Episode 322: Drs. William Cromwell and Nick Norwitz

Welcome to the Low Carb MD Podcast. No one is beyond help.

No one is beyond hope.

As we have always said, we are bringing you medical information and cutting-edge science, but none of this is medical advice. Please seek out input from your own doctor. Welcome to the Low Carb MD Podcast. Brian's going to be so upset that he missed this one. I have an absolute luminary, well, two absolute luminaries, I should say. Uh, I have Dr. William Cromwell, who, uh, if you've been living under a rock, he's a well-known lipidologist, and we'll do a formal introduction of him in a minute. And we have Dr. Nick Norwitz, uh, who you all know from our podcast now several times coming back and joining us. We're going to be talking about two very interesting topics. We're going to be talking about statins and Oreos, and we're also going to be talking about the new meta-analysis that just came out very recently, which I think is relevant to this topic. Just a bit about Dr. Cromwell. He is a board-certified family medicine doctor. He started his undergraduate medical education at LSU, and he did his family medicine residency in Kentucky, went on and did his lipid fellowship at Washington U. He is one of the co-inventors of the LP-IR score, or the, you know, the NMR lipids that we order now every day. He started a lab which was eventually acquired by LabCorp that had over 17 million samples. So, and he's had 700 research papers, and now he's the CMO of Precision Health Reports. He has a clinic in Raleigh, North Carolina, and he has decided to bring his level down to the mere internist level here at Low Carb MP podcast and educate us about lipids. And coming back, we have Dr. Nick Norwitz, who's a PhD from Oxford and is now finishing his MD. And if you've been living under a rock, you know, now probably more publications on the ketogenic diet in the past 3 years than any other researcher I know. Is that accurate, Nick? That's probably—

um, you probably don't know a lot of researchers then, but I'd like to think I've been reasonably productive. During my time in school.

Well, is there somebody more productive than you? I doubt it, right? So I, I literally have an alert on all the, you know, all the, you know, studies that, that come out related to ketogenic diets. And, um, and you're definitely very active in that front. Uh, but let's get started. Let's, um, maybe what we'll do is, um, Dr. Cromwell, you can start, um, by just talking about this new meta-analysis that, uh, that came out that was very interesting. It was a meta-analysis, and maybe Nick, you can give it just a brief sort of summary of that meta-analysis, uh, which basically looked at the relationship between saturated fat intake, somebody's BMI, and their LDL on a low-carb diet, which I think confirms some things that, that, you know, our listeners are probably well aware of. So Nick, do you want to just give a one or two line about that, that meta-analysis, and then maybe we can get Dr. Cromwell's opinion on it.

Sure, I'm gonna need more than one or two lines, but, um, we documented before some preliminary data about the potential causes of increased LDL cholesterol on low-carb diets. Conventional wisdom would have you believe that things like saturated fat are what are really boosting LDL to the sky-high levels we see in some people. Now, saturated fat can impact LDL cholesterol to some extent. But what we're seeing in what we call the lean mass hyper-responder phenotype, which we can dig into in a minute, are very, very high levels, which I don't think there's much reason to believe based on current literature that saturated fat would be sufficient to drive LDL to like 300, 400, 500. You need something else going on. Now, in a paper, um, that you were on, Drew, back in 2021, we had a cohort that we looked at, and we actually found this interesting inverse association between BMI and LDL cholesterol change and LDL levels on low-carb diets. In other words, leaner people had higher LDL when they went low-carb, and in a dose-response manner. Now, those initial data were kind of limited in the way they were collected and the nature of the cohort. So, um, what a colleague of mine, Dr. Adrian Sotomayor, who has been on here before as well, I think, um, thought of doing was what's called a meta-analysis of randomized controlled trials. So when you think about the perceived hierarchy of study types in medicine. You think about these randomized controlled trials near the top, right? You take people and you randomize them to different interventions, and then you see what happens with respect to, you know, various outcomes. Well, what you can do is a meta-analysis where you cluster a bunch of these and analyze them as a group. So it's a very rigorous type of study. And what we did, there were kind of 3 main findings. One was, did meta-analyses for each WHO BMI class. So lean or healthy weight under BMI 25, overweight 25 to 30, class 1 obesity 30 to 35, and class 2 obesity. And what we found was only in those studies with lean BMI were there actually increases in LDL. Overweight, class 1 obesity, no increase in LDL with low carb. And class 2 obesity, there was actually a decrease. We also looked, and there was an inverse association between BMI and LDL cholesterol. Again, Point being, the leaner you are, the higher your LDL raises on a population level. And then finally, using some individual participant-level data, we looked at the effects of saturated fat and could compare them to the effects of lower BMI. And the punchline is that the lower BMI really did dominate as a major driver of increased LDL on low-carb diet. So this is consistent with the lipid energy model that we've been talking about for a while. We can dig into that. I know you've had Dave on to talk about that, etc., etc. But Really, the cool thing about this study is, is that it confirms this phenomenon whereby leaner people are the ones susceptible to the larger LDL increases, which is really provocative mechanistically and also raises questions about the potential consequences of these shifts, whether they're, you know, natural, quote unquote, metabolic response, and if that has any bearing, which we don't know, on risk profiles. So that was a new meta-analysis It was just a really beautiful study, and I give a lot of props to the team, particularly Adrian, for putting this together. It came out and the reception's been great.

So I want to, I want to take this to the clinical level, and I think Dr. Cromwell, maybe you can give us insight. So, you know, walk us through the last 30 years of knowledge about saturated fat and changes in cholesterol, and then help us understand this new emerging data and this new meta-analysis in relation to your, you know, uh, you know, watching 3 decades of sort of lipid research unfold, how does this change anything, if at all, in your opinion, on how the lipidologists like Dr. Dayspring and other people should— Dr. Weiss and, you know, many other sort of luminaries in this field in lipidology— um, how should they assimilate this data into, you know, modern lipidology?

