Psychiatry & Psychotherapy Podcast - Ketogenic Diet Review and Update with Dr. Matt Bernstein

Episode Date: April 11, 2025

Explore the transformative potential of the ketogenic diet for mental health with expert Dr. Matt Bernstein, a pioneer in metabolic psychiatry. Learn how targeting metabolic dysfunction can profoundly improve psychiatric outcomes in bipolar disorder, schizophrenia, depression, and epilepsy, with fewer side effects than traditional treatments. Discover actionable insights on diet, metabolism, and brain health backed by the latest scientific research. Tune in to revolutionize your understanding of nutrition's role in mental wellness! By listening to this episode, you can earn 1.75 Psychiatry CME Credits. Link to blog. Link to YouTube video.

All right, welcome back to the podcast. I am joined today with Dr. Matt Bernstein. He is a psychiatrist in, let's see, multiple roles at this point, but you were mostly private practice for most of your career. Actually, I started off as an inpatient doctor at McLean Hospital on the schizophrenia bipolar inpatient unit for about eight years. And I've always throughout my career taking care of people with serious illness. It's a real passion of mine. Okay. So you did inpatient for eight years, then you went outpatient. I work at this program that is a team-based approach with people with complex illness called Ellenhorn,

and I'm the chief medical officer there, and I also have a private practice. And I also now have opened a program called Accord, which is focused on metabolic treatments for mental health. Okay. And so conflicts of interest, we're talking about this. Is it a conflict of interest that you're part of metabolic mind as an advisory board?

I think I want to mention that. I think you've been given that position because you are bringing patients into ketosis in your practice and you have some expertise on this. But you're not like there's no drug companies, there's no supplements that you sell. You don't have any other than your private practice and your roles there.

That's right. And metabolic mind is a nonprofit focused on education around metabolic and mental health. Okay. Yeah. And so today we're going to be talking about the ketogenic diet It's going to be kind of an update. I had Chris Palmer on here a couple years ago when this book came out, and there have been

quite a bunch of studies since then, and I wanted to go through each one of those with my audience. I also wanted to kind of pull from you, just the practical things that you've learned. You've been putting people into ketosis for a little over three years, you said, as kind of part of the practice of what you do with severe mental illness, with bipolar, with schizophrenia, and with other types of mentalness as well. Correct, yeah. And just to give my audience kind of like an idea from what I learned from you

is when you started three years ago, you were doing this really by yourself, you would put people in intermittent fasting, you would see if they responded positively that, and then maybe move them to ketosis later on. And then in the last year, you've started this kind of impatient setting, not traditionally like what we would consider like a partial program, but people come, they learn

how to, they get into ketosis. Is that correct? Yes. Well, just to clarify a little bit, when I was doing in private practice, I would often combine intermittent fasting with MCT oil, which is medium-chain triglyceride oil. It's a derivative of coconut oil, and it's extremely ketogenic as a fat. So people may have heard of this drink called a bulletproof coffee where they put butter and MCT oil in it or heavy cream and MCT oil actually tastes a little better. So if someone's fasted overnight for 12 hours and they have one of those drinks in the morning with the MCT oil and the fat, but nothing else, and then they can, you know, wait another four or five hours before they have their first meal, they often can get some ketones

just from doing that one intervention, not high levels, but a little bit of ketone, a little bit of ketosis, maybe a BHB, beta-hydroxybutyrate level of, let's say, you know, 0.3 or, point five. But that's enough for a lot of people to feel a difference in many ways. They often feel some cognitive clarity, some extra energy, they don't feel hungry. And so that's a little taste of sort of what ketosis might start feeling like for them. And many people try that and love it and feel energized. And they say, well, you know, they come back to me and say, what was that? That was amazing. That wasn't just like coffee. That was super amazing. And then they're really open to learning more about this whole science and potentially,

going further in their metabolic therapy. Really cool. Okay. You know, I first got introduced to the ketogenic diet, actually through, I was at a conference giving a talk on psychogenic seizures. I was running a program for people with psychosomatic issues. And so I was at this neurology conference. And it was put on by some ketogenic foundation. And I was like, what is this?

Like, is this really being used for people with seizures? Like, how does that work? And so tell me, like, a little bit of that history just to kind of prime us for this idea. Sure. Yeah, I mean, we can go back really far in history to, you know, pre-agriculture times, and we can go back to Hippocrates where they were using fasting to treat seizures in 5th century BC. But, of course, fasting's not a very sustainable treatment to maintain ketosis. I mean, eventually you have to eat, and you'll break your ketosis unless you're eating.

ketogenic. So that didn't really get figured out until 1921. A doctor at the Mayo Clinic published a small case series of some kids that he put into a what he called a starvation mimicking diet or a ketogenic diet. They knew about ketones at that point. And he figured out how to create a diet where the body will make high levels of ketones. And he reversed epilepsy in three kids, I think it was. And then published that and published the things. theory behind it, and then it really took off. People started publishing rapidly in the 1920s about using ketogenic diets to treat seizures. And that literature just kept growing, where eventually there's now 12 randomized controlled trials showing that ketogenic diets work in both kids

and adults with epilepsy. And it not only works, it works with people who are drug-resistant. So that's really important point because some of these kids and adults are on, you know, three different seizure medications at a time. They've tried, you know, eight or ten of them. They still can't get control. And then the ketogenic diet can still work in those really treatment refractory situations in a good percentage of the time. Yeah, I'm curious, what level of the Bedihydroxy butyrate are they getting to get into a place where it's actually helpful for their seizures?

Well, interestingly, you know, the literature is pretty old and people weren't measuring beta-hydroxy B-rate levels daily through most of these trials. You know, often they're measuring it, you know, once a week or even less. And so we don't even know in the epilepsy literature that well how, you know, high the level of BHB you need to get most of the time. They more measure it based on the ratio of the diets. That was usually the sort of how they measure the intervention. So in a ketogenic diet, we think, you know, think of a ratio, the classic one in 1921, was a four to one ratio of calories from fat compared to calories from protein and carbohydrates combined. And yeah, almost no one does a four to one anymore. Ever. I mean, that's like a really, really extreme diet. You're getting, you know, almost 90% of your calories from fat. And you think about that.

You're going to be protein deficient on that diet, almost certainly. You're even below probably the RDA, US RDA, protein levels, which are too low for most people anyway. And so what they learned also, you know, as they were doing these studies in epilepsy, is that you could reduce the ratio and still get the same benefits in almost everyone. And so they ended up with more like 2 to 1 ratios or even 1.5 to 1 ratios. There was even a study that was really interesting at the Mass General epilepsy clinic. they had people on a waiting list to get into the ketogenic diet program. And they put them on, you know, just on their own, the families,

they just told them eat this low glycemic index diet, which is really more liberal, much more liberal than even like a one-to-one diet, you know. But all the carbs are from whole foods, and they're still low carbohydrate. You're not eating anything that's going to raise your blood sugar rapidly. And those kids did almost as well as the kids on ketogenic therapy. And they didn't even, they weren't able to correlate

it with the beta-hydroxybutyrate levels. So, and we'll get to this in one of the studies, you know, we are really still early in trying to figure out the science of, does it, do people really need a BHB level in a high range? And when I say a high range, I mean above two or 2.5, that's, you know, that's really deep ketosis for someone. But could there, you know, could people get some of the same benefits, at least a subset of people at a much lower BHB level. We don't really know all the answers to that yet. Okay. Yeah, that's good.

This is good practical information because, you know, as we're thinking about our patients with mental health, we want to monitor, like, compliance. I know compliance is to some degree related to the success of if this is going to be successful. And interestingly, it's a treatment where we actually can measure the level of the treatment much better than with most medication therapies. So I can see if someone's, you know, their BHB level, I like to check them daily for people,

especially in the first few months. And we really can see, is the treatment causing this biomarker to change that I want to change, as opposed to, you know, if I prescribe any antidepressant and most antipsychotics and mood stabilizers, I can't know even what the level of the treatment is in their blood, whereas in this I can, which is cool. Okay, so let's talk a little bit.

bit about the bi-directional relationship between metabolic issues and brain disorders. And I almost recognize, because I'm so much on Twitter and there's so much anti-psychiatry there, there's this kind of sense that, you know, schizophrenia isn't even a brain disease. Depression isn't even a brain disease. So we're going even further than that and saying, actually, no, and metabolic stuff influences this to a high degree. Tell me about the bi-directional relationship here. Yeah. Yeah, well, of course, these are brain diseases. I mean, that, that I think, you know, is without a doubt. There's so much data. But what we are saying, though, is that, you know, the mind and the body are intricately connected and that the brain is influenced

by the metabolism, what's going on all throughout the body and the metabolism in the brain itself. So, you know, there's multiple studies. I think the one that I really find interesting is this one And it's Perry from JAMA psychiatry. I think it's 2021. And they measured kids longitudinally. So this not only was showing a bi-directional relationship, but showing before and after a little bit. So these kids, they were measuring their metabolic health and mental health,

and they followed them for a lot of years. And the kids with the worst metabolic health in childhood had much higher risk of severe mental illness when they got into their 20s. And so they looked at something, I think the data showed that the highest quartile of insulin levels at age nine, compared to the lowest levels of insulin at age nine, there was a threefold increased risk of schizophrenia in their 20s. So, I mean, that's a really big change just from insulin levels. And we know from other studies that insulin resistance is involved in all these mental illnesses, including schizophrenia. Yeah, that doesn't surprise me.

