This Sugar Science Blew Our Minds! (ft’ Dr. Andrew Koutnik)

Psychological stress forces the body, particularly the liver, to make sugar.

This is the beautiful part of science. If you read a paper like this, it's incredibly impressive.

How do these findings like shed new light on that phenomenon?

It's one of the challenges of type 1 diabetes because with type 1 diabetes, there's going to be a whole species of diabetes that will be enriched in, say, people taking developing the ability to turn on and off parts of your brain. This paper elucidates something incredibly novel. We have known for some time there's different forms of diabetes.

What this paper illustrated, we have our first return guest with us on the platform this morning, Dr. Kutnick. So welcome back, thanks for joining me.

Hey Nick, it's an honor to be here. I'm excited to talk about some unique new papers that have recently come out that are of interest to, to both of our, uh, scientific, uh, fields.

So the first paper I want to discuss, I want to jump right into it, was published just recently in Nature. Basically, high level, it shows that stress, psychological stress, forces the body, particularly the liver, to make sugar. Make new sugar and churn it out. So my questions to you are twofold. First, when you read this paper, what was your immediate reaction as a scientific mind? And second, what is new about the mechanism discovered in this paper?

Yeah, so my first reaction was apprehension, Nick. I thought, okay, is it really going to elucidate a novel model or fascinating new, new finding? And when I actually read it, I was blown away, actually. It was, it was quite fascinating because they discovered a direct mechanism by how stress induces not just this breakdown of stored glucose. This is how we think of stress-induced glucose elevation. We think, okay, it causes stress, your body starts breaking down glucose immediately, spilling into the bloodstream to mobilize nutrients for, uh, let's say if you're running away from a lion or a dinosaur, whatever the analogy might be, you want to mobilize those nutrients immediately so that you could have it for your ability to survive that moment, right? That's what your body is built. But this mechanism specifically shows a novel way in which it does it, where the stress specifically triggers a key part of the brain. They called it the medial amygdala neurons with a connection to another part of the brain called the hypothalamus. But what's different about this mechanism, unlike the other mechanism, is that it actually directly almost sends electrical signal, kind of like turning on a light switch of stress direct to the liver that flips on not just the breakdown of these stored glycogen molecules, which is how we typically think of stress-induced glucose elevation, but this mechanism actually shows that it can cause the production of glucose. So it seems that this, this novel mechanism is showing us something very unique, that our ability to mobilize nutrients during stress, typically for us to— as a survival mechanism, if we were running away from, again, a lion or a dinosaur or whatever it was— is actually twofold. Both the breakdown of glucose, but also this unique mechanism that it also facilitates the production of glucose from a direct brain-to-liver, almost electrical signal, uh, nerve-based, uh, synapses from the top of the brain all the way to the liver, mobilizing glucose. And I think the implications of this are seen all over in the real world.

Yeah, thanks for the summary. And the other, uh, people probably gonna overlook the joke about the, the dinosaur, but that would be a nice discovery if we were running from dinosaurs. That aside, yeah, I, I, I do wanna double-click on just how fascinating this is because I feel like the, the clinical real-world applications can, which we're gonna get into, can get separated from what are just so fascinating scientifically. So to elaborate on what you said, like, yeah, we have this brain region that signals stress and then it communicates to another brain region. We typically think about it as changing the hormonal status of the body, activating the adrenal glands to release cortisol. What this paper found is something entirely different. So we won't necessarily go into all the details of the, uh, tools they used. I'll explain them in a newsletter I link below, but they were effectively able to show, like you said, a direct neural line from the brain to the liver that forced the liver to make sugar completely separate from like cortisol, adrenaline, and other, uh, glucoregulatory hormones. And if people just want a little bit of a glimpse into the methodology and how scientists dissect these pathways, it's super cool. So you can do, for example, um, genetic manipulations of animals where you express very particular receptors on these neurons, so you can specifically activate them and see what happens. So that's how you can separate it from, say, cortisol and adrenaline. And then you can do really cool things if you want to determine the neural line from the hypothalamus to the liver. You can do experiments with engineered viruses. Some viruses, in this case, they, I think they use the pseudorabies virus, can move up neurons kind of like they're climbing up a ladder. It's called retrograde axonal transport. So it's kind of blew my mind. I didn't really even realize we could do this, but you can tag a virus with a fluorescent protein, inject it into the liver, and then if there's a direct neural line to the brain, you can look in the brain for a glowy, glowy green signal. So this is the kind of methodology they use in this paper to unpack— yeah, there is a direct neural line from the brain to the liver, and then during stress, it tells the liver, hey, change your gene expression to make more glucose de novo. So not even just breaking down the stored glycogen, but synthesizing it from proteins, circulating glycerol, lactate. So literally, this isn't an exaggeration, stress makes the liver make new sugar and then spill it into the bloodstream. Totally new mechanism, blew my mind too. Any further reactions on mechanism before I want to ask you about some of the real-world implications from your, uh, clinical and personal experience?