That's the right question to ask. And as I was listening to Nick recount the, you know, the meta-analysis as it was done, I was thinking to myself, you know, wow, what would my 20-year-old self say to my 60-year-old self regarding what I just heard? You know, so if you go back to the late '80s when I got started, our understanding of lipoprotein metabolism was basically production and clearance. You make, you clear.. And if you have a lot of something, you ever made too much, you didn't clear it, or both. And we, we put things into these categories. And in those categories, you think of diet as something that affects primarily the production axis. You think of insulin resistance, metabolic syndrome, as affecting both production and clearance. And you think of various genetic disorders as being primarily a clearance exercise, like FH. And so you kind of put things in these buckets. The second thing is that we historically have thought of lipid changes in bulk over long periods of time, right? And this is in part because the way clinical trials are done, uh, you have people who come, they are evaluated for baseline lipids or lipoproteins, something is done, and 3 or 6 months later you look again, and we kind of have this rinse-repeat every 3-month window of opportunity to see what's happening.. And that translates into clinical practice. We see patients, we usually have them back about every 12 weeks, so that's kind of our time frame for thinking that lipids are changing. So that is totally different than the work that Nick and Dave and others have helped us understand, is that there's something that is much, much more dynamic and much more robust going on under various physiologic circumstances. And that was one of the first things that I appreciate about working with Dave was the, the fact that he had such a large amount of information on his physiology and could demonstrate just how quickly things were changing for, at that point, as yet determined reasons. So we, we went from this standpoint of thinking, um, diet has something to do with it, it's bulk change over time, and we're looking at turning a carrier in the sea. That's basically kind of what we're doing with lipids and lipoproteins. As you know, saturated fat got a black eye for a variety of reasons. The Seven Countries Study was one, others where it was vilified. You know, saturated fat is bad. There's an association in some populations, but not all, that a certain amount of saturated fat intake might correlate with increased LDL. And then, you know, as you know, totally different conversation, we went into this very, very hard line of Avoid fat, limit fat, avoid saturated fat. That's heart healthy. At the same time, for other reasons, we're becoming obese, insulin resistant, we're, we're carb overloaded and driving worse pathophysiology with our carbohydrate intake. And, and lost to many people is the idea that maybe we got it wrong in the first place. Maybe fat isn't all bad, and maybe saturated fat isn't the enemy. Now, that's heretical in some circles, right? So to even utter something like that was just shot down for a long time. But then you have to step back and say, well, what's the quality of the data that informed us to begin with, and what are we learning now? Mechanistically, one of the things I think is interesting, there are certain animal models out there where you can do a lot of things that are not very physiologic for humans, but you can push systems to an extreme.. And if I push certain animal models to an extreme with high saturated fat, I can actually downregulate LDL receptors. I can, I can get things to a point where the usual expectations of LDLs going up because numbers of receptors are going down, I can demonstrate that. But to get there, you have to use a very high saturated fat intake and an animal model that isn't what we're talking about. But nonetheless, that's kind of informed our thinking, right? So saturated fat at population level could be associated with higher LDL. In animal models, I can increase saturated fat to the point that I downregulate LDL receptors. Uh, all seems to be consistent in the world. Then through the work of a lot of people, we, we kind of have our eyes opened that, you know, maybe carbs are a problem, at least for some people. And we start picking out these insulin-resistant metabolic syndrome people diabetics as really not served well by a low-fat, high-carb approach. And we, we begin to try on for size the idea that carb capping or carb restriction may actually have value. And then we begin to transition to, well, actually, for other reasons, there are nutritional advantages to a keto diet in neurologic disease, in GI, in other places. And so there have been, I think, camps that have been contributing information, and we're just now beginning to synthesize it into something that's not just a one-off here and there. It's a growing body of understanding. And as that happens, we have to do things like meta-analyses because, you know, we don't have one large 20,000-person trial to try to definitively inform us. And when you do these meta-analyses, now we get to what Nick was talking about, and that is some insights that there are dose relationships based on your lean mass or obese status. And, um, what I would take away from it and what I do in my clinical practice is to say, uh, we need to know who you are, you know, what's your physiology, what are the things that are potentially in play, and how can we tune our therapeutic lifestyle to optimize the challenges that you have. If you are someone who has insulin resistance, metabolic syndrome, you do not need to be on a low-fat diet. That's a bad idea. You need to be on some type of carb-restricted diet, and then we can talk about what that looks like. And then the question comes, well, what's the potential downside? If, if I were to adopt a low-carb diet, would everybody have a high LDL, like lean mass hyperresponder? No. Do all lean people who adopt a keto diet have a high LDL? No. Do some people have a lean mass hyperresponse phenotype? Yes. Uh, and what are the characteristics that set that up? We know from this meta-analysis and other work that you guys have been involved in, it has a lot to do with, with leanness. The lipid energy model would say would have a lot to do with carbohydrate stores and whether we are carbohydrate depleted And now we're trying to maintain normal physiology and energy expenditure based on energy that's coming from free fatty acids. And since our diet in that setting is going to be fueled by free fatty acids, how are we going to traffic that? Um, we can borrow from other pathophysiology and talk about later on is to— when things hit the portal circulation, uh, there is a downstream effect. Free fatty acids hitting the portal circulation will result in more VLDL and more export of lipid from the liver. And in the lipid energy model, that trafficking is critical to maintaining energy that we need for daily activity. But what the lipid energy model anticipates that is new is that we can really increase VLDL production and rapidly hydrolyze triglycerides through lipoprotein lipase from VLDL, which means we're pretty dynamic. I, I call this an adaptive model. Our body is able to adapt. We're adapting to being lean, we're adapting to low carb stores, we're adapting to energy available to us in the form of free fatty acids, and we're able in really quick succession to change our energy utilization so that we have adapted, we're maintaining and we're healthy. And it should be that way. If we get stuck on a desert island, we don't want to think that if we don't catch a fish, we're going to die, right? We got to have some way to survive. And maybe that's what we're getting to. I think at the end of all this is clinically, it is more complicated than we thought. There are opportunities that don't have to be low fat. There are opportunities that or not options, low carb, but there are options that, that are, are not, um, fat dependent in terms of having a low-fat, low-saturated diet. And at the end of the day, what we don't know— and Nick and you guys both know this— what we don't know is what is the physiologic consequence of some of these very interesting phenotypic shifts, because we haven't systematically studied them. We're only now beginning to systematically study them. Um, we believe that major risk factors present in abnormal levels for a long time have the potential for risk. But what happens if a normal person who's physiologically healthy and happens to trip into a phenotype where their LDL goes up a good bit, but that's not who they usually are in the first place? The clock has started now that I have a much higher LDL. Is that a problem? When is it a problem? How long does it have to be that way to be a problem? Is it a year? Is it 5 years? Is it 10 years? Are there factors physiologically that would buffer my reaction to this high ApoB load? Is it true that everybody with the same ApoB load has the same potential atherosclerotic consequence? These are provocative questions, and I think they're the right questions to ask. And we're just now, uh, through a lot of good work with a lot of people, beginning to systematically get data to answer them.