It kind of makes me think about how there's a lot of data on increased rates of schizophrenia and lower SES groups. And I think about how lower SES groups often, just by the nature of what's available, are kind of consuming more highly processed foods, increasing insulin, increasing metabolic disease. And maybe that's a big part of it. Probably there's also just increased risk of different types of traumas, neglect, that kind of stuff as well. but okay so this is this specific study it's called longitudinal trends in childhood insulin levels and body mass index and association with risks of psychosis and depression young adults right yeah the other the depression part i think it was at puberty that the kids who gained the most weight had a fivefold increase risk of major depression in their 20s yeah these odds ratios are very high

um so this is an adjusted odds ratio of with psychosis and psychotic disorder 3.2. So, wow, very high. A puberty onset major increase in BMI was associated with depression odds ratio 4.46, but not psychosis. So people who had gained weight right around puberty, that was more depression.

So that's an interesting one because there's some before and after effect. So, you know, sort of suggestion of causality, and certainly not proof. But, you know, some of these other studies look at rates of obesity leading to higher rates of depression. And if you're depressed, you have a higher risk of becoming obese afterwards as well. So that was a separate study. And similar things with diabetes. And, you know, there's some work by Cynthia Calcan.

We'll talk about one of our other papers where she looks at much worse course in bipolar disorder if someone has. insulin resistance, pre-diabetes or diabetes, they have a much more chronic and tractable course of their bipolar illness as well. So there's a lot of things in the literature about this bidirectional relationship. Part of me wonders with that if the medications are part of that difficulty, right?

Because let's say someone has more severe mania, they may need higher dose of medication, those higher doses of medication may naturally come with increased obesity, like if they're on lithium, zyprexa, you know, and they're still not getting better, and then you increase the zyprexa, and you know, now they're, they've gained maybe 20 or 30 pounds in the next six months. Yeah, yeah, you're bringing up a super important point. You know, these medications, you know, I worked on an inpatient unit. I appreciate greatly that these medications are wonderfully important and effective in acute illness. And they save lives. I mean,

I've seen medications save lives dozens and dozens of times when people are acutely ill. I've also seen, you know, ECT save lives. And, you know, I'm a big believer in medical treatment for severe illness. But what I think I'm coming to is that they're really good in the acute phase. And they may actually impair recovery. They may impair recovery. They may impair long-term outcomes if we continue to use them at high doses in maintenance and that that's that's a real concern and of course I have many people on maintenance medications and you know as I'm practicing you know psychopharmacologists so but it's really concerning to me that I think the data is now starting to really show us how dangerous they are

metabolically and how important metabolism is for long-term outcomes right and so you put this all together and I think as a field we have a real problem to contend with. Yeah, you know, I'm an outpatient psychiatrist too, and I treat people long term, and it's like, it's kind of like you get them stable. They need to be on the antipsychotics. If someone's psychotic and they're stomping the shower because they see spiders and they're screaming, you know, this is a traumatic state to leave them in if we don't treat them. And I think people or anti-psychiatry who just think, oh, why don't you just use natural means? Why don't you just

tell them to exercise? It's like, no, you don't understand severe mental illness if you think you can get someone like that to exercise, period, right? You need to get them out of the psychosis. But then once they're stable, it's like, okay, how do I get them to exercise? How do I change their diet? And then how do I carefully monitor them? There's a lot of nuance there and like decreasing their meds. Some people are super risk averse, and so they never decrease the meds ever. I'm a little bit more willing to decrease the meds and then try to catch early on if they're decompensating. But not everyone's like that. Yeah, I mean, I think there's a good opportunity to talk about this paper by Lex Wondering from 2013, if we can go into that now because it's exactly on this topic. And,

You know, this is not a new paper. This came out, you know, 2013. It's a group from the Netherlands. But I think it's incredibly important for everyone to know about. This is super influential in my career, treating people with antipsychotics. So they were doing a dose reduction study in people. They took all the people who had first break psychosis in their catchment area, and they offered to be in this study about dose reduction and discontinuation of antipsychotics. So they got them, stable, first episode psychosis, six months of stability to enter the study. And then they put them into randomized into two groups. So one was dose reduction, dose discontinuation in collaboration between the patient and the doctor. So that's really important. It wasn't just a schedule that

was preordained. You know, the doctor and the patient both could input into what the right timing was of dose reductions and even to stop dose reductions if they wanted to. But their goal was to try to get the dose as low as possible and even towards zero if they could. And the other arm of the study was maintenance treatment sort of by APA guidelines, you know, a low dose of an antipsychotic, but not changing, not trying to reduce the dose towards discontinuation. And they kept them in that, you know, randomized setting or, you know, different arms for 18 months, I believe. And in that first 18 months, as we would imagine, there was more relapse in the dose reduction group, actually more than double the relapse rates. So, you know, like 43% relapse in that first 18 months versus, I think,

21% relapse. So, you know, if you stop there, you say, you know, why should anyone be reducing medication? You know, this is a bad idea. But they didn't. They did a follow-up five years later. So it was seven years after the first episode, five years after they were first in that, in that randomized treatment groups. So essentially after the 18 months, let's say, you know, they, they were allowed to do whatever they wanted. So it was just sort of, you know, go and live out in the wild, basically. You're not in a study anymore. But then they contacted them again, five years later, and they did this very thorough evaluation on like 80% of the people that were the original study. So they ended up studying over 100 people. It was like 103 people at the seven-year

mark from their first episode of psychosis. And they had tremendously improved outcomes. So they were measuring this very fancy functional recovery scale, and they were measuring the pans. Those are sort of two things that they were evaluating. And both were much better in the dose reduction group at seven years. So that was really fascinating. And in the end, they did a, you know, a Kaplan-Myer plot of relapse rates, and the lines ended up crossing. So you had more relapse in the dose reduction group in those first years, but by about year three or four, they were having fewer relapse rates than the people who are in the maintenance treatment group in the first 18 months. So really, really interesting. And just to give, you know,

the listeners a little bit of the data, in terms of symptoms on the PAN scale at the seven-year mark, they were almost the same. In terms of symptoms. symptomatic recovery. You know, about 69% in the dose reduction group, 67% non-significant change in terms of symptomatic remission. But the functional remission was way different. So the dose reduction group at seven years had a 46% functional remission compared to a 19.6% functional remission in the maintenance group. And if you add those two things together, symptomatic and functional remission, they had this term called recovery, had to have both symptomatic remission and functional remission to meet the criteria for what they called true recovery.

And true recovery rates in the dose reduction group were 40.4% versus 17.6%. It's a huge difference. The P value is, you know, 0.004. And so, you know, that's an amazing study. It shows that, you know, if we try to reduce doses in those first periods of time, maybe not in the first six months, but if you wait for some stability, and then in the next 18 months, you really try to reduce the dose, you're going to get better long-term remission rates.

Yeah, that's really interesting. There's so many questions I have from this. One is, are you reducing people with borderline per sali disorder? Because sometimes they can present early on with something that looks like psychosis, suicidality. You know, because we know from other data that those patients don't need to be on antipsychotics, right? Yes, although, you know, first episode psychosis, I don't think there's too many borderlines lumped together in a first episode psychosis study. I mean, you know, working on an inpatient unit, someone comes in, if they're really psychotic.

I mean, they're hearing voices and they're having these sort of first-ranked Schneiderian psychotic symptoms, you know, thought broadcasting, thought withdrawal. that's not just borderline personality disorder. And so, you know, I do assume that they did a pretty good evaluation. You know, this is a pretty serious academic group that did this study. So, yeah, maybe a few, you know, slip in, but I don't think that's a big portion of it. I think that the key takeaway from here is that we may have good reason to be a little bit willing to try the dose reduction. Yeah. and to monitor and then to also, you know, think about other ways of treating people long-term.

Yeah, I agree. I think it's important that the psychiatrists recognize that the patients are often coming in and saying, you know, can I be on a lower dose of this medication? You know, I don't like the way it makes me feel. I'm gaining all this weight. I'm sluggish. I'm tired. I can't think. And the psychiatrist is thinking to themselves often, well, I'm taking a huge risk to lower this dose. We have a risk of some relapse. And that's seen in the study, too, of course, there is that risk. But there's also reward potential in the long term for this person. And that's just, that's an important just other side of the ledger for people to be having their minds that it's not, you're not just doing this because someone might feel better in the short term, have fewer side effects. They really have a much better chance of recovering, you know, five, seven years later

if we try to get their dose down. And there's real science there. And there's been some follow-up studies that have looked at this same question. and come up with some similar things, but no one has done the full seven-year study, as far as I can tell. It's really the length of time that mattered because the relapse rates were so much higher

in the first few years. If you just stop there, you just think, you know, this is a terrible idea. And so that was the really important part of this study is how long the follow-up was. This is good. I think there's been a sense in the psychiatry community that,

you know, if someone has schizophrenia, they probably need to be on meds the rest of their life. Right. This is not the other extreme, where it's like just take them off. This is you have a provider who knows the patient who's taking educated moments of reduction. And so I think that that is part of the study as well

that we need to pay attention to. It's not just you're listening to this, you're a patient and you're like, okay, I'm just going to get off this medication. It's like, well, you're going to find a psychiatrist or a provider who's maybe willing to go down monitor, go down, monitor, you know, in a very sort of mindful way. Yeah, and there's a whole, you know, art and science to medication reduction as well, which I don't think we have time to

