could, we could sort of talk a little bit about that. And, and, um, and again, I think this is not a time to be overly simplistic, right? I think this is a time for nuance, and it is a time for, um, putting the finer point on the similarities and differences of these models. So, so maybe just start by explaining... Well, p- pick the one you wanna start with, and let's, let's kinda walk through them.
- SGStephan Guyenet
Yeah. So the two models are the carbohydrate-insulin model and the, um, energy balance model. And the carbohydrate-insulin model, I just wanna get a little more specific with that because there are different versions of this, and, uh, so this is the one that has been, uh, promoted by David Ludwig and particularly in a recent review paper that he published with, along with some other researchers. And, uh, the energy balance model in this case is being represented by Kevin Hall, but I would say he was pretty close. It has very deep roots in models, similar models that go back decades. Um, carbohydrate and insulin model, um, in its most recent incarnation is a lot more complex than previous inclinations, so I'm gonna do my best to kind of summarize it and, and hit the key points. Uh, but essentially, it's the idea that, um, there are things in the diet and in the environment that impact insulin signaling, and insulin signaling impacts body fatness. And then that fattening process of insulin signaling on adiposity then downstream leads to elevated calorie intake and elevated, uh, and possibly a decline in metabolic rate. So it proposes a reversal of the relationship. This is, this is one of the key aspects I wanna highlight. Proposes a reversal of the relationship between, um, energy balance and body fatness. Basically, the energy balance phenotype is downstream of that fattening process instead of being upstream. And that, that really, I think, is a, a key difference. Um, when we go to the energy balance model, the energy balance is upstream of the fattening process. So basically, we have all these things happening in the environment and physiologically in our bodies. Those signals are impinging primarily on the brain, and that, um... And, and, and then energy balance is, um, a result of primarily that brain activity, and then that is feeding into adipose tissue. So in that context, adipose is, um... Body fatness is kind of receiving the excess energy. It's not really the driver of this process, but when excess energy enters the body, it's what mops it up. So it, it's kind of, uh... That's, that's kind of the, the difference between those two models.
- PAPeter Attia
So let me say that again just to make sure we've, we've got this right. So in the former model, in the carbohydrate-insulin model, the idea is that the, the primary cause is the adipose tissue increasing in its fatness, right? The adipose tissue wants more energy that's driven by the external factors, both carbohydrate and otherwise. [lip smack] So in a drive to increase the influx of fatty acid into adipose tissue, you see a reduction in circulating metabolic fuels, which drives an increase in appetite. So intake goes up to accommodate the reduction in circulating metabolic fuels, which is being caused by a drive towards fatness. The conventional model basically says that's happening in reverse. It's saying that the input of fuels into the system leads to an increase in circulating metabolic factors that is now driving energy balance into the fat cell. Is that ... safe overview?
- SGStephan Guyenet
Yes. Um, you said conventional model though, and I think-
- PAPeter Attia
Is that the-
- SGStephan Guyenet
Do you wanna-
- PAPeter Attia
Do you wanna call that energy... Is that the energy balance mo- I, I, I've heard both terms used
- SGStephan Guyenet
Yeah. Okay. Um, so sorry, just to... I think this is worth a moment to clarify. So if you look at, um, David Ludwig's paper, he contrasts the carbohydrate-insulin model against what he calls the energy balance model, um, which is basically calories in, calories out. So it's like it's not really re- none of this is really regulated. It's just however many calories you happen to passively eat or how much you decide to exercise, and then your, your fat tissue is a result of that. So, um, the energy balance model, in contrast, is, um, acknowledging all this brain regulation of body fat, brain regulation of appetite, and saying actually body fat is a regulated process. However, it's, um... body fatness, I should say, is a regulated variable. However, it's regulated by the energy intake and expenditure via the brain.
- PAPeter Attia
Got it. Okay.
- SGStephan Guyenet
Yeah. So the, the conventional model, I, I will say, is not a model that really any research- any obesity researcher is currently, um, ascr- uh, ascribed to, at least the obesity researchers who are actually studying the mechanisms of, uh, body fat regulation.
- PAPeter Attia
Yeah, and one of the things that I think Kevin pointed out on this podcast to, to go further on that, um, down that thread is the energy balance model does not consider all calories identical.
- SGStephan Guyenet
Yeah, it does not presuppose that they're all identical. That's right. In, in terms of their impact on, on body fat.
- PAPeter Attia
Correct. So not only do they potentially have different thermogenic effects, but y- they also might have different regulatory effects on compensatory appetite, right?
- SGStephan Guyenet
That's right, and you know, I think where we really see this emerging is in animal models. Um, and I, I don't know how relevant this is for humans, but I'm just using it as a general, like, proof of principle that it can happen. You can see in animal models where you can change their diet composition. Um, David Ludwig has shown this for carbohydrate quality. It's been shown for dietary fat as well, and you can actually produce animals that will gain fat independent of calorie intake. So you can actually clamp them at their former calorie intake, and they will nevertheless gain fat. So, um, so there, at least in principle, those types of effects are possible.
- 1:58:41 – 2:07:26
Body weight set points: a hypothetical comparison of two individuals
- PAPeter Attia
Yeah, I believe Rick Johnson described an experiment like that on my recent podcast with him, which was an isocaloric swap to a very high fructose diet where the animals didn't gain weight, but they fuel partition differently. They got fatter. Um, this was over a long time.
- SGStephan Guyenet
Mm-hmm.
- PAPeter Attia
This was over nine months. So nine months for a mouse, right, is an eternity.