I think it's important for the audience to appreciate how impressive this work is because they went so far as to essentially put put nerve-based stimulation through many of the ways you describe, but also almost— it's essentially like developing the ability to turn on and off parts of your brain. Not just parts of your brain, but very specific regions of the brain where you can say, okay, I turn this on, we see this consistent, uh, effect, and we can turn it off. And then they went so far as to actually block other pathways that interact with it to very clearly show, hey, there's this elevation in glucose and we know these factors are going to interact with it. Let's take each factor out one by one and show how this mechanism directly induces this effect. This is the beautiful part of science. You read a paper like this, you— it's incredibly impressive, uh, to watch how they go about proving the impact, uh, of this part of the brain directly signaling glucose release. It's, it's amazing and really well done.

I agree. And now I want to make things a little bit more concrete and clinical for people because, as we kind of alluded to but didn't say directly, this does explain or adds, you know, a layer of depth and resolution to things we see in humans. However, the methods they used, you just can't do these experiments in humans. You can't chop off human heads to stain brains, you know, after an experiment. So necessarily these pathways are dissected using animal models, but they do give, like I said, new flavor and resolution to what we see in humans. So now I want to ask you, you know, first, as a type 1 diabetic, yourself, do you find stress increases your blood sugar? And if so, how do these findings like shed new light on that phenomenon?

It's one of the challenges of type 1 diabetes because with type 1 diabetes, you're— for your audience, you don't produce insulin anymore. So everything you eat, all the changes in glucose, whether it be from a mechanism like this through stress, isn't immediately cleared by an automatic insulin release, like an automatic, uh, like a computer algorithm that immediately sees a glucose elevation, immediately brings it down with insulin. So a lot of people may never know stress is causing their blood glucose to go up and then also their insulin to go up. But in type 1 diabetes, the only way to fix a blood sugar elevation, in this case stress, is you have to administer insulin. So it gives you this unique ability to give insulin when glucose goes up or not, to just see how the direct impact of something like stress can actually elevate glucose, removing the direct automated impact of insulin. And I've seen this a number of times. So, uh, one key example is I remember I was going to give this really about 10 years ago, uh, maybe it's not even 10 years ago, it's coming up on it. I was preparing for a TED Talk. There's a— this is the last time this has ever happened to me in a public speech format, but this is very common for people doing public speech or some type of social environment where they fear something, whether it be social reaction or something along these lines, this mechanism. And I remember standing in front of, it wasn't even on the day I did the TED Talk, it was actually just standing in front of peers in a room acting out what I was going to say. And it just felt so awkward, so uncomfortable. You know, you feel almost like I'm being judged, which I wasn't, but either way, so you feel this. I remember my face got, felt the stress response, my heart rate went up, I started to feel the tingles in my face, all these clear signs that I would always feel with, with high levels of stress. And I watched my CGM precipitously rise 100 milligrams per deciliter over the next 30-45 minutes. Um, so a clear sign of this social stress. But I've also seen, um, one of the telltale signs of this is in exercise. So very max effort sprints can elucidate a similar response where you're giving max stress onto the body in the form of physical exercise. And it's very common in kids with type 1 diabetes. Parents will often reach out and say, well, what do I do? Because, uh, my son goes up 100 to 200 milligrams per deciliter the second they walk out to perform their sport. Like, it has nothing to do— they didn't do any exercise. The anticipation of it alone, or when they do a max effort sprint, they can see this. I've actually had a real-world example of this on myself where I was on a, on a very low carbohydrate diet, so my glycogen levels were already hypothetically lower. I was 20 hours into an overnight postprandial fast. I went for a 60-minute run. So in theory, my glycogen levels are pretty low here, Nick. So not probably zero, but very low. And I went for this run and I could see at the end of the run my blood glucose started to drop. And I didn't have blood sugar on me, blood sugar tablets on me, which are what you should always do. But I didn't. And I was about 20 minutes away from my house and I'm like, oh man, I'm in trouble. So I actually used this mechanism, not knowing how it was actually working. And I sprinted. I went for a max effort sprint, 3 bouts of it, 30 seconds max effort to the point where I was like, just like, like dying, gasping for air., and lo and behold, it elevated. Now, it didn't elevate 100 to 200 milligrams per deciliter, but it went up around 50 milligrams per deciliter.