I have so many questions sort of bubbling up. When I was exposed to my first, uh, lean mass hyperresponder, it was probably 2017, 2018, and, uh, you know, right around then. And I, I found, uh You know, Dave Feldman, really, he deserves a lot more credit than I think he's gotten for being such a calm force and bringing this, you know, further and further forward. And yeah, Nick, I was happy to work with Dave and you and Adrian to sort of like bring this into the literature, because when I saw it, I was like, doctors need it on paper. And so what we did was you know, at the time, 2017, '18, I'm like, what is this phenomenon? Right? And I have amazing New York City lipidologists who I won't name, who I actually called and I said, hey, I want you guys to see, you know, a patient. And they're seeing, you know, really high levels, you know, of cholesterol increases, you know, as they've lost weight or when they're really, really lean.. And I had done a little bit of digging and I saw that anorexia literature sort of hinted at what was there, that, you know, the more anorexia somebody has, the more the LDL increases, the more sort of severe the anorexia. And then what reverses the phenomenon was actually just eating, right? Eating reverses the phenomenon, the lipid phenomenon. And there were some genetic studies into that. And then I looked into the fasting literature, very similar, similar thing. Ramadan fasting stratified by BMI sort of showed a very similar lipid change. And I, and I asked this New York City lipidologist who's very, very well known, and I said, you know, you must have seen this before in anorexia and fasting. And he goes, and there was no resistance at all. He says, yeah. That's my second most common consult is this fit young female. That's how he described it, with an eating disorder who I have to sort of reassure and get them to eat. And so it hit me that it's not that this field hasn't seen the lipid energy model, it's that they're resistant to sort of applying it in a high saturated fat diet perspective. And, um, and this was the reason for that, that case study we did, or that case series we did. We showed that just a tiny bit of carbohydrate reversed the phenomenon. And yet here I am, 5, 6 years later, uh, still finding that the modern lipidology world is still very resistant to this this information. Can you help me understand why you think there's this resistance? What is it? It's not that I don't think they've seen it. So I want your insight because you have decades of experience in this field. You've gone to lipidology meetings, you go to the— you're sort of the luminary in this field. So what is the barrier to adopting this new information? I would say it's old information, really, you know.

Well, I think some of it has to do with the potential harm that we believe would happen from persistently high ApoB/LDL. So we can start from there and work our way backward. So, you know, as clinicians, as physicians, we don't want to do anything, and if we don't know, in the absence of data, we ask ourselves, what's the worst thing that can happen here? I have somebody with an LDL of 500, what's the worst thing can happen? I can grow a plant. I can take somebody and put them into a situation which is unhealthy and potentially highly risky, and I don't want to do that. So our default in certain cases is to anticipate the worst and try to, you know, manage against that. So if we just start there, that, you know, there is a causal link between ApoB and atherosclerosis, But I believe Peter Attia said it correctly, it's causal but not sufficient. That doesn't mean that we can explain everything by a high ApoB. However, we have FH, we know in FH kids, they're born with poor LDL receptor function, they have very high LDL from birth, they have a consequence from that which is unfortunate with lots of atherosclerosis and early events. So there's the model. And so we sometimes pull back from what we know and say, I see a lot of bad things happening from persistently high LDL. I don't want that to happen to my patient, therefore I'm going to bridle against the idea that that's okay. I'm going to default into being safe and treating this person so I don't leave them in a risky situation. So put it, put in its best light, I think that's part of it, is that people don't want to put people at risk, they don't want to do things that would be harmful, or potentially dangerous. So that's part of it. Another part of it is the way that we explain what we think is happening. You know, anorexia is a great board exam question. You have somebody who comes in with characteristics of anorexia and high LDL, and the traditional explanation is they've downregulated their LDL receptors, because LDL receptors are very energy intensive to make, they're very energy deficient, they're going to have to cut corners physiologically somewhere to maintain critical function, LDL receptors are down, and therefore some of their LDL elevation is due to downregulation of LDL receptors. No discussion of a potential lipid energy model in that, just going back to, you know, this is our working understanding, and, and there are data to support that. Are they sufficient? Different conversation, right? So we go back to what our models were for basic physiology, what our consequences are for increased numbers of atherogenic lipoproteins, And now I think we're doing a better job of asking more questions to put context around the individual. So now non-invasive imaging is a very common extra step. Do you have plaque? Now if you have plaque, that's a different conversation of what we're going to do with your high LDL than if you have no evidence of plaque. And this is a recent phenomenology, that's a different conversation. So now we're, we're not trying to do a one-size-fits-all, we are better trying to understand characteristics that would change the tone and tenor of the conversation. And I think we're also getting to the point where we're trying to have an informed patient-physician conversation rather than us telling the patient, this is what you're going to do. It's a better conversation to say, this is what I'm seeing. These are the things that make a difference. Let's have a conversation. What do you think about this? Do you think this is something that is, you know, do we need to do something different with your diet? Do we need something different with exercise? Do we need to add medication? Do we need to have watchful waiting and leave you the way you are? And if we do have watchful waiting, what does that look like? What are we going to follow? How are we going to at some point find if things are turning badly? So all of these things, I think, contribute to the idea. And it just, you know, last thing I'll say is, um, in the LDL world, it's a very hard thing to imagine that individuals with the types of elevations that we're talking about, 300, 400, 500, don't have consequence. It's very hard for that to resonate with people. It's got to be consequential. Because it's so high. And when it's that high, our experience is that there, there are consequences to that. And, and what I would say just as a simple statement that may help to open the conversation is, how long has it been that way? That's the first question I ask patients when I see them in consultation. They come in with bad numbers. When did that start? How long has it been that way? If it's been that way for a long time, that's a different conversation. If it just happened, that's a different conversation. Not to take us off track, but I had a 70-plus-year-old lady came to me. I'm here for you to tell me I don't need a statin. Who told you you needed a statin? My doctor.

Why?

My LDL is 210. How long has it been that way? You're the first doctor to ask me that. She pulls out a binder, flip, flip, flip, flip, flip, flip, flip. 5 years ago. Prior to that, it was perfect. She had a TSH of 50. We have a hypothyroid patient.

Yeah.

Okay, now we have unmasked part of the problem. CAC score is zero. You don't need a statin. You need to get your thyroid axis fixed, right?