- SGStephan Guyenet
Yeah, that's a very long time. Yeah, absolutely. That has been shown in a number of contexts that that can happen in rodents. But they also... I think it's worth pointing out that rodents have... Their energy expenditure is more plastic than ours.
- PAPeter Attia
Yeah. Uh, by the way, I wanna go back to that 'cause we-- this is something I-- this'll tie into what we're about to talk about. But you have a person who weighs 200 pounds, person who weighs 160 pounds. They're the same height. 100 and-- The 200-pound person loses 40 pounds. They're now 160 pounds. The other person's always been 160 pounds. On the surface, they look identical. In fact, let's pretend they're siblings, but one was obese, and he's now post-obese. The other was never obese. Let's pretend that that one that lost the 40 pounds has really kinda dialed it in and doesn't yo-yo. He manages to stay there. It's three years later. Are they the same person yet?
- SGStephan Guyenet
No, probably not. They're... So Rudy Leibel has done studies where I think they've had people out to a year, maybe two years, where they have them weight reduced, and the starvation response, this leptin-dependent starvation response, he hasn't seen any sign that it goes away. So could it maybe be possible? [clears throat] Could it maybe be possible under some circumstances? Maybe. But, um, the evidence that I've seen suggests that it is at least not typical. Um, I do think... So I wanna specify that I think that the, uh, set point around which the lipostat regulates can change based on dietary and environmental variables. And so, for example, you know, just to give a, an example that you'll be familiar with and others probably listening will be familiar with, if people go... If you take someone on a typical diet and put them on a low-carb diet, you don't have to tell them to reduce their calorie intake. That will occur spontaneously, and they will lose, lose fat and end up t- in the typical person comfortably being at a lower weight. They're not experiencing the starvation response, and you can see this on other diets as well. So I think there are things we can do to change the set point, however That person, that doesn't mean that they are cured. So if they went back to their other diet, if they just went back to how they used to be eating so they're not maintaining this attempt for weight reduction anymore, they generally will go back to where they were. So it's not that there's a durable resetting of the set point to like flipping a switch and resetting, like restarting your, you know, computer. It's more like the set point has been modified because it's in a different environment, and as long as you maintain that change, you can maintain the effect. But if the change goes away, then the effect goes away.
- PAPeter Attia
Where do we think is the greatest window of vulnerability for someone? Just going back to these two hypothetical individuals who... And let's just now make it even more, let's take the genes out of it. Let's pretend they're identical twins, uh, born in the same household. Uh, clearly they both can, they both possess the genetic traits that would allow them in the right environment to become obese. Um, [clears throat] but one of them, you know, let's just say had an injury in high school that kinda kept him home from playing sports, and he, he, he ended up playing more video games and kinda eating more. The other one was more active. So that explains why when they're now 40 years old, one's 40 pounds overweight, the other's not. Are there windows in a person's life when they are more susceptible to that resetting of a set point, a higher and higher set point, which it sounds to me like it never goes down. It's a monotonic crank, right?
- SGStephan Guyenet
I don't really know. I honestly don't know. I mean, I think certainly there is a substantial potential for most people to gain weight at almost any point in life. Um, so I'm not really sure. I think, I think one... And, and, and by the way, let me say that I think there are other people who could probably answer this question better than me. There are people who have studied, uh, the trajectory of weight gain over the lifespan. So part of the issue here is I'm just not that well-versed on this literature. Um, but one thing I'll point out that is potentially interesting is there may be an influence, uh, of the intrauterine environment, so what's going on as you're developing, um, inside the uterus. So there is evidence, I wouldn't call it strong, but there is evidence that women who undergo bariatric surgery for obesity and lose a lot of weight, their children are at a lower risk of developing obesity than women of similar weight who did not undergo bariatric surgery, and the effect size is large.
- PAPeter Attia
Wait a second. Meaning two women of the same weight have children. One of them was, is that weight naturally, and the other one is weight reduced at that weight secondary to gastric bypass?
- SGStephan Guyenet
No, no. So, um, think about two women with severe obesity.
- PAPeter Attia
Ah, ah, okay, okay.
- SGStephan Guyenet
Yeah, yeah.
- PAPeter Attia
Yeah, yeah. I think-
- SGStephan Guyenet
One of them has gastric bypass.
- PAPeter Attia
Yep, okay.
- SGStephan Guyenet
After that surgery, they both have children.
- PAPeter Attia
But starting from different weights, obviously, 'cause the gastric bypass-
- SGStephan Guyenet
Correct
- PAPeter Attia
... one is weight reduced. Okay.
- SGStephan Guyenet
Yes.
- PAPeter Attia
Mm-hmm.
- SGStephan Guyenet
And the children of the woman who had the surgery and had previously lost weight before getting pregnant will have a lower risk of obesity. And again, I wouldn't call the evidence strong.
- PAPeter Attia
And what if, what if she achieved, what if she achieved that weight loss without gastric bypass? So what if you had two women who were overweight and one of them lost weight through diet and nutrition, and then they got pregnant?
- SGStephan Guyenet
I don't know. Um, I am not aware of data on that. I would... If, you know, if you wanted me to guess, I would say it would probably be similar. Um, but is gastric bypass, does that have, you know... I- is that really the same physiologically as, um, w- the physiological situation that you get from diet and weight loss, diet and lifestyle?