I just, I wanna, sorry to interrupt, but I wanna double-click on apparently paradoxical but sensible what you did was. So effectively what you're saying was you as someone with type 1 diabetes was in a tricky scenario where your blood sugar was dropping and you were running back to your house. So one might think, oh crap, what do I do? It's a catch-22, rock and a hard place. If I continue to run, I'm gonna cut through my glycogen stores, and then I'll be in, uh, deep shit, so to speak. But I can't stop. So what? So what you did was you sped up, and what that did is it created a stress response to make your body make sugar, which you can do through, you know, using alanine and other amino acids or lactate. So just the increasing lactate from your muscle glycogen stores, it's called the Cori cycle. You probably had a lot of muscle glycogen stores, which can't defend blood glucose. But if you use them, you can boost lactate, and the lactate can be turned via this, you know, stress pathway in the Cori cycle into glucose. So it was a paradoxical kind of ingenious move to get yourself out of a really tricky and dangerous situation, if I'm representing this correctly. And then just to emphasize, you know, why this is so important, particularly in you, but, but the clinical relevance here is when your blood sugar rises, and this is shown in the study, from stress— could be psychological or physical stress— glucose goes up for the reasons we already explained, but insulin follows because a healthy human body will monitor blood glucose and try to keep it in this tight range. So this was actually shown in the study where, you know, in the mice, when they stressed them out, the glucose went up and then the insulin kind of chased it. But type 1 diabetes is a pretty brilliant natural experiment, so to speak, because you don't have naturally the pancreas responding with insulin. You need to titrate it in. You need to have the knowledge to do that. So to bring in like a personal example, we did an experiment, um, first year at Harvard Med where a bunch of us wore continuous glucose monitors, and I found what spiked me more than anything was actually being late to class. It created a stress response. I kind of ran over, so I had this little spike, but it was very acute. It would go up and come down, and it would come down presumably because my body's like, hey, all right, glucose in the bloodstream, let's give some insulin, get it back into the muscles. But if you have type 1 diabetes, or even if you have particular flavor of type 2 where you have beta cell dysfunction, you won't have that autocorrection, and a little stressor can end up with a huge spike that is then prolonged. So this is incredibly clinically relevant.

Absolutely. And to even piggyback off that even more is that, you know, we know that, uh, you know, there's a huge number of people who now are categorized as pre-diabetic, you know, actually I looked at data from the, um, Center for Disease Control over 10 years ago, and it was showing that over 50% of adults had prediabetes. That was 10 years ago, and things have certainly not gotten better on the general population level. So there's— not only is it Type 1 and Type 2, do we see this example, but it also seems like people— those are even at risk is over half the population. And so yeah, it's, it's, it's incredibly relevant for the majority of Americans. Uh, and we all know that stress can be— have negative impacts on our health, but we usually think about that from a psychological perspective, the burden it places on us. It's almost like this invisible backpack that's on our back all the times, and stress just puts more weight into that backpack for everything you do throughout the day. But it actually has a metabolic effect, as this paper shows. And this metabolic effect clearly illustrates that not only are you breaking down some of the, the stored materials your body has already for energy, but also it's, it's making new, uh, forms of that same material in the form of glucose. And so the real-world, uh, implications of this are real for the majority of people out there, and the, the health implications are also very real. I I imagine one of the simplest and most powerful take-homes is to reduce our overall stress, right? But I think—

easier said than done. I'm sure we can both attest. But I mean, it's, it's a nice visualization if you're doing some sort of meditation or breathing, like you are literally changing the direct wire from your brain to your liver saying, hey, tamp down the glucose generation. So it's not just like fluffy, it's biology.