So, you know, I want to come back to that. There's often genetic issues that I have to look into, hormonal issues, like you said, thyroid. And I want to, you know, quickly make one or two points and then segue to Nick So I would have thought that what you discussed was that leveraging non-invasive imaging to reduce harm, not being paternalistic, sharing in the decision-making. But I would have thought the first thing I told Dave we need to publish is the solution. I was like, people are going to freak out and they're going to see this problem. I was like, we need to show them the solution. Right. So like my whole thing was I want to get this case series, you know, of carb reintroduction showing that the LDL comes down so that, you know, we're like sort of massaging the LDL, you know, fanatics, let's say. Or maybe that's a strong word, but, and, and yet there was still this hesitation. It's like, guys, we're showing you what to do, you know, we're showing you how to massage. So I don't know if that If it's just this idea of there's harm, you know, we gave them the tool in the literature to at least, you know, reverse the phenomenon. And just to come back, you know, we just did a practice-wide, you know, a correlation analysis. We found that in our clinic, CAC was inversely related to LDL. There was no relation. Maybe a slightly negative correlation, right? So I'm here unable to reconcile this data. And so I'm looking forward to everything that Dave is working on and Nick's working on. And maybe this is where we can segue to Nick. Nick, can you talk about what you want that lipidologist out there to get from the meta-analysis you worked on, and then maybe talk about what you, you know, people have asked me, I've talked to 2 reporters already about the statinversarios, right? Who they may or may not have talked to you. And they said to me, what does Nick want to accomplish with this? Because it's, I know, I think I know, but I think out there people wanna know what is the statinversario mean, right? What does it mean? And so can you give your side of it? What do you want the lipidologists out there to know, you know, from that meta-analysis? And what do you want them to understand? Maybe you can give an idea of this experiment, this really cool experiment, you know.

So from the meta-analysis and all the other work associated with the lipid energy model and lean mass hyperresponders, what I want academics to get out of it and clinicians first and foremost is curiosity. Curiosity that, you know, is required to push the science forward and curiosity that in my humble opinion as an early-stage researcher and a medical student, is the best thing in the long term for patients. I think Dr. Cromwell articulated really well why I think people might be resistant to this line of research because it seems to be instigating, let's say, inappropriate behavior by patients. I don't see it that way. I see it as we're asking pertinent questions that are at the boundaries of the unknown. In order to echo echo Dr. Cromwell's point, you know, wrap context around the individual, we need to pursue this literature. We absolutely do not know the answer to say, you know, what is the consequence of having a lean mass hyper-responder phenotype over many years? That is a complete unknown. And that's exactly what we're saying. We're asking the question, you know, is this a high-risk profile? And we need to refine the data we have in order to answer that question so that we can better treat individual patients. Now, that doesn't mean in the meantime we can't treat conservatively. In general, I'd advocate for that. Now again, it's a patient-physician conversation to have, but this is just about asking questions that, you know, if I put my scientist hat on, are crazy fascinating. Like an inverse relationship between BMI and LDL change when you're low carb, that's interesting. Um, and then when I put my med student hat on, trending towards clinicians, like, well, what does this mean for patients? And I don't see how you cannot be curious, especially in the context of the rise of ketogenic diets for a whole host of things. Now we have like RCTs on autosomal dominant polycystic kidney disease. We know it's good for epilepsy, mental health disorders, neurological conditions, the weight—

obesity, diabetes reversal, right? Obesity, right?

Well, I was, I was listening, I was listing things that weren't necessarily related to obesity and diabetes reversal because those patients would be at lower risk for the LDL changes. So what I'm saying is the population of patients who could be therapeutically treated with expanding clinical use cases of ketogenic diets is expanding. So we'd expect to see this population of lean mass hyperresponders expand as well, which I think we are. And so the boogeyman, so to speak— maybe it's an appropriate boogeyman— still remains the LDL on low-carb diets. We need to understand what's going on there so that when patients find themselves in a rock and a hard place situation, we can figure out, well, how hard is that hard place? And we can better, you know, inform decision-making. So With respect to the meta-analysis, again, meta-analysis of RCTs, that's the kind of rigorous data that I think, you know, academics should look at and really, as you said, assimilate into their knowledge base and consider what this means.

And has that happened?

And has that happened with, in your opinion, with the, with the RCT, the meta-analysis that just was released?

It was just released, so I can't really evaluate that. It's a couple of days old. I, I'm hopeful. I'm hopeful. But what, as I was saying, transitioning, I want to capture more than just Jade Tower academics with respect to curiosity around this phenotype. I am realizing that practically speaking, there is power in the grassroots movement and public curiosity. It's a really weird time, honestly, to be a scientist and medical student, uh, 2024, like This is a bizarre time where the public has incredible access to information through the internet and through social media. So they're going to get information. And the question is, you know, are you going to be one of the content producers? And if so, you know, what influence, what voice are you going to have? Um, I am really inspired by things like, I guess I would call it the citizen science movement, where people are really investing and becoming curious about their own health. And I think overall that's a fantastic, fantastic trend. And the fact that we can, you know, on scale give information to people to empower them to take control of their own health, I think is really, really cool. Now, with respect to lean mass hyperresponders and the lipid energy model, I guess what I would say— I'm going to cast as broad a net as possible because I was asked this question by a reporter as well who posed it as this. She said, all right, you know, how is this relevant to your average person? Because most people aren't low carb. Most people who are low carb might not even be lean. So the population of people this might actually be pertinent to, this whole Oreo versus statin thing, which I'll dig into a minute, might actually be really small clinically speaking. So, you know, why would I write a story on this? Why is this interesting to your average person? Well, aside from the headline, which is, you know, Oreo lowers cholesterol more than statin in Harvard medical student, which definitely, you know, clickbait potential there. For me, it really is about pursuit of curiosity, and I think that's something that we can all get on board with. This is something that I am so awestruck by every day, and I wanted a way to translate that scientific awe— I've been calling it nerdthudie, nerdthusiasm— to like the general person, to give them something, an observation, a legitimate observation, even if it's an unequal one, where they can't help but just stare and say, what is going on there? And then hopefully chase down the rabbit hole to join us in this scientific pursuit. So in the broadest possible way, that was my intention. Narrowing it a little bit, and maybe the more honest answer, as Dr. Cromwell said, productive provocation. And as you said, there's been a tendency to want to ignore this topic. I really tried to push for, I think, pretty nuanced conversation with people who have very big platforms. I won't name names. You probably can know who I'm thinking about. Putting together very carefully worded responses and inviting to have discussions with me. That has not happened. Uh, maybe it will happen in time, but I'm by nature an impatient person. So I wanted to design something that would catch attention and force people to talk about it via basically just dropping a little bit of a media nuclear bomb. And this, I thought, Oreo versus statin would be that tool. So to speak. So about half a year ago now—

did you contact Invisco yet? I'm just curious.