- PAPeter Attia
[lip smack] Uh, see, I don't, I don't think it is. I, I think, I think gastric bypass is a unique situation where provided a person doesn't take in liquid calories, it's, it's, it's v- quite durable. Uh, the, the rue and why. Um, obviously liquid calories can completely disrupt the feedback mechanism there. Um, kind of similar, by the way, to a GLP-1 agonist. This actually kinda gets back to I've seen patients who take semaglutide who don't lose weight, and the way you can cheat semaglutide is to drink your calories. If you-
- SGStephan Guyenet
Hmm
- PAPeter Attia
... if you drink massive amounts of calories. Again, this is anecdotal. I don't think, I don't know that this has been studied. I'm just saying this based on observing a number of patients. But, um, you know, if you continue to drink a lot of alcohol, if you continue to drink juices and things like that, um, you, you're, you, you can sort of bypass some of the, the, the GLP-1 effect on the brain, it would seem. That, that's my only explanation for why I see that. Um, and, and we do see that definitely with, with, with gastric bypass. So who, who knows about, about what that would look, what that would look like.
- 2:07:26 – 2:16:34
Takeaways for people who want to lose weight and keep it off
- PAPeter Attia
I mean, what, what advice or what insight comes from this as it pertains to a person who's listening to this? And by definition, half the people listening to this are probably at a body weight above where they wanna be. What's the takeaway, right? Is it-
- SGStephan Guyenet
In terms of pregnancy?
- PAPeter Attia
No, no, just in terms of overall weight loss. Like, how, how do we get a, you know... We're sitting here in this environment that is you know, almost deliberately trying to put weight on us. We're not going to get any help from our ancestors because the reality of it is our ancestors didn't care if we gained weight. Quite happy to have us gain weight, actually. They just want to make sure we don't starve. Um, and so what can they do? And more importantly, how do they keep it off? Because as you said, most people can lose weight, but the keeping it off is really, uh... It, it poses a challenge.
- SGStephan Guyenet
Yeah. So I think that I'm, I'm gonna take this as a question about on the individual level, which I assume is what you meant.
- PAPeter Attia
Yeah.
- SGStephan Guyenet
Um, so I think that there are a couple of different things to think about. Um, I think that for people who have, um, for people who have obesity, uh, body mass index 30, 35, I think it's worthwhile to consider medical treatment, so something like semaglutide, for example. The tools that we have now are just way better than what they used to be. That's a separate topic we could talk about. Um, but semaglutide, as far as we can tell, it's a very safe drug. It causes something like 18% weight loss, um, which is much better than the typical effect you're gonna see in diet and lifestyle, um, strategies. But like diet and lifestyle, it's something you have to maintain. So I think at this point, now that the tools are getting better, particularly now that the tools are getting better, I would recommend seeing an obesity medicine specialist, I think for people who are, you know, experiencing su- substantially impaired quality of life or are really concerned about the health impacts. Um, about, you know, if we switch the focus to people who might just want to lose a few pounds or who are overweight, where they're not in as, uh, perhaps not as serious a situation, I think the brain... So your appetite and your body fatness are very much regulated by your brain based on inputs that your brain is receiving, and so-- and a lot of that is non-conscious. And so the approach that I like to take is to try to give the non-conscious brain signals that are going to be more consistent with your goals, signals that are going to tell your brain to regulate things in a, in a more slimming direction. And that way, you're not relying on heavy exercise of willpower all the time, which I don't think is really sustainable or effective for most people. You are, instead of setting up a scenario where you're, you have these non-conscious urges that you're having to fight with your conscious brain, you're addressing the non-conscious urges directly, so there is no, is no fight. That's the, that's what I prefer. And so I think controlling these signals that your brain is receiving is, is really important, and there are different ways to do that. One of them is to control your food environment, so the sensory cues in your environment that your brain is exposed to, whether there is food in your immediate vicinity, how tempting that food is, um, how hard you have to work for it. So, like, if you can just grab it and put it in your mouth, that's not as good as if you have to, you know, walk into a room and then peel an orange before you can eat it. Um, just little effort barriers like that. And then, um, and then with the types of food we're eating, there are, um, there's a wide variation in the number of calories that it takes to feel satisfied at a meal depending on what foods you're eating. So typically, a typical person sits down and eats food until they feel satisfied, and then they stop eating. That is the, um, intuitive, typical, natural, easy way of interacting with food. But depending on what's on your plate, that point can be reached with vastly different numbers of calories. So if you're eating-
- PAPeter Attia
So, so is the, is the proximate sign of satiation more, um, a gastric distension function in that immediate, uh, cessation mode?
- SGStephan Guyenet
Oh, that's, that's a great question. Um, I don't know. So okay, I-- That is important. I, I can say that it's important. But if I were to, like, assign what percentage of the effect is attributable to that, I don't know what percentage I would put on it. Like, is that more than 50%? Um, I think it's a very complex system. The brain is receiving a lot of signals. Some of them are stomach distension. Some of them are signals from the small intestine about what the nutrient composition is. Uh, some of them are simply sensory detec-- or a sensory detection of food properties and stimulating, uh... Your brain knows based on the sensory properties what the nutritional composition of the food is based on prior experience that it has stored. So there's a lot of stuff going on that contributes to ultimately that sensation of, uh, satiety, satiation would be the, the proper term for it, and, uh, and satisfaction that causes us to end the meal. Certainly, stomach distension is, is a biggie. Yeah. So anyway, so that relates to calorie density, which is an important determinant of the satiating and satiety-promoting properties of, of food. So in other words, how many calories are there per gram or per volume of this food? If you have a food that has more volume per calories, fills up your stomach more, it stimulates those stretch receptors more, that goes up to your brain stem, and that's a signal that opposes further food intake. And then, uh, protein, so more protein is more satiating per calorie. Um, palatability, the better something tastes, the less it fills you up per calorie. And I'm actually not sure how to disentangle that from, um, calorie density. I'm not sure how much of... Like, what's the independent variable there, and what's the dependent variable, or is it some of both? I don't know the answer to that. Um, but they're both strongly correlated with, uh, lower satiety.