Yeah. And to piggyback off that, the psychological part of this is huge because you can't just drug this effect down, right? So there's things called beta blockers, which block the adrenal, adrenaline, or the epinephrine response specifically at receptors in the liver, which is the mechanism that we traditionally know, but it never completely abolished the effect. Okay, and everyone's wondering why not. I mean, if it's directly blocking the mechanism we know to be working, why is it not abolishing this effect altogether? And this paper helped explain in part why that might be, because the beta blockers were only addressing one part of the mechanism that we traditionally understood, which is the AP axis. But this new mechanism explains why there's still an effect that the beta blocker doesn't completely address. And we know beta blockers are commonly used for, for, for a number of therapeutic treatments, uh, but actually a lot of people with type 1 diabetes use beta blockers to block this stress-induced effect, um, around, around stressful events or around exercise. And, uh, this paper actually helps illustrate why it doesn't completely abolish the glucose-based effect. So it's quite quite fascinating.

I think we're gonna keep piggybacking off each other. I have one more comment that I wanna go onto our next paper, but as you were saying that, I was thinking about how complicated and beautiful this gets. And so another factor to bring in is something like a metabolic state of obesity. One thing we actually know is that, at least from animal models, that obesity fat tissue releases certain hormones like leptin, which can remodel the blood-brain barrier around the hypothalamus. Which is the feed-in point from the medial amygdala and root to the liver, and then in that way change nervous system tone. So you could be amplifying the negative metabolic impacts of stress via this mechanism by having excess fat tissue and the hormones that fat tissue produces. So it is kind of like a moment when you start to like elaborate on what this mechanism could really mean. But with that, I do want to get on because we have so much to cover. Another study published this year, 2025, in Nature Medicine was looking at the glycemic responses, um, to different foods. Now that's one thing. What I actually wanted to ask you about is glycemic mitigators. So what are glycemic mitigators? Fat, protein, fiber. These are things that in general you can eat before having a certain glucose load that will lower your blood sugar spike. So if you eat fat, say, you know, before eating some amount of carbohydrates, say you have like you know, buttered eggs before you eat a sweet potato, the spike will be lower. And it's actually the same for fiber. But one of the really interesting results of this study that I wanted to pick your brain about was how this interacted with pre-existing insulin resistance. So one of the things they found was that in people with insulin resistance, they don't get the same blood sugar protective effects from things like fiber. So if you're insulin resistant and you have fiber before a carb load, the fiber quote-unquote benefit on blood sugar will be less Whereas for someone who's insulin sensitive and healthy, the fiber, probably say before a bowl of grapes, would reduce the spike. So it's kind of a one-two punch because you're already insulin resistant. The carbs are going to like spike your sugar an amount, and then the insulin is going to fall and you're going to have hyperinsulinemia. But in this case, the quote-unquote protective food variables aren't as protective. What do you think about that?