I have not. Um, about half a year ago now, I kind of envisioned a crossover experiment that I do on myself, um, based on principles of lipid energy model, based on n=1 experiments others have done, including Dave on himself, based on data like from your patient case series, whereby I reintroduce carbs and lower my LDL. Now, what hasn't been done before is a pure addition. So I'm not swapping out carbs for fat, a pure addition, and using intentionally a perceived to be unhealthy food. What is the one food that everybody knows and is just like, this is not a healthy food? Oreo cookies. So I'm like, all right, I'm going to try to lower my cholesterol with Oreo cookies. Then just because that's not enough, I want to actually compare it to something. What do I compare it to? Statin therapy. Of course, standard of care therapy. So the study design was I have a running ketogenic diet, continuing basically my normal ketogenic diet, then have about 2 weeks— that ended up being 16 days— of 12 Oreo cookies per day, have a washout period to return to my baseline BMI and cholesterol levels, and then engage in 6 weeks of statin therapy. Dr. Cromwell helped me decide on 6 weeks specifically of 20 milligrams rosuvastatin daily. So that was the protocol. Now the results— here's the drumroll, and this podcast will go out after the paper drops— the Oreo cookies were about 2 times as potent in about 1/3 the time frame as the statins. So the Oreo cookies lowered my LDL from 384 to 111 milligrams per deciliter. That's a 71% drop in 16 days. And the statins lowered my LDL from 2— sorry, 421 to 284 at nadir. So it was a 6-week trial. The lowest point was actually at week 4. And even that lowest point was a 32.5% drop. So the Oreo cookies in relative terms were twice as potent, and also about in absolute terms twice as potent. So that's pretty stunning. Literally, Oreo cookies doubled the effect of high-intensity statin therapy and work quicker. So that's the kind of thing where people should really— I mean, there's your, you know, your excerpt, your clickbait, everything I just said. That's something that everybody should stop and be like, is this real? Is this a joke? What's going on here? And then engage in the conversation that leads to the lipid energy model and lean mass hyperresponders. It was a difficult thing for me to do in terms of perceived ethics, let's say, because I don't want to make a joke of this because this sounds almost like a joke. That is not my intention. I'm trying to force a conversation. This is actually a very serious matter. And so one thing I'm contemplating, and I won't know until this comes out and we see what the response is, but— and I posed the question this way to Professor Walter Willett. I was at his house interviewing him the other day, was I've expressed my intention, provoke productively conversation around Lee Messheper's lipid energy model to push forward research that needs to happen. But I'm also, you know, starting a fire, so to speak. I think I am. Am I going to lose control of this fire? Is there a narrative here that's going to spin out of control that I, you know, can't control? And will the net impact be positive or negative? Now, I wouldn't have done this if I wasn't pretty sure the net impact would be positive. That said, I want to be forward about the fact that I can't possibly know that until after the fact. So I'm going to be, you know, definitely trying to caveat this as much as possible, definitely trying to make my voice as loud as possible in terms of, you know, saying things like— and we say this everywhere— this is a metabolic demonstration. Oreo cookies are not a health intervention. But what people really need to, I think, acknowledge is what happened here, even if it's only one patient, primary clinician treating this one patient, is you have a circumstance where a clearly unhealthy intervention resulted in a dramatic, quote, improvement in a biomarker that's presumed to be bad. So how do you hold those two things and reconcile them? And I'm not telling you what the answer is. That's the point. I'm not telling you what the answer is. I'm posing an implicit, uncomfortable question, which I think everybody needs to grapple with and just realize this is an uncomfortable question. That science needs to lean into, not shy away from.

Well, can we hear from Dr. Cromwell? Dr. Cromwell, can you help me understand? You know, because, because you and I are going to deal with these patients, right? We already do, right? We do. They come into our office. And I in no way think that you need to think about the narratives that are going to come off. Nick, I'm going to disagree with you. Your job is to elucidate information and spread knowledge as a scientist and a doctor. When you have a one-on-one clinical relationship with a patient, that's our job. Our job is to come up with what is going to apply to the person in front of us. And I disagree with Walter Willett that we need to be thinking about that before we give a message. We need to be thinking about—

That was me. That wasn't Walter. I posed that question. That was me.

Yeah, no, I wouldn't even think about that. Your job is to spread awareness and knowledge, right? And spread truth.

You can't— we—

I don't think we should decide for others what they should make of information, right? And so I don't know, Dr. Cromwell, tell me, like, hearing this experiment— and I, I guess you knew about it, you were, you know, made the recommendation on the statin, um, how does this change how you view the patient who comes to you with a high cholesterol? Does this change at all? Does this change anything for you?

Well, it does, um, but as I was listening to Nick describe this, and, and this is something I've held fast for a long time, and that is, uh, we don't do a good job placing people into risk buckets in a traditional sense and then believing everybody in the risk bucket is the same. This is the old exercise of what are your risk factors, what's your calculated risk score, you're in the high-risk bucket, everybody in the high-risk bucket gets the same stuff because you're all high-risk. That's fundamentally not true. Similarly, uh, if you play that in reverse, what do we know the consequence of the phenotype as it is being encountered in somebody who hasn't been that way in a long time? So what this does for me is it reinforces the need to make sure that we're having a broad, comprehensive conversation one patient at a time. I believe for a long time, as the guidelines are now beginning to more clearly articulate, it's your clinical history, it's all your risk-enhancing factors, it's your biometrics, it's your biomarkers, it's the harmonization of 6 national and international guidelines that inform the conversation of where you are and what needs to be considered. And that's the whole genesis of Precision Health Reports. That's why we put those reports out there, so that people can answer a few simple questions, get their blood work done that's highly focused on this, and then get all the details informed by non-invasive imaging, et cetera, that allows you to say, this is you. And now we need to describe or consider what is the impact of this lean mass hyper-responder phenotype, number one. Number two, the thing this taught me is If you are bothered by the LDL, simply reintroduce carbs and your LDL will go down. So the simple solution, Tro, like you said, was give them the solution. The solution is I can augment your carb intake. And if we're trying to get a keto effect, what I don't know— Nick would have to tell me— you know, how many carbs can you take in and maintain the benefit of keto and at the same time blunt the LDL response? We're trying to kind of thread a needle here, right? So if I'm bothered by the LDL and I want to get it down, I refeed with some kind of carb. And I try to do this in a way that blunts the bad guy, quote unquote, but at the same time leaving you in a state where you're benefiting from the keto, right? So, so that I think is an interesting conversation to have with somebody, especially to say, listen, I've got a condition and a keto diet has helped me. My epilepsy is better. Something important is better in my life. I want to stay better. Excellent. And let's say I did my due diligence and you had a CAC score of 500, and I'm bothered by this super high LDL response. Can we agree that an ideal scenario would be your epilepsy stays great, you stay on your keto diet, and your LDL isn't 500? Okay, that sounds like a win-win. And can we get there by introducing a few carbs incrementally to see how well we can do both? You stay healthy on a keto diet your LDL stays somewhere off the moon. And we're trying to, you know, kind of navigate down thoughtfully a road which isn't obligating people to a statin, it's not obligating people to, you know, this type of thought. And it is nuanced because if this person is physiologically healthy with normal arteries and this has just started, they could talk me into, you know, I'd like to watch this Okay, when we watch this, we're watching this with our eyes wide open, and we can't predict the future. We don't have a crystal ball. We have to consider what the potential pros and cons are. So if you're good with it, and we're both knowing what we're going to be watching, and we're checking in regularly, let's do this. Let's go ahead. And now we're, we're in that informed patient conversation mode, right? This is where I, as a clinician, uh, would like to see people end up at the end of the day, their curiosity stimulated, data produced, which is missing, helping to fill in our understanding of the boundaries of this conversation. And then it turns back to a better informed patient conversation with some simple solutions that, and if you don't like that LDL, how about we change your macros and see if we can't both improve the thing we worry about as well as maintain the benefit that you're seeking.