- PAPeter Attia
And there, I've also seen these experiments where people are drinking from a bowl or a cup, and it's being refilled constantly versus one where it just kinda runs out. I mean, what are the differences in how much people consume based on just... I mean, I notice this in myself. Like, I'm eating, I put a little too much in the bowl, but I eat it anyway 'cause it's like, "Oh, I gotta finish this thing." And if I had finished it 10 bites earlier, I would've been totally happy.
- SGStephan Guyenet
Mm-hmm, mm-hmm.
- PAPeter Attia
Again, I don't know how much that matters, right? Does that, does that type of behavior factor in the long tail here, or is that just an acute thing that is sort of irrelevant in terms of, uh, you know, optimal weight maintenance?
- SGStephan Guyenet
I think it's very plausible, but the problem is that stuff comes from Brian Wansink, and so it pretty much got blown up.
- PAPeter Attia
He was the guy at Cornell that kind of... Did he falsify a bunch of data or something?
- SGStephan Guyenet
I don't think there is clear evidence that he falsified. I think there... Or at least I-
- PAPeter Attia
But he P-hacked a lot of it pretty badly.
- SGStephan Guyenet
Oh, yeah, really badly. And, and there are some data where it's not clear where they came from, and they're very implausible. Uh, so I, I don't, I don't know how strong the evidence is that there was actual fabrication. I think there may have been some evidence of that. Um, I don't know where that landed, but basically, there were a bunch of problems, and yeah, he got blown up. So, um, I would say that anything that has his name on it [chuckles] at this point is pretty suspect.
- PAPeter Attia
Yeah.
- SGStephan Guyenet
And the, the soup, the refilling soup bowls was one of his classic experiments. Um, and I'm just gonna disclose that I did cite one of his studies in my book, so I wasn't immune fr- uh, from getting taken in by some of
- 2:16:34 – 2:26:54
Evidence that favors the energy balance model of weight gain
- SGStephan Guyenet
this stuff.
- PAPeter Attia
One of the things I, that going back to the sort of the, um, energy balance model versus the carbohydrate-insulin model, [clears throat] one of the arguments in favor of the carbohydrate insul- insulin model is other examples of growth that are regulated from the hormones out to the intake of energy. And I was thinking about this the other day because I measure my kids. I have three kids, and every three months, each of them gets a little tick on their closet door. And I don't know if you have kids or if you've done this, but any, any, I think anybody listening to this who measures their kids at regular intervals, and I don't know why I picked three months, but four times a year. Um, the non-linearity of this is unbelievable, right? They'll go little, little, jump, jump, slow down, jump. Like, it just, it's pretty intense. And that growth correlates with how much they seem to eat during that interval period. Like right now, my youngest, who's not yet five, I think he eats more than the other two combined.
- SGStephan Guyenet
[chuckles]
- PAPeter Attia
Uh, and I'm probably not being facetious.
- SGStephan Guyenet
Hmm.
- PAPeter Attia
He goes to a preschool or a pre-K where they give them breakfast there. He eats a breakfast at h- he eats two breakfasts at home. Then he goes there and still eats more than all the other kids. I mean, the kid's just an eating machine.
- SGStephan Guyenet
[chuckles]
- PAPeter Attia
And he's growing commensurate with that. I think most people would agree he's not growing because of how much he's eating. He's eating that much because of how much he's growing, so he's responding to growth hormone and all these other things. And, and I think that's basically the central thrust of this carbohydrate insulin model, right? Which is whether it be sleep disturbances that increase, you know, or decrease insulin signaling, or foods that stimulate insulin, they're driving that hormonal environment that is driving the increase in food intake. But I think experimentally, what can we say about the differences in these models? I, I, a- and I guess, I think experimentally, you have to think about this in terms of animals and humans, right? It seems to me that the balance of experimental data would suggest the energy balance model is easier to explain. Is that, would you agree with that?
- SGStephan Guyenet
I think so, although I do want to acknowledge, uh, the fact that they're not mutually exclusive. So, um, I think it has not been ruled out that there could be a contribution from that type of a model.
- PAPeter Attia
Why is this so important? Is it so important? Is it important to understand this? Um, I, I, I, I, I suppose the implications have to come down to how we treat the condition, right?
- SGStephan Guyenet
I think that it is important in the sense that if you understand the mechanism of something, it makes it easier to address. And it, you know, it... If you look at the history of obesity drugs, weight loss drugs, I should say, most of them were discovered in entirely haphazard ways, like dinitrophenol was a high explosive that was used in World War I, and somebody figured out if you take it, it makes you lose weight, and it does so by increasing your energy expenditure. So some people managed to cook themselves and like literally from the inside out. And then you look at other drugs, and most of them are psychiatric drugs. They are just repurposed psychiatric drugs that just happen to cause weight loss.
- PAPeter Attia
Mm-hmm.