Number one, I think anytime you see something like this, the first thing you assume is why is this happening? Because just like the prior paper we're talking about, if we understand why, it might help us understand how to apply it, um, with all these interactive mechanisms happening at the same time. And historically, we would think, okay, that might be a microbiome effect, right? Because the microbiome places this almost barrier in the gut between the food and the intestinal absorption of it. And so it places this kind of interactive barrier that processes various components of the food in particular ways. And that has been kind a lot of the recent focus because it's emergence and how fascinating the microbiome is and all this emerging evidence on and, and all sorts of model systems of how powerful it could be on overall health. But what one thing you mentioned that's very important is that often has been brought up at the expense of understanding the metabolic physiology or the metabolic status of the individual themselves as well, uh, as a key determinant factor. Because we've even run— I mean, actively we're running a study for the Department of Defense right now uh, where we, we recruited a very consistent cohort— all males, all similar BMI. Um, the, the, the tightness of this cohort was they're 21 years of age average, mean 21 years of age, standard deviation 2 years. Super tight cohort. Yet if you look at their basal insulin levels, they differ almost sixfold. So even though you look at someone and think, oh wow, they're probably similar equivalent health, there could be wildly different metabolic effects in the body, and insulin is just one example. But what's really cool about this study is they took it a step further and actually used a tool called continuous glucose monitor and was able to utilize that real-world tool, uh, something you— anyone could in theory go buy right now, as long as you— they sell them at your local pharmacy, or you can get a prescription if you happen to have diabetes. And they found these unique differences across all these, these individuals, those who are at risk for or maybe have diabetes. And what I find interesting about this paper, um, is for me at least, my interpretation of it was yes, there was this unique effect of the lack of, or the, the reduction in protective effects once someone has diabetes. They were showing that on the average, someone with type 2 diabetes may not see the same level of protection from fiber or the same little protection from other nutrient ingredients that we associate with health, or you can co-consume them with other foods and they'll blunt maybe some of the negative effects of some, some food compositions or some, some dietary strategies. What's the actual implication of this, Nick? I think one of the most important implications of this is that what would someone do if they had diabetes? Because we are told to eat fiber, we're told to take some foods and maybe mix them together with other healthier foods like, let's say, protein. Or maybe swap forms of food for other forms of food to improve our overall health. And I don't think this paper says not to do that. I think what this paper essentially illustrates is that there are categories or subgroups of individuals that are, are particularly at risk for higher glucose spikes, even though you're adding some of these tools and strategies into them. But what I think this paper also showed by illumination of these fasting tools called continuous glucose monitor, is that you can monitor your own glucose response as an individual in a one experiment and see your personal response. Because this is probably a bit too controversial, although you, you do a good job of balancing that out well, Nick. No one got diabetes from a flat glucose curve. Okay. Like, no one gets diabetes with a flat glucose curve. And so if you're seeing that you have these wild, crazy spikes with certain foods and not others, you can lean into the foods that you seem to have better glucose responses to. And it doesn't mean that you're going to get diabetes because you have a normal glucose response to certain foods. But if you have a wild spike, that's— and some of these individuals were going north of, you know, 200 milligrams per deciliter, that would be concerning. Those, those levels are associated with adverse effects.

So I'm going to first emphasize the clinical importance, which you so correctly pointed out, and then another, like, deeper level of, of the science here. First, neither you, I, or the researchers are saying, hey, fiber is quote unquote bad if you have diabetes. I think what we're trying to combat, if you don't mind me, hopefully not speaking for both of us, but towards speaking for both of us, is that the platitudes that are put out for the general population, like, oh, fiber is better, you know, oatmeal is okay because yes, it has carbs, but there's also the fiber that balances it out, is not serving public health because it's very contextual. So people should know, hey, if you have insulin resistance, I'm not saying oatmeal's bad, be-all end-all. But this idea of, oh, it has fiber and that will protect you, it changes based on your metabolic circumstance. So that's clinical implication 1. Clinical implication 2 is actually just has to do with the overall individuality and heterogeneity of blood sugar responses to the same foods. And this isn't new research. This has been documented at least for 10 years where, you know, there's research showing if one person eats a banana and the person eats a cookie they can have totally different glycemic responses to it. So there's a lot of individuality, but you can dissect that with continuous glucose monitors, which is a great tool that's now available over the counter. And then the third element is the level, uh, scientific level deeper is, well, then how can we use those clinical tools to feed back on the science? And one of the things this research group is doing that is really brilliant is using AI and machine learning to find signals in the personal data that inform the underlying pathophysiology. So one example that I'm sure you picked up on, but we should just give it because it blew my mind, was the potato to grape ratio. So basically what they found was that, huh, people with insulin resistance seem to preferentially spike more to potatoes relative to— in this case, grapes were a big spiker, so you could have, uh, uh, you know, grapes as a control. So basically, and they actually gave me a protocol which I'll, I'll link below, you could wear a CGM and then measure your peak spike to potatoes on one day versus grapes on another day, ratio those, and those will give an indication of your level of insulin resistance, which is scientifically just freaking cool.