Yeah, I worry right now. So you won't be recommending Oreos anytime soon then? Yeah, but I worry—

I hope a carb is a carb and they have many choices.

Well, actually, interestingly, and maybe you probably know this better than I do, I think fructose has been implicated in raising LDL and insulin resistance. So probably the one to avoid, but the carb to avoid. But I want to chime in on one little key point here. Over the last 5 years, and this is really thanks to Nick Norwitz, we've been doing genetic testing on every— we started with the LabCorp's 7 or 9 gene panel, and we've expanded to GB HealthWatch's. No affiliation with either one.. And now we have a huge repository of genetic data. We found there's ways to sort of even augment this CARBRA introduction. Some people have that ABCG8 genes that maybe are associated with plant sterols, and maybe some phenotypes or genotypes that respond to increased fiber with meals. So we've, like, along the way, you know, the amount of nuance I've learned has just been phenomenal. But I think what— sorry, that's my watch— what really amazes me about these results from the Oreo experiment is the magnitude of LDL drop juxtaposed to very expensive medications like bempopoic acid and PCSK9 inhibitors, and which are very, very, you know, costly and, you know, put a burden on our healthcare system. One of the problems I've had, and I want your opinion on this, is when you have some drugs like statins have a pretty clear all-cause mortality benefit in some studies, bempopoic acid seems to— in primary, there may be a hint towards an all-cause mortality reduction. Okay, how do we reconcile that we can lower an LDL more potently with an Oreo, right, versus these medications that may have an all-cause mortality benefit in certain populations, right? So how— when do you decide to say, now's the time for the drug, you know, and drop the nutrition approach, or How do you, as you're assimilating this information, how do you go about making this clinically applicable? You talked about that patient with a CAC score of 500. So maybe we can start right there.

Oh, I think the simple answer is that it doesn't have to be just one approach. We can have both oars in the water. I tell some of my patients, If you're in a canoe and you're rowing, you want to go straight. That means you have to put the oar on both sides of the canoe and row. If you only row on one side of the canoe, you're going to go in a circle. All I need is drugs, I'm going to go in a circle. All I need is diet, I'm going to go in a circle. I need to fix the thing which physiologically is wrong with me as best I can. And I ask myself, and how are we at that point? So I've optimized therapeutic lifestyle. I've treated comorbidities. The person has done as well as they're going to do. And my question is, what does it look like at that point? Are we talking about people who have had events? Have they had interventions? Do they have subclinical disease? Who and what are we talking about? And then when we find ourselves in the position where we have optimized the non-pharmacologic things and we have individuals with real disease, then the medications that have been shown to give us the best likelihood for hard outcome benefit are reached at that point. That's what we're doing. We're trying to affect the natural history of the disease and optimize the potential for outcome improvement. But that to me, I think, is, is the easy one. The harder one are the people who are not yet declared clinically or subclinically have anything.

Right?

We don't have an event, we haven't been intervened upon, we don't see anything subclinically as far as imaging is concerned. And they have numbers that didn't improve when we did all the right therapeutic lifestyle things, or they improved but not to the point where we're satisfied. And this again is where we step back and we start integrating the larger picture. What are the risk-enhancing factors do we have? Oh, I've got an Asian Indian ancestry individual with a high Lp and rheumatoid arthritis, and I start adding more and more things to the equation that independently drive vascular risk, that's going to be, I think, a conversation where we're reaching for a pharmacologic solution potentially more so than somebody who has none of those things. So again, we go back to the individual.

I'm going to make it even more, you know, you have that person with the CAC of 500, his lipid level is 400, and you say, here's your Oreo for an LDL of 100, or here's a statin for your LDL of 220, right? So, so that's like, that's a real clinical conundrum here because we're used to following a biomarker, and our guidelines are becoming increasingly reverting sort of unfortunately back to, uh, number-based sort of guidelines. At least in Europe now, it's less than 50, I think, right, is the— so what the European lipid, uh, uh, sort of benchmarks have gone in secondary prevention. So, you know, if I use an Oreo to get to that guideline, right, versus a, you know, drug that's been shown to reduce all-cause mortality in primary prevention and high-risk or secondary prevention, you know, help me understand, what do I— do I toss these numbers out and Or am I going with the, you know, long-term studies that, you know, have showed all-cause mortality benefit? This is the question I'm left with, right?