- SGStephan Guyenet
Some psychiatric drugs cause weight gain, some cause weight loss, and the ones that cause weight loss, we just said, "Hey, can we repurpose these? Can we combine them to accentuate the effect?" And that, that was kind of most of the history of weight loss drugs. And now, for the first time, we have drugs that are safe and effective, FDA-approved... I should say we have a drug, Wegovy, AKA semaglutide, that is safe and effective and was developed for this purpose based on mechanism from the bottom up. So it wasn't just a haphazard discovery. So I think we are in a new era now where we are actually designing weight loss drugs based on mechanism. And so I think based on an understanding of the biological mechanisms of, uh, of regulation. And so I think because we're out of the haphazard era and into a more targeted, refined era, I think that's... We're in a place where it becomes really important to understand mechanism. And, you know, in cardiovascular medicine, I'm sure you recognize this, like there are incredible insights that have been coming out of the genetics that have resulted in therapies like the PK-
- PAPeter Attia
PCSK9 inhibitors, yeah
- SGStephan Guyenet
... PS... Thank you. Thank you. I always mix the letters up. Um, that's a great example of it, where we first understood the biology, and then we came out with the therapy, and it-
- PAPeter Attia
Yeah
- SGStephan Guyenet
... works awesome. And so, um, I think that's the era we are getting into now with obesity. So I think it actually is really important to understand the mechanism.
- PAPeter Attia
I think m-my, my prediction is over time these models will have more and more in common. Um, I mean, one of the arguments against at least the way the carbohydrate-insulin model is typically played out is actually given by semaglutide, which raises in-insulin in the short run. So one of the things we always, you know, see when we put patients on this drug is they're gonna lose a ton of weight, and their insulin levels go up slightly. Now, this tends to resolve over time. So over a long enough period of time, we tend to see insulin come down, but in the short run, for about three months, we see elevations in insulin. Now, we also know that it's increasing insulin sensitivity, so they're getting really the double effect. But it's hard to reconcile that with a model that would state insulin must go down for weight to go down, for fat to go down.
- SGStephan Guyenet
Yeah, I mean, I think that model is just not even plausible at this point. I think, you know, if you wanna say, "Is it a factor?" I think that's still in play. But to say like this is the determinant, I just don't even think that's plausible at this point.
- PAPeter Attia
Yeah.
- SGStephan Guyenet
Um, and yeah, I mean, this class of drugs was identified based on its ability to increase glucose-stimulated insulin secretion. That is what GLP-1 does. It's an incretin hormone. And so that was the original purpose and why it was used in type 2 diabetes management, 'cause it gives people more insulin around meals when they really need it. 'Cause, you know, if you just inject insulin, that's a really kind of crude way to manage your blood glucose. It's not time-specific. So GLP-1 gave it that much needed time specificity and also had some other beneficial effects, and it was only after that that they figured out that it has this big impact on food intake and, and body weight. So yeah, I mean, absolutely, I agree with that. Um, and yeah, there's just, there's just a lot of other literature. I think, you know... I, I wanna, I wanna talk about... I wanna like throw the hypothesis a bone though. Um, I think that if you look... So if you look at genome-wide association studies on body mass index, they're all about the brain, largely about the brain. If you look at genome-wide association studies on body, uh, fat distribution, so you're getting rid of the BMI, you're controlling for BMI, and you're saying, "Where is that fat on the body?" Those have more of an insulin signal. So insulin pops up pretty prominently in those. And so I think there's still a world, and, you know, David Ludwig-
- PAPeter Attia
In other words, the distribution of fat on the body seems-
- SGStephan Guyenet
Where it is
- PAPeter Attia
... more related to insulin signaling.
- SGStephan Guyenet
Correct.
- PAPeter Attia
The total amount of fat on the body seems more related to e-energy intake, if I remember what we talked about really at the beginning, was it's not just regulated by the brain, but it's on the, it's more on the intake side of the equation.
- SGStephan Guyenet
Yeah. And I would say, you know, even to add a little bit of extra nuance on top of that is we're really talking about body mass index. So, you know, if you consider this idea of energy partitioning, which the carbohydrate-insulin model is all about, there could be some of that flying under the radar of body mass index. And so I don't think the door is closed to that, and the fact that it's showing up in body fat distribution... And now, you know, David Ludwig is publishing studies suggesting that there could be, uh, correlations between baseline insu- insulin secretion and how much, what proportion of fat loss is lost as fat versus lean tissue. I think, you know, there's a world in which there could be some energy partitioning effect. I just don't think it explains obesity, because obesity is not just energy partitioning. You have a bigger body, you're eating more energy, you're burning more energy, you have more lean mass, more fat. That phenotype is not explained by energy partitioning. But could there be some subtle energy partitioning phenotype that is operating also? Maybe. So that's kinda like my, uh, like, view of how there could be a way that this has validity.
- 2:26:54 – 2:34:46
The synergistic effect of fat and carbohydrates and observations that a low-fat diet or a low-carb diet can cause weight loss
- PAPeter Attia
Yeah, it's interesting, and you said something earlier about, um, the more you restrict carbohydrate or the more you restrict fat, typically the more weight you're gonna lose. The sweet spot, if you wanna gain weight, is to have lots of both of them. Is that... How much of that do you think comes down to the hedonic component of how good ice cream tastes, the ubiquity of food choices, or do you think there's something very unique physiologically going on? Um, y- you know, you've probably heard of the potato diet. I mean, all these diets just sound so stupid, but-
- SGStephan Guyenet
[laughs]
- PAPeter Attia
... I mean, again, it's sort of one of these things, like, y- if you talk to somebody who just mainlines potatoes all day, they lose weight like crazy, right?