What's cool about this is it actually is trying to tie kind of these advanced laboratory-based assessments into real-world applications that almost gives the power back to the individual, right? Because a lot of times you're, you're beholden to going to a doctor, getting a single test at one time point, maybe hopefully every quarter, maybe only every 6 months. And then for some patients, you're lucky if you get it once a year. But this gives you insights every 5 minutes over 24 hours, over weeks and weeks and months if you, if you continue. So, and that type of resolution gives you insights that are incredibly empowering, um, and guides decision-making that's so individualized to your personal response, you know. And it also illustrates that just because, you know, someone— something is quote-unquote healthy doesn't mean it is for you, you know. I think that that's important. Important to appreciate. And again, we're not wildly saying that potatoes are inherently unhealthy or anything like that, but what it is illustrating that on the average, you— if you have diabetes, you may fall into one of these buckets more often on average. But if you were to take this away and try and be actionable with it, you don't want to assume that. You want to actually find out, hey, look, if I'm— if I respond to certain foods, by assessing it for yourself, that's going to be your best indicator of how you respond, because we're all slightly different humans. I mean, look no different than me, me and Nick. I mean, he's got this amazing beard. Mine's not quite as good. He's got different color hair. You can't tell, but he's actually much taller than me as well. I think he's like 6'5", 6'7". You know, all these individual differences that come out there. Uh, and so I think at the end of the day, it— it—

how are you really? I don't know if people are gonna be missing this joke. I'm like 5'7", 5'8". You're what, 6'3"?

Uh, maybe 6'5", you know, give or take, you know. It's so, uh, you know, but I, I— it really—

would you get in person and take a side-by-side photo for the thumbnail?

I don't know if we do have one. We'll have to— we'll have to definitely get one.

Um, no, we have— uh, have we met in person?

Yeah, we had Metabolic Health Summit some years ago. Yeah, I mean, so it, it definitely shows that, you know, there's clear sub, uh, group differences, but you can test for yourself. You can be your own experiment, see how you respond to foods, and really eliminate the uncertainty of, of what's best for you using tools like a CGM. You don't— and it also helps illustrate in my mind that your metabolic status does predict better or worse responses on the average. And, and finding that out might be very helpful. You can get things like, uh, you know, ask for fasting insulin, ask for fasting glucose, maybe get a HOMA-IR score, look at your HbA1c. These aren't sufficient alone, but obviously this paper illustrates, and other papers that we're, we're going to talk about do. But essentially what they show is that, yeah, foods like fiber and protein, they tend to do better for most people, but it doesn't mean that it's always universally applicable. And that goes to show where the science has been going for quite some time, which is the personalized medicine, which is getting to the individual level and helping people empower themselves, which this paper has shown with tools that you can actually apply in the real world, which I think is really, really powerful.

I have one more question on this line of research, actually a separate paper, and then I'm going to save the rest of the goodies for part 2. But so this was a 2025 Nature Medicine paper. Prior to that, there was a precursor paper on this population published in Nature Biomedical Engineering in 2024.. And what they did in this study is they used continuous glucose monitors, um, and machine learning to decode 4 flavors, 4 causes of, uh, uh, in this case type 2 diabetes, although it overlaps with type 1. So I'll list them. There was muscle insulin resistance, um, liver glucose dysfunction, uh, incretin dysfunction, which is like GLP-1 hormones, uh, and then beta cell dysfunction, which actually overlaps with type 1 diabetes. So there's a little bit of a Venn diagram going on here, but What I found very fascinating was this bucket of incretin dysfunction. Now, again, incretin, GLP-1 hormones. These are hormones that are important for glucose homeostasis, and they're also the basis for this generation of weight loss drugs. Ozempic, Wegovy, they mimic GLP-1. They do it at a much super physiological dose. But the point is they act on this axis. Now, putting two and two together with other research. So other research has shown that certain medications that are commonly prescribed— in this case statins, the most common statin being atorvastatin or Lipitor— can in human trials plummet GLP-1 levels. So this is all new research, but you can see, maybe people are putting two and two together already, it begs the question: if there are particular causes of diabetes that we can now dissect at the individual level, and human data on how certain pharmacotherapy including statins can target this pathway, then one would reasonably hypothesize, hey, there's going to be a whole type, a whole species of diabetes that will be enriched in, say, people taking statins or other medications that lower GLP-1 levels. So I had a conversation with the senior author of this research, who himself, um, uh, had diabetes or prediabetes. He had the beta cell dysfunction type. But I asked him, do you agree this population of people with diabetes who have a GLP-1 defect that's driving the diabetes might be enriched in, say, statin users, and how does that implicate care? That was me on my soapbox for a little bit, but as someone with diabetes and a fellow researcher, I wanted to get your perspective on this possibility and what it means for clinical care.