go into all of that right now. But, you know, it's important to look at certain things. A couple things I would say is, you know, make sure the person's not using substances if you're thinking about those reductions. Make sure they've been stable for a period of time. Make sure there's some family support and other eyes on someone, not just your own, to catch things. early and it's really important to catch things early and potentially go back up on the dose, especially if it's getting serious. Although we may tolerate a little bit of symptoms, that's part of the art of this is, you know, when we first reduce, we might get a little more symptoms, but is it a amount of symptoms that is tolerable and containable without it turning into a full

episode? That's sort of the art of it and how well you know your patient and things like that. Yeah, that's good. Okay. So let's talk about patient selection for metabolic intervention Yeah. What are some of the things that jump out at you as like, this is the person that you're thinking about using these things? I wish I knew a lot of science behind this. Right now, it's really, we're still in the early stages of figuring out who this is going to be best for. So, I mean, the biggest criteria in my mind as a clinician is,

can I get the motivated to do it? Because if they don't want to do it, we're not going to get anywhere. And so if someone's willing to listen to me about some of the data and some of the, the anecdotes and some of the potentials of benefits for them, if they're willing to understand how to do this, are they organized enough to be able to do this at home, or do they have enough support around them where family and a team that can help them do this,

those are end up becoming, those practical things are probably the most important criteria right now. Do they still have symptoms that are bothering them? Do they have a reason? Do they want to be on a lower dose of their medication? are they having symptoms that are still getting in the way? Are they having side effects that are getting in the way of their full recovery? You know, what's motivating them to do this, to do something different?

Yeah, I think since, you know, I've been talking about exercise diet for a decade with my clients, not everyone's ready for it, right? Right. But I slowly talk about it when they have pain points, when they come in, you know, like, hey, can we try, instead of changing your med, can we try to address the pain point with exercise or with diet? It's like they're not always amendable to it. Absolutely. Sometimes they are. And I think, you know, it's really another thing that I look at and try

to use as a motivational point is, you know, do they have another metabolic condition going on that we can also treat with this intervention? And that may be the thing that motivates them the most, actually. And it's okay if my patient's main motivation is to lose the 40 or 50 pounds that they gained on their medications, even if my main motivation is to improve their symptoms and reduce their side effects and improve their brain health, you know, we can sort of meet in the middle and, you know, come up with this and agree on the same plan, even though our motivations might be, you know, somewhat different. Sometimes they, they just say, well, if it's going to help me lose weight, great, I'll do it. And they don't really believe me about the mental health benefits, but then a couple

months later than they believe me. Yeah. Yeah. Yeah, I find myself pulling up some of the studies and just showing them the data, you know, or I'll send them the studies or send them little summaries of the studies. Like I had an OCD patient the other day, and I was like, look, in this one case study, that there's this person that had a benefit. We could try this and see, you know, if it has a benefit. I also agree side effects. You know, one of the things that people have a lot when they get into outpatient psychiatry is they have side effects, sexual side effects, you know, reduce libido on SSRIs on different medications. And so like they may want, they may not care about that when they're in the midst of a suicidal, you know, place. But when they get more stable,

they start caring about these kind of things. They start caring about their weight again. Absolutely. Yeah. Yeah. They start, you know, people, and they realize that their quality of life as not as good as they want it to be, even if their symptoms are better. They don't have the functional recovery that they're hoping for. And that's always what I have my eye on. It's not just reduction of symptoms. So many studies are all just about looking at, you know, these scores on these scales, which have to do with symptoms, and that's important. But, you know, it's really what matters is people live their life in the way that they want to. And often the side effects are really getting in the way of that. The other thing, I don't know if you felt this, but with

patients who are pre-diabetic, they have metabolic disease. These patients sometimes can be a little bit more motivated. And I tell them, like, look, if you get to a point where you need to be on insulin, you will gain 15 pounds in the first couple months. I remember reading those studies where they started people on insulin and they gained weight. And I was like, wait, why are they gaining weight right away? And it's because the insulin is shoving the extra glucose into fat. That's right.

That's what happens when you inject insulin. Well, and not just insulin. Most diabetes treatments are actually making the disease worse over time. They're reducing glucose, which is great, and reducing some of the secondary effects. But they're ultimately causing more insulin and gaining weight and more insulin resistance, which is essentially furthering the fundamental process of diabetes, which is insulin resistance. And I can't not mention the fact that ketogenic diets have really. reverse type 2 diabetes as well. And that's been shown in multiple long-term studies now.

And that's a whole separate topic of, you know, how effective ketogenic diets are in these really severe conditions like type 2 diabetes and epilepsy. And yet they are not considered first-line treatments at all in the medical community. I don't want to get on too much of a soapbox. But, you know, that's something to really ponder, I think, for people. We have these treatments, dietary treatments that reverse these, you know, really significant diseases. And yet we don't offer them even to people. Most of the public doesn't even know the fact that ketogenic diets reverse type two diabetes.

Yeah. Most doctors don't even know that. Yeah. Yeah. Yeah, I was looking into this before it became a thing for psychiatry. I remember I have a couple textbooks that I got on ketogenic diet when I was looking at it. It's like really interesting that it. I think it's hard to get people on it.

And so I think we'll eventually get to a part of this talk where it's like, let's talk about the practical side. It's really going against the grain of, or not to use a pun, but against the grain of how we eat and how we've been taught to eat by our government and how the medical community teaches people how to eat. This is really going against the grain.

So, yeah, it does feel like a radical thing to do to go on a diet like this. Let's take some time going through some of the key studies that have come out in the past couple years on ketogenic diet and mental health. And by the way, if you're listening to this and you're getting lost in the studies and you want to look at it deeper, we are going to have a good handout. We're 24 pages in right now. So we'll have that on the website, Psychiatryopodcast.com.

Let's start with Culkin at all 2022. Yeah. Yeah. So really interesting study, they call this the Trio BD study, treating insulin resistance with metformin as a strategy for treatment-resistant bipolar depression. So they took a group of people with treatment resistance, you know, very well-defined treatment resistance. They had to be failed at least two standard treatments for bipolar depression at the right doses for the right amount of time. Most of them had been on at least eight treatments lifetime for their bipolar

and still were stuck in depression pretty chronically. So these are pretty ill people. There were 45 of them, both bipolar 1 and bipolar 2, and all of them also had insulin resistance measured by this score called the Homa IR, which is kind of complicated. But it's an easy blood test to order, and I order it on everyone now. And I think that's a good clinical pearl we can get into later.

What is the Homo IR? So it's a fasting insulin and a fasting glucose, and they do a calculation. based on a formula, and that tells you essentially how insulin-resistant someone is. It's not the best measure of insulin resistance. It's just the easiest one to get. The real best measure of insulin resistance is a glucose-tolerance test, which is a real bear of a test to order and for anyone to do. So we generally don't do that, and we do this shortcut test called the Homa IR. So it's just a single blood draw. It doesn't cost that much.

You can even just get a fasting insulin without the fasting glucose, or you can do the calculation yourself and save a little money. So it's a really important test, I think, because it's essentially that's the root cause of pre-diabetes. That's the step before pre-diabetes, even if someone's hemoglobin A1C is nice and low, 5.1 or 5.2, and you say, great, this person looks metabolically healthy, they still might already be insulin resistant. Because that's, insulin resistance starts, you know, sometimes 10, 15 years before the diabetes process. And insulin resistance is highly correlated with brain illness and mental illness. And in this study, we'll see how important it is

for bipolar depression. That's the important point about this study. So they randomized these 45 people, 25 on placebo, 20 on metformin, high-dose metformin, which people are probably familiar with. It's a diabetes medication that does have some effect on insulin resistance.

And so of those 20 who are on metformin, 10 of them converted to being insulin sensitive on the Homa IR. And then one of the people in the placebo group actually converted to being insulin sensitive. And then they followed these people using the madras and other scales as well. The madras was the primary outcome at week 14. And they had a really significant difference in the people who converted to being insulin sensitive versus the people who remained insulin resistant in terms of their response on the madris. And on these other scales, like Hamilton Anxiety Scale, the CGI clinical global impression for bipolar, and the GAF scale, all were significantly better in the people who were converted to being insulin sensitive compared to the people who remained insulin resistant.

Wow. Yeah, this is really interesting stuff. And this is just metformin. Just metformin. We're not talking about diet. We're not talking about exercise. We're just talking about metformin. Yeah, just metformin. Now, metformin is, I mean, even in the study, you see, is not that powerful at reversing insulin resistance. It only was able to do it in 50% of the people, and they were using high-dose metformin, which, as some users probably know, does have some side effects. There's a lot of GI side effects when you get the dose of metformin up to 2,000 a day, which is generally what they're trying to do in this study. Some people stayed on 1,500 a day. But, and, you know, even at that high dose of metformin, only half of them could convert to being insulin sensitive.