- SGStephan Guyenet
Yep, absolutely. And, and this has been shown really clearly in animal models, which have the advantage of you can get really tight control for a large proportion of the animal's lifespan. Um, John Speakman published a study that is the best one that's been done in animals, where 29 different diets, I believe, and five different strains of mice, they systematically varied the carbohydrate to fat ratio in the diet, and they said, "How does that interact with body fatness?" And what they saw was if you start with animals that are on a low-fat, um, high-carbohydrate diet, and you start replacing that carb with fat, they get fatter and fatter and fatter and fatter and fatter until you hit about 60%. And then you, if you keep increasing the fat and decreasing the carbohydrate, they get slimmer again. And there are studies published, mice lose weight on a ketogenic diet, just like humans do. You can put mice on the diet, you can put rats on the diet, they lose weight. So it's really in the middle that the problem is, and-
- PAPeter Attia
Which is ironic, 'cause that's where the standard American diet is. If you just-
- SGStephan Guyenet
Absolutely
- PAPeter Attia
... walk into the grocery store and just eat without any filtering, you will eat that wrong combination of fat and carbohydrate.
- SGStephan Guyenet
Exactly, and if you look anywhere in the world where people are rich enough and industrialized enough to eat whatever they want, that's generally what you're gonna see. At least after a couple decades of, you know, cultural sh- adaptation, you're gonna see pretty equal proportions of fat and carbohydrate. That's what you see in most parts of the world. And so, um, so why is that? Well, I think there are some hypotheses that can be considered, and, and, and let me just be clear here that this is speculative, so I don't wanna present this as the answer, 'cause I don't think we really know. But I think this is primarily a empirical observation that we're trying to explain. Um, but some explanations, you could say maybe it's a physiological effect. So if you're not eating much carbohydrate, your body has to work a little bit harder to synthesize glucose, for example. So there are some physiological ways in which you're kind of increasing the demands a little bit, your metabolic demands. Same with very low, um, fat diets. You're slightly increasing metabolic demands. You have to synthesize... You're gonna synthesize more fatty acids. And, um, but I don't know. I don't think that's, I don't think that's a great explanation because those metabolic demands are very small.
- PAPeter Attia
I- in Speakman's experiments, did... These were all ad libitum, I assume.
- SGStephan Guyenet
Yes.
- PAPeter Attia
So were they just eating... Were they also significantly eating less at the extremes in terms of total energy intake? And were they controlled for protein?
- SGStephan Guyenet
Yes. Yes. I think the answer is yes to both of those if, if I'm recalling correctly. Um, and, uh, although, you know, I do wanna put a little asterisk on that, that sometimes there is not a perfectly tight correlation in rodent studies between energy intake and fat gain. So you can get some results where it's not fully explained by that, or in some cases not explained at all. So, um, yeah, where was I? What was I saying? Lost my train of thought there.
- PAPeter Attia
[laughs] Well, I think we're still trying to reconcile why-
- SGStephan Guyenet
Oh, yeah, yeah, yeah
- PAPeter Attia
... at the extremes you're-
- SGStephan Guyenet
Yeah
- PAPeter Attia
... what's accou- Is it, is it all being driven by less intake, or is there some increased metabolic cost of living at the extremes?
- SGStephan Guyenet
Yeah. So the physiology would be one, and then the other would be just the neurobiology and the, the food intake. And I think that's the most satisfying explanation we have right now is that it is simply more appealing to eat food, more motivating to eat food that has both carbohydrate and fat. I'm not saying I have strong evidence that that's the explanation, but that's kind of the only thing I can think of that explains it. I mean, why else... Uh, the, the metabolic cost of being at the extremes should be pretty modest. It can't explain effects of hundreds of calories a day Which are, which is what's observed-
- PAPeter Attia
That's right
- SGStephan Guyenet
... when you put people on a very low fat or a very low carb diet. Their energy intake declines by hundreds of calories a day right away, automatically. And so it's... Yeah, that's the only thing I can think of is that essentially the food has less implicit value to reward regions of the brain because of how our motivation is determined for certain types of food properties. The food is intrinsically less motivating, and so we eat less of it. That's my best guess as to the main reason.
- PAPeter Attia
And it's interesting. I mean, I think it's not very difficult to take your carbohydrate intake down to 5 to 7%. Like a ketogenic diet will do that, and it's a pretty easy diet to adhere to, especially today. Harder 10 years ago when there were fewer sort of food choices geared towards it. But if, I mean, I'm not necessarily saying it's a overly pleasant diet, but it's not difficult. You don't have to put a huge amount of effort to eat 5% of your calories in carbohydrates today. I don't even know how one would go about getting only 5% of their calories from fat. That, that is a much harder thing to do, is it not?
- SGStephan Guyenet
Yeah, I think it would be very difficult to get that low. When you look at, um, studies that test low-fat diets, some of the lowest fat diets I've seen were in the kind of 10% fat range. Um, I mean, even foods like, you know, whole wheat and corn have a fair amount of fat in them. Not a lot of fat, but I don't remember exactly what it is, but not far off from 10%, just from those foods that we would call starch foods.
- PAPeter Attia
Yeah.
- SGStephan Guyenet
Um, yeah, so I think it is quite challenging to get that low in fat. You can do it with, um, more refined diets like semi-purified diets. Um, it's not so hard to do that in rodent studies. But to design a diet that someone will actually eat as a human in a, you know, randomized controlled trial or something at that level of fat intake I think is pretty challenging.