This paper elucidates something incredibly novel. We have known for some time there's different forms of diabetes. What this paper illustrated was, you know, for, for different stuff, because we traditionally bucked bucket them in 3, and this illustrated 4, as you, you really well articulated. And of the first 2, the muscle and hepatic insulin resistance accounted for around, I think, 34%. And then the, the incretin-based and the beta cell dysfunction-based ones around 40%. So actually a higher percentage of the overall group that then place these individuals at risk for the development of type 2 diabetes. And this leads to this idea because we've known for a while, you know, and, you know, controversy aside, that statins can place a certain number of individuals who consume them at risk for diabetes. It's a known risk factor of statins. You know, the idea of statins, of course, most of your users will certainly probably know this, is that lower— you know, the idea is that would lower your cardiovascular risk. It's believed to do that through LDL. But there's also other mechanisms through inflammation pathways, etc., that are hypothetical. There's a lot of nuance there, but that's the reason people would often prescribe them in a medical setting for that logic. But we also know that there's evidence that that in some scenarios, particularly this paper you highlighted, that it can increase, you know, glucose intolerance, uh, and diabetes risk, uh, in some humans. And as you pointed out, this illustrated a very powerful— I mean, powerful, it was actually really clear, uh, effect on, on GLP-1s, um, and an increase in a subsequent HbA1c, which is a 2 to 3 month average of blood glucose levels. And why that's important is because when we think about overall risk, uh, for long-term, you know, disease like cardiovascular disease, We think about them, at least I do, in hierarchy. You know, what, what matters most to overall risk? And we know that diabetes, or the elevation— and usually defined by the elevation in blood glucose in almost all these analysis, and in this case HbA1c— is by and large a much more powerful risk factor, multifold larger than anything else typically is, um, as far as hierarchy and risk for the number one cause of death was cardiovascular disease. This paper illustrates that there, that there are certain drugs that come with— honestly, everything does— with a risk-to-benefit ratio. There's, there's no decision in life that doesn't come with a positive-negative ratio. You always hope to maximize the reward and minimize the risk in every scenario, but it clearly illustrates that there's, there's no free pass entirely in life, and there usually never is when you take something that directly blocks particular pathways, like a statin does. And so you'd expect some type of off-target impact. Now, what I think this illustrates, though, is that this research group, once again in this paper, also illustrated the ability to subgroup phenotypes based on CGM profiles. This again allows for it just by monitoring with these, these wearable tools at home. Obviously, if you have their computer algorithms, which I'm sure will eventually get to light of day for people to utilize, will allow them to understand their potential type of risk, and then they can make more informed decisions. And I thought another aspect of this that was particularly interesting, um, is the relationship with the microbiome. And, you know, we often think of the microbiome as kind of this, uh, very complex, novel, very interesting area. But it also— there's a lot of things we, we know on the average tend to improve what, quote unquote, good microbiome health, which is kind of a quotes around it because it's kind of hype. It's all subjective to the outcome you want. But ultimately what we're targeting here is precision medicine. And the best way to do that more than anything isn't to just rely on average outcomes. This is just telling us, oh man, if I take a statin, I need to be careful about my glycemic control. Or if I make that decision with, with doctor-patient relationship environment. But ultimately self-testing and responsiveness is a critical part of this.. And if you are gonna take something that puts you at risk, understanding that it appears to be through a mechanism that might be microbiome related, maybe you want to improve your overall microbiome health and kind of consider the implications of, of, of things like, I don't know, fiber, like exercise, like good sleep, all things interact.