But it is something to know for someone, you know, who isn't willing to do big dietary changes, but they're willing to take metformin. You know, this is something to consider, measure their insulin resistance, see if they're insulin resistant. If they are, you know, put them on a high dose of metformin and see if they can convert to being insulin sensitive and see if that improves their bipolar depression. You know, I think this is a really well-done study, you know, quadruple-blind, placebo-controlled, you know, this is a rigorous scientific study. So, you know, this is the kind of thing. It's small, of course, you know, only 45 patients. But I think there's enough there to say, given how safe metformin is, that it's a reasonable thing to try for people who aren't willing to do diet. Yeah, so, okay, just a couple things to reiterate here. Homo IRR is how they define insulin resistance,

and we can order this fasting insulin, fasting glucose, and we can put this in calculators online. It's like you multiply them times each other, and you divide it by 405. Right. Well, it depends on what units. You have to pay attention to the units that you're laughing. Pay attention to the units.

These are using not the units. from the US, they're the Canadian units, which are different. Or at Quest Diagnostics and probably the other big labs around, you can just order the Home IR and they'll do it for you. And there's actually a little bit of a fancier one that Quest Diagnostics does. They send it to a specialty lab, and they use the C protein that's part of insulin as well, and it makes it a little more accurate.

Oh, that's cool. Yeah. I also think that sometimes ordering a lab like this and showing a patient, like, hey, you are insulin resistance. Right. Hey, we need to start treating this. We have a target.

We have a number. There's something about that that can increase. It's not like we're just talking about diet and exercise. And we're talking about something physiologically going on that we can intervene, especially early, before it becomes a bigger issue. It's an incredibly important point. Yeah.

People really believe. lab work, like something about that feels so real to my body here. I need to fix it. And, you know, no matter what we say about diet, because it will help your mental health, they may not believe it, but when you start seeing the lab work, that really gets people's attention. I found that to be true almost universally. Yeah. Okay, let's talk about Sethy et al-2024. This was a four-month pilot trial involving patients with bipolar disorder and schizophrenia. Phrenia, let's go through this one. So yeah, so Shabani Sethy has a metabolic psychiatry clinic at Stanford, and this is her pilot trial.

It's important to say right off the bat, the rest of the trials we're talking about do not have control groups. So these are uncontrolled trials, and so, you know, that's just a caveat we have to put out there. And I think that's the state of where we're at today, which is important that this is just, the beginning of the field. But I think it's important to educate my audience on this, despite they're not being randomized trials, we're maybe a couple years away from randomized trials. I'm guessing if we were to do an update in two years, we'll have two or three of them. How many do you think we'll have in two or three years? I agree with you. Yeah, a few. But there's probably about five or six randomized controlled, big randomized control trials going on right now.

Some of them will probably take a little longer than two years to come out. But they're bigger trials, they're randomized, controlled trials, and almost all of them have a really cool basic science component. We're going to get to one of the pilot trials has an MRS component. All of these ones that are going on now have metabolomics and genomics and MRS or functional MRI. So we're going to see a lot more science coming out in the next couple of years, which I'm incredibly excited about. But yeah, the state of the field were very early in terms of the clinical data.

I mean, we're really telling a whole story here about not just the power of these diets, but how important it is to think about reducing medications. How can you reduce medications with something else to take its place? A whole story about having biological markers to look at as psychiatrists when we're doing interventions. But yes, the clinical trial data is in its infancy and, you know, most of the of it as non-controlled, as we said. Okay, so in these participants, a large proportion, 76% were bipolar, 24% some form of schizophrenia.

They had a high adherence to ketosis, 80 to 100%, which is really good. Yep. And just to say this was a 1.5 to 1 diet as well without patients. So not a heated ratio. and they weren't measuring ketones daily. They were measuring them weekly. And they only said adherent was greater than a 0.5 BHB, which is not very high from a clinical perspective.

So we're not even talking about getting people into deep ketosis here. We're just talking about kind of a mild ketogenic diet that people could follow. Good point. Good point. It's really helpful. And I think that's helpful because, you know, when I put patients on this type of

diet what I do nowadays is I have like the keto mojo yeah and for a hundred bucks they can get a bunch of test strips they can get a little device they prick their finger they don't have to do it every day but I want to see you know are they getting into ketosis and so this according to this study 0.5 would be getting into far enough to get the benefit which is interesting yeah yeah but you know people may get more benefit if they get the you know I see this clinically they may get more benefit they get their levels above one or you with more severe illness like schizophrenia, bipolar, schizoaffective, or severe major depression, I actually try to shoot for above two a lot of the time, especially in the beginning,

to try to really see a robust response, really robust response. Okay. And neither of us have any ties to these companies that are selling these devices. Just we're not promoting what we financially are benefiting from here. Right. You know, it's so interesting because, like, there's no drug reps for the ketogenic diet, right? Like, when I started talking about exercise a decade ago, you know, doing podcast episodes on it six years ago, articles before that is like, there's no person coming to a psychiatrist's office talking about exercise. But at the biggest psychopharm conference, out of Nevada, it's one of the big ones,

they spent the whole time talking about lifestyle stuff. Amazing. This last year. And I was like, oh my gosh, the field is changing. And that's really exciting to me because it's like six years ago, people thought I was crazy for talking about strength training and mental health. So we've come a long ways. Okay, but let's get into what this study actually found.

Yeah. Yeah. So they looked at some different outcomes. The one that I thought was interesting was the CGI, the clinical global impression. The average improvement was 31% on the CGI, which is significant. And there were 33% of people were considered recovered at baseline, up to 75% people recovered based on the CGI at the end. And then really cool thing, too, was that there was a difference between the adherent group and the semi-adherent group. So the people who followed the diet better and had higher beta-hydroxybutrate levels,

had a 92% recovery by the end versus only a 50% recovery at the end for the semi-adherent group. So there's a sort of suggestion of a dose response that we're seeing in that. There's also a significant improvement in life satisfaction, 17% improvement, another 17% improvement in the global assessment of functioning, the GAF. These are all highly significant. 19% improvement in sleep, 32% reduction in the BPRS for the people with psychotic disorders, and the metabolic outcomes were really good, too. All the people who had metabolic syndrome when they started

were taken out of metabolic syndrome. So, I mean, in every one of these studies that we're going to go through, we can talk about mental health outcomes and metabolic outcomes. The metabolic outcomes are really nice in all of these studies, like really strong metabolic improvements. 36% reduction in visceral adiposity, which I think is very interesting, Because there's a lot of links to visceral adiposity and future issues.

Absolutely. I mean, that's the dangerous kind of fat. That's the fat that's hanging around your organs. Not, you know, subcutaneous fat. There's a lot of people who are metabolically healthy who have a lot of subcutaneous fat. But if you have visceral fat, by definition, even if you're not overweight at all, you can be have a normal BMI, have visceral adiposity, and you are metabolically very unhealthy. We have 30% of people in America have fatty liver disease right now, almost exclusively from metabolic causes.

It used to be all from alcohol, you know, 50 years ago. Now it's all from metabolic issues from food. And that's very dangerous. This is a sign that, you know, you're going to have multiple problems going forward. You know, diabetes and everything that comes from that, heart disease, cerebrovascular disease, all that's going to happen if you have got a lot of visceral adipose tissue. Yeah. Other things, I did a big dive on this the other day that can decrease that as vigorous exercise. Yeah.

But 36% is a lot. It's a lot of reduction. Yeah. And I see this all the time because I measure visceral adipose tissue at our clinic. We have a machine that measures that. It's one of the things that we always see benefits, as well as in the home I.R., we always see benefits when people are doing ketogenic diet. So just to go back over the number, 32% reduction in psychotic symptoms in the brief psychiatric rating scale.

That's cool. Yeah. That's cool. This is impressive. And this is only a four-month pilot trial. You know, this is only four months. That's right. And you're getting those huge reductions.

That's right. And yeah, these interventions are meant to be long-term. And what I say to people is, you know, with most medications, you're getting. most of the benefits in the first couple of months. But that's not true with this. You get benefits accruing month after month after month. The longer someone is doing these changes like ketogenic diets and exercise, the benefits just keep accruing over time.

You know, a year later, two years later. I mean, you know this. You see people keep exercising. They keep getting more and more benefits as they keep doing it. Metabolic health is physical health. you know, like how well your mitochondria are functioning. Absolutely. It relates to how your brain is going to function.

I've been looking at studies on creatine recently in mental health, and there seems to be a really nice link there as well. It's helping the mitochondria function better. Yes. So, okay, let's go to Dannen-At-All 2022. Yeah, so this one I think is really fascinating. So this is a doctor in Toulouse, France, who towards the end of his career, learned about ketogenic diets. And then he actually convinced his sister's family, I believe, to put his nephew on a ketogenic diet who had both seizures and autism.

And this boy had an incredible response to the ketogenic diet. For both seizures and autism is my understanding. And he was just blown away. And he said, all right, I want to try this in my inpatient unit. And so he was an inpatient doctor's whole career. He had all these patients who were coming in and out of the hospital, and he offered this intervention to 31 inpatients. Some of them were admitted just to do this intervention.

Some were just getting admitted for other reasons. But these are people with chronic, severe, you know, sort of intractable illness, schizoaffective disorder, severe major depression, and bipolar disorder. And he put him on a Whole Foods ketogenic diet. He added some vitamins, not high doses. I don't think the vitamins were all that significant. in this, but it's important to add.