Yeah. Well, I think the other thing we can put into this is the aggregate data that we now understand, that the lower we maintain LDL and ApoB, the more impact we have on plaque stabilization, plaque regression, and events. And this is why the European guidelines, followed by other guidelines, have emphasized a threshold and not a target. There's not a point where we stop, there's a number we stay below to optimize outcome benefit, and as far below as possible. So if I have a 500 CAC score in somebody with real disease, I'd like to prevent new plaque formation, I'd like to stabilize existing plaque, I'd like to create a scenario where all in, we have a decreased likelihood of events going forward. Part of that's going to be reversing, um, some of the drivers that made the LDL go up. So I'm going to refeed you with some carbs so we can get that axis to tone down to something which is not as adaptively out of range, right? We want to get that back down in range, but that doesn't solve the problem of your plaques natural history, new plaque formation, or plaque stabilization. So now we're going to consider what does that look like to maintain LDL levels sufficiently low that we can stabilize and regress plaque and put you in a milieu where you're most likely to have the potential benefit for decreased events down the road. And if you happen to have a great response to all the above and have an LDL cholesterol of 40, you stay there. Well, can't I get off of some of my stuff and be at an LDL cholesterol of 70? You can, but you're giving away the potential for plaque stabilization, you're giving away the potential for plaque regression. Those are things that scale in a linear fashion all the way down to an LDL cholesterol of 20. So we haven't— and this, this 10 years ago was an interesting conversation because 10 years ago we did not have convincing data that we knew what happened when you had LDL cholesterol less than 50. As you went down to 40 and 30 and 20, was there better event reduction? Was there better plaque stabilization? Was there better plaque regression? We had no idea. Some people felt, well, you know, you get to some point and it would attenuate, and further reduction might not translate to any of those things. And yet, with existing agents and the trials that have been done, it continues down. Now, the flip side of that is, is it true that people with very low LDL in coronary disease still have events? Yes. I didn't say it fast enough, yes. So optimizing LDL is important, but there are many other things that we need to be aware of, and I think we do ourselves and our patients a disservice by, as important as LDL can be, making it the only solution to a complex problem. And that's where I think we have to step way back and not be so LDL-centric that we lose lots of opportunities.

I like that. You know, we've heard a lot about the glucocentric model, and I think we need to start thinking about that LDL-centric model and moving away from it. Certainly we know that SGLT2 inhibitors increase LDL, decrease cardiac events. We know that Vascepa, or, you know, prescription-grade EPA, seemingly reduces soft plaque volume over the serial CCTAs. We know exercise reduces those plaque volumes, especially those in the crack cores. And when I was trying to do my own little systematic review on plaque reduction, I was most compelled by Vasipa, right? And I found that it's like the least embraced by lipidology, maybe because of mineral oil, maybe because of not, but it's like the least embraced. So I find this idea and concept of plaque volume, you know, and plaque stability— again, I find like this cognitive dissonance in the lipidology. And I'm not, you know, gearing this towards you, but in the field of lipidology, I, I just, you know, between the lack of embracing of imaging and the lack of support of EPA and the lack of curiosity around, you know, this lipid energy model, I'm left disenfranchised from the lipid community. But you give me hope. You give me hope.

I'm with you in that. I think the VASIPA data is stunning. When you, when you ask two questions, number one, these individuals had state-of-the-art medical care, right? They had statins, ACE inhibitors, ARBs, other things that are disease-modifying drugs. Their LDL cholesterol was reasonably controlled. They had triglyceride of 150 to 499. They were randomized to placebo or 4 grams of piricosapentethyl, and the number needed to treat to prevent any event— not just the first event, the number needed to treat for any event— was less than 10. We have never seen an intervention in lipidology with a number of— treat less than 10 on a background of statin therapy. You ask yourself, well, what's it doing? It's not affecting your lipoproteins in a way that describes the benefit. As you know, due to chain length and position of double bonds, it becomes a part of the membrane. It decreases cholesterol rafts, inflammation, lipoprotein oxidation, foam cell formation, plaque rupture, thrombosis. It works at the site where all the action's happening., and it, you know, does not allow this complicated pathophysiology to go unchecked. It is absolutely complementary and helpful to other approaches. And so hopefully SGLT2s, icosapent ethyl, EPA, we got to think broadly. And Nick, you know, we got to think broadly about ascribing that we know what we don't know, right? That's why we have to be intellectually curious, and we have to consider, can my lipids actually change a lot in 14 days? Can I actually drop my LDL 70% in 2 weeks? And what model accommodates that? Is that my LDL receptor? What did it just do in 2 weeks to do this? Well, You can affect an LDL receptor quickly, I will tell you that, right? So PCSK9s, if I give you a shot of Repatha, 4 hours later you have no circulating PCSK9, it's undetectable. And in daily studies, I can tell you that 7 days, actually 5 days after an injection, you are steady state. So that is the fastest pharmacologic LDL effect I've ever seen. This isn't that.

What's the average effect size? The relative effect size of Repatha after 5 days where it stabilizes in percent?

50% reduction in LDL.

And the Oreos were 70%.

Yeah, so different mechanism, right? And that's the point is that if, you know, and this is something that, you know, again, you and Dave and others have helped me understand. And Dave will tell you, when he and I started talking about this 3 years ago, I was unconvinced. I mean, I was interested, but I was, you know, coming from where I came from with my, you know, usual way of thinking in lipoprotein physiology. And the first thing I had to understand was things that rapidly transition and why they rapidly transition. And I think the lipid energy model has a lot of value in helping us to accommodate the changes that you see in this very unique population of lean mass hyperresponders. And so this, this helps me clinically say, isn't it great that we can do things that are very dynamic in a short period of time and we can be nuanced? We can do a little bit of carb reintroduction and it's not going to take a statin. As a matter of fact, in your case, Nick, is kind of a home run response, right? 30% reduction with a drug that—