- PAPeter Attia
Yeah. I, I think 10% is... I mean, even 15% is not easy. I mean, you've ... You're really working hard to do it. And, you know, I'm not convinced that there are great health benefits to that either. Um, probably better than being in the, in the messy middle, but, um, it, it's pretty hard to do.
- 2:34:46 – 2:48:43
Red Pen Reviews
- PAPeter Attia
So last thing I wanted to ask you about was Red Pen Reviews. So, so tell folks a little bit about how long you've been doing that, and what's the frequency with which you guys, uh, put these reviews out?
- SGStephan Guyenet
Yeah. So Red Pen Reviews is a 501 [c] [3] nonprofit that publishes the most informative, consistent, and unbiased reviews of popular nutrition books available. And the thing that really makes us unique is that we have developed this semi-quantitative, uh, review method, structured review method that we apply to each book that yields numerical scores for scientific accuracy, reference accuracy, and healthfulness. And essentially, what this does is, um, it allows us to apply the same rigorous method to all books such that you know, um, where it's coming from, where the numbers are coming from, and you can compare in an apples to apples way between different books. So you could say, "I wanna know the best book on topic X," and you can literally compare the scores of two books apples to apples and choose the one that has the highest, let's say, scientific accuracy. And, uh-
- PAPeter Attia
So-
- SGStephan Guyenet
Go ahead.
- PAPeter Attia
Oh, sorry. Go ahead. No, I was gonna say, how often do you do... This is a pretty labor-intensive process, so how many do you bang out in a year?
- SGStephan Guyenet
It takes us about 40 to 100 hours per book. And, um, we have, we've been operating since 2019, and we have 14 reviews. I will say that our pace, um, was quite slow last year due to COVID related time challenges, so we're on target to have probably six to eight reviews published this year. And, um, I think we have some things going on behind the scenes that could potentially greatly accelerate that pace. We're seeking funding.
- PAPeter Attia
How m- how many folks do you have that, uh, review the books?
- SGStephan Guyenet
We have a total of, I believe, eight reviewers right now.
- PAPeter Attia
Wow. Okay.
- SGStephan Guyenet
Yeah.
- PAPeter Attia
So each reviewer might do one a year on average.
- SGStephan Guyenet
On average, yeah. But the way it turns out is that some people do the majority of reviews, and then other people, um, only rarely do a review.
- PAPeter Attia
And how do you guys select books for review?
- SGStephan Guyenet
We are trying to maximize our impact on public health knowledge and public health, so we really try to pick the books that are most impactful right now. So we're looking for books that are selling the most. We're looking for books that are having the most, uh, social media engagement. We're looking for books from authors that are particularly influential. So we're, we're really trying to, um, give people information about the things that they are already interested in.
- PAPeter Attia
So what are some of the recent ones that you guys have published? Which books? Obviously, we talked about "The Carnivore Code" as one of them.
- SGStephan Guyenet
Yeah. So that's one of the recent ones. Let's see. I'm just gonna pull it up. Redpenreviews.org. Um, the most recent one we did was "The Volumetrics Diet," "Ultimate Volumetrics Diet" by Barbara Rolls. Um, before that, we did "The Carnivore Code." I was the primary reviewer on that. We have a primary reviewer and a, a peer reviewer Before that, "Eat, Drink, and Be Healthy" by Walter Willett. Uh, before that, "Eat Fat, Get Thin" by Mark Hyman. So that's a taste of some of the ones that we've been doing.
- PAPeter Attia
And, um, uh, given that you've been doing this for three years now, anything surprise you so far? Were there any that you went in thinking, "Yeah, this seems like it's gonna be a pretty good book," and you came away thinking, "No, this... They really didn't get this right," and vice versa, where you went into it thinking, "This is gonna be nonsense," and you came out thinking, "Actually, they've changed my view on something"?
- SGStephan Guyenet
Yeah, absolutely. I think one of the things that has really come into focus, um, through the course of this process is that credentials are not a reliable correlate of information quality. So there are people that have MDs and, you know, like David Perlmutter, he's a board-certified neurologist, and his book, "Grain Brain," got the lowest scientific accuracy score of any book we've reviewed. And I'm not trying to pick on anybody in particular, but that's just an example of the credentials not lining up with the scientific accuracy, and we've seen that in many cases. So it-- I would say, you know, probably there is some correlation there, so I think credentials, they're not completely meaningless, but once you get into people who are highly credentialed, it's just highly variable. Some of their books do really well. Some of their books do really poorly. So that's, that's been somewhat of a surprise to me. I mean, like, you know, we, we-- I think most people who are educated in this sphere know that there's a lot of low-quality information in this sphere, right? But I think you can't help but be shocked sometimes at just how bad the situation is. Like, even people-- and, you know, we know, like, in science, there's a replicability crisis, so scientists are not infallible. There's problems like this even in this peer-reviewed literature. But once you get outside of the sphere where there's accountability and you're in the public sphere where there's very little appu- accountability, it's like a free-for-all. And many people, even those who are well-credentialed, share information that is very low quality, and there's very-- most people have very little ability to detect it. And even, you know, somebody like me, who I consider myself knowledgeable, at least in some areas, I can get taken in, too, and maybe I don't-- I might read a book, and it seems compelling. I'll tell you, "The Carnivore Code," it did worse than I expected. I-- as I was reading it, I actually-- there were parts where I was like, "Hmm, yeah, this is interesting. I'm gonna, I'm gonna look into this. This is pretty-- This is kinda making sense to me." And it wasn't until I started checking the citations and doing scientific literature searches where I was just like, "No, this is really not the-- this is not the best interpretation that a r- that, you know, an unbiased person would come to looking at this, this body of evidence." And so, yeah, I mean, I think, um, there have been many surprises, and most of them were updates in the direction of thinking that the state of popular nutrition books is actually even worse than I thought.