This is also a non-controlled trial, but they mentioned in the paper that these individuals could all serve as their own controls to some extent because they had all been admitted to these hospital units before and never had these kinds of outcomes before. But they didn't have measurements of those previous studies, so they just sort of mentioned that as a throwaway. But I think it's interesting. So they measured adherence just by doing urinary ketones, which is not the best way to measure ketones, but you can see them in the urine. They only measured it once during the study. So really,

we don't have a good sense of whether people were truly in ketosis or not, but they just measured it based on compliance with the diet, which was done by a dietitian. And, you know, they came in like meal boxes that they were just given, essentially, on the unit. And interestingly enough, they were only on the unit six days a week. And in France, they get to go home on Sunday, even if you're an inpatient, and you just go back to the unit on Monday. So they weren't, you know, they weren't having the diet actually even seven days a week. And so they said excellent compliance in 39% good compliance in 43% fair compliance in 18%. And then they measured rating scales. So the Hamdi, 23 people got that, the Madras, 21 people got that, the people who had depressive symptoms.

And the people with psychotic symptoms, there were 10 of those, had the pans done as well. And then they did the CGI on 27 out of 28. Only 28 people completed the study out of the 31. three people just, you know, couldn't do the diet. And you had to do the diet for 14, only 14 days. Some of them did it for much longer, but 14 days minimum was considered someone who did the diet and had their outcomes measured at the end. And the reductions in these rating scales are like mind-blowing. It's almost, you almost can't believe how well these people did.

So the Ham D went from 25.4 average to 7.7 average, which is, which is basically remission. I mean, that's like huge, that's a huge reduction. That's going from depressed to not depressed. Right. The effect size, they give a Cohen's D for effect size,

3.1 on the Ham D. I mean, it's just like mind-blowing. Yeah, it's ridiculous. Right. The Madras, similar, even better, 29.7 to 10.1, the Cohen's D was 3.6. And then in the Pans,

91.4 to 49.3, huge reduction. psychosis. The Cohen's D was 3.5. And then the CGI, clinical global impression, 4.9 to 2.0. Now, that's also enormous, you know, 3.8. And of course, that's a, that's a radar scale. That's not a self-report scale. So, you know, there's all kinds of ways this study can be criticized from a scientific point of view. You know, there could be bias in many different directions. However. Sure. Patient selection. Yeah. These are the patients who are willing to do it, you know, so they're like healthy enough to choose to do it, right?

Trusting of their doctor to go through this. Right. The belief of the doctor could have a huge impact. But he's been treating them. It sounded like for years. Yeah, most of them are around years care for more than 10 years. Right. Some of them long.

Yeah. So, you know, it's like, yeah, this isn't a randomized trial. I'm not buying the self-randomization type thing. But at the same time, for me, it's like, I look at this and I'm like, hey, okay, this is a big, this is a big impact. We should pay attention to this. This is really interesting. Yeah, and some of the other things about the study were interesting. So 64% were discharged on less medication than they came in. Only one person out of all of this was on more medication than they came in. The rest were either had the meds change and it was somewhat equivalent. So, you know, most people were on less medication despite these incredible improvements. So that's really incredible. And great metabolic health improvements as well, all super significant improvements in weight, fasting blood glucose, hemoglobin A1C, blood pressure, liver function test, total cholesterol, even the LDL went down, which we don't always see with ketogenic diets. That's a whole separate topic we could get into, but triglycerides went down very significantly,

which we always see in ketogenic diets. And most people said it was well tolerated. They did a little follow-up and, you know, 46% were still having good adherence in the follow-up. 18% were having partial adherence and 21% had totally stopped doing the diet when they followed them up later because it's hard, much harder to do it at home, especially if you didn't get a bunch of education on how to do it at home and you just were, you know, given this, you know, it's not, you did, people need education to learn how to do this. So yeah, huge improvements in metabolic health. And generally, you know, people are interested in can keep, you doing it when they left. And I agree with you that even though it's not control, even though there's

all kinds of scientific criticisms of this study, which are valid, it's still really impressive to see those kinds of numbers. I worked on an inpatient unit. I almost never, ever, ever saw those kinds of improvements in an inpatient stay, you know, unless it was, you know, some kind of, you know, acute manic episode that's, that resolved the meds or, you know, major depressive episode that that someone got ECT and they had that kind of improvement. But otherwise, you just, you don't see those kinds of improvements. And these are not the kind of illnesses that are super, you know, placebo responsive, I would say. Like, we don't even do placebo trials in schizophrenia.

You know, so, you know, these are not, you know, mild to moderate depression where placebo makes a big difference. These are different kinds of illnesses than that. Yeah. The other interesting things to me in this study was the systolic blood pressure decreased from 134 to 123 so that's like 10 points 11 points yeah triglycerides dropped from 204 to 137 I tend to see you know triglycerides tend to increase with more carbohydrate so that makes sense to me yeah weight dropped about 10 pounds so you know so the there's multiple ways to think about like as you talk to patients about this to

to get them to think about trying this, right? Yeah. You know, I've heard from some people this fear of LDL's increasing, so it's interesting that LDL's dropped 10 points in this study. And we also know that there's a difference in types of LDL, which you could probably speak to. Yes, yeah. The LDL topic is very complex, and it's very controversial.

I mean, you can do a whole episode on low-carbohydic diets and LDL, but just the summary points that I think about is, you know, overall cardiac risk goes down on ketogenic diets. That's been seen in big trials. If someone's LDL is high to start, it often stays the same or goes down. Especially if it's someone's overweight and they're losing weight on the ketogenic diet, LDL tends to go down. Where you see the LDL's going up are usually in leaner, fitter people whose LDL is starting at a low point. and then you put them on a ketogenic diet, the LDL sometimes do go up. But still, the overall cardiac risk, even when the LDL goes up,

the overall cardiac risk goes down because there's so many other factors that go into cardiac risk calculators, blood pressure, triglycerides, HDL, total cholesterol, you know, all the insulin resistance, all those are very important factors. So even when LDL isolated goes up, overall cardiac risk goes down. There are this subgroup of people called lean mass hyper responders where the LDL goes really, really high. And the triglycerides go really low, and the HDL goes really high.

That's this triad called lean, which constitutes a lean mass hyper-responder. But they've even done studies on that. These people in ketogenic diets whose LDL are on average, like astronomically high, like 300, 350, some of them are even 500 in the study that they did. And they had absolutely clean cardiac angiograms compared to a control group that had the same profile,

except their LDL was lower and they weren't on a ketogenic diet. The people who had these super high LDLs said a suggestion of actually less cardiovascular disease on the gold standard test, the cardiac angiore. So that's a study of about 100 people. That literature is still coming out and really interesting. But it's really changing this whole story about LDL being so important to cardiac risk. You know, LDL, different subtypes, you know, they're really delivery molecules for energy in many ways too they're not just atherogenic molecules and some of them are atherogenic and some of them are not and so we just measure ldl and all that's all we care about we're

missing like you know a huge part of the story right and things like visceral adiposity right things like triglycerides things like inflammation right that have other impacts plus i mean if you're in us like you know for people with mental health issues you know there's there's a um an inflammatory state often associated with acute illness. So if you're able to get someone out of psychosis, that's going to decrease cortisol, that's going to decrease stress hormones, you know, all sorts of stuff.

Like imagine being paranoid and psychotic, getting you out of that state is very good for the brain long term. Absolutely. Oh, yeah, for the brain and the body. I mean, you know, inflammation we know is correlated with cardiovascular events. Okay. Okay. So this is a really good study. Anything else you want to talk about on this one before we move on? No, I think we can move on. Okay. Let's go to Calabrese et al-2024.

Oh, you know, I actually, I didn't skip this one. Yeah, I didn't take notes on that one. That's a small sort of case series. But I think the main point on that one is that, you know, people getting into remission, you know, a few people, you know, just, you know, complete remission. And that's the kind of, there's, there's lots of case reports. coming out now, Calabrese. There's another couple of ones by Nicole Laurent, who's a psychologist in Washington State with both depression, and she had two people with schizoaffective disorder put in complete remission. I mean, literally CGI is down to zero off of medications. And so, you know, there's those types of case reports and anecdotes and stories that are now getting published and many of them online. I mean, there's just, there's lots, there's dozens and dozens of people who are being put into remission coming off of all their medications just by being on ketogenic diet, exercise, and maintaining a metabolic regimen. So just fascinating. Yeah.

The thing I found interesting about this one was the, it's only three adults, right? Right. But OCD symptoms improved and binge eating disorder symptoms improved as well. Oh yeah. Okay. Let's go to Campbell et al 2025. This is probably one of the most interesting to me. Yeah, this one's really cool. So Ian Campbell is a researcher at University of Edinburgh. He's very public about the fact that he got into this because he has bipolar disorder and went on a ketogenic diet himself to lose weight and realized totally by happenstance that it completely stabilized his bipolar disorder. And then he got super fascinated in it. And, you know, he's a PhD researcher, really, really wonderful guy. And so he has a whole group now at University of Edinburgh. There is this

whole group that they're doing lots of studies. This is, they came out initially with a clinical study in 23. This is the same clinical study, but now they're reporting in 2025 on some more outcome measures that are really interesting. And this is all people with bipolar disorder. So, and once again, this is a pilot study. This is not a randomized trial. Correct. Six to eight weeks. Interesting, he describes them as euthymic bipolar patients. Right. So it's not like they're acutely ill, acutely manic. But the thing that I like about this one is how he looks at the brain. What's going on in the brain?

Yeah, I mean, he's a real scientist. So, yeah, he's got this MRS component to this study and also some other really cool aspects. So let's get into it. Yeah. So 27 were recruited. 26 began the study. 20 of them completed the study.