Sorry to interrupt, but they get quite specific in the, the statin paper where, you know, like you said, the microbiome is a little bit of a black box.

Yes.

But you can look at the metabolome of the microbiome. So rather than saying, you know, what the gut bug is, which isn't necessarily that informative because every ecosystem is different. What are the metabolites generated? Because these are the arsenal, these are the action, like, you know, points, the places to intervene because you can measure them in the bloodstream. What is the microbiome making? In this paper, they actually showed it was a secondary bile acid called UDCA, which they could supplement back and basically reverse the negative effects of statins. As you were saying, like, this is really important clinically. And I think to demonstrate that, just like, imagine where medicine could be in 5 years. So I know the Snyder Lab, the ones that did this research, are trying to make this publicly available. So now imagine you have a CGM on and you have an app on your phone. You are at high cardiovascular risk. Maybe you've had a heart attack, so your doctor puts you on a statin for a medication. The app could detect not just an increase in glucose, but a particular shape of that glycemic curve that could indicate, hey, not only are you developing insulin resistance or type 2 diabetes, but you're generating it of this particular flavor of diabetes. And that could be a contraindication, or at least inform future action— changing medications, taking, you know, um, compensatory supplements like a TU-DCa or UD-DCa. And I'm not making medical claims here, I'm just saying like, that is why this research is so important to talk about, because we are shifting from these platitudes of, hey, lower is better, more fiber is better, yada yada, to what's going on in you. And these tools, I think, are like really around the corner. We already have the tools. It's just like the CGMs, we're just like folding them in with large population datasets and using machine learning and AI to make this interpretable and actionable at the clinical and individual level. I think that is really exciting. Not to put value judgments, broad value judgments on any food or medication, but like you said, risk-benefit analysis. And the art here is that at an individual level, the risk-benefit analysis doesn't just line up with large population data sets on either side or large population findings. So we are refining that, and it will give people the power to make much more informed individual choices and iterate on them in real time, which I think is pretty cool.

So I couldn't agree more, Nick. And I also think it's going so far as to potentially inform, uh, medicinal therapies in a clinical setting, because, uh, knowing the form of, or flavor of diabetes, as you put it, uh, directly impacts historically how we know people respond to certain treatments like metformin, like lifestyle such as exercise, uh, and beyond. And but there are some interventions that kind of sweep the whole gamut, like some of these like diets like very low carbohydrate ketogenic diets. But there, if you're going to do a medicinal therapy like metformin or physical activity, the type of diabetes has historically illustrated a likelihood to be responsive to those treatments. And so understanding why, which I think you do a great job, Nick, of illustrating for those out there trying to learn more, empowers you. There's no way around it. It empowers you as an individual when you understand why things are happening so that when you, you, hey, I want to try this, this, your doctor says, hey, do you want to try statins? Or hey, do you want to try metformin? Are you someone who your flavor of diabetes, uh, would, would put you as a potential beneficiary of that? Because there's always a risk with everything you do, um, no matter what. Even if you exercise, there's a risk, right? And so, uh, that— I, I could not— I know as someone living with diabetes for almost 2 decades now, the— you want to be empowered. You don't want to feel like your disease is just in control of you, or that even that someone else is making all the decisions for your health. You want to feel like you're empowered at an individual level. I think almost every healthcare provider out there would also want their patients to take the bull by the horns and take action using helpful and evidence-based resources. So I think you do a great job, Nick, of doing that with your platforms.

Appreciate that. Well, speaking about platforms, I'm going to push the rest of our discussion into part 2, where I'm going to tease— we're going to talk about what I thought was the coolest fiber paper. We were shitting on fiber before. Now we're going to give it some praise. The coolest fiber paper I've ever read. Some potential risks about long-term ketogenic diets, maybe, maybe not, and some more practicals as well. So before we sign off on platforms, where can people find more about you and follow you?

I, I'm on X and Instagram. X, I'm on @akoutnik and then Instagram, Andrew Koutnik, PhD. I also have a website, andrewkoutnik.com. And if you want to follow for information, there's Substack and YouTube as well. Where you can find a lot of helpful resources.

Thanks so much. Everybody come back for part 2, which will actually be my 4th interview on this channel with Andrew.