32.5% reduction. Yeah, it was 32.5% at the nadir, and then it came up a little bit because I exercise more. But yeah, it's something. And it does allow— I mean, the open-mindedness about this allows for toggling of variables that I played in with myself to create some pretty interesting protocols. You asked before about kind of, you know, you had the kid with epilepsy, and as soon as you said that, you know, how you kind of walk down the line, I was coming up with protocols in my mind, which I've never thought about about before, but there are really interesting things you could do. So for example, actually calling back to, to some of the data we're going to report in the Oreo versus Statin, I think the driving force is hepatic glycogen depletion. And there was a, a suggestion of this in the study because what I was doing is I was trying to control ketosis by dosing exogenous ketones, but I can only dose exogenous ketones when I'm awake. So there's a period of time in the morning when I could actually measure what my endogenous ketosis was, which is kind of a weak man's proxy for hepatic glycogen stores. And it was actually after 1 week of the Oreos, my LDL only dropped to 308. It was in the 2nd week where there was a huge drop. And interestingly, my morning ketones before I gave the exogenous ketones, so in the kind of fasted state to kind of again get maybe a sense of my glycogen stores, they were averaging like 0.5. So I was actually in nutritional ketosis the entire 1st week. So probably the 100-gram dose was taking time to fill up my glycogen stores. Then when they filled up, that's when my cholesterol dropped like a rock. So maybe the effect could have been faster at a higher dose. I bring this up because if that is indeed the driving force, then it informs what we need to do in order to, um, let's say titrating carbs in order to kind of find that, that ideal place between controlling the seizures and say, um keeping your LDL at a desired level. It's definitely harder in a kid like, you know, a person with epilepsy or who needs therapeutic ketosis as compared to someone with, say, diabetes or obesity. Because in that case, add 100 grams or 150 grams of like low GI carbs, your insulin is going to stay low. You're still going to get the weight loss benefits and the glycemic control even at slightly higher carb doses, even though you won't be in ketosis. But if you need ketosis, what are things you can do? One, if you can afford exogenous ketones like a free BHB, You can use that to augment. You can use MCTs. Here's actually something that was occurring to me, and I was smirking about it because it's a little ironic. Polyunsaturated fats, but not because PUFA lower LDL. PUFA are more ketogenic than saturated fat. So if you need therapeutic ketosis, shifting your diet to one that is higher in PUFA will actually raise your ketone levels so you have more of a buffer to bring them down to your therapeutic threshold. While titrating in the carbs. I've played with this myself. I know everybody's a little bit different, but if I eat higher saturated fat keto diet but same macros, my ketones can be like at, you know, 0.8, 1.0. With certain PUFAs, particularly for me, I don't know why, sesame oil, I can get my BHB to over 6 in 20 hours. That's how ketogenic they are. I've done this a few times. The highest it's ever been is 6.8 in 22 hours. But again, Just opening your mind to the possibility of things like the lipid energy model and how potent carb reintroduction can be for lowering, it kind of inspires one in a clinical mindset to start thinking about what creative things can we do to, to find the sweet spot. And in this case, it literally could be you shift from saturated fat to polyunsaturated fat. Maybe that has some direct effect on LDL, but it also gives you more of a ketone buffer in, say, a child with epilepsy to titrate in more carbs. Just as an example of like, we can get pretty creative, um, with protocols.

You know, I have a lot of patients who would love that conversation and who do love that conversation. They, they want to hear a lot more about nutritional intervention and a lot less about what pill do I pick up at the pharmacy, you know. And, and I don't blame them. And so as, as we are, you know, learning these things, as you have this information, this is very, very helpful. Because now it increases the tools in my kit, right? This gives me an opportunity to have a quality conversation with somebody that is, you know, again, more therapeutic options, not necessarily pharmacologic options, but they're therapeutic options.

I have one parting question before we end this, and you guys share how people can contact you, or maybe Nick, you can also share how people who are listening who want to make sure this next round of research happens where they can sort of help. Um, but you know, one last question for you, Nick. It's which felt worse?

Oreos.

Oreos felt worse?

Oh, way worse. Wait, way worse? Yeah, the statins, I had a little bit of myalgias. They were pretty tolerable. I was keeping Dr. Cromwell informed. My, my CPK went up a tiny bit, but it was a pretty mild elevation, if anything. The Oreos, the stomach pain sucked. I didn't sleep well. My emotions were like all over the place.

Yeah, there goes the Nabisco sponsorship. So I don't know.

But I mean, that obviously, I used to love Oreos as a kid. I thought I'd really maybe get a little bit of hedonic pleasure out of it. But after being low carb, sugar free for like almost 5 years, 4 and a half years, you throw Oreo cookies as an addition into your diet every single day, your stomach is not ready for that.

I remember you sharing pictures of you sort of way back when and eating foods recommended by a dietitian and you telling me how miserable you were. So I didn't know sort of which one was going to be worse for you.

It was a fun conversation.

Part of me wishes you said statins, but, you know, okay.

I know everybody wants me to say statins. This is not the first time I've been provoked by that question, but the honest truth is The Oreos were more unpleasant for me. Now, if you talk to 12-year-old me, yeah, that wouldn't have been an issue. But I'm not 12 anymore, so I'm sorry, Oreo. Also, if you're listening, Oreo, my girlfriend says she wants you to bring back Dulce de Leche flavor. So if I can pull that, you get me a lot of brownie points.

Yeah.

Or we should call it Oreo points. We don't want to, you know, keep the sponsorship potential gone.

Brownie points. Dulce de leche Oreos, my request.

How do people find you if they're interested in getting a Precision Health Report? How do people find you if they need lipidology help? And, uh, Nick, maybe you can guide people afterwards.

Yeah, yeah. So one of the easy ways is precisionhealthreports.com, all strung together as one long thing. So precisionhealthreports with an S dot com, that gets you to the website that will describe, um, what we do to try to make sure that you have that informed data that's very personalized for the decision you need to have with your care team. With Nick's help and Dave's help, I'm slowly having a presence on Twitter/X. I can't say that I'm social media savvy, but I make a few contributions here and there. So I do have some content that is out there on that. Appreciate your podcast and a few podcasts that I've been invited to be on. That's another way. My clinical practice right now is really part-time. My full-time gig is really more research and teaching. The patients that I see primarily are those that are already mine. I see selected patients on selected days during the month. So I'm still clinically active, but the bandwidth I have for new patients is fairly low. I don't mind doing second opinions. So SecondMD, for example, is a platform where people who have various conditions can get linked up with specialists. And it's an opportunity to kind of do some one-off conversations of things that might need to be considered, what workup has been done, what hasn't been done, what are the considerations you need to talk to your doctor about. So happy to do it that way as well.

Awesome.

Thank you.

Nick?

Well, I think people listening probably know where you can find me on Twitter. I'm @NickNorwitz, N-I-C-K-N-O-R-W-I-T-Z. So check out my Twitter when this drops. We will have a lot of coverage. I'll be posting all the podcasts we do, obviously the paper. There will be a video abstract to the paper, which we're hoping to have some degree of virality. So if you check that out, which you can find via Twitter or via my YouTube, and maybe, Trey, you can link it below, please watch that, leave a comment, like it, share it around because it's really beautifully done. Dave did the editing and he basically killed himself over it. Um, but it's, it's, it's brilliant and you'll really love it. It's, um, I think it's engrossing. And then the most important thing you can do, um, this is really my call to action is, Joe is going to include the link to the paper below. I'm going to include it on Twitter. It'll be associated with the video abstract. Take the link to the actual paper and copy it into an original tweet and tweet it out. It can be a quote retweet of me or someone else or a completely original tweet. The key is to include the link. This helps with paper attention metrics, and it just provides incentive to promote doing more research like this. So, and for journals to publish research in this area. So again, take the link to the actual paper and include it in a tweet or a quote retweet. That really actually has a big impact. Each individual can really make a contribution that way. So thanks for having us on. This was a really fun conversation. And I look forward to seeing how this drops.

All right, guys, thank you.