They had to be euthomic for three months. They excluded people with severe hyperlipidemia. they were all assigned a dietitian. They were put on also not an extreme form of the diet, but it was a real ketogenic diet, 60 to 75% of the calories from fat, 5 to 7% from carbohydrates, and the remainder from protein.

They did measure ketones daily in this one, and blood glucose daily, which is really important for this. They ended up doing these ecological momentary assessments. So essentially they were texting people and asking them some questions every day, and they made correlations between their ketone numbers and their responses to these questions

that ended up becoming super interesting. They had a dietitian that met with them weekly and was able to adjust the diet to achieve ketosis and weight loss if desired. So this is also really important. Instead of just being given this program and just let go, they had some really nice dietitian support

for these people, and they were able to adjust to make sure that they were getting into ketosis and measuring it daily. So the moods remained euthymic throughout the study, except for one person who had a little bit of hypomania that resolved quickly. And we do see that sometimes with bipolar people who go into ketosis.

There is sometimes a little bit of hypomania and a little bit of insomnia that you have to pay attention to as a clinician. If someone's thinking about using these things clinically, that's a really important clinical point, and they did see it in this study. So on these ecological momentary assessments or EMAs, these daily reports, they only had that on a subset of 14 people because I think their technology wasn't working

in the first part of the study and they finally got it working. So it was sort of the last 14 people that they were able to do it for. So they had really nice positive correlations between mood and ketones, positive mood. The P value was less than 0.001. And there's nice little plots on the study

if you want to look at these. They're nice to look at. Positive correlation between energy and ketones, also P less than 0.001. Negative correlations for impulsivity in ketones. And that actually had the biggest effect size of them all. They had the best reduction in impulsivity with, you know, correlation with ketones of all these measures that they did. And also a reduction in anxiety and ketones, all these less than 0.001. They did not have a

correlation between racing thoughts and ketones that ended up becoming, that was not significant. Now, the R values for these, you know, 0.2, 0.3 for impulsivity, negative 0.3 for impulsivity. negative point two for anxiety. These aren't huge. These are not huge. This is not like groundbreaking. Is it groundbreaking for you? I mean, I just like to see these correlations.

Like anything with dose response, you know, it's exciting. Yeah, I mean, it's not groundbreaking. But of course, these are also people who are starting off and remaining euthymic. These are not symptomatic individuals. But I think what you're emphasizing here is that they're looking at like, like a spot of like this is one day and this is their ketone level. Right. And as their ketones are higher, their symptoms are all what we would want them to be.

Exactly. Right. And so that's probably the most interesting thing from those numbers. Exactly. That's what's interesting about it. Exactly. That there's this correlation between the levels of the treatment and the nice things we want to see. And so if someone's not feeling as well,

well, we can say, all right, let's try to get those keto numbers a little higher. I think that's what this says to me. And you'll probably feel a little better. So that's nice. And it's just so nice to have a relevant biomarker in psychiatry. I mean, we do psychiatric treatments all the time. And unlike the rest of medicine, we don't have biomarkers to follow that are really relevant. And it's really cool to have one that seems to have some relevant. Here's the other thing. And I'm going to just pull up actually the study here, for those of you we were watching this on YouTube, so you can see this actual thing,

and you can pull this study up at home. But when I look at data, I look at, does the data represent a linear line or not? Yeah. Okay. Because often there's a step, like once we get to a certain point, we see a huge reduction.

And so I actually think if I were this researcher, or what I would say is like, it seems like there's a step in some of these things. And what I mean by that is when you get past, for example, three or 2.5 for the ketones, all of a sudden there's a drop in impulsivity. And there's like a huge drop. Okay. Do you see that?

Yeah, exactly. Or for thoughts and for anxiety, when you get past the two and a half mark, it seems like there's a drop in anxiety, a drop. Yeah, specifically anxiety. There's a drop. Do you see what I'm talking about here where it's like, all of a sudden, boom, it's like every number is low. Yes.

And so I think as well, when we think about a small correlation of like a point three or a point two, we miss the overall picture unless we look at one of these nice plots where we see like, no, there's maybe the data is not best modeled in a linear line. Maybe it's best modeled in like a stepwise approach. And so that was one of my takeaways. Yeah, I really appreciate you doing that. And it's also, you know, to think about that in the context of this being such a short study, too, and I was making that point earlier that, you know, getting the higher levels of ketones in the earlier stages is probably really important.

And we see these really nice improvements when you get above like two and a half or three. And I see that clinically all the time. And someone really gets into deep ketosis. They're like, wow, this really feels amazing. That doesn't mean they have to stay in deep ketosis, you know, at month eight or month 24. Or it may be that they don't have to. We see this in the epilepsy literature in the kids that most of them who respond early,

they're able to stop the diet after a few years, and their illness not only doesn't come back, it keeps getting better later on, even when they stop the diet. Stabilize the brain. See, I had a close friend whose kid was in, had sort of like intractable seizures, and I convinced him to get their kid on a ketogenic diet and they did it for a couple years

and they've kind of moved away more into like, I think more, a lot of whole foods, but not as ketogenic and the kid's doing a lot better. Yeah, amazing. Yeah, so it makes sense to me that if you can stabilize the brain from having seizures for a period of time and you have a kid that's growing up now, every year it's like their brain is growing up their brain is developing their brain is changing right

and so um all these epigenetic things all these new pathways uh that makes sense to me okay but the thing that i love first okay okay if we reverse that's that cerebral insulin resistance as well right then you know we don't have to do that heavy lifting in the beginning we're doing this heavy lifting of reversing insulin resistance you know getting the mitochondria to be healthy again And at a certain point, you have enough healthy mitochondria, you may be able to add to that, but the biggest changes may happen in the first six months, a year, but you still continue to get benefits as you keep going

with a healthy lifestyle. And some amount of ketones are providing that signaling to the mitochondria on an ongoing basis. That's great. So there's this one study that I really like, I'll share with you. I don't think you've probably seen this one, but they looked at three groups of people

they looked at people who are more elite level athletes they looked at people who are average people and they looked at people who have some sort of metabolic dysfunction and they looked at the amount of watts that they could produce on a bike for a given level of lactic acid and then they looked at the mitochondrial health in these three different groups and what they found was that yeah there's a there's a big difference in the mitochondria health. Let me see if I can pull this study up. I want to show you. It's really cool.

I think this is, I think this was the most interesting one right here. The relationship between blood lactic acid and exercise output and the three different groups, the people with metabolic issues, the normal people and the athletic people. And so it's another picture of this like, you know,

mitochondrial health, you know, the ability to produce watts at different, more watts at lower lactic acid threshold, right, is mitochondria health. And I think it's kind of feeding into this whole holistic picture

that we're giving of how that then can make an impact on our mental health in multiple domains. Yeah, absolutely. Yeah, and if those of your listeners who heard Chris Palmer before or read his book, you know all about mitochondria having all these other important functions

besides making energy as well. in the brain. And so mitochondrial health is so important for mental health, you know, because of energy, but because of all these other reasons as well. And lactic acid's super interesting, too, that, you know, there's now, I think, a lot of data showing that lactic acid improves brain health. And so, of course, when we're exercising, we use it as a measure of how healthy our mitochondria are. But if we exceed our lactic acid threshold and make lactic acid frequently, that's going to improve brain health. That's directly correlated to increasing BDNF,

and that's one of the mechanisms. We want to exercise and make some lactic acid some of the time, and some of the time go exercise right at the threshold as well. Right, vigorous, vigorous exercise, good for BDNF. Okay, the thing that I love about this study is the MRS data, because I think this really shows, you know, what types of the brain. are actually impacted, like the anterior cingulate cortex had an 11.6 decrease with, you know,

there's a decrease in glutamate and glutamine. Yeah. Yeah, I think it was 13% in the combination of glutamine and glutamate, I believe. And that's really significant. That's big. Yeah. Because, you know, first of all, we know that glutamine and glutamate get lowered in anti-convulsin treatment of bipolar disorder. So this is a known mechanism by which things like, you know, balk acid and lomotur gene help in bipolar disorder. This is one of the mechanisms by which those anticonvulsome medications work. They lower brain glutamine and glutamine.

These are excitatory neurotransmitters, neuromodulators. Yeah. Actually, interestingly, chlozapine, Clozapine increases glutamate neurotransmission. which, you know, I'm not quite sure how it relates to this area, specifically having decreased, because I think you have, you know, sometimes different things going on. But clozapine indirectly improves NMDA glutamate receptors.

It blocks 5HT2A receptors, which inhibits glutamate release. so indirectly boosting glutamate release in the prefrontal cortex. Okay, so maybe that's where it is. For an area. So if we're boosting in the prefrontal cortex, that maybe has a different impact than in this study when we're seeing it in the anterior cingulate cortex decrease. Okay, so that makes sense to me. It's because in the anterior cingulate cortex, you have emotional regulation,

you know, you have cognitive control. So having maybe a blunted there, we have some increased emotional regulation, right? Maybe less rumination, less self-like heightened emotional processing. Yeah, it sort of makes me think about this question that is still puzzling me. And so, you know, we know that antipsychotic medications are, you know, obviously really helpful for psychosis. that they're doing something very different to metabolism than what ketogenic diets are, which are also, we're now seeing, are helpful for psychosis. So how do we explain that?