- PAPeter Attia
[chuckles] Well, it's interesting. You know, I, I'm in the process of trying to finish my book, and one of the things that is so daunting is the fact-checking process. Uh, and I don't know what, [clears throat] what an author relies on because [lip smacks] I could certainly never rely on my publisher to fact-check my book. It's too technical. Like, they, they don't have people with the, uh, knowledge to do the fact check, so I have to have analysts who work for me fact check, but they have to be analysts who had no help in doing the research for the book because you have to get fresh eyes on it. So I think that means my book will come out more accurate than it probably would otherwise, and yet that's very difficult. I mean, that's-- I mean, I just, I just know there's gonna be something we get wrong, right? We're just-- we're gonna either, you know, incorrectly cite something, or we're gonna have misinterpreted it or something like that. And I feel like we're, we're in as good a position as almost anybody can be given the size of our team, um, and yet it's still very daunting. Uh, there's something about a book that is daunting in the sense that, you know, from the-- you, and you know this, you've written a book. From the time you put your last, last, last edit into a book until it hits the shelves is about eight months. Something's gonna change in those eight months. Ev- so even if it was perfect the day you finished it, which I think is impossible, eight months later it's not, let alone two years later.
- SGStephan Guyenet
Yeah, absolutely. I agree with that. And, you know, every book's gonna have mistakes, but, and the process that you described is, I guarantee, far more rigorous than most books that are being published in this sphere. And as you said, publishers do not impose a filter on the contents of these books, I mean, very little of one. They just simply do not view it as their job to police the claims of authors. So if somebody comes in with an MD, for example, their position is... And, and this is literally what my publisher told me. They're like, "You're the expert. We're not experts. You know, we are not here to check your work." And so for me, my process was I sent each chapter out to experts in the field and had them look at it. Does that mean my book has no mistakes? Absolutely not. [laughs] Uh, I have, uh, you know, I've been cataloging my mistakes on my mistakes page on my website, the ones that I know about.
- PAPeter Attia
I think that's, I think that's inevitable. Um, and I, and I've sort of accepted the fact that that's going to be the only way I'll sleep at night is just acknowledging 5% of this stuff is gonna be wrong. Let's collectively figure out what it is, and let's create a repository where we can put the updates.
- SGStephan Guyenet
I think that's absolutely the, the best attitude, 'cause not only is that a truth-seeking attitude, but you are putting yourself in a position where you're not, you're not presenting yourself as someone who has to be, you know, right to be rational. You're presenting yourself as someone who's trying to get towards the truth, and your audience can help you and help, you know, all together to get closer to the truth. And that's kind of the, the attitude that I like to cultivate as well. And, um, yeah, you know, thinking about... One thing I wanna mention in this context is that the method that, that we developed for Red Pen Reviews is available on our website for authors to, to see. And it is-- Part of the reason why we do that is 'cause we're trying to help authors m- write better books. Like, there are-- We do random citation checks. That's one aspect of it. Um, we have certain criteria for healthfulness. We have certain criteria for scientific accuracy, and it can't be gamed. Like this-- I don't think our method can be gamed. Like, if you look at our criteria and you write a book that fulfills, you know, s- that you think would score well, that's gonna be a good book, uh, from an evidence standpoint. I don't think it can be gamed. And so, like, when I think about me and my book, which came prior to Red Pen Reviews, I would've loved to have this resource, and I think if I had had it, I, uh, I would've written a book with higher evidence quality. So, you know, just like knowing that there's accountability and having a method that kind of helps you turn that into something concrete to, to improve what you're doing, I like to view our organization as not just being, you know, finger-wagging at, at people, you know, who make mistakes but more providing, also providing a resource for, you know, to help the information quality be good from the beginning.
- PAPeter Attia
Tell me, is the-- are the reviews available to everyone or only to donors?
- SGStephan Guyenet
Yeah, absolutely. They are freely available to anyone. Um, we love receiving donations, but, um, access to our resources does not depend on that.
- PAPeter Attia
Excellent. Well, Stephan, this has been super interesting. Uh, I always enjoy, uh, interacting with you and talking about, uh, all of these things. Um, I think it is kind of amazing how much we still don't know about something that is so ubiquitous and so important. Um, but I also get the sense we're kind of converging, and I, and I do think that these, these good faith debates that exist between people like you, Kevin, David Ludwig, I think they're really good for the field, um, because I think it's forcing people to be sharper in their thinking, uh, and ultimately I think getting us closer to theories that are aligning better with experimental evidence. I know my thinking on this has changed quite a bit, and I certainly find now the balance of the evidence more on the energy balance side of the equation. But, um, I constantly enjoy just trying to understand both sides of this, and it's, uh, it's very complicated. A-again, suggesting that there's more overlap than we probably appreciate.
- SGStephan Guyenet
Yeah. Well, I appreciate you having me on. It was fun to discuss all this stuff.
- PAPeter Attia
Yeah. Thanks, Stephan.
- SGStephan Guyenet
Thank you. [outro music]
Episode duration: 2:48:43
Install uListen for AI-powered chat & search across the full episode — Get Full Transcript
Transcript of episode 9G3iLbQCIHI