You know, how do we explain why antipsychotics, even though they're harming metabolism, why are they helping with positive symptoms of psychosis, especially in these acute settings, which we know that they do? I mean, clearly they do. And we just don't know the answer to that question. I've asked it to Chris Palmer, and his answer was interesting. He says that, you know, when the, when the neurons running out of energy,

it can either get hyper-excitable or hypo-excitable. You know, it can either, like, you know, start doing too much or get, you know, hypo-active. And that when we replace the energy or fix the energy in the neuron, it's now going to do its job the way it's supposed to. You know, it's not going to become hyper-excitable or hypoactive. And so, you know, it has the ability to sort of get the neurons to do what they're supposed to be doing in different regions.

And so it's not like, you know, we're just, just, you know, turning something on or turning something off. You know, medications are kind of blunt in that way. They're just doing one thing. Whereas this has many mechanisms of action that have been shown in the epilepsy literature. And, you know, what we just don't know yet in mental health, all these mechanisms. But we're seeing a nice suggestion here about lowering glutamine and glutamate in the anterior signalet cortex and bipolar.

Yeah, and I think that's a good point. It's like what is the mechanism, we're not sure. Also in the study, they showed there was a decrease in total coline in the anterior singular cortex and the posterior singular cortex. Yep. And it's another one, myoenocetal as well, which is a secondary signaling molecule. I don't know if you know some people take myoanocetal as a supplement for various things. It seems to help with severe OCD and anxiety.

There's some very small studies looking at that. that, but I do know a lot of people who take that supplement, actually. There was no significant metabolic change in the right dorsolateral prefrontal cortex, which is interesting. Yeah. But, you know, we're looking at, we're just scratching the surface here. I think it was what you're saying as well. It's like, and when you look at diet studies and you look at big changes like this,

it's not just one upregulation of one protein. Right. There's probably 10,000 things going. on. Exactly. Which one of those is the thing that's going to make the biggest impact for the patient? We're not sure. Just like an exercise, there's tons of things going on. Exactly. Exactly right. Yeah. We're doing something that has such a profound effect on physiology that, you know, it's going to be hard to find exactly what the one thing is. And that's what they found doing all this research on ketogenic diets and epilepsy that they've come up with about, you know,

seven or eight different mechanisms that various researchers think are really important and the most important one, no one can figure out what the one mechanism is by which it works so well in epilepsy. And I think it's going to end up being the same. And therefore, you can't, you know, you can't create a pill that's going to do what exercise does or what a food diet does for the brain. We'll never have a pill. Yeah. And you can't just probably take, like the bulletproof coffee type of thing. You know, like that's probably not going to be enough because you're, and just, okay, I'm going to take this supplement

and then go on eating a lot of highly processed foods and foods that are proclamatory. You won't have a good of an effect. Although interestingly, they've done these studies on MCT oil that are fascinating. People with mild cognitive impairment, they do a mini-male status exam on them, and then they give them MCT oil in a fast state,

and then they do a mini mental status exam on them an hour later, and they get significant improvement. That's from one dose of MCT oil, just because they made some ketones. And for those folks, you know, this mechanism of, you know, brain insulin resistance is probably a really important one, and that immediately gets bypassed by making some ketones, because ketones don't need insulin to get into neurons,

whereas glucose does need an insulin signal to get into neurons. And so you bypass that blockade, right away and you get an immediate boost and an immediate benefit. But I totally agree with you that that's not enough for most people to have a profound, you know, lasting mental health change, you know, just doing some MCT oil is not going to do it. Yeah, and I think the other thing that I've been thinking about with this is like, I'm not just, I'm not going to only do this with patients, right? This isn't the only treatment I'm thinking of exercise as well.

Oh, yeah. Some of the, you know, other things. all these various important lifestyle things. So ketogenic diet or even a low-level ketogenic diet or even just this low-glycemic index diet, if that's all someone will do, just a clean, healthy diet that doesn't have sugar and processed flour

and we'll make a little bit of ketones. Exercise, especially if we can get them to do some resistance training and some high-intensity work as well, mindfulness training, and then some circadian rhythm work as well is super-powering. for a lot of our folks.

And I talk about that, you know, kind of all day long sometimes. I feel like the importance of getting out and seeing morning sunlight and blocking blue light at night and getting outdoors during the day, eating meals at regular times, waking up at regular times. That is incredibly powerful for mental health and physical health. And it's so against the grain of what most people are doing that you can get huge benefits just from making these small interventions. Yeah. Yeah. Yeah.

I think we should kind of get into the practical stuff as much as possible at this point. Sure. So when someone is getting into ketosis, what are some things that you think they should know? Like let's say someone's listed in this and they're like, okay, I'm going to try this. Or I'm going to get my son on the ketogenic diet. Like what are the things that they should consider in that process? are in that process? Yeah.

Well, first of all, a couple of absolute contraindications. You can't do a ketogenic diet if you're taking an SGLT2 inhibitor, these drugs that are used for diabetes. So, you know, most people are not on those medications, but that happens to be an absolute contraindication.

You can get ketoacidosis from that. Otherwise, you know, there's very few absolute contraindications. If someone's really severely underweight, I wouldn't do it. it is done quite often clinically with people with restrictive eating disorders. We could go into that a little bit, but that might shock people. It sort of shocked me when I first heard about trying this and someone with a restrictive

eating disorder. So the things to know are start gradually. It's not a great idea unless you're really under a team of professional care to do this all at once. So I generally like people to start with cleaning up your diet, get reshugure try to get rid of some processed flour, do that for a couple weeks, keep eating veggies. I like people eating animal protein, but make sure you're getting adequate protein all the time. Those are really important factors.

And then just start gradually reducing carbohydrates. So maybe start with one ketogenic meal that has very low carbohydrate for a week. And then the second week maybe add in a second meal of the day with very low carbohydrates. And then, you know, by the third or fourth week, you can do all three meals. And when you're getting rid of all these carbohydrates, so when I'm, you know, just practically speaking, you know, bread, rice, potatoes, you know, we just aren't sugar. We're not eating any of that. Unless you're getting a really, you know, well-made ketogenic version of bread, which can be done, but it's not easy. We have to replace those calories with healthy fat calories. So what do we mean by healthy fat calories? So there is some controversy in this world. I won't get too. much into the weeds, but everyone agrees that mono-unsaturated fats are healthy. There's really no controversy about mono-uncaturated fats. So high in things like olive oil, extra virgin olive oil, avocados, avocado oil, nuts and seeds

in their whole form, not seed oils, but whole seeds and maybe minimally processed seeds like tahini is like ground sesame seeds, for example. So generally higher in mono-insaturated fats and those foods, all very very, very much-edged seeds, all very safe to eat. And then, you know, animal fats, you know, some amount of animal fats are probably healthy and important. They have more saturated fat. They still have the other fats in them as well. But, you know, if they're well cared for animals, there may be some benefits to that, higher omega-3s and grass-fed beef versus not, and grass-fed dairy also is higher in omega-3s. Dairy is one of the ways that sometimes people get more of their fats in a ketogenic

diet. It tends to be easier and more satisfying. If someone tolerates dairy and is willing to eat cheese and cream and things like that. And it's, you know, it often makes things taste delicious. So, you know, for people who do well with dairy, that's a, that's a good trick. And then, you know, I do generally advise people to avoid a lot of seed oils because they're highly processed and they oxidize very easily. So saturated fats are more stable than these polyunsaturated fats. So replacing with healthy fats and because you're going to be eating more fats, you've got to pay attention to your fat, fat content. Olive oil, love it.

Okay, so here's my thought. Dark chocolate, okay? Yeah. So 80%. How much of that can I eat? Well, okay, so I like chocolate, too, by the way. And I've found that you can get chocolate that's made with a sweetener called alulose. Alulose, okay.

New sweetener that most people, people are, you know, heavy-duty scientists are now agreeing is probably the best of the artificial sweeteners, of them all. Alulose is a naturally occurring sugar similar to fructose that exists in foods like tapioca and corn at very low quantities. So they just purify it out of these other, you know, vegetables. And then you can get it in powdered form. It tastes almost exactly like sugar, but our body doesn't metabolize it. So it ends up being no calories. and it tastes sweet, and it doesn't seem to raise insulin or blood glucose at all.

Some of the other artificial sweeteners actually raise insulin levels, and that's problematic, because the only way we make ketones is the lower insulin level. So alulose doesn't seem to do that, and it actually seems to have some GLP1 activity, which is also fascinating. So some people take high doses of alulose instead of a GLP1 inhibitor if they can't afford it. Wow, interesting. Okay. Yeah, interesting.

Um, okay, so I'll use. So you can get chocolate made from out with alulose or stevia or just the highest percentage cacao that you can. So there's very little sugar in it. So I mean, there's one company that makes like 95% of cow. Oh, yeah, there's like there's 100% of it. 100% unpalatable. Yeah, 100% maybe palatable, but 95% from a good company that really knows what they're doing

can be delicious, I think. Okay, okay. Nice. Okay. Okay, what's the hardest thing that you find most patients, like the hardest step? Yeah, yeah, before I get to that, the one of the caveat to say to people is, if they are going fast into ketosis, they have to really pay attention to drinking tons of fluids and taking electrolytes. Because you have this diureasis when you first transition into ketosis that can be, you know, really uncomfortable. There's this idea of keto flu.

Some people try this on their own, and they say, oh, I hate the way I felt because they went really fast. They didn't replace their electrolytes. They got headaches and low energy and low blood sugar, and they felt terrible. So that's why I advocate a slower approach. It's much more likely to feel good. Some amount of measuring ketones is a really good idea. We mentioned blood measurements with a device. There's also breath devices that are pretty accurate now.

And so if someone really doesn't want to park their finger, they can get a breath meter. And I found that if someone's not shooting for a real, you know, without trying to be super accurate about their ketone levels, they're just more casual about it. That's another approach that we can take. Let's use the breath meter. Okay. Okay. So with the electrolyte supplementation, common question. Like how much? What do they do? Like, what's your, what's your, what's your take on that? Like three to five grams of salt? So you could just use sodium, and that generally can work. We're also losing potassium and magnesium as well. So ideally we'd replete those too. There is a salt that's made that's half sodium, half potassium. That's an interesting product that you could use in the beginning.

And just really, because now you're making your own food. You're not eating these processed foods, right? You have to really salt your food. And then additionally, you might put some of these electrolytes in your water bottle once or twice a day in the first couple of weeks. And like I said, just drink lots and lots of water, you know, you know, six, eight liters of water for sure. Maybe more.

Yeah. Wow. Okay. If you're going fast, especially. Because there's this big diureesis that happens when you go into ketosis and it takes away a lot of your electrolytes and water and you can get, you know, really dehydrated otherwise. Okay.

I'm going to have to watch that because I do sauna as well. I'll lose like, I'll lose four, five pounds in the sauna. Me too, yeah, yeah, I do that as well. You should definitely be taking in lots of electrolytes and really pound the water. Yeah, if you're going to do ketosis and sauna at the same time, especially. Yeah, you really got to be careful. Yeah. And make sure to get some magnesium, because you're losing all these electrolytes.

Magnesium potassium. Do you do potassium pills ever or do you tend to do the salt? I generally get these like little electrolyte packets. Okay. A couple companies that make them without any sugar at all. And those are, they taste good. And they're just really easy and you get a nice blend of those three electrolytes. I also, for some people, it's not a bad idea to take some trace minerals in the early stage as well.

You can get a trace mineral supplement because you're losing all these electrolytes too. Especially if you saw a lot, I would recommend a trace mineral supplement. Because you lose, you know, all these important things, you know, zinc and selenium and iodine. It's all coming out in your sweat. as well as toxins, but, you know, sweat brings out everything. I sweat a lot, so I think about this. I think I need a sweat mentor here.

I need you to send me your mineral recommendations. Yeah. But they can just get a little pill that has all these trace minerals in it, and then you just get the bigger quantity minerals in an electrolyte pack. It's pretty simple. Yeah. Yeah.

Yeah, that's good. Okay. Any other practical recommendations that you normally tell people? Let's see. I mean, probably not the best time to, like, if you're really, especially if you're going fast and you're, you know, it's not the best time to try to get your personal best and your exercise routine. If you're really into exercise, you're not going to want to push yourself to the max in the

first couple of weeks. Your body is adapting to a whole new way of producing energy. and it's working hard to do that. You're turning on genes that have basically been turned off. We all have the genes to do this. But, you know, these genes are our survival genes that exist because we needed to survive in food scares and food scarce environments, our entire evolutionary history.

All those genes have been essentially, you know, quieted down through epigenetics, through our diet and lifestyle. So we have to turn all these genes back on and it takes some weeks to adapt. to that. So I wouldn't, you know, go for personal bests in your, you know, in your training in the first few weeks. And then even after that, you know, it may take some time to get back to, especially in endurance, you know, if you're trying to train in ketosis. It's really great in the end

for endurance. A lot of endurance athletes are ketogenic. It's, you know, because you're essentially teaching your mitochondria to burn fat better. And so, you know, beta oxygen is improved over time in ketosis, but definitely not. In the beginning, there's some adaptation that happens. Yeah, I've heard about people training in ketosis and then competing with some sugar. Yeah. So it's like they've taught their body to mostly do ketosis. But you don't need that much sugar, actually.

The recent study came out that, you know, people who are these ketogenic athletes, they only needed to add back, I think it was like seven grams of sugar per hour in these long distance, endurance things, to get all the benefits of having some glucose around. So we don't need to add that much. And our body's making glucose all the time. That's another really important point for people to know. Our brain is additionally using ketones as fuel, but we never stop using glucose as fuel in our brains and everywhere throughout our body, even if we're in the deepest ketosis and we eat zero carbohydrates. Our body's making glucose through buhouniogenesis all the time

and using it all the time. But we're just making the amount that we need instead of this excess that we need to store as fat. Really good. Okay, we're kind of coming to a close here. Any final things you wanted to kind of bring out before we wrap things up? Yeah.

I mean, just trying to put it all together a little bit. You know, we looked at the wondering study about dose reductions of antipsychotics, having better rates of full recovery than people who didn't do a dose reduction. We know about all the metabolic problems with a lot of these antipsychotics and mood-stabilizing medications. And just thinking about the long-term health of people with severe mental illness. It's just, you know, we know that people with severe mental illness die younger, you know, on average, you know, about 15 to 20 years younger than people without it. There's many causes of that, but metabolism is a big cause of it. And I think, unfortunately,

the medications are a big cause of that. So, you know, thinking about people's long-term future, you know, we want to think about this as, you know, how do we heal and get to people to recovery long-term? I think this really could have a role for a lot of people out there. And I don't want to happen what's happened in epilepsy and in diabetes where there's been great, randomized, controlled, long-term trials showing that ketogenic diets work, but doctors don't even mention them to people until maybe never, or maybe like five or ten years into their illness, when people should be given an option. I mean, I just really believe that we, you know, even if you don't, you know, want everyone to do this or you don't want to do it yourself with your patients, I think people

deserve to know about this treatment. Yeah. I think through this process of, of looking at this paper with looking at these papers with you and kind of rethinking through this. It's like what I think I would like to do is, I'd like to find a nutritionist in the different states that I'm licensed in, Florida, Texas, California, and someone I can trust, you know, that I can kind of have like a partner

with some of these patients because it's almost like, you know, I need someone that can kind of hold someone's hand or the family's hand. I need to find that person. Yeah, I have a suggestion for you. There's a practice that's nationwide of ketogenic dietitians, registered dietitians, all have experience with ketogenic diets. And now half their practice is mental health, actually. Yeah. Great. What's the name of it? It's advanced ketogenic therapies run by Denise Potter. Oh, okay. Yeah, who's been doing these diets for, you know, in 20 years with people with epilepsy. But in recent years, it's really, getting into the mental health space

and she has wonderful programs online for people. And then I also think your program is a great option. I'll just plug that at the end there. You know, just a place to send people. I think there are people that are a good fit for that who are willing to go and travel and learn the ketogenic diet. And so I think that that is a really cool resource. It's called Accord.

Yeah, Accord. Yeah, and it's really helpful because there are some people and some families that even with dietitian support can't seem to get into ketosis at home. Right. It's hard, you know, and I think they're doing everything right, but, you know, they come to us and we just immerse the minute. We've got a chef and a dietitian, delicious food,

exercise, mindfulness, circadian rhythms. People get, you know, this huge benefit in a month or two, and then they're really excited to keep doing this when they get home. So it's sort of like an immersive jumpstart for people. Yeah. Cool. And then, yeah, I think the final thing is I'm going to be following these randomized trials. And maybe we'll have you, maybe we'll have Chris Palmer come back on.

Maybe we could, you know, when one of these studies comes out, you and me could interview the person that did the study. We could have them on together and talk about their specific study and kind of just continue to educate ourselves. But thank you so much, Matt Bernstein, for coming on. It's been a pleasure. I think you have a lot of practical knowledge. you're obviously, you know, someone who's balanced, a psychiatrist, who's been practicing for a long time. And that's the kind of person I like to have come on to kind of just give your wisdom and your experience. And this is just an added sort of thing that we could think about with each of our patients.

Like, is this right for them? We give them, you know, mention it. Look at the studies ourselves, educate ourselves. Try the diet ourselves so we can learn about it. and don't make the same mistake. I made the first time I did it. I did not take enough electrolytes in. You lose like six, 10 pounds in the first couple weeks.

I continue lifting. I do power lifting. And I would have these horrible spasms after. And my friend Darcy Trankel, shout out to her. She was into the ketogenic diet for mental health issues a decade ago. Oh, wow. She was all about it. And called her up.

And she said, yeah, you need to drink a good leader of bone broth a couple times a day and you'll be fine. And so, you know, a lot of electrolytes and bone broth, right? Yes, yeah. Well, yeah, I really appreciate you having me and being able to talk about this topic I'm so passionate about. And it was great to go through it all with you. Yeah, thanks. Cool. So we'll have on our website, Psychiatrypodcast.com, tied to this episode, we'll

have all of your contact information. We'll have all of the articles that we talked about today. You know, this is free. It's not behind a paywall. So please, you can download the PDF. You can share it with people. You can share it with your patients. We're just trying to give information out for free across the world. I know people access these kind of things and they find it very helpful. And so if this is something you want to dive into deeper, please go there. If there's more actually there than we had time to discuss, we've probably discussed about 30 to 40% of what we have on that document. So there's a lot more to dig into,

and I'm going to be editing the next couple weeks, looking at each thing, and we talk about different groups of people that in particular may be of advantageous benefit to this from this diet. So, yeah, we'll leave it there for